Mar-apr 2008

Volume 25 Number 2 March - April 2008

IETE TECHNICAL REVIEW The Institution of Electronics and Telecommunication Engineers PRESIDENT S Narayana VICE-PRESIDENTS A K Agarwal

P N Chopra

Anita G Dandekar

PUBLICATIONS COMMITTEE Chairman M L Gupta Co-Chairman M C Chandra Mouly Members H O Agrawal

S S Agrawal

Smriti Dagur

M Jagadesh Kumar

Surendra Pal

Giridhar R Joshi

T K De

T S Rathore

S K Kshirsagar

Coopted K M Paul

K S Prakash Rao Special Invitee

S C Dutta Roy

P Banerjee EDITORIAL BOARD Chairman Dilip Sahay Members

H O Agrawal

A K Bhatnagar

R G Gupta

S S Motial

Neeru Mohan Biswas

H Kaushal

Secretary General

Dy Managing Editor

V K Panday

A P Sharma

The IETE Technical Review invite articles preferably readable without mathematical expressions, state-of-the-art review papers on current and futuristic technologies in the areas of electronics, telecommunication, computer science & engineering, information technology (IT) and related disciplines. In addition, informative and general interest articles describing innovative products & applications, analysis of technical events, articles on technology assessment & comparison, new & emerging topics of interest to professionals are also welcome. While all the papers submitted will go through the same detailed review process, short papers and Practical Designs will receive special attention to enable early publication. Manuscripts may please be submitted in triplicate to the Managing Editor along with a soft copy on floppy/CD/e-mail. Detailed guidelines to authors may be seen on IETE Website : http://www.iete.org under the heading ‘Publications.’ Annual Subscription : Subscription and Advertising rates are available on request and also on website: iete.org Address for correspondence : Managing Editor, IETE, 2, Institutional Area, Lodi Road, New Delhi 110 003, Telephone : +91 (11) 43538842-44 Fax : +91 (11) 24649429, email : [email protected]; [email protected], Website : http://www.iete.org; http://www.iete.info FREE TO IETE CORPORATE MEMBERS : (Cost of Production: Rs.19.00)

COPYRIGHT IETE Technical Review is published bimonthly by the Institution of Electronics and Telecommunication Engineers. All rights of publication are reserved by the IETE. Copyright and Reprint permission : Abstracting is permitted with credit to the source. Libraries are permitted to photocopy for private use of readers. It is the IETE policy to own the copyright of the scientific and technical papers it publishes on behalf of the authors and their employers, and to facilitate the appropriate reuse of this material by others. Authors are required to sign an IETE copyright transfer form before publication. This form is available online in IETE website:www.iete.org. IETE retains the authors’ and their employers’ right to reuse their material for their own purposes.

47

IETE TECHNICAL REVIEW Published bimonthly by the Institution of Electronics and Telecommunication Engineers March-April 2008

Vol 25

No 2

CONTENTS

48

SCAN

IETE – R S KHANDPUR GOLD MEDAL AWARD LECTURE

Dilip Sahay 81 49

59

Telephone Caller-ID Signal Sending Over Internet

Binaural Dichotic Presentation of Speech Signal for Improving its Perception to Sensorineural HearingImpaired using Auditory Filters

Aleksandar Lebl and •arko Markov

D S Chaudhari

Issues in Mobile Ad hoc Networks for Vehicular Communication S S Manvi and M S Kakkasageri

INVITED ARTICLE 73

91

Dielectric Parameters as Diagnostic Tools and Indicatrix of Disease — A Microwave Study V Malleswara Rao, B Prabhakara Rao and D M Potukuchi

MediaFLO™ - The Ultimate Mobile Broadcast Experience Sachin Kalantri

Note : The Institution of Electronics and Telecommunication Engineers assumes no responsibility for the statements and opinions expressed by individual authors.

Vol. 25, No. 2, March-April’08

I E T E

T E C H N I C A L

R E V I E W

48

SCAN The second issue of Technical Review for the year 2008 contains five articles including an invited paper and a lecture delivered during an IETE Award function. First article by Aleksandar Lebl and Zarko Markov, describes calling subscriber number transmission over Internet to the called subscriber. The paper describes how real time protocol is used for this purpose. The paper discusses two ways of transmission of CLI over internet. According to the authors the two methods were developed for the telephony requirements of Telecom Serbia. In the second article, S S Manvi and M S Kakkasageri, explain “Vehicular Ad hoc Networks” (VANETs) for inter vehicle and vehicle to road side communication. The paper describes various characteristics and applications as well as a few issues that needs to be taken into consideration with VANETs. The paper broadly mentions the wireless technologies that can be utilized but misses the adaptation of available mobile access technologies including PMRTS. In the invited article, Shri Sachin Kalantri describes about multicast service on Mobile TV. The author has elaborated about MediaFLO platform to deliver quality content with roughly half the spectrum and less than half the infrastructure required. According to the author, Mobile TV service is going to be quite popular and mobile broadcasting using mediaFLO would offer high quality service to the customers. The fourth paper is an IETE-R S Khandpur Gold Medal Award lecture delivered by Prof D S Chaudhari on “Binaural Dichotic presentation of Speech Signal for improving its Perception to Sensor neural Hearing Impaired using Auditory Filters” on the theme “Connecting Persons with Disabilities: ICT opportunities for All” set by International Telecommunications Union (ITU) for this year’s theme. In the last and fifth article, V Malleswara Rao, et all describe X-band microwave (9-10 GHz) study on Dielectric Parameters as diagnostic tools and indicatrix of disease. This article demonstrates the capability of MW dielectric Measurements. I take this opportunity to convey to the readers about the sincere efforts that have gone to present these selected articles by the staff of the publication division of IETE. I also would like to bring to your notice the sincere efforts to make some changes in the cover design of this journal by IETE Secretariat. I hope that this cover design gels with the serious and interesting reading of this journal and any suggestion in this regard is welcome. I also take this opportunity to request and invite authors of article for state-of-the art review papers on current and futuristic technologies and products in the areas of computer science, engineering, Telecommunications, Information Technology etc and request them for their suggestions for improvement of the journal. Dilip Sahay Chairman, Editorial Board

Vol. 25, No. 2, March-April’08

I E T E

T E C H N I C A L

R E V I E W

49

Telephone Caller-ID Signal Sending Over Internet ALEKSANDAR LEBL

AND

•ARKO MARKOV

ABSTRACT This paper describes two methods of calling subscriber number sending over Internet to the called subscriber who is connected to the classic telephone exchange by Internet. One, faster, method uses two event codes, while the other method uses only one event code, but it is significantly slower.

1.

INTRODUCTION

In the process of integrating packet and traditional telecommunication, different interworking methods are developed. One interworking method consists of connecting traditional user equipment (telephone, fax) and classic telephone exchange, or interconnecting telephone exchanges, across Internet. The basis for this interconnection is transmission of traditional telephone network signals (dual tone multifrequency signals (DTMF), line signals, fax signals, interexchange signaling) using procedures adopted in Internet. These procedures were standardized in RFC2833 [1], in year 2000. Considering its latitude, the authors separated RFC2833 into three recommendations [2-4]. In these recommendations, possibilities for Internet transmission of almost all signals known in modern telephone network are described. One of the rare exceptions, not presented in recommendations, is Caller Identification Signal (CIS), which local exchange (LE) sends to the called subscriber at the connection beginning. This paper presents some solutions for CIS signal transmission.

2.

THE TRANSMISSION OF SIGNALS OVER INTERNET

CIS

The methods for CIS transmission over Internet, which are suggested in this paper, can be used in the setup phase of connection, shown in Fig 1. In the setup phase of connection, classic TE is connected to the Local Exchange (LE) using Internet. Gateways GW1 and GW2 are located at the points of contact of telephone network and Vol. 25, No. 2, March-April’08

Internet. These GWs make telephone speech signals and signaling information suitable for transmission over the Internet and enable their regeneration in telephone format at the receiving end. One of the Common Control Unit (CCU) functions in LE is to manage the operation of CIS signal generator (CISGEN). The generated CIS signal is sent from the LE towards GW1. The demodulator (DEM) in GW1 demodulates CIS signal, and the packetizer (PCK) forms packets of CIS signal coded as telephony event in accordance with the Internet standards. At the Internet egress port, depacketizer (DPC) in GW2 depacketizes messages, and the modulator (MOD) modulates CIS signal. In this way CIS signal is put back to its original shape, and sent towards TE.

3.

METHODS FOR TELEPHONE SIGNAL TRANSMISSION OVER INTERNET

Real Time Protocol (RTP) is used for telephone signal transmission over Internet [5]. Two methods can be used: (1)

Transmission of telephone signal parameters (TSP) (frequency, modulation frequency, level, duration) ([2], section 4. RTP Payload Format for Telephony Tones), This method of transmission is shown in the Fig 1a;

(2)

Transmission of event code (EC) ([2], section 2. RTP Payload Format for Named Telephone Events), as it is shown in the Fig 1b. The Named Telephone Event means that it is necessary to detect the type of the signal (for I E T E

T E C H N I C A L

R E V I E W

50

(a)

(b) Fig 1 The transmission of telephone signals over Internet

example: ring signal, DTMF digit, etc.), and to send special code for the detected signal.

bits are volume field describing the signal power level, and 16 bits are the event duration field.

The characteristics of the first method are faster transmission, simpler gateway, but also the possibility to transmit invalid signal. For example, if the original signal is attenuated more than it is expected, its level will be measured and the parameter expressing this attenuated level will be sent across Internet.

RTP packet, shown in the Fig 2, is the payload of the UDP datagram.

The second method requires a more complex gateway. This gateway, standing at the point of connection between telephone and Internet network, needs more time to recognize the type of telephone signal at the Internet ingress port, but the (tone) signals are replayed in standard shape and level. Figure 2 presents the RTP packet structure when signals are transmitted using event code and the position of RTP packet in Internet packet. After the usual RTP header, consisting of 12 octets, RTP payload follows. In the example, represented by the figure, one event is defined. The payload format for each event consists of one 32bit word, In this word, 8 bits at the beginning are the event field for event code, E bit indicates if the packet contains the end of the observed event, 6 Vol. 25, No. 2, March-April’08

4.

ABOUT CIS SIGNAL AND ITS TRANSMISSION IN CLASSIC TELEPHONE NETWORK

The procedures for CIS signal transmission in classic telephone network are defined in [6]. According to this recommendation, CIS signal is sent associated with first ring (or ringing) signal using two methods. In the first method, data transmission occurs during the first long silent period between first two ring patterns. In the second method, data transmission occurs prior to first ring pattern and in that case Terminal Equipment Alerting Signal (TAS) is sent before data. TAS is used to signal to the Terminal Equipment (TE) that data transmission is to be expected. The TAS shape and the characteristics are defined in [6]. Timing relations of CIS signal to ringing signal, or to TAS, are accurately defined. For example, if the CIS signal is transmitted during the first long silent period between two ring patterns, its transmission must start in the time interval between I E T E

T E C H N I C A L

R E V I E W

51

Ether. Header (14 bytes)

IP Header (20 bytes)

0

UDP Header (8 bytes)

RTP Header (12 bytes)

RTP payload

PT

Event

E R

31

Volume

}

)

duration

RTP header RTP payload

Fig 2 RTF packet structure when signals are transmitted using event code

tdmin = 0.5s and tdmax = 2s after the first ring pattern, as it is shown in Fig 3. The various timing parameters in Fig 3 are as follows:

CIS signal consists of sending Frequency-Shift Keying (FSK) signal at the rate of 1200 baud [6,7]. The signal of frequency 1300 Hz represents binary 1 (mark bit) and the signal of frequency 2100 Hz represents binary 0 (space bit).

-

R - ring signal;

-

CIS - CIS signal;

The parts of the CIS message are successive, without interruption in sending:

-

tRING - the duration of a ring signal;

-

-

tm - the duration of a CIS signal;

Channel Seizure Signal: the block of 300 continuous bits of alternating “0”s and “1”s;

-

tdmax - maximum time delay between a ring signal end and a CIS signal start;

-

Mark Signal: the block of 180±25 or 80±25 mark bits;

-

tdmin - minimum time delay between a ring signal end and a CIS signal start;

-

Message type;

-

Message length;

R

tRING

CIS

tdmin

tm

CIS

R

tm

tRING

tdmax

Fig 3 CIS signal transmission between two ring patterns

Vol. 25, No. 2, March-April’08

I E T E

T E C H N I C A L

R E V I E W

52

-

Presentation Layer Message (contains different parameters: date and time of the call, calling line identity, etc.);

The use of dynamic PT is explained in [2]. The structure of the RTP payload in this case is as follows:

-

Checksum, followed by 1 to 10 mark bits.

-

telephony event code: CIS signal transmission after ring (code CIS) - separated code, which is not defined in [2] and [3] - 8 bits. Our application only deals with CIS signal transmission between two ring patterns, while the coding according to other scenarios, explained in [6], is planned for future;

-

the length of the RTP content which codes CIS signal - 8 bits;

-

the number of space-mark pairs in Channel Seizure Signal of CIS - 8 bits;

-

the number of mark bits in Mark Signal of CIS - 8 bits;

-

the n bytes of CIS message content (from Message Type to Checksum) - nx8 bits.

Every byte representing Message type, Message length, Presentation Layer Message and Checksum is enveloped by start bit (space) and stop bit (mark).

5.

CIS SIGNAL TRANSMISSION USING SIGNAL EVENTS

CIS signal presentation in the packet format, as the one explained in Fig 2, is not mentioned in [2] and [3], but is possible. Such a coding is too complex and requires great number of bytes. This coding will be realized by presenting every bit of CIS information using 4 bytes, included in the RTP payload shown in Fig 2. In order to avoid this, we formed the new format for telephone event coding. This format is presented in Fig 4. Some facts simplify the formation of a packet:

This RTP payload has only 4 bytes more than CIS message. According to recommendation [8], the maximum expected length of the Presentation Layer Message is 75 bytes (including Message Type, Message Length and Checksum, it is 78 bytes). The Message Length in the CIS message has 8 bits, thus enabling the length of the Presentation Layer Message to be 255 bytes. The value mentioned here and in [8] is significantly smaller.

1)

The CIS signal transmission level is defined in [6], and it can be known, a priori in GW2. That’s why it is not necessary to send it in RTP packet;

2)

Each bit duration is also defined in [6], and it is not necessary to send it in RTP packet;

3)

The sequence of CIS message parts is defined and known in GW2, so must not be particularly coded. For example, it is known a priori that the CIS message begins with channel seizure signal consisting of alternating “0”s and “1”s, followed by the mark signal, consisting of a block of mark bits.

Alternatively, only CIS signal can be represented by the telephony event code, and the ring signal can be represented by the separate packet and the separate telephony event code, mentioned in [2].

There are two methods for CIS signal transmission. In the first one, one telephony event identifies ring signal and CIS (ring signal & CIS). The second one uses transmission of different codes for ring and CIS.

For sending CIS associated with TAS, as defined in [6], the separate telephony event codes can be used for each scenario of sending. The RTP payload format in this case may be the same as the one in Fig 4.

5.1. The first method The structure of RTP packet, which transmits CIS signal coded as telephone event, is presented in Fig 4. The payload type (PT) is included in RTP header. For our application, we chose dynamic PT. Vol. 25, No. 2, March-April’08

5.2. The second method

According to [1], the telephony event code 89 is reserved for ringing. After this code, the codes 9095 are free for assignment. That’s why the codes are configured in the following way: -

90 - the unique telephony event code representing ring signal & CIS;

I E T E

T E C H N I C A L

R E V I E W

53

0

PT

event 1. byte in CIS message

31

length

01 pairs num. is seizure

bit 1 number in mark

2. byte in CIS message

3. byte in CIS message

4. byte in CIS message

. . . (n–1). byte in CIS mess. n. byte in CIS message

}

}

RTP header

RTP payload

Fig 4 RTP packet structure when the CIS signal is transmitted using telephone event code

-

91 - telephony event code representing only CIS, while the ring pattern is represented by the code 89.

After this, codes 92 to 94 can be used for CIS signal transmission associated with TAS, according to scenarios defined in [6].

6.

THE CHANNEL CAPACITY NEEDED FOR CIS SIGNAL TRANSMISSION

The method, which is suggested in this paper for CIS signal coding by telephony events, makes great saving in channel capacity. Let us suppose that Presentation Layer Message contains only date and time of call (8 bytes) and the calling line identity (6 bytes in our example, but more bytes may be used). The message coded by this method has 25 bytes. This is significantly less than that in the case of sample coding according to the recommendation G.711. The CIS message, with the Presentation layer containing 8 bytes for date and time of call and 6 bytes for calling line identity, will have 700 bits coded using FSK (the duration of every bit is 0.833 ms). For coding samples of this signal, 4666 bytes are needed. The saving is great also compared to the case of coding each bit of FSK signal by 4 bytes, as telephony events are presented according to [2]. For this coding, 2800 bytes must be used.

Vol. 25, No. 2, March-April’08

7.

TIME RELATIONS IN THE REPRODUCED CIS SIGNAL

Important elements in CIS signal transmission are the time relations between a CIS signal and a ring signal. These relations in the reproduced signal on the receiving part of the connection GW2 → TE must be the same as in the original signal on the sending part of the connection LE → GW1 (refer Fig 1). If a CIS signal is transmitted using the first method (section 5.1), by coding ring signal & CIS with the code 90, the ring signal reproduction will start after a delay as shown in Fig 5. The timing parameters in Fig 5 are as follows: -

tp - the delay between a reproduced ring signal start and an original CIS signal end;

-

tD1 - the delay of a reproduced ring signal in relation to an original ring signal. Other abbreviations are adopted in Fig 3.

Signals in the direction LE → GW1 are called original signals. Signals in the direction GW2 → TE are called reproduced signals. The original ring signal and CIS signal are shown in Fig 5a. The reproduced ring and CIS signal are shown in Fig 5b. The ring signal reproduction, according to Fig 5b, starts after the

I E T E

T E C H N I C A L

R E V I E W

54

R

CIS a)

tRING

tm tdmax

R

CIS

b) tp t RING tD1

Fig 5 (a) Original ring and CIS signal LE →GW1 (b) Reproduced ring and CIS signal GW2 → TE

original CIS signal is completely received. The delay of the reproduced ring signal after the end of original CIS signal (tp) is the sum of the message propagation time over Internet and time for CIS signal end detection. Thus, the maximum delay of ring and CIS signal reproduction is: tD1 = tRING + tdmax + tm + tp

(1)

When the longest expected message, having Presentation Layer Message of 75 bytes, is sent, its duration is tm ≈ 1080 ms. The ring signal duration is tRING = (1±0.1) s, and the delay in ring and CIS signal reproduction, neglecting tp, is: tD1 ≤ 4180 ms

(2)

As ring and CIS signals are sent using unique telephony event code, even in the case the CIS signal is not sent after ring pattern, it is necessary to wait until the time tdmax, to see whether CIS signal succeeds ring pattern or not. If CIS signal does not succeed ring pattern, GW1 sends the telephony event code 89 - “only ring signal”. In this way, the ring signal delay in reproduction, when only ring signal is sent, is: tD ≥ tRING + tdmax ≈ 3s Vol. 25, No. 2, March-April’08

(3)

If CIS signal follows ring pattern, after CIS message reception and its decipherment, the telephony event code 90, which defines ring signal & CIS, is sent. The second method (section 5.2) can be used for CIS signal transmission, whereby separate packets are used for telephony event ring (packet with event code 89), and for telephony event CIS (packet with event code 91). In this case, the delay in reproduction is shorter than in previous case (given in section 5.1) when ring and CIS are sent by unique event code (packet with the event code 90). The delay can be explained using Fig 6. Original ring and CIS signal are presented in Fig 6a. Reproduced ring and CIS signal on the GW2 side of connection are presented in Fig 6b only for the case of maximum CIS signal start delay. First, the ring signal, which is sent as the event code 89, is reproduced. It starts with the delay tp after the original ring signal end. The reproduced CIS signal starts with the delay tp after the original CIS signal message end. This time is the sum of message propagation time over Internet and the time needed for CIS signal end detection. Due to this, the maximum delay between the start of reproduced CIS message and reproduced I E T E

T E C H N I C A L

R E V I E W

55

R

CIS a)

t RING

tm tdmax

R

CIS

b) tP

t RING

tP tD2

R

c)

t DRING t RING

CIS

t D3=t D2 –t DRING

Fig 6 (a) Original ring and CIS signal LE → GW1 (b) Reproduced ring and CIS signal GW2 → TE (c) Required implemented delay of ring signal

900 ms, we have:

ring signal is: tD2 = tdmax + tm – tRING

(4)

Considering the requirement defined in [6], to satisfy the condition tD2 ≤ tdmax, we need to have tRING ≥ tm. If this condition is not satisfied, the start of the reproduced CIS signal could be delayed by more than tdmax: = 2s. The most unfavourable situation is when the longest message is sent and the reproduced ring signal has the shortest duration. The Presentation Layer Message of this CIS signal has 75 bytes, and the message duration is tm ≈ 1080 ms. For the shortest ring pattern of tRING ≈

Vol. 25, No. 2, March-April’08

tD2 ≈ 2180 ms

(5)

In order to prevent this situation, the delay must be introduced in the start of ring signal reproduction, represented in Fig 6b. The delay of the ring signal start in this case can be seen in Fig 6c. Its value is: tDRING > tm – tRING

(6)

and with concrete values tm ≈ 1080 ms and tRING ≈ 900 ms, we obtain: tDRING > 180 ms I E T E

T E C H N I C A L

(7)

R E V I E W

56

Perror CIS

10-1

10-2

10-3

10-4

10-5

10-6 10-8

10-7

10-5

10-6

BER

Fig 7 Probability error in CIS signal transmission

8.

THE ERROR PROBABILITY IN CIS SIGNAL TRANSMISSION

When CIS signal is transmitted using telephony event code, the wrong identification may happen as the result of error in transmission over Internet. As Vol. 25, No. 2, March-April’08

CIS signals are sent using RTP, they are sent without checking and retransmission and may be corrupted. The message will be wrong in the case of error on the part of message whose accuracy is verified I E T E

T E C H N I C A L

R E V I E W

57

by the part of message called frame check sequence (FCS). This part of message, except RTP payload, includes Ethernet Header, IP Header, UDP Header and RTP Header and its length is, according to Fig 2, LNHeader = 54 bytes (or LHeader = 432 bits), where LNHeader and LHeader are the lengths in bytes and bits of all headers which are checked by the FCS. RTP payload can be at most LNRTPpay = 82 bytes (or LRTPpay = 656 bits) for CIS signal transmission in our case. The packet error probability (i.e. probability of packet loss) can be calculated as: PerrorCIS = (LHeader + LRTPpay) . BER (8) where BER is Bit Error Rate, i.e. the probability that a bit is corrupted.

ACKNOWLEDGEMENT

This paper is the result of research supported by Ministry of Science of Republic of Serbia. REFERENCES 1.

H Schulzrinne & S Petrack, RFC2833: RTP Payload for DTMF Digits, Telephony Tones and Telephony Signals, May 2000.

2.

H Schulzrinne & T Taylor, RFC 4733: RTF Payload for DTMF Digits, Telephony Tones, and Telephony Signals, December 2006.

3.

H Schulzrinne & T Taylor, RFC 4734: Definition of Events for Modem, Fax, and Text Telephony Signals, December 2006.

4.

H Schulzrinne & T Taylor, Definition of Events For Channel-Oriented Telephony Signalling draft-ietf-avtrfc2833biscas-05, June 2007.

5.

H Schulzrinne, S Casner, R Frederick & V Jacobson, RFC 3550: RTP: A Transport Protocol for Real-Time Applications, July 2003.

6.

ETSI EN 300 659-1 Vl.3.1, Access and Terminals (AT); Analogue access to the Public Switched Telephone Network (PSTN); Subscriber line protocol over the local loop for display (and related) services; Part 1: On hook data transmission, 2001-01.

7.

ITU-T Rec. V.23, Data Communication over the Telephone Network: 600/1200-baud modem standardized for use in the general switched telephone network, Fascicle VIII. 1, November 1988.

8.

ETSI ETS 300 778-1, Public Switched Telephone Network (PSTN); Protocol over the local loop for display services; Caller Display Service - Terminal Equipment Requirements; Part 1: Off-line data transmission, July 1996.

The result of this calculation is shown in Fig 7.

9.

CONCLUSION

The paper presents two methods of calling subscriber number transmission over Internet to the called subscriber. These methods are developed for the telephone network requirements of Telekom Serbia. The first method uses only one telephony event code from the group of codes, but, unfortunately, produces great delay of ring signal. The other method is efficient, because ring signal delay is small, but the method uses two telephony event codes and the more complicated GW on the user side.

Vol. 25, No. 2, March-April’08

I E T E

T E C H N I C A L

R E V I E W

58

Authors Aleksandar Lebl, born 1957, BSc 1981, MSc 1986 is employed from 1981 in the Switching Department of Research and Development Institute IRITEL in Belgrade, Republic of Serbia. During years he worked on the project of Digital Switching System for Serbian Telecommunication Industry. ^

Address: Aleksandar Lebl, IRITEL d.d, Batajnicki put 23, 11080 Zemun, Serbia, Europe. E-mail:

•arko M Markov, born 1946, BSc 1969, MSc 1975, PhD 1976 is a scientific counsellor in IRITEL, Institute for Electronics and Telecommunications, Belgrade, Serbia. Area of work: Switching technics, Teletraffic theory, Network signalling. Author or co-author of hundred papers and six books. At the University of Belgrade, School of Electrical Engineering, Dr Markov is a professor at the course of Switching technics and Network signalling. ^

Address: • arko Markov, IRITEL d.d, Batajnicki put 23, 11080 Zemun, Serbia, Europe.

Paper No 143-B; Copyright © 2008 by the IETE.

Vol. 25, No. 2, March-April’08

I E T E

T E C H N I C A L

R E V I E W

59

Issues in Mobile Ad hoc Networks for Vehicular Communication S S MANVI

AND

M S KAKKASAGERI

ABSTRACT Vehicular ad hoc networks (VANETs) are a specific type of Mobile Ad hoc Networks (MANETs) that are currently attracting the attention of researchers around the world. With pervasiveness of mobile computing technology and wireless communications, VANETs could be a key networking technology of the future vehicle communications. VANETs can make a possible wide-range of interesting applications focusing on vehicle traffic safety, entertainment in vehicles, cooperative driver assistance, sharing traffic and road conditions for smooth traffic flow, user interactions, information services, etc. Key characteristics that distinguish VANETs from other networks are time-varying nature of vehicle density, high mobility, and time-critical safety applications. Hence, devising protocols for VANETs may not be successfully accomplished by simple adaptation of protocols designed for wired networks and MANETs. This paper outlines the current research issues in VANETs, which may benefit the researchers to design and develop protocols for VANETs.

1.

INTRODUCTION

Vehicular Ad hoc Networks (VANETs) are an envision of the Intelligent Transportation System (ITS). Vehicles communicate with each other in two ways: (1) Intervehicle communication and (2) Vehicle to roadside infrastructure communication. VANETs are based on short-range wireless communication between vehicles. Unlike infrastructure-based networks such as cellular networks, these networks are constructed on the fly (self organizing). VANETs are special case of Mobile Ad hoc Networks (MANETs). The key differences as compared to MANETs are following: components building the network are vehicles, restricted vehicle movements, high mobility and time-varying vehicle density [1]. One advantage of VANETs over MANETs is that most of the vehicles provide sufficient computational and power resources, thus eliminating the need for introducing complicated energy-aware algorithms [2]. The optimal goal of VANETs is to provide safer and more efficient roads in future by communicating timely information to drivers and concerned authorities. The prominent evolution of wireless communication witnessed recently has sparkled the interest of the automotive industry. Many Vol. 25, No. 2, March-April’08

manufacturers have already developed system prototypes, allowing vehicles to communicate with their surroundings using wireless media [3]. This has motivated the research community to design and develop protocols and standards for VANETs. A typical VANETs scenario is as shown in Fig 1. Vehicle to vehicle and vehicle to roadside base station/gateway communication is possible for providing safety and other information services to vehicle users. Group of vehicles together may form a cluster to disseminate information among themselves as well as to other clusters and base stations. VANET may integrate networking technologies such as WiFi (Wireless Fidelity Standard IEEE 802.11 b/g), WiMAX (Wireless Metropolitan Access Standard IEEE 802.16) and Bluetooth (IEEE 802.15). WiFi can be used for vehicle to vehicle as well as vehicle to base station communication. WiMAX can be used for forming wireless backbone connecting different base stations. Bluetooth can be used for intra-vehicle communication as well as communicating with nearest neighbors. In a VANET, each vehicle in the system is equipped with a computing device, a short-range wireless interface and a GPS (Global Positioning System) receiver. GPS receiver provides location, speed, current time and direction of the vehicle. I E T E

T E C H N I C A L

R E V I E W

60

Fig 1 A typical VANET scenario

Manufacturers are already enhancing cars with sensors that help drivers to park and provide GPS compasses as standard equipment on luxury cars. Within a decade, full integration of on-board software and hardware computing facilities with wireless communications and environmental sensors can be achieved. Each vehicle stores information about itself and other vehicles in a local database. The records in this database are periodically broadcasted. A record consists of the vehicle identification, position in the form of latitude and longitude, current speed of the vehicle, direction, and timestamps corresponding to when this record was first created and when this record was received. The vision for VANETs includes applications such as route planning, road safety, e-commerce, entertainment in vehicles, cooperative driver assistance, sharing traffic and road conditions for smooth traffic flow, user interactions, information services, etc. [4-9]. VANETs have unique requirements with respect to applications, types of communication, self-organization and other issues Vol. 25, No. 2, March-April’08

such as media access, security and routing. In order to meet these requirements, the structuring of functionalities into protocols and their interaction must be re-thought. A number of research projects show great interest in VANET, which aim at development of VANET protocol architecture, connectivity for intervehicle communications using roadside infrastructure and secured routing of critical events [10-12]. This paper briefly outlines research issues in a VANET and also focuses on the adoptable wireless technologies for deployment of VANETs.

2.

UNIQUE CHARACTERISTICS OF VANETs

In order to suggest a protocol stack suitable for VANETs, we should pinpoint the fundamental characteristics that differentiate VANETs from other networks.

•

Geographically constrained topology: Roads limit the network topology to actually I E T E

T E C H N I C A L

R E V I E W

61

one dimension; the road direction. Except for crossroads or overlay bridges, roads are generally located far apart. Even in urban areas, where they are located close to each other, there exist obstacles, such as buildings and advertisement walls, which prevent wireless signals from traveling between roads. This implies that vehicles can be considered as points of the same line; a road can be approximated as a straight line, or a smallangled curve. This observation is quite important, because it affects the wireless technologies that can be considered. For example, since the packet relays are almost all in the same one-directional deployment region, the use of directional antennas could be of great advantage.

•

Partitioning and large-scale: The probability of end-to-end connectivity decreases with distance; this is true for one-dimensional network topologies. In contrast, connectivity is often explicitly assumed in research for traditional ad hoc networks, sometimes even for the evaluation of routing protocols. In addition, VANETs can extend in large areas, as far as the road is available. This artifact together with the one-dimensional deployment increases the above probability.

•

Self-organization: The nodes in the network must be capable to detect each other and transmit packets with or without the need of a base station.

•

Unpredictability: The nodes (or vehicles) constituting the network are highly mobile. Because of this reason, there is also a high degree of change in the number and distribution of the nodes in the network at given time instant. The nodes must be constantly aware of the network status, keep track of the hosts associated with the network, detect broken links and update their routing tables whenever necessary.

Since vehicle mobility depends on the deployment scenario, the movement direction is predictable to some extent. In highways, vehicles often move at high speeds, while in urban areas they are slow. In addition, mobility is restricted by the road directions as well as by traffic regulations. Assuming that these regulations are obeyed, there are lower and upper speed bounds, and restriction Vol. 25, No. 2, March-April’08

signs that obligate drivers to move on specific roads and directions. Hence, mobility models can now include some level of predictability in movement patterns. Car manufacturing companies have already implemented such models for testing mechanical parts.

•

Power consumption: In traditional wireless networks, nodes are power limited and their life depends on their batteries - this is especially true for ad hoc networks. Vehicles however can provide continuous power to their computing and communication devices. As a result, routing protocols do not have to account for methodologies that try to prolong the battery life. Older network protocols include mechanisms such as battery-life reports for energy-efficient path selection, sleep-awake intervals, as well as advanced network crosslayer coordination algorithms. These schemes cannot offer any additional advantages to vehicular networks.

•

Node reliability: Vehicles may join and leave the network at any time and much more frequently than in other wireless networks. The arrival/departure rate of vehicles depends on their speed, the environment, as well as on the driver needs to be connected to the network. In case of ad hoc deployments, the communication does not easily depend on a single vehicle for packet forwarding. This occurs because of non-coverage of communication range between communicating vehicles. Thus there is a need to take help of intermediate nodes for packet forwarding to destination vehicle. Intermediate nodes must be reliable to forward the packets efficiently.

•

Channel capacity: The channels in VANETs over which the terminals communicate are subjected to noise, fading, interference, multipath propagation, path loss, and have less bandwidth. So high bit-error rates are common in VANETs. One end-to-end path can be shared by several sessions. In some scenarios, the path between any pair of users can traverse multiple wireless links and the link themselves can be heterogeneous. So smart algorithms are needed to overcome of fluctuating link capacity in networks.

•

Vehicle density: Multi-hop data delivery through vehicular ad hoc networks is I E T E

T E C H N I C A L

R E V I E W

62

complicated by the fact that vehicular networks are highly mobile and sometimes sparse. The network density is related to the traffic density, which is affected by the location and time. Although it is very difficult to find an end-to-end connection for a sparsely connected network, the high mobility of vehicular networks introduces opportunities for mobile vehicles to connect with each other intermittently during moving.

example, a vehicle can check the status of up front vehicles status (speed, brake applied, road blocks, etc.).

•

Co-operative assistance systems: Coordinating vehicles at critical points such as blind crossings (a crossing without light control) and highway entries.

•

Safety services: Safety applications include emergency breaking, accidents, passing assistance, security distance warning, and coordination of cars entering a lane. Furthermore, sensors embedded in the car engine and elsewhere could be used for exchanging information, either with the onboard computer of the vehicle itself or vehicles with sophisticated computing and communication abilities, for diagnostic purposes. Also, safety applications are time sensitive and should be given priority over non-safety applications. This could facilitate preventive maintenance and minimizes road breakdowns.

•

Traffic monitoring and management services: In such type of services, all vehicles are part of a ubiquitous sensor system. Each vehicle monitors the locally observed traffic situation such as density and average speed using an onboard sensor and the results are transferred to other vehicles via wireless datalink through the network.

To deal with disconnections in sparse ad hoc networks, carry and forward, (where nodes carry the packet when routes do not exist, and forward the packet to the new receiver that moves into its vicinity) method may be applied. Work given in [13] explains that there is a high chance for moving vehicles to set up a short path with few hops in a highway model.

•

3.

Vehicle mobility: Since the nodes are mobile, the network topology may change rapidly and unpredictably and the connectivity among the terminals may vary with time. VANET should adapt to the traffic and propagation conditions as well as the mobility patterns of the mobile network nodes. The mobile nodes in the network dynamically establish routing among themselves as they move about, forming their own network on the fly. Moreover, a user in a VANET may not only operate within the network, but may also require access to a roadside infrastructure. Hence there is a need of strong mobility patterns in VANETs.

APPLICATIONS OF VANETs

Some of the important applications of VANETs are as follows:

•

Message and file delivery: This application focuses on enabling the delivery of messages and files in a vehicular network to the target receivers (group communication) with acceptable performance.

•

Internet connectivity: This application focuses on connecting the vehicles to the Internet using roadside infrastructure and intervehicle communications to facilitate browsing, send/read e-mails, chatting, etc.

•

Communication-based longitudinal control: Exploiting the look- through capability of VANETs to help avoiding accidents. For

Vol. 25, No. 2, March-April’08

Other applications are more related to multimedia communications like entertainment and non-safety information for example information download at gas stations or public hotspots, and car-to-car information exchange, etc. Some of these applications will be free, while others would require a service subscription or a one-time payment.

4.

ISSUES IN VANETs

VANET raises several interesting issues in regard to media access control, mobility management, data aggregation, data validation, data dissemination, routing, network congestion, performance analysis, privacy and security.

4.1. Media access control (MAC) Design of VANET MAC protocols should give I E T E

T E C H N I C A L

R E V I E W

63

more significance to fast topology changes and types of services rather than power constraints or time synchronization problems. Moreover, VANET MAC protocols have to reduce the medium access delay and increase the reliability, which is important in case of safety applications. An IEEE working group is investigating a new PHY/MAC amendment of the 802.11 standard designed for VANETs, which is known as Wireless Access in Vehicular Environments (WAVE), also referred as IEEE 802.11p [14]. In terms of MAC operations, WAVE uses CSMA/CA (Carrier Sense Multiple Access / Collision Avoidance) as the basic medium access scheme for link sharing and should probably use one control channel to set up transmissions. ADHOC MAC [15] is a MAC protocol conceived within the European project CarTALK2000 with the purpose to design novel solutions for VANETs. ADHOC MAC works in slotted frame structure, where each channel is divided into time slots. The ADHOC-MAC protocol is devised for an environment in which the terminals can be grouped into clusters in such a way that all the terminals of a cluster are interconnected by broadcast radio communication through slotted channel. Such a cluster is defined as One-Hop (OH). The access mechanism of ADHOC-MAC is Dynamic TDMA (Time Division Multiple Access) and channels are assigned to the terminals according to terminal needs. Directional antenna transmission has a promising place in VANETs, in particular for MAC issues. In VANETs, node’s movement is limited by roads and driving rules (e.g. opposite driving directions on the same road). Hence, directional antennas would surely help in reducing interference and collisions with ongoing transmissions over parallel neighboring vehicular traffic. A MAC protocol that uses directional antennas in an ad hoc network where the mobile nodes do not have any location information is proposed in [16]. The protocol uses RTS/CTS (Request to send / Clear to send) exchange similar to that in 802.11 for enabling the source and destination nodes to identify each other’s directions. The nodes transmit as well as receive data packets using directional antennas, thereby reducing the level of interference to other nodes as well as to themselves. Vol. 25, No. 2, March-April’08

A MAC protocol that adapts Dual Busy Tone Multiple Access (DBTMA) for use with directional antennas is discussed in [17]. By transmitting busy tones directionally, in addition to the directional transmission of RTS/CTS and data frames, protocol avoids collisions in a much finer grain in terms of spatial reuse and thus increases the channel capacity significantly.

4.2. Mobility management Since vehicles are highly mobile and change their point of network attachment frequently while accessing Internet services through gateways, it is advisable to have some mobility management schemes that take care of vehicle mobility and provide seamless communication. Mobility management has to meet the following requirements: seamless mobility (communication must be possible irrespective of vehicle position), low handoff latency, support IP V6, scalable in terms overheads. Ad hoc routing protocol extensions are unsuitable for the mobility management of VANETs. They highly depend on the routing protocol deployed in the ad hoc network, and they do not support Mobile IPv6. Applicationspecific enhancements are an interesting approach for the mobility management of VANETs. They support the mobility of nodes and they are independent of the ad hoc routing protocol. A mobility management protocol called MMIP6 is based on the principles of Mobile IPv4, but is designed to support IPv6-based mobile nodes organized in ad hoc networks [18]. MMIP6 uses foreign agents (FAs) like in Mobile IPv4, which are located at the Internet Gateways (IGWs). The FA (Foreign agents) represents the vehicle located in the VANET; this way, it hides the multihop capability of the VANET and the vehicles appear as common mobile nodes. A very important feature is that MMIP6 relies on globally routable and permanent IPv6 addresses to identify the vehicles. With the use of FAs, all vehicles participating in the VANET form one logical IPv6 subnet, where the IGWs act as transition points between the VANET and the Internet. The IPv6 addresses can be assigned statically to each vehicle, i.e. they are preconfigured in the communication hardware shipped with the vehicles. In contrast to Mobile IPv6, a vehicle does not receive a valid IPv6 careof address when entering a foreign network. MMIP6 I E T E

T E C H N I C A L

R E V I E W

64

avoids link local addresses when a vehicle is located in a foreign network.

affects the performance of data dissemination is given in [20].

4.3. Data aggregation

A probabilistic validation of aggregated data in VANETs is given in [21]. An aggregated record comprises of following: individual records from vehicles, random number, time stamp, signature; regular record (one of the individual record selected by using “random-number mod number-ofindividual-records”), time stamp and signature. When a vehicle receives an aggregated record, it first verifies the signature and certificate of the sender and then verifies individual records. Vehicle uses random number to generate an index for set of records. Later, it checks for regular record matching with the record in set of records as per the index. If there is a match, then the record is valid.

The vehicles have to pass on the data sent by the neighbors to other neighbors of its coverage area. This increases the number of packets to be sent by a vehicle. Therefore, data aggregation techniques are applied to reduce such overheads. Data aggregation is an interesting approach, which reduces the number of packets transmitted drastically by combining several messages related to the same event into one aggregate message. For example, the records about two vehicles can be replaced by a single record with little error, if the vehicles are very close to each other and move with relatively the same speed. The CARISMA road traffic simulator served as a means for modeling the mobility of the network nodes and tests the data aggregation [19]. In simulations, a rain area was placed randomly within the scenario and all vehicles start to send a warning message if they detect it. After the aggregation, the aggregate is not broadcasted directly, however, a timer is started to wait for further messages. A revocation message will be generated if a vehicle detects no rain. This revocation message is included in the aggregate to adapt the hazard area information.

4.4. Data validation A vehicle may send the data it has observed directly (assuming that a vehicle always trusts the data it has gathered itself) to its neighbors. Sometimes, malicious vehicles may send the incorrect information to confuse the users. For example, a malicious node may send the false accident information and divert all the vehicles on other roads, which may some times lead to traffic congestion. In such situation, data validation techniques must be applied before passing on the received information to other nodes. The vehicles test the validity of data received from other vehicles. This is done by correlating the data from different vehicles and cross-validate it against a pre-defined set of rules. A formal model of data validation and dissemination in VANETs and study of how VANET characteristics, specifically the bi-directional mobility on well defined paths, Vol. 25, No. 2, March-April’08

4.5. Data dissemination Data dissemination can be defined as broadcasting information about itself and the other vehicles it knows about. Each time a vehicle receives information broadcast by another vehicle, it updates its stored information accordingly, and defers forwarding the information to the next broadcast period, at which time it broadcasts its updated information. The dissemination mechanism should be scalable, since the number of broadcast messages is limited, and they do not flood the network. VANET characteristics like high-speed node movement, frequent topology change, and short connection lifetime especially with multi-hop paths needs some typical data dissemination models for VANETs. This is because topological transmission range needs to maintain a path from the source to the destination, but the path expires quickly due to frequent topology changes. A successful VANET data dissemination model needs to handle issues such as sparse network density, interfering environment, long path length, latency, etc. The transmission power signal level of a vehicle may be too strong or too weak during certain times of the day and in certain city environments. When the transmission range is too strong, it creates interference and lowers the system throughput. When transmission power signal level is too low, the vehicle cannot reach other vehicles. Smart algorithms for data dissemination that adjusts according to the transmission power signal level are needed. I E T E

T E C H N I C A L

R E V I E W

65

The dissemination mechanism can either broadcast information to vehicles in all directions, or perform a directed broadcast restricting information about a vehicle to vehicles behind it. Further, the communication could be relayed using only vehicles traveling in the same direction, vehicles traveling in the opposite direction, or vehicles traveling in both directions. In order to design a data dissemination model, we have to consider some of the following issues:

•

For one way, and two way in the context of traffic, a system for scalable traffic data dissemination and visualization in VANETs is needed. Vehicles moving in both directions may yield the best performance. But vehicles in the opposite direction needed better model to increase the data dissemination performance.

•

How to make efficient usage of available bandwidth consumed by each vehicle?

•

To limit the number of re-transmissions due to collisions.

•

In dense networks, such as cities or major highways with a large portion of equipped vehicles, the data load on the channel should be controlled in order not to exceed the limited wireless bandwidth. In contrast, in sparse networks, channel saturation is not a critical issue. Moreover, messages should be repeated since equipped vehicles are most likely out of wireless radio range of each other; vehicles inside the area of influence of a hazard, but not reachable at the time they are detected, should also be notified. Note that in case of experiencing a dense network, the forwarding strategy is required to be very efficient in terms of overhead while ensuring high reliability to priority messages with the most important payload, i.e., safety-of-life.

•

Safety information must be kept alive: Safety hazards can be associated with a time duration and geographical area while/where they can potentially affect vehicles safety state. The distribution of some state information will be repeated (e.g., periodically or at detection of a new neighboring vehicle) for a defined duration of time while being inside a specific geographical area. The specific strategy to

Vol. 25, No. 2, March-April’08

optimize this repetition process is to be developed. A multihop information dissemination scheme in inter-vehicle networks is discussed in [22]. Model for information Dissemination in mobile ad hoc geosensor networks is given in [23].

4.6. Routing Since the topology of the network is constantly changing, the issue of routing packets between any pair of nodes becomes a challenging task. Most protocols should be based on reactive routing instead of proactive routing. Multicast routing is another challenge because the multicast tree is no longer static due to the random movement of nodes within the network. Routes between nodes may potentially contain multiple hops, which situation is more complex than the single hop communication. Improving proactive routing in that of VANETs with the movement prediction framework is given in [24]. A route is composed of several communication links (pair of vehicles) connected to each other from the source to the destination. By knowing the movement information of vehicles involved in the routes (including source and destination), we can predict their positions in the near future in order to predict the lifetime of the link between each pair of vehicles in the path.

4.7. Network Congestion Congestion control in VANETs is a challenging issue. The Internet is based on an end-to-end paradigm, where the transport protocol (e.g. TCP) instances at the endpoints detect overload conditions at intermediate nodes. In case of congestion, the source reduces its data rate. However, in VANETs, the topology changes within seconds and a congested node used for forwarding a few seconds ago might not be used at all at the point in time when the source reacts to the congestion. Due to the mainly broadcast/geocast oriented communication and the highly dynamic network topology, conventional mechanisms such as perflow fair queuing are difficult to apply. So an appropriate model is needed for VANET where each node locally adapts to the available bandwidth. Congestion control for VANETs has not been I E T E

T E C H N I C A L

R E V I E W

66

studied thoroughly so far but this feature will be extremely necessary for VANET applications and network performance. The work given in [25] performs congestion control by using clusters depending upon the vehicle density. The cluster size is reduced as the vehicle density increases. Due to the high mobility and the resulting highly dynamic network topology, congestion control has to be performed in a decentralized and selforganized way, locally in each VANET node.

4.8. Performance analysis The cost of deploying and implementing designed schemes for VANETs in the real world is high. Hence there is a need for development of effective simulators to evaluate the performance of protocols for VANETs before deployment. Most research in this area relies on simulation for evaluation. Key component of simulations is a realistic vehicular mobility model that ensures conclusions drawn from experiments will carry through to real deployments. A VANET simulation platform should provide support to vehicle to vehicle as well as vehicle to base station communications in various road conditions, traffic conditions and mobility patterns. Simulation may also use collection of mobility traces and network statistics to experiment on a real vehicular network. To facilitate users to rapidly generate realistic mobility models for VANET simulations following mobility models may be considered [26]: Straight freeway model or Highway mobility model, Random Waypoint (RWP), and Real-Track model (RT). In Straight freeway model or Highway model nodes move following a certain path in certain direction. Nodes are not supposed to change their direction. The velocity in case of Highway mobility model is very high in three lanes (slow, medium and fast lane). Vehicle can change lanes as they do in real life situation i.e. from slow to medium lane but not to fast lane directly. Nodes can overtake each other. The Random Waypoint model implementation is as follows: At every instant, a node randomly chooses a destination and moves towards it with a velocity chosen randomly from minimum to maximum allowable velocity for every mobile node. After reaching the destination, the node stops for a Vol. 25, No. 2, March-April’08

duration defined by the ‘pause time’ parameter. After this duration, it again chooses a random destination and repeats the whole process again until the simulation ends. Real- track model are derived from the streets of the actual maps. The grouped nodes must move following the constraint of the tracks. At the switch stations, which are the intersections of tracks/ streets, a group can then be split into multiple smaller groups; some groups may be even merged into a bigger group. Such group dynamics happen randomly under the control of configured split and merge probabilities. Nodes in the same group move along the same track. They also share the same group movement towards the next switch station. In addition, each group member will also have an internal random mobility within the scope of a group. The mobility speeds of these groups are randomly selected between the configured minimum and maximum mobility speeds. One can also define multiple classes of mobile nodes, such as pedestrians, and cars, etc. Each class of nodes has different requirements: such as moving speed etc. In such cases, only nodes belonging to the same class can merge into a group. The random trip model is a generic mobility model that generalizes random waypoint and random walk to realistic scenarios, which gives a realistic flavor to simulations [27]. Some of the simulators used for evaluation of VANETs are as follows. VanetMobiSim, Glomosim, CORSIM, QualNet, NS-2, OPNET, PARAMICS, CORSIM and VISSIM [28-32]. But none of them are standard simulators. Thus there is a scope for designing VANET simulators.

4.9. Privacy and Security VANETs demand a thorough investigation of privacy related issues. On one hand, users of such networks have to be prevented from misuse of their private data by authorities, from location profiling and from other attacks on their privacy. On the other hand, system operators and car manufacturers have to be able to identify malfunctioning units for sake of system availability and security. Further, wireless link characteristics introduce also reliability problems because of the limited wireless transmission range, the broadcast nature of the wireless medium (e.g. hidden terminal I E T E

T E C H N I C A L

R E V I E W

67

some cases the vehicle real identity will be required for service usage. In such a case, it is obvious that the communication partner has the identity anyway, so the identity must only be protected from neighbors overhearing the communication. A communication protocol that keeps the identity of the vehicle hidden from third party observers is needed.

problem), mobility-induced packet losses, and data transmission errors. In the automotive market, customers can choose among a large variety of products and there is a strong competition among automakers. Customers concerned about a new technology would probably pick products that reflect their concerns. It is therefore a vital interest of all car manufacturers promoting vehicle to vehicle communication technology, to pay close attention to security and privacy of such systems. The huge number of vehicles registered in different countries and traveling long distances, well beyond their registration regions, requires a robust and scalable privacy and security management scheme. Some of the security and privacy related issues are as follows.

•

•

Secure Positioning: Position is one of the most important data for vehicles. Each vehicle needs to know not only its own position but also those of other vehicles in its neighborhood. GPS signals are weak, can be spoofed, and are prone to jamming. Moreover, vehicles can intentionally lie about their positions. Hence there is a need for a secure positioning system that will also support the accountability and authorization properties, frequently related to a vehicle’s position. A very dangerous and often ignored fact about privacy is that innocent looking data from various sources can be accumulated over a long period and evaluated automatically. Even small correlations of the data may reveal useful information. And once privacy is lost, it is very hard to re-establish that state of personal rights.

A data verification system, which helps to prevent the forging attacks, has to be developed. This can be achieved by a data correlation mechanism that compares all collected data regarding a given event.

•

•

Existing solutions such as frequency hopping do not completely solve the problem of jamming. The use of multiple radio transceivers, operating in disjoint frequency bands, can be a feasible approach. In scenarios where a vehicle communicates with a dedicated partner, we assume that in

Vol. 25, No. 2, March-April’08

•

A good thing about mobility is that real (communication) traffic analysis would probably be hard to do, since nodes usually move at high speed and in large geographic areas. But nevertheless, an attacker might use the properties of communicating vehicles as an aid for tracking a specific car. A strong solution is needed to overcome this problem.

•

Vehicular networks lack the relatively longlived context. Hence password-based establishment of secure channels, gradual development of trust by enlarging a circle of trusted acquaintances, or secure communication only with a handful of endpoints may be impractical for securing vehicular communication. Particularly for VANETs security concerns, the issues for identification and addressing have to be resolved.

Some of the efficient architectures that can really manage above mentioned privacy and security related issues are discussed in [33-36]. A group signature based secure and privacy preserving vehicular communication framework is given in [37]. This scheme achieves authenticity, data integrity, anonymity, and accountability simultaneously. It utilizes a group signature scheme, in which members maintain only a small number of secret key/group public key pairs. Privacy is provided due to the fact that signers are anonymous within the group from which they sign. Additionally, not only are signers anonymous within their group, but two messages signed by the same individual are not linkable, that is to say, one cannot determine if two messages came from the same member of the group, or two different members of the group. Digital signatures are a good choice because safety messages are normally standalone in VANETs [38]. Because of the large number of network members and variable connectivity to I E T E

T E C H N I C A L

R E V I E W

68

authentication servers, a Public Key Infrastructure (PKI) is a good way to implement authentication. Under the PKI solution, each vehicle would be given a public/private key pair. Before sending a safety message, it signs it with its private key and includes the Certification Authority (CA) certificate. Because of the use of private keys, a tamper-proof device is needed in each vehicle. This is where the secret information will be stored and the outgoing messages will be signed. To lower the risk of compromise by attackers, the device should have its own battery and clock. The clock should be capable of being resynchronized when passing by a trusted base station on the side of a road.

5.

WIRELESS TECHNOLOGIES FOR VANETs

Wireless technologies suitable for VANETs are Wireless Metropolitan Area Networks (WMANs), Wireless Local Area Networks (Wireless LANs/ WiFi) and Wireless Personal Area Networks (Wireless PANs), Dedicated Short Range Communications (DSRC)/WAVE (Wireless Access in Vehicular Environments) together with their ad hoc mode of operation [39,40].

5.1. WMANs A WMAN (Wireless Metropolitan Area Network) can interconnect distant locations. Two kinds of WMANs exist: back haul and last mile. Back haul is for enterprise networks, cellular base station communications and Wi-Fi hotspots. Last mile setups can establish wireless as an alternative to residential broadband modems. WMAN connections can be PTP (Point-to-Point) or PMP (Point-to-Multipoint). One of the most interesting recent developments is the standardization of WMANs in the form of WiMAX IEEE 802.16. Finally, the WMAN category also includes the GSM/GPRS (Global System for Mobile communications/ General Packet Radio Service) Cellular infrastructure networks. The WMAN type of technology could be employed in infrastructure - based vehicular networks alone, or in coordination with Wireless LANs or Wireless PANs (and their ad hoc multihop types) as last-hops. It provides a high-potential solution for vehicular networks, even for distant highway environments. Another possibility is to Vol. 25, No. 2, March-April’08

maintain permanent direct links from vehicles to cellular base stations, without the direct communication among vehicles. However, from the cellular network perspective this will probably result in a relatively low throughput. Features of WiMax are given in Table 1.

5.2. WLAN/WiFi WiFi is another possibility for vehicular networks. An IEEE 802.11 transmitter has a 250 mts omni directional coverage range, which is potentially sufficient enough to maintain a level of multihop connectivity in both highway and urban regions. In addition, extended-vicinity antennas (umbrellas) could be employed in base stations, for covering larger distances. A lot of research has been done for the popular IEEE 802.11 wireless protocol, mostly for the MAC (CSMA/CA) and network layers. However, this research cannot be taken off the shelf for use in vehicular networks. This is because of the unique characteristics of VANETs that we have described in section 2. Features of WiFi standards are given in Table 2.

5.3. WPAN Wireless Personal Area Networks are used for short-range wireless communications (IEEE 802.15 or Bluetooth). Even though the data rates offered by WPAN are tempting, the short transmission range (maximum 10-20m) restricts the applicability of this technology to only dense urban-area vehicular networks. Features of Bluetooth are given in Table 3.

5.4. DSRC/WAVE Dedicated Short Range Communications (DSRC) was conceived to provide architecture for nodes within a vehicular network to communicate with each other and with the infrastructure. In DSRC, subsequently specialized as WAVE (Wireless Access in Vehicular Environments, also referred as IEEE 802.11p), GPS-enabled vehicles are equipped with on board units, which can communicate with each other to propagate information through Vehicle-to-Vehicle communications. DSRC/WAVE operates in the 5.9 GHz band (U.S.) or 5.8 GHz band (Japan, Europe) and has 75 I E T E

T E C H N I C A L

R E V I E W

69

TABLE 1: WMAN standards 802.x

Characteristics

Advantages

Applications

16 a

Operates in 2 to 11 GHz range. WiMAX, Range up to 50 Kms. Data rate per sector is between 60 and 70 Mbits/s. Support modulation methods from BPSK (binary phase shift keying) up to 64 QAM (Quadrature amplitude modulation).

Very long range. High data throughput. Thousands of users per site. MIMO (multiple-input and multipleoutput) support and turbo code support.

Broad band wireless access for rural and metropolitan areas. Backhaul for WLAN hotspots/hubs.

16 e

Operates in 2 to 6 GHz range. Mobile WiMAX, typical range is likely to be between one and three miles. Data rate is up to 15 Mbits/s. Support for adaptive modulation.

Support of low latency data, video, and real time voice services for mobile users at upto pedestrian speed. Backward compatible with 16 a base stations. MIMO and Enhanced LDPC (low-density parity-check) coding support.

Offers regional roaming

20

Operates in licensed bands between 500 MHz and 3.5 GHz. Mobile broadband wireless access, range up to 15 Kms. Data rates of 1 Mbit/s per user. Modulation rates from BPSK to 64QAM.

Support for mobile users at very high speeds of up to 250 Kmph. Also supports voice over internet protocol. Global mobility and roaming. Supports both convolutional and turbo coding.

Mobile users in motor vehicles and trains.

TABLE 2: WLAN/WiFi standards 802.x Characteristics

Advantages

Applications

11 a

Operates in unlicensed 5.725-to-5.850 GHz. Optimum range is 50 feet. High average throughput speed 6 to 54 Mbps. OFDM (Orthogonal frequencydivision multiplexing) modulation technique.

No interference.

Multimedia applications like home and business.

11 b

Three channels in the 2.4 GHz unlicensed frequency. Range up to 150 feet. Throughput speeds of 2 to 11 Mbits/s. DSSS (Direct sequence spread spectrum) modulation technique.

Low hardware price. Compatible with 11 g.

Home, campus, factory and office networking.

High average throughput speeds. Backward compatible with 11 b.

Multimedia applications like Home and office networks.

11 g Three channels in the 3.4 GHz unlicensed frequency. Range upto 150 feet. Throughput speeds of 6 to 54 Mbits/s. OFDM modulation technique.

Vol. 25, No. 2, March-April’08

I E T E

T E C H N I C A L

R E V I E W

70

TABLE 3: WPAN standards 802.x

Characteristics

Advantages

Applications

15.3 UWB (Ultra wideband)

Operates in 2.4 – 2.4835 GHz frequency. Data throughput speed of 110 Mbits/s at a range of 10 mts and 480 Mbits/s over 1 mt. OFDM, optional TDD (Time division duplex) modulation.

Can penetrate walls. Uses less power.

High speed home and office networking. Specialized imaging.

15.4 ZigBee

Operates 27 channels in three unlicensed frequency bands: 2.4 GHz, 902 to 928 MHz and 868 to 870 MHz. Covers up to 30 mts range. Peak data rate is 250 Kbps. CSMA-CA (Carrier Sense Multiple Access/Collision Avoidance), optional TDD.

Uses very low power.

Personal-areanetwork (PAN) for monitoring and controlling devices in home networks.

MHz of bandwidth allocated for vehicle communications, which are based on line of sight with a range of up to 1 km and vehicle speeds of up to 140 km/h.

6.

2.

Emanuel Fonseca & Andreas Festag, A Survey of Existing Approaches for Secure Ad Hoc Routing and Their Applicability to VANETS, NEC Technical Report NLE-PR-2006-19, NEC Network Laboratories, March 2006.

3.

Jun Luo & J. P. Hubaux, A Survey of Inter-Vehicle Communication, Proc. Embedded security in CarsSecuring current and Future Automotive IT applications, pp 164-179, Springer-Verlag, October 2005.

4.

Murat Caliskan, Martin Mauve, Bernd Rech & Andreas Luebke, Collection of dedicated Information in Vehicular Ad Hoc Networks, Proc. 12th World Congress on Intelligent Transport Systems 2005, San Francisco, U.S.A., November 2005.

5.

HolgerFler, MarcTorrent-Moreno, MatthiasTransier, RolandKrger, HannesHartenstein & Wolfgang Effelsberg, Studying Vehicle Movements on Highways and their Impact on Ad Hoc Connectivity, Proc. ACM MobiCom 2005, Cologne, Germany, August 2005.

6.

SaschaSchnaufer, HolgerFler, MatthiasTransier & WolfgangEffelsberg, Vehicular Ad Hoc Networks: Single-Hop Broadcast is not enough, Proc. 3rd International Workshop on Intelligent Transportation (WIT 2006), Hamburg, Germany, pp 49-54, March 2006.

7.

SaschaSchnaufer, HolgerFler, MatthiasTransier & WolfgangEffelsberg, TrafûcView: Trafûc Data Dissemination using Car-to-Car Communication, ACM Mobile Computing and Communications Review (MC2R), vol 8, no 3, pp 6-19, July 2004.

CONCLUSIONS

VANETs combine short-range communications, with the scalability and mobility of classic ad hoc networks, in order to support a number of applications aiding in the safety, entertainment and simplification of everyday driving. Emerging wireless technologies are expected to enhance the better models in vehicular networks, if both the automotive manufacturers and the research community show great interest in creating efficient, interoperable standards. This article described the characteristics of VANETs and raised several issues such as media access, routing, congestion, performance analysis, security, etc., for implementing intelligent transportation system and services. The article has also reviewed the ongoing progress in the area of VANETs and the technologies to be adopted for VANET deployment. The article has given enough coverage to understand the research problems in VANETs. REFERENCES 1.

Applications, Proc. Radio and Communication, Linkoping, Sweden, pp 375-380, 2005.

M Rydstrom, A Toyserkani, E Strom & A Svensson, Towards a Wireless Network for Traffic safety

Vol. 25, No. 2, March-April’08

I E T E

T E C H N I C A L

R E V I E W

71 8.

9.

10.

11.

S S Manvi, M S Kakkasageri, Jeremy Pitt & Alex Rathmell, Multi Agent Systems as a Platform for VANETs, Proc. Autonomous Agents and Multi Agent Systems (AAMAS), ATT, pp 35-42, Hakodate, Japan, May, 2006. S. S. Manvi, M. S. Kakkasageri & Jeremy Pitt Information Search and Access in Vehicular Ad hoc Networks (VANETs): An Agent Based Approach, Proc. International conference on Communication in Computing (CIC-2007), The 2007 World Congress in computer Science, Computer Engineering and Applied Computing, Las Vegas, Nevada, USA, pp 23-29, June, 2007. H Fubler, M Moreno, M Transier, A Festag & H Hartenstein, Thoughts on a protocol Architecture for Vehicular Ad hoc Networks, Proc. 2nd International Workshop in Intelligent Transportation (WIT 2005), Hamburg, Germany, pp 41-45, March 2005. M M Artimy, W Robertson & W J Phillips. Connectivity in inter-vehicle ad-hoc networks, Proc. Engineering Canadian Conference on Electrical and Computer, vol 1, pp 293-298, May 2004.

Distributing Inter-Vehicle Warning Messages, Proc IEEE conference on local computer networks (LCN), Florida, USA, 2006. 20.

Tamer Nadeem, Pravin, Shankar & Liviu Iftode, A Comparative Study of Data Dissemination Models for VANETs, Proc. The 3rd ACM/IEEE Annual International Conference on Mobile and Ubiquitous Systems: Networks and Services (MOBIQUITOUS 2006), San Jose, California, July 17- 21, 2006.

21.

Fabio Picconi, Nishkam Ravi, Marco Gruteser & Liviu Iftode, Probabilistic Validation of Aggregated Data in Vehicular Ad-hoc Networks, Proc 3rd international workshop on Vehicular ad hoc networks, pp 76 - 85, 2006.

22.

Timo Kosch, Christian Schwingenschlgl & Li Ai, Information Dissemination in Multihop Inter Vehicle Networks - Adapting the Ad-hoc On-demand Distance Vector Routing Protocol (AODV), Proc IEEE 5th International Conference on Intelligent Transportation Systems, Singapore, 2002.

23.

S Nittel, M Duckham & L Kulik, Information Dissemination in Mobile Ad-hoc Geosensor Networks, Proc. Third International Conference on Geographic Information Science (GIScience 2004), College Park, Maryland, October 2004.

12.

Stephan Eichler, Florian Dtzer, Christian Schwingenschlgl, Javier Fabra & Jrg Eberspcher, Secure Routing in a Vehicular Ad Hoc Network, IEEE VTC 2004 Fall, Los Angeles, USA, September, 2004.

13.

Vinod Namboodiri, Manish Agarwal & Lixin Gao, A study on the feasibility of mobile gateways for vehicular ad-hoc networks, Proc First International Workshop on Vehicular Ad-hoc Networks, Philadelphia, USA, Oct 2004.

24.

Menouar, Hamid Lenardi, Massimiliano Filali & Fethi, Improving Proactive Routing in VANETs with the MOPR Movement Prediction Framework, Proc. 7th International conference on ITS, Sophia Antipolis, France, pp 1-6, 2007.

14.

Stephan Eichler et al., Performance Evaluation of the IEEE 802.11p WAVE Communication Standard, Proc. 1st IEEE International Symposium on Wireless Vehicular Communications (WiVeC), Baltimore, USA, September 2007

25.

15.

F Borgonovo, A Capone, M Cesana & L Fratta, ADHOC MAC: a new MAC architecture for ad hoc networks providing efficient and reliable point-to-point and broadcast services, ACM Wireless Networks (WINET) Journal, July 2004.

M Al-kahtani & H Mouftah, Congestion control and clustering stability in wireless ad hoc networks: Enhancements for clustering stability in mobile ad hoc networks, Proc 1st ACM Workshop on Quality of service and security in wireless and mobile networks, 2005.

26.

Niranjan Potnis, Atulya Mahajan, Mobility models for vehicular ad hoc network simulations, Proc. 44th annual Southeast regional conference, Florida, pp 746-747, 2006.

27.

J Y Le Boudec & M Vojnovic, Perfect Simulation and Stationarity of a Class of Mobility Models, IEEE Infocom 2005, Miami, Florida, 2005.

28.

Fiore, Harri, Filali & Bonnet, Vehicular Mobility Simulation for VANETs, Proc 40th Annual Simulation Symposium (ANSS’07), Norfolk, USA, pp 301-309, 2007.

29.

http://www.webs1.uidaho.edu/niattproject/ corsim.html

30.

http://www.paramics-online.com

31.

http://www.csie.ncku.edu.tw/ klan/move/

16.

A Nasipuri et al., A MAC Protocol for Mobile Ad Hoc Networks Using Directional Antennas, Proc. IEEE WCNC 2000, Chicago, vol 1, pp 1214-1219, September 2000.

17.

Z Huang et al., A Busy Tone-Based Directional MAC Protocol for Ad Hoc Networks, Proc IEEE MILCOM 2002, Anaheim, CA, October 2002.

18.

Bechler, M & Wolf, L, Mobility Management for Vehicular Ad hoc Networks, Proc. Vehicular technology conference, vol 4, pp 2294-2298, 2005.

19.

Stephan Eichler, Christian Merkle & Markus Strassberger, Data Aggregation System for

Vol. 25, No. 2, March-April’08

I E T E

T E C H N I C A L

R E V I E W

72 32.

Owen, L E, Yunlong Zhang, Lei Rao & McHale, G, Trafûc Flow Simulation Using CORSIM, Proc. Simulation Conference, Orlando, USA, pp 1143-1147, 2000.

36.

M Jakobsson, Xiao Feng Wang & S Wetzel, Stealth Attacks in Vehicular Technologies, Proc. First International Workshop on Vehicular Ad-hoc Networks, Philadelphia, USA, Oct. 2004.

33.

Jong Youl Choi, Markus Jakobsson & Susanne Wetzel, Balancing Auditability and Privacy in Vehicular Networks, Proc. 1st ACM international workshop on Quality of service and security in wireless and mobile networks (Q2SWinet ’05), October 2005.

37.

J Guo, J P Baugh & S Wang, A Group Signature Based Secure and Privacy Preserving Vehicular Communication Framework, Proc Mobile Networking for Vehicular Environments (MOVE) workshop in conjunction with IEEE INFOCOM, Anchorage, Alaska, 2007.

34.

Klaus Pll, Thomas Nowey & Christian Mletzko, Towards a Security Architecture for Vehicular Ad Hoc Networks, Proc. The First International Conference on Availability, Reliability and Security (ARES 2006, IEEE Computer Society Conference Publishing Services, pp 374-381, Los Alamitos, 2006.

38.

M Raya & J P Hubaux, The Security of Vehicular Ad Hoc Networks, Proc Third ACM Workshop on Security of Ad Hoc and Sensor Networks (SASN), New York, USA, 2005.

39.

Y F Ko, M L Sim & M Nekovee, Wi-Fi based broadband wireless access for users on the road, BT Technology Journal, vol 24, no 2, pp 123-129, April, 2006.

40.

C Laurendeau & M Barbeau, Threats to Security in DSRC/WAVE, Proc 5th International Conference on Ad-hoc Networks, Ottawa, Canada, pp 266-279, 2006.

35.

Amer Aijaz, Bernd Bochow, Florian Dtzer, Andreas Festag, Matthias Gerlach, Rainer Kroh & Tim Leinmller, Attacks on Inter-Vehicle Communication Systems Analysis IT 2006, Proc 3rd International Workshop on Intelligent Transportation, Hamburg, Germany, March 14-15, 2006.

Authors Sunilkumar S Manvi completed his PhD from Indian Institute of Science, Bangalore. Presently he is serving as a Professor of Department of Computer science and Engineering, REVA Institute of Technology and Management, Bangalore. His areas of research include wireless/wired networks, AI applications in network management, Ecommerce, Grid Computing and multimedia communications. He has published over 25 papers in referred national/international Journals and 60 papers in referred national/ international conferences. He has coauthored books “Communication Protocol Engineering” and “Computer Concepts and C Programming” published by PHI. He is a reviewer of several reputed international/national Journals. He is a member of IEEE USA, Fellow of IETE, India, Fellow of IE, India and member of ISTE, India. He has been included in Marqui’s Who’s Who in World and International Biographies of Cambridge, London in the year 2006. Address: Department of Electronics and Communication Engineering, Basaveshwar Engineering College, Bagalkot 587 102, India. Email:

Mahabaleshwar S Kakkasageri completed his MTech from Visvesvaraya Technological University Belgaum, Karnataka. He is pursuing his PhD in the area of Vehicular Ad hoc Networks (VANETs). Presently he is serving as a Lecturer of Department of Electronics and Communication Engineering, Basaveshwar Engineering College, Bagalkot, Karnataka. He has published 03 papers in referred national/ international Journals and 06 papers in referred national/international conferences Address: Department of Electronics and Communication Engineering, Basaveshwar Engineering College, Bagalkot 587 102, India. Email: [email protected]>

Paper No 146-A; Copyright © 2008 by the IETE. Vol. 25, No. 2, March-April’08

I E T E

T E C H N I C A L

R E V I E W

73

INVITED ARTICLE

MediaFLO™ - The Ultimate Mobile Broadcast Experience SACHIN KALANTRI ABSTRACT The exponential growth in wireless penetration and advancement in technology has accelerated the development of new and exciting wireless services. Given the mass appeal for video and multimedia content, technology providers have debated the feasibility and economical viability of large scale delivery of high-quality multimedia content to a wide range of wireless subscribers. Although delivery of this type of content is technically feasible over today’s existing unicast networks such as 3G, these networks cannot support the volume and type of traffic required for a fully realized multimedia delivery service (many channels delivered on a mass market scale). Offloading multicast (one-to-many) multimedia traffic to a dedicated broadcast network is more efficient and less costly than deploying similar services over 3G networks. Multicast services, such as the FLO mobile broadcast platform, are built ground up to address the market demand for mobile media and provide the critical link between technical feasibility and economic viability. Designed to work in concert with existing cellular data networks, FLO effectively addresses the issues in delivering multimedia content to a mass consumer audience. Unencumbered by legacy terrestrial or satellite delivery formats, this technology offers better performance for mobility and spectral efficiency than other mobile broadcast technologies, offering twice the channel capacity. FLO is a globally-recognized, open technology standard with a broad-based licensing program. The FLO Forum, with 90+ active members, including Huawei, LG Electronics, Motorola, Samsung and Sony Sharp, is driving the global standardization of MediaFLO Technology.

MOUNTING DEMAND FOR MOBILE TV In 2007, worldwide mobile telephone subscriptions reached 3.2 billion – equivalent to half the global population. India’s mobile subscriber base totaled 281.62 million at the end of January 2008 [1]. The exponential growth in wireless penetration and advancement in technology has accelerated the development of new and exciting wireless services. The mobile phone has become indispensable in India today, increasingly providing consumers with access to targeted and personalized content. The advent of mobile TV is one of the most exciting of all the latest capabilities transforming the mobile phone today. Research suggests that mobile TV service is around four times more appealing than mobile gaming [2,3]. That revelation, coupled with the fact that there was a 15 percent increase in the worldwide sale of mobile handsets last year [4], highlights the potential for a truly a mass-market technology. Spurred by the strong appeal for video and multimedia content, technology Vol. 25, No. 2, March-April’08

providers have constantly debated the feasibility and economical viability of large scale delivery of high-quality multimedia content to a wide range of wireless subscribers.

ADVANTAGES OF A BROADCAST NETWORK

DEDICATED

Although delivery of multimedia content is technically achievable over today’s existing cellular networks, these networks cannot support the volume and type of content traffic required for a fully realized multimedia delivery service (many channels delivered on a mass market scale). The best mobile TV experience is delivered over a dedicated mobile broadcast network, which aggregates programming and prepares it for transmission to handsets. 2.5 G or 3G telephony is configured for one-to-one or “unicast” network connectivity – this enables streaming of live content to mobile handsets but the quality of the broadcast will deteriorate as the number of viewers increase. I E T E

T E C H N I C A L

R E V I E W

INVITED ARTICLE

74

Fig 1 Forecast showing significant growth for mobile TV [2]

By establishing a dedicated broadcast network for mobile TV, operators can prevent any degradation to existing voice and data services. A dedicated mobile broadcast network allows pay TV providers to deliver a range of different channels

and services, while maintaining a very high quality user experience. In other words, they can provide a compelling mobile viewing experience that mimics what consumers have become accustomed to after more than 70 years of conventional television.

Note: Based on server cost for 1,000 to 3,000 simultaneous users is $40,000 (Source: Bakhizen & Horn, 2005, and Interview with Wireless Carrier by Robert Hale & Associates). Fig 2 Vol. 25, No. 2, March-April’08

I E T E

T E C H N I C A L

R E V I E W

INVITED ARTICLE

Figures 2 & 3 show how content server costs via unicast technology rise based on the number of users; however, mobile broadcast technology costs remain fixed for unlimited users.

MediaFLO - The Technology with an Edge The MediaFLO mobile broadcast platform enables the delivery of high-quality media to countless mobile subscribers, with more efficient coverage and higher channel capacity than alternative technologies. A key component of the MediaFLO mobile broadcast platform is the FLO air interface, a globally-recognized, open technology standard which is purpose-built to efficiently broadcast rich multimedia content over both singlefrequency and multiple frequency networks with no impact on the capacity of the cellular network. This allows for the efficient distribution of mobile television in the most cost–effective manner. As a result, existing cellular networks can be preserved for core voice, SMS, data, and other value-added services. The MediaFLO platform supports layered modulation and source coding which extends the geographic coverage area while providing a graceful degradation of service. Consumers can receive signals where reception would not otherwise be possible, and this efficiency provides better coverage along with higher quality services. Network coverage is more predictable, and that adds up to a better quality of service. MediaFLO technology can deliver superior service with roughly half the infrastructure as compared to legacy broadcast technologies. Alternatively, it supports twice as much capacity for content programming within the same amount of spectrum and geographic coverage area. Greater capacity equates to a dramatic increase in revenuegenerating services and optimizes operational costs while delivering a compelling end-user experience. This clearly differentiates the MediaFLO platform from alternative technologies. Eliminating wait times associated with downloading and buffering content, MediaFLO technology provides a more immediate, immersive, interactive experience. Channel switching takes approximately two seconds, replicating the channel surfing experience that consumers have grown accustomed to while watching TV at home. Vol. 25, No. 2, March-April’08

75

Invented for the mobile environment, the MediaFLO platform optimizes power consumption on the mobile device (4 hours of view time on a standard battery). This technology provides superior carrier to noise performance that very favorably impacts a network’s infrastructure cost. A factor of two reductions in transmitter costs is a reasonable outcome for similar bit per second Hz capacity modes – based on current independent field test results using typical log 30 propagation. MediaFLO achieves superior throughput and spectral efficiency in 5, 6, 7 and 8MHz channels. For example, in a single 8 MHz RF channel, MediaFLO can support up to 30+ streaming channels of QVGA-quality video and AAC+ stereo audio as well as multiple Clipcasting™ downloads per day.

Live mobile TV for the masses MediaFLO provides the delivery platform for a compelling mix of content and services, including broadcast TV, which help reach a mass viewing audience in the millions. Wherever they go, subscribers can watch their favorite sports, ‘mustsee’ shows, and live newscasts. By multiplexing channels, MediaFLO technology can pack more than twice as many services into available spectrum than competitive technologies at the same level of quality. And because it uses an efficient transport mechanism with less overhead, there is additional capacity for more tailored content such as short-format video clips and streaming data services.

Clipcasting™ Media Consumers have a nearly insatiable demand for personalized, timely, and informative content. The MediaFLO system allows them to tap new revenue with Clipcasting media – short-format video clips delivered on a scheduled basis. From news and individual sports highlights to movie moments and comedy clips, consumers can subscribe to a wide variety of Clipcasting media that is broadcast and stored on their devices.

IP Datacasting Applications The mobile lifestyle is all about staying informed and connected, and the MediaFLO system can help by instantly delivering a large number of data and I E T E

T E C H N I C A L

R E V I E W

INVITED ARTICLE

76

services. This persistent stream of information can provide constant real-time updates to applications such as news tickers, sports scores, financial reports, weather, and traffic. By combining the delivery of live TV with value-added services such as Clipcasting and IP datacasting, MediaFLO is providing the most compelling mobile media solution available with no compromise to the user experience (Table 1).

Media FLO vs Competitive Technologies A number of technologies address, at least partially, the requirements of mobile multimedia. These technologies are mostly variants or derivatives of an existing digital television broadcast format. The Table 1 lists a number of significant features of the individual formats and the implications for the user. By providing an ideal balance of technical performance parameters, MediaFLO provides a superior user experience. The ability to change

channels quickly is always important to the user. Watch time should be comparable to talk time, if not longer, so as to not compromise the functionality of the device. The capacity of the system is optimized when per application QoS is available in a network. A combination of both real-time and non-realtime media provides the best overall content mix. The delivery of non-real-time content allows immediate access to content such as weather or news summaries by topic, while real-time streaming services support live events such as sports. The ability to support both wide-area and local content within a single RF carrier allows an operator to maximize the value of the available spectrum through the flexible allocation of channels.

Why Service Provider Should Choose Media FLO The selection of a mobile broadcast technology can have a strong influence on network deployment costs. Several factors help determine the cost:

TABLE 1: Service experience and features Format

Average Channels Switching

Time Video Watch Time With 850 mAhr Battery

Per Channel QoS [4]

File Download

Local- and Wide-Area in Single RF Channel

ISDB-T

~1.5 sec

unknown

Yes

No

No

T-DMB

~1.5 sec

~2 hours

Possibly

Possibly

No

S-DMB

~5.0 sec

~1.2 hours

No

No

No

DVB-H

~5.0 sec

Goal ~4 hours Demo ~2 hours with 1600 mAhr bettery

No

Possibly

No

MediaFLO

1.5 sec

~4 hrs with standard 850 mAhr battery

Yes

Yes + integrated Clipcasting solution with memory management, conditional access and subscription model

Yes

Vol. 25, No. 2, March-April’08

I E T E

T E C H N I C A L

R E V I E W

77

INVITED ARTICLE

•

Number of infrastructure sites that are required.

A GLOBAL ECOSYSTEM

•

Total spectrum required to support a defined channel line up.

•

Total number of transmitter assemblies required to achieve a service line up.

The introduction of mobile television services in India presents a major opportunity for the entire mobile TV value chain including: service providers, wireless operators, broadcasters, handset manufacturers, infrastructure providers, technology enablers, and content providers (Fig 3).

The table below shows the relative costs of utilizing the various technologies (Table 2). This comparison assumes that each system has the same link margin, which forces the capacity constraints. The table attempts to target 20 realtime services at 300kb/sec per service; however, due to structural limitation, some formats cannot achieve the desired link margin at the specified bit rate. In those cases, the product of average bit rate and number of services is held constant. This analysis shows that, due to the superior efficiency of the FLO air interface in the areas of Packet Error Rate (PER) performance, protocol efficiency, and the application of layered service and modulation, MediaFLO technology can deliver equivalent or superior service with roughly half the spectrum and less than half the infrastructure required. The implications for the user and service operator are significant relative to the cost and breadth of services that can be delivered.

Wireless Operators For wireless operators, new revenue streams are possible through subscriptions and mobile advertising. Wireless operators can also benefit from exclusive access to popular content, programming, or value-added services – such as Clipcasting, IP datacasting, interactive services – or by exclusive rights to sell certain handsets.

Content Providers and Programmers For content providers, mobile TV presents an exciting opportunity to increase revenue and extend their services into the mobile space by leveraging existing content assets and developing new mobile content. Content providers and programmers may also be able to negotiate for a share of advertising revenues.

TABLE 2: Required infrastructure for comparable service Format

Channels Per Transmitter

Infrastructure Costs for 20 Channels

Channels per MHz

Required Spectrum for 20 Channels

ISDB-T

13 channels, 6 MHz ~ 230kbps each

~2X

~2

12 MHz (26 lower quality channels)

T-DMB

3 channels, 1.5 MHz ~ 250kbps each

~4-6X

~2

10.5 MHz

S-DMB

~20 channels, 25 MHz

Broadcast satellite plus terrestrial repeaters

<1

25MHz

DVB-H

9 channels, 6 MHz ~ 300kpbs each

~2X

1.5

12MHz

MediaFLO

20 channels, 6 MHz ~ 300kbps each

Reference (1X)

>3

6MHz

Vol. 25, No. 2, March-April’08

I E T E

T E C H N I C A L

R E V I E W

INVITED ARTICLE

78

Fig 3 The mediaFLO ecosystem bringing all players in the value chain together

Handset manufactures Handset manufacturers are able to sell advanced products to wireless operators which may exportable to other countries that have also chosen MediaFLO as their preferred mobile TV technology platform.

Other stakeholders There are many other key players in the MediaFLO ecosystem For example, infrastructure vendors are very critical to the success of the MediaFLO platform. They fall in a variety of business categories including transmitters, encoders, and Conditional Access Systems (CAS). There are numerous opportunities for companies to supply their mobile TV infrastructure in support of the MediaFLO ecosystem.

CONCLUSION While the mobile TV market is still in its infancy, it is expected to be a key driver in the sale of mobile

Vol. 25, No. 2, March-April’08

devices and the creation of new multimedia services. Currently, India is fourth in global wireless penetration and the consumers here are increasingly looking at their mobile phones as ‘allpurpose’ devices. This trend, coupled with our huge appetite for entertainment on the television hints at the future success of mobile TV in the country. MediaFLO technology was designed specifically to address the global market demand for mobile multimedia services, making them more economical, efficient, and accessible than ever before. REFERENCES 1.

TRAI January 2008.

2.

Informa – August, 2007, Juniper – September, 2007, TeleAnalytics – November, 2007(this compilation shows that Mobile media subscribers are expected to grow significantly across the globe).

3.

FICCI-PWC report 2006- 2007.

4.

Q3 mobile handset sales: Gartner (source: http:// www.thehindubusinessline.com/2007/11/28/stories/ 2007112851720400.htm)

I E T E

T E C H N I C A L

R E V I E W

INVITED ARTICLE

79

Author Sachin Kalantri is a senior staff engineer, Qualcomm, India. Sachin is responsible for promoting MediaFLO Technology in the South Asia region and is actively involved in field trials and product deployment for MediaFLO. In the past, Sachin has been the Chair of the Test and Certification Committee under the FLO Forum, a US based organization that is responsible for the standardization of FLO Technology and has worked extensively on QSEC 800, BREWChat, QChat, MediaFLO and Digital Cinema. Prior to joining Qualcomm, Sachin has worked with leading companies such as IBM, Bellcore, TATA Consultancy Services and P&O NedLloyd. He has been associated with MTNL, Indian Air Force, and DRDO on various projects. Sachin has experience of more than 17 years in the telecom industry, in areas spanning wireless systems, residential broadband, mobile broadcast, video on demand, VoIP and push-to-talk systems. Sachin holds a Bachelors degree in Electronics and Telecommunications Engineering from the Government College of Engineering, Pune and a Masters degree in Electrical Control Systems from the Indian Institute of Information Technology, Kharagpur.

Email: <[email protected]> Paper No 172-B; Copyright © 2008 by the IETE.

Vol. 25, No. 2, March-April’08

I E T E

T E C H N I C A L

R E V I E W

IETE – R S KHANDPUR GOLD MEDAL AWARD LECTURE

81

Binaural Dichotic Presentation of Speech Signal for Improving its Perception to Sensorineural Hearing-Impaired using Auditory Filters D S CHAUDHARI ABSTRACT Sensorineural hearing loss relates to spread of spectral masking resulting in reduction in frequency resolving capacity. Such hearing-impaired persons find difficulties while identifying consonantal ‘place’ feature cued by spectral differences. In binaural dichotic presentation scheme splitting of speech signal in real-time into two signals with complementary short-time spectra using filters with magnitude response based on two auditory filter banks with linear phase was implemented and evaluated. Filter banks corresponding to eighteen critical bands over 5 kHz frequency range were used. Listening tests were performed on subjects with ‘mild’ to ‘very severe’ bilateral sensorineural hearing loss. The usefulness of the scheme for better reception of spectral characteristics was evident as the results indicated improvement in speech quality, response time, recognition scores and transmission of ‘place’ feature particularly.

INTRODUCTION Sensorineural hearing loss entails increase in hearing threshold, dynamic range reduction, and increase in temporal masking and hence degradation of temporal resolution and increase in spectral masking due to degraded frequency resolution. Amplitude compression and frequency compensation are incorporated in many hearing aids. Signal processing schemes such as spectral transposition and speech enhancement using the properties of ‘clear’ speech form the basis for several techniques that have been investigated. These techniques are supposed to enhance the performance of hearing aids for the persons with residual hearing. Besides, they are also likely to enhance the performance of other sensory aids like cochlear prostheses and vibro-tactile aids used by profoundly hearing-impaired. Widening of auditory filters along the cochlear partition characterise the increase in hearing loss [1]. The problem of degradation in speech perception due to broadening of critical bands corresponding to auditory filters is not adequately dealt with by above-mentioned processing schemes. Due to spectral masking, recognition of transition of formants and frequency bands of the noise bursts becomes difficult for a person.

Vol. 25, No. 2, March-April’08

Binaural dichotic presentation, in which speech signal is split into two complementary spectra, would solve above problem. In this, the signals corresponding to two neighbouring bands that are likely to mask each other get presented to different ears. Humans have ability to perceptually combine the binaurally received signals from the two ears for improving speech perception under adverse listening conditions [2]. Among several schemes used for binaural dichotic presentation, one employed synthesis of vowels with the alternate formants presented to the two ears. Possible fusion of the information at the higher levels in the auditory system resulted in proper perception of vowels. Some researchers tested the scheme of splitting speech using 8channel constant bandwidth of 700 Hz for binaural dichotic presentation. The scheme was experimentally evaluated by finding the signal-tonoise ratio (SNR) that satisfied 50% correct speech (word) recognition. An overall improvement of about 2 dB in SNR for the dichotic condition over diotic was indicated [3]. The proposed scheme used critical bands corresponding to auditory filters based on psychophysical tuning curves as described by one of the researchers [4]. Eighteen critical bands over

I E T E

T E C H N I C A L

R E V I E W

IETE – R S KHANDPUR GOLD MEDAL AWARD LECTURE

5 kHz frequency range were used. The magnitude response of each band is an approximation of an ideal filter with the bandwidth of critical band having linear phase response. The implementation was done using the off-line processing of speech signal on five normal hearing subjects with simulated sensorineural hearing loss of varying degrees in the age group of 21 to 40 years and ten hearingimpaired subjects in the age group of 18 to 58 years with ‘mild’-to-‘very severe’ bilateral sensorineural hearing loss [5,6]. The usefulness of the scheme was indicated by improvement in speech quality, response time decrease, enhancement in recognition scores. The better reception of consonantal ‘place’ feature without affecting the other features was realised by noting transmission of various speech features. Based on the above encouraging results, the scheme was implemented for real-time processing of speech signal. For comparison of the dichotic presentation of the processed signal with diotic presentation of the unprocessed ones, listening tests were carried out on subjects with bilateral sensorineural hearing loss.

1. METHOD 1.1. Subjects The bilateral sensorineural hearing-impaired subjects who participated in test were from different parts of India and had no difficulty in clearly recognising the test stimuli. The six right-handed hearing-impaired subjects were familiar with English and they had ‘mild’ to ‘very severe’ bilateral sensorineural hearing loss. The pure tone threshold averages (PTAs) are given in Appendix - A. Subjects’ PTA difference between right and left ear was from 4 to 30 dB.

1.2. Stimuli Nonsense syllables were used for stimuli for minimising the contribution of linguistic factors and maximising the acoustic factors. The twelve consonants /p, b, t, d, k, g, m, n, s, z, f, v / were used with vowel /a:/ as in father in vowelconsonant-vowel (VCV) and consonant-vowel (CV) contexts. In order to conveniently accommodate them on subjects’ screen in computerised test administration system, the number of stimuli was Vol. 25, No. 2, March-April’08

82

restricted to twelve. Speech stimuli were acquired and analysed using a PC based set-up. The signal from microphone goes to an amplifier, low pass filter (fp = 4.6 kHz, fs = 5.0 kHz, pass band ripple < 0.3 dB, stop band attenuation > 40 dB) and then through ADC of data acquisition board interfaced to a PC. DAC of data acquisition board was used for testing the stimuli. Syllables were spoken by a male speaker and each syllable was recorded a number of times. They were played back after spectrographic analysis. Out of these recordings, the syllables with most normal sounding were chosen to be stimuli.

1.3. Procedures Two DSP boards based on 16 bit fixed-point processor TI/TMS320C50 were used for real-time speech processing [7]. Each board comprises a processor along with analogue interfacing circuit (AIC) with 14-bit ADC, DAC with low pass filter and a programmable timer used for setting sampling rate. The processing set-up consisted of an input low pass filter, two DSP boards operating with sampling rate of 10 k samples/s and two audio amplifiers. In the off-line processing implementation cascade combination of band-reject filters [5] was used. Being computationally intensive it was found unsuitable for real-time implementation. An FIR filter with comb filter magnitude response was used for real-time implementation as shown in Fig 1. The frequency sampling technique of an FIR filter design was used and magnitude response was approximated with 128 coefficients. The filter program and coefficients were loaded into the program RAM on the DSP chip using serial port interface. After loading, the serial port was disconnected while keeping DSP boards powered. However, there was no data transfer between the two DSP boards. The magnitude of all filter bands was kept constant in real-time processing of speech stimuli. The magnitude response of each filter bank was obtained by applying sine waves of constant amplitude from 100 Hz to 5 kHz (∆f = 20 Hz) and plotted as shown in Fig 2 (pass band ripple < 2 dB, sideband attenuation > 28 dB, transition bands < 90 I E T E

T E C H N I C A L

R E V I E W

IETE – R S KHANDPUR GOLD MEDAL AWARD LECTURE

Filter 1

1

3

Filter 2

2

4

.........

83

17

s1(n)

18

s2(n)

s(n)

.........

Band

Passband frequency

Band

Passband frequency

1

-- – 0.20

2

020–0.30

3

0.30–0.40

4

0.40–0.51

5

0.51–0.63

6

0.63–0.77

7

0.77–0.92

8

0.92–1.08

9

1.08–1.27

10

1.27–1.48

11

1.48–1.72

12

1.72–2.00

13

2.00–2.32

14

2.32–2.70

15

2.70–3.15

16

3.15–3.70

17

3.70–4.40

18

4.40

Fig 1 Schematic representation for splitting of speech signal using two comb filters. The filter magnitude response is shown in each block (table shows 3-dB cut-off frequencies)

Hz). The magnitude response was verified by obtaining spectrograms using a spectrographic analysis set-up [8].

1.4. Listening tests Experimental investigation aimed at evaluating the effectiveness of the scheme in reducing the effects of spectral masking. The implementation had constant gain for all filter bands and listening tests were conducted on bilateral sensorineural hearing-impaired subjects. The stimuli were presented at the most comfortable listening level of an individual subject [9]. For each listening condition, subject did presentation level selections and it was kept constant for all the tests under a particular listening condition.

Vol. 25, No. 2, March-April’08

To find confusion among the set of 12 English consonants, listening tests were performed. Automated computerised test administration system was resorted to avoid the repetitiveness and time consuming nature of the tests [8]. The tests were administered for (a) an unprocessed speech diotically presented and (b) processed speech dichotically presented. The subjects were briefed about the test procedure.

1.5. Evaluation schemes Various researchers used different methods for performance evaluation of speech processing schemes. Intelligibility test and perceived sound quality judgement have been generally employed. In intelligibility test subject listened to a list of

I E T E

T E C H N I C A L

R E V I E W

IETE – R S KHANDPUR GOLD MEDAL AWARD LECTURE

84

(a)

(b)

Fig 2 Magnitude response of the filters used in real-time processing of speech signal: (a) left ear (b) right ear

standard words (e.g. spondees, phonetically balanced words, central institute for the deaf wordlist) and the correct responses were noted. The parameters like clarity, loudness, etc. were considered for perceived sound quality judgement. Though speech intelligibility test is well established, judgement of sound quality has been used for comparing hearing aids [10]. The recognition score versus SNR (15, 20, 25 and ∞ dB) for both types of hearing aids were plotted for comparing the performance. The evaluation scheme in which the processed speech is mixed with noise and listening tests were carried out to find the SNRs Vol. 25, No. 2, March-April’08

for 50% correct word recognition was also employed in some studies [10,11]. Some researches used this method with a variation of SNR in 3 dB steps [3]. The features responsible for the relative improvement cannot be assessed from intelligibility scores. Earlier subject responses were noted for 16 syllables in consonant-vowel context to the stimuli and scores were recorded in the form of confusion matrix. For studying the various features the stimulus-response cell entries were converted to stimulus-response confusion probabilities that I E T E

T E C H N I C A L

R E V I E W

85

IETE – R S KHANDPUR GOLD MEDAL AWARD LECTURE

were then subjected to information transmission analysis. The probability of correct responses is called recognition or articulation score. It can be obtained as the sum of probabilities in the diagonal cells. Recognition scores may have influence of the subjects’ response bias. A measure of covariance between stimuli and responses, employing mean logarithmic probability (MLP) measure of information was furnished in information transmission analysis [12]. The response time statistics shall help in comparing the speech processing and presentation schemes besides the information available from stimulus-response confusion matrices. In case of same recognition score or information transmitted as a result of two schemes, one with less response time can be considered as superior. In present study for evaluating the benefits of speech processing for dichotic presentation, the stimulus-response confusion matrix for the close set of speech stimuli was used and the response time statistics was obtained. For obtaining recognition scores and information transmission matrices were analysed. In order to study the contribution of various speech features, the cell entries in the matrix were used to obtain matrices by grouping stimuli with the same feature. The aim

of the scheme was to study the effect of spectral smearing due to the loss of spectral resolution. Hence improvement in the reception of place feature without adversely affecting the reception of other features was desirable. This required appropriate selection of set of stimuli. The nonsense syllable stimuli with 12 consonants formed two stimuli sets in VCV and CV contexts with vowel /a:/ were used for studying the reception of consonantal features of voicing, place, manner, nasality, frication and duration [9].

2. RESULTS AND DISCUSSION Listening tests were carried out on six hearingimpaired subjects in order to evaluate the scheme for dichotic presentation. The evaluation was done by qualitative assessment of the stimuli followed by assessment based on response times, recognition scores and information transmission for various features. The results from listening tests conducted with six hearing-impaired subjects in VCV and CV contexts are presented here. The speech stimuli were presented for comparing diotic presentation of an unprocessed speech with the dichotic presentation of processed speech with constant

Fig 3 Recognition scores in VCV and CV contexts : (a) for subject SG (b) averaged for the six subjects. US: unprocessed speech, PS: processed speech Vol. 25, No. 2, March-April’08

I E T E

T E C H N I C A L

R E V I E W

IETE – R S KHANDPUR GOLD MEDAL AWARD LECTURE

86

Speech Features (a)

Speech Features (b)

Fig 4 Percentage relative information transmitted for subject SG: (a) VCV and (b) CV contexts. OV: overall, DU: duration, FR: frication, NA: nasality, MA: manner, VO: voicing, PL: place

gain filter implementation. A compilation of subjects’ qualitative assessment about the test stimuli for ascertaining the improvements in speech quality was carried out. Subjects indicated definite preference for processed speech over unprocessed.

Vol. 25, No. 2, March-April’08

In order to compare the effectiveness of processing scheme in terms of load on perception process, average response time was used. For obtaining recognition scores stimulus-response confusion matrices were used, which were subjected to information transmission analysis in

I E T E

T E C H N I C A L

R E V I E W

IETE – R S KHANDPUR GOLD MEDAL AWARD LECTURE

87

Speech Features (a)

Speech Features (b)

Fig 5 Percentage relative information transmitted averaged for the six subjects: (a) VCV and (b) CV contexts. OV: overall, DU: duration, FR: friction, NA+ nasality, MA: manner, VO: voicing, PL: place

order to obtain a measure that is not affected by subjects’ response bias. The matrices resulting after combining the twelve stimuli in-groups were analyzed for reception of the consonantal features of duration, frication, nasality, manner, place and voicing.

Vol. 25, No. 2, March-April’08

In response time analysis, decrease in response time was observed due to processing for all the subjects. As seen by t-test, decrease in response time was statistically significant for most of the subjects. Highly significant decrease (p < 0.005) in both the contexts were indicated from

I E T E

T E C H N I C A L

R E V I E W

IETE – R S KHANDPUR GOLD MEDAL AWARD LECTURE

paired t-tests across the subjects. Most of the hearing-impaired subjects showed significant decrease in response time for processed speech. This indicated an improvement in listening condition with processing. The recognition scores for a subject and averaged across the subjects for both the contexts are plotted in Fig 3. The scores for processed speech (PS) implementation were higher than unprocessed speech (US) for all the subjects. The relative improvement in recognition score (R.S.) was calculated by following formula. [(R.S.)PS – (R.S.)US]/(R.S.)US Percentage scores ranged from 9.2 to 23.6 in VCV and 14.4 to 19.2 in CV contexts. Averaged across the subjects, the percentage improvement in the scores was 14 in VCV and 16.3 in CV contexts. For testing the statistical significance of improvements in scores due to processing, the recognition scores were subjected to t-test. Highly significant (p < 0.005) improvement in both contexts was observed for all the subjects. For testing the significance of the improvement due to processing paired t-test was also carried out across the subjects. And the improvements were highly significant (p < 0.0005) for both the contexts. The usefulness of the scheme was evident as all the subjects showed highly significant improvement in recognition score due to processing. The confusion matrices were subjected to information transmission analysis. The overall information transmitted as well as information transmitted for specific features were obtained for all the subjects. Figures 4 and 5 show information transmitted for a subject and averaged across the six subjects for VCV and CV contexts. Better reception of almost all the six features of duration, frication, nasality, manner, place and voicing contributed to overall improvements in speech reception. These improvements were observed to be higher for the features of manner, voicing, and place in both the contexts. Maximum improvement in place feature was observed for all subjects as well as averaged across the subjects. Averaged across the six subjects, the relative improvement for place feature was 34 and 41 % in VCV and CV contexts respectively. The contribution of all the six features of duration, frication, nasality, manner, place and Vol. 25, No. 2, March-April’08

88

voicing to overall improvement was indicated by information transmission analysis of the stimulusresponse confusion matrices. Maximum improvement was observed for place feature for almost all the subjects. Since the place information is linked to frequency resolving capacity of the auditory process, it can be said that the implemented scheme has reduced the effect of spectral masking without adversely affecting the reception of the features cued by amplitude and duration.

3. CONCLUSIONS The promising strategy for improving speech intelligibility for hearing-impaired listeners was implemented in which speech signal was split into two signals with complementary spectra employing critical bandwidth corresponding to auditory filters. The use of signal processing strategy resulted in improvement in overall speech reception quality in the listening test carried out on sensorineural hearing-impaired listeners with real-time processing of speech for dichotic presentation. The tests recorded remarkable fall in average response time for most of the subjects indicating reduction in load on perception process. Significant improvement was recorded in recognition score suggesting the corresponding enhancement in listening condition. Results indicated reduced effect of spectral masking since better reception of consonantal ‘place’ feature was observed implicating possibility of implementing the strategy in binaural hearing aids for persons with ‘mild’ to ‘very severe’ levels of binaural sensorineural hearing impairment. ACKNOWLEDGEMENT

The author wishes to the authorities of Department of Science and Technology, New Delhi for providing supports. REFERENCES 1.

J R Dubno & D D Dirks, Auditory filter characteristics and consonant recognition for hearing impaired listeners, Journal of the Acoustical Society of America, vol 85, pp 1666-1675, 1989.

2.

B C J Moore, An Introduction to the Psychology of Hearing, Academic, London, 1997.

3.

T Lunner, S Arlinger & J Hellgren, 8-channel digital filter bank for hearing aid use: preliminary results in I E T E

T E C H N I C A L

R E V I E W

IETE – R S KHANDPUR GOLD MEDAL AWARD LECTURE monaural, diotic and dichotic modes, Scandinavian Audiology Journal, S 38, pp 75-81, 1993. 4.

E Zwicker, Subdivision of audible frequency range into critical bands. (Frequenzgruppen), Journal of the Acoustical Society of America, vol 33, p 248, 1961.

5.

D S Chaudhari & P C Pandey, Dichotic Presentation of Speech Signal with Critical Band Filtering for Improving Speech Perception, Proceedings of IEEE International Conference on Acoustics Speech and Signal Processing, Seattle, Washington, paper AE3.1, 1998.

6.

D S Chaudhari & P C Pandey, Dichotic Presentation of Speech Signal using Critical Filter Bank for Bilateral Sensorineural Hearing Impairment, Proceedings of 16th International Congress on Acoustics (ICA), Seattle, Washington, 1998.

7.

Texas Instruments, User’s Guide, TI-TM320C5X Digital Signal Processor Products, Texas Instruments, USA, 1993.

89

8.

D S Chaudhari, Dichotic presentation for improving speech perception by persons with bilateral sensorineural hearing impairment, PhD thesis, IITB, Mumbai, 2000.

9.

C Simon, On the use of comfortable listening levels in speech experiments, Journal of the Acoustical Society of America, vol 64, pp 744-751, 1979.

10.

A Gabrielsson, B N Schenkman & B Hagerman, The effects of different frequency response on sound quality judgments and speech intelligibility, Journal of Speech and Hearing Research, vol 31, pp 166–177, 1988.

11.

D B Hawkins & W S Yacullo, Signal-to-noise ratio advantage of binaural hearing aids and directional microphones under different levels of reverberation, Journal Speech and Hearing Disorder, vol 49, pp 278-286, 1984.

12.

G A Miller, P E Nicely, An analysis of perceptual confusions among some English consonants, Journal of the Acoustical Society of America, vol 27(2), pp 338-352, 1955.

APPENDIX – A

Hearing thresholds for the hearing impaired subjects Subject Code (Sex, Age)

SG (M, 27)

SSN (M, 31)

KRV (M, 49)

BAS (M, 58)

SAV (M, 46)

LDM (M, 52)

Ear L: left R:right

Hearing thresholds (dB HL)

PTA

Frequency (kHz) 0.25

0.50

1.0

2.0

4.0

6.0

L

25

45

75

100

120

120

73

R

25

60

70

100

120

120

77

L

80

70

80

75

75

75

75

R

65

60

70

65

85

85

65

L

50

60

60

60

60

65

60

R

40

45

50

60

65

75

52

L

50

40

30

30

40

40

33

R

45

50

35

30

30

30

38

L

50

45

45

45

35

40

45

R

60

70

65

65

85

95

67

L

65

65

50

40

40

75

52

R

70

80

85

80

80

95

82

PTA: pure tone average hearing threshold level (dB HL), test frequencies: 0.5, 1 and 2 kHz.

Vol. 25, No. 2, March-April’08

I E T E

T E C H N I C A L

R E V I E W

IETE – R S KHANDPUR GOLD MEDAL AWARD LECTURE

90

Author Devendra Chaudhari, obtained BE, ME from Marathwada University, Aurangabad and PhD from Indian Institute of Technology, Bombay, Mumbai. He has been engaged in teaching, research for period of about 21 years and worked on DST-SERC sponsored Fast Track Project for Young Scientists. Presently he is working as faculty member in Department of Electronics and Telecommunication Engineering at Government College of Engineering, Amravati. Dr Chaudhari published research papers and presented papers in international conferences abroad at Seattle, USA and Austria, Europe. He worked as Chairman / Expert Member on different committees of All India Council for Technical Education, Directorate of Technical Education for Approval, Gradation, Inspection, Variation of Intake of diploma and degree Engineering Institutions. As a university recognised PhD research supervisor in Electronics and Computer Science Engineering he has been supervising research work since 2001. He has worked as Chairman / Member on different university and college level committees like Examination, Academic, Senate, Board of Studies, etc. He held chair position for one of the technical sessions of International Conference held at Nagpur. He is fellow and life member of various national, international professional bodies. He is recipient of Best Engineering College Teacher Award of ISTE, New Delhi. He has organized various Continuing Education Programmes and delivered Expert Lectures on research at different places. His present research and teaching interests are in the field of Biomedical Engineering, Digital Signal Processing and Analogue Integrated Circuits.

Address: Department of Electronics and Telecommunication, Government College of Engineering, Amrarali 111604. Email: Paper No 186-B; Copyright © 2008 by the IETE.

Vol. 25, No. 2, March-April’08

I E T E

T E C H N I C A L

R E V I E W

91

Dielectric Parameters as Diagnostic Tools and Indicatrix of Disease — A Microwave Study V MALLESWARA RAO, B PRABHAKARA RAO

AND

D M POTUKUCHI

ABSTRACT An X-band microwave, (9-10 GHz) technique to determine the dielectric blood parameters as an indicatrix for the severity of disease is reported. Relevance of dielectric parameters as the main indicators in the area of diagnostic tools is discussed. An experimental set-up and the relevant procedural details for the measurement of microwave blood parameters for the typhoid and diabetic disease is presented. Microwave (MW) dielectric parameters viz, dielectric constant εr*(ω), wave velocity ν (ω) and the impedance z(ω) are measured for collected blood samples (from hospitals) and compared to the clinical values reflecting the disease severity. The dielectric parameters are identified to exhibit similar trends as exhibited by clinical parameters for disease severity. Parametric relations are obtained to address the correlation between dielectric and clinical parameters reflecting the severity. The capability of MW dielectric measurements as the diagnostic tools and severity indicators is demonstrated.

INTRODUCTION At MW frequencies, the changes in the dielectric properties of tissues are closely related to frequency and to the amount of water present. The MW method of determining the lung water content is known [1] to utilise the changes in dielectric properties. The method is based on a continuous monitoring of the reflection / transmission coefficient to indicate changes in the permittivity of the lung tissue. This method has the advantage of using highly penetrating MW signals rather than ultrasonic signals, the later being highly attenuated and dispersed in the lung. Radiometry technique originates [2] from the fact that all bodies above absolute zero temperature emit energy in the form of electromagnetic radiation. The use of energy in the MW spectrum provides a method of controlling the rate and uniformity of heating of deep-frozen materials. MW thawing [3] techniques are known to recover the deep frozen organs from low temperature storage banks. In MW biology studies, waveguide systems are preferable as the fields are known. Recent investigations [4-6] have revealed that effect of non-ionising EM radiation on human body may not be restricted to thermal effects only, but it may help to explain some of the unsolved important biological activity. The propagation of electromagnetic (EM) radiation in a fluid needs to be characterised in Vol. 25, No. 2, March-April’08

terms of characteristic constants of the media, viz., the propagation constant γ ( i.e expressed in terms of attenuation constant α, and phase shift constant β). Needless to say that depend upon the nature of the media, one can estimate the media constants by measuring these parameters by propagating the EM radiation in the fluid. Measurements of MW region parameters [7,8] viz., complex dielectric constant εr*( ω), attenuation constant α, Phase shift constant β, electron (or ion) density ηe and collision frequency ‘ρ'’ provide the necessary information to estimate the characteristic constants of the components of the fluid under test. Finally, these measurements represents a family of diagnostic tools in the field of MW EM radiation. In the present research work, it is proposed to perform diagnosis by propagating EM radiation in blood plasma. An attempt is made to study the diversity of the blood parameters using MW EM radiation. The present work is based on the fact that the changes in electrical properties are caused by the prevalence of disease at varied severity in the human anatomy.

EXPERIMENTAL SET UP FOR MEASURING THE BLOOD PARAMETERS The block diagram of the experimental set-up [9] consisting of a MW bench used for measuring I E T E

T E C H N I C A L

R E V I E W

92

the blood parameters during the present investigation is presented in Fig 1. The dielectric constant of the collected blood sample is determined [10,11] with the experimental set up shown in Fig 1. The MW bench is operated at a frequency of 10 GHz.

RESULTS AND DISCUSSION The blood sample (collected from the patient) is divided into two equal parts. While the clinical test is administered ( by the hospital clinics) over the first part, the second part of the sample is exposed to microwave radiation as to further determine the dielectric parameters, i.e., that the sample is treated as the media of propagation. The clinical test on the first sample is known to come out with a parameter value, whose magnitude reflects upon the severity of the disease. As such, more is the clinical parameter, the severe is the disease attributed (or the blood sample of the patient is severely inflicted). The dielectric parameters of the blood samples (of patients with diversity of disease

collected from hospitals) is presented for the following case studies. 1) Typhoid 2) Diabetes The observed variation of MW blood parameters (collected from patients of different diseases) are presented in Fig 2 and 3 for typhoid and diabetes respectively.

1. Typhoid It is observed that for values obtained for Dielectric constant εr*(ω): (a) The dielectric constant εr*(ω) is found to be slightly greater than 2.95 for normal blood sample, the sample not affected by the typhoid disease. The corresponding value of clinical parameter is less than 80. (b) The dielectric constant εr*(ω) is found to vary between values 2 and 2.9, for the case of marginally affected typhoid sample.

VSWR meter

Klystron Power Supply

Klystron Oscillator

Crystal Detector

Isolator

Attenuator

Slotted Waveguide Section

Load

Fig 1 Experimental set up for the measurement of dielectric parameters of blood sample Vol. 25, No. 2, March-April’08

I E T E

T E C H N I C A L

R E V I E W

93

However, the corresponding value of the clinical parameter is found to be 80 for that marginally effected case. (c) The dielectric constant εr* (ω) is found to be less than 2, for the affected blood sample, while the corresponding value of the clinical parameter is found to be more than 80.

2. Diabetes It is observed that for values obtained for Dielectric constant εr*(ω): (a) found to be a value between 1.915 and 4.9 for the normal blood sample, the sample not affected diabetes. While corresponding value of the random sugar is found to be in between 60-160. (b) found to be greater than 5 for the diabetic affected blood sample. While the corresponding value of the clinical parameter (random sugar) is greater than 160 (mg/dL)

Statistical Analysis for parametric relations and correlation between MW dielectric and clinical parameters An overview of the collected data of clinical parameters and the observed dielectric parameters (at the various levels of disease severity) seems to maintain a strong correlation between them to address the problem of severity of disease and its

determination by MW dielectric method. The observed variation of MW dielectric parameter values with severity of disease is found to be analogous to the variation of clinical parameters (supplied by clinics) with severity of disease. As such, the possible correlation between them are estimated for different samples. The value of the correlation coefficient [12] between MW measurement and clinical measurement is estimated. The values of these coefficients are found to be in between 0.96 6 to 0.996 to imply that there is a strong correspondence between the two methods used to estimate the severity of disease. Further a meticulous and in depth analysis is carried out for a possible parametric dependence between the clinical measurement and MW measurement of blood parameters. An overview of variation of observed dielectric parameters (Tables 1 and 2) with clinical parameters (reflecting the severity of disease) seems to follow a third order polynomial dependence. A non linear least square method is used to fit the data to the equation f(x) = a + bx + cx2 + dx3

(1)

Where x is dielectric value The data in Tables 1 and 2 is fitted to the equation (1). The goodness of the fit [13] is demonstrated through the corresponding t-test and the p values ≥ 0.995. The back estimated values are superposed as solid lines in the Figs 2 and 3 for

TABLE 1: The values of clinical parameters and dielectric parameters for typhoid disease S.No.

Clinical lab Parameter

Dielectric Constant

Velocity (108)m/s

Impedance (Ω)

Confirmation

1

640

0.7486

3.4673

435.73

Positive

2

620

0.917

3.1328

393.69

Positive

3

160

1.315

2.6161

;328.76

Positive

4

88

1.495

2.4536

308.33

Positive

5

80

2.065

2.0877

262.35

Marginally effected

6

80

2.864

1.7727

222.77

Marginally effected

7

20

7.005

1.1335

142.44

Negative

8

10

9.905

0.9532

119.79

Negative

Vol. 25, No. 2, March-April’08

I E T E

T E C H N I C A L

R E V I E W

94

TABLE 2 : The values of random sugar and dielectric parameters for diabetes Random Sugar (mg/dL)

Dielectric Constant

Velocity (108)m/s

Impedance (Ω)

1

63

1.915

2.1678

272.43

Negative

2

68

2.065

2.0876

262.35

Negative

3

71

3.155

1.6889

212.25

Negative

4

77

3.275

1.6577

208.32

Negative

5

152

4.965

1.3463

169.19

Positive

6

177

6.225

1.2024

156.10

Positive

7

231

7.255

1.1137

139.97

Positive

8

240

7.705

1.0807

135.82

Positive

9

292

8.625

1.0215

128.37

Positive

10

307

10.285

0.9354

117.55

Positive

11

328

10.675

0.9182

115.39

Positive

NORMAL

150

—— Polynormial x Experimental

100

x

x

50 x

0 0

0.5 Dielectric Constant

x

1

Clinical Parameters

Clinical Parameters

S.No.

ABNORMAL

800 600 400

200 0 –200

Confirmation

x

x

—— Polynormial x Experimental

x

0.2

x

x

0.4 0.6 0.8 Dielectric Constant

1

Fig 2 Variation of the clinical parameter with dielectric constant of the blood sample for typhoid disease for both in normal and abnormal range of disease

—— Polynormial x Experimental

150

x

100 50

x x

0.2

xx

0.4 0.6 0.8 Dielectric Constant

1

Random Sugar

Random Sugar

NORMAL 200

350 —— x 300 250 200 150 100 0.4

ABNORMAL Polynormial Experimental x

x x x

x

x

0.6 0.8 Dielectric Constant

1

Fig 3 Variation of the random sugar with dielectric constant of the blood sample for diabetes disease for both in normal and abnormal range of disease

Vol. 25, No. 2, March-April’08

I E T E

T E C H N I C A L

R E V I E W

95

different diseases. An overall study of relative magnitude of the polynomial coefficients reflects upon highest value adopted by the linear coefficient term and its predominance for the strong correspondence (or correlation) between dielectric parameters and clinical parameters. This observation is suggestive of using MW dielectric parameters as an equivalent and potential tool of medical diagnostics.

CONCLUSIONS In this paper, the estimation of severity of ailment of typhoid and diabetes diseases are considered by measuring the dielectric constant, velocity and impedance of the blood sample. This is carried out by propagating electromagnetic waves through the blood sample. The trends of results and investigations are fruitful as the meticulous statistical analysis presented in the paper supports the claim that the microwave dielectric parameter can be used as a powerful tool to estimate the severity of disease. The above discussion for the observed trends in clinical and dielectric parameters followed by the meticulous statistical analysis is suggestive of

•

Subtle and strong implied relationship between the MW dielectric and the clinical parameters.

•

MW dielectric parameters can also be treated as bench markers to diagnose medical diagnostics.

•

Representing a reliable estimate of the severity of disease (affecting the biological system under consideration).

REFERENCE 1.

Magdy F Iskander & Carbh Durney, Electromagnetic Techniques for Medical Diagnosis: A Report, Proc IEEE, vol 68, no 1, Jan 1980.

Vol. 25, No. 2, March-April’08

2.

Om P Gandhi, Medical applications of Electromagnetic fields-Part V, IEEE Trans Microwave Theory Tech, July, 15, 1981.

3.

R Paglione and et al, 27 MHz Ridged Waveguide Applications for Localized Hyperthermia Treatment of Deep-seated Malignant Turners, Microwave Journal, vol 24, Feb 1981.

4.

S Gabriel, R W Lau & C Gabriel, The dielectric properties of biological tissues: III parametric models for the dielectric spectrum of tissues, Phy Mod Biol, vol 41, 1996, pp 227-2293.

5.

Yuri Feldman & Irina Ermolina, Time domain dielectric spectroscopy study of biological systems, IEEE Transactions on Dielectrics and Electrical Insulators, vol 10, no 5, 2003, pp 728-748.

6.

Liquid dielectric property determination using mono pole probes operating at microwaves frequencies, IEEE Instrumentation and Measurement Technology Conference 2006, Sorrent, Itley.

7.

Edward C Jordan & Keith G Balmain, Electromagnetic Waves and Radiating Systems, PHI, New Delhi, 2004.

8.

Rajeswari Chatterjee, Microwave Engineering special topics (A post graduate text book), Affiliated East West Press, New Delhi 1988.

9.

V Malleswara Rao & B Prabhakara Rao, The Role of Microwaves in the measurement of Blood Parameters, Journal Institute of Engineers India, pp 61-63, Jan 2006.

10.

V Malleswara Rao, B Prabhakara Rao, Blood Sugar Estimation Through Measurement of Dielectric Constant By Using Microwaves, 62nd ARFTG IEEE Microwave Measurements Conference Differential Measurements, Dec 2-5, 2003, Hotel Boulderado Downtown Colorado USA, pp 301-305.

11.

M L Sisodia & G S Raghuvamshi, Basic Microwave Techniques and Laboratory-Manual, New age International Pvt Ltd, New Delhi, 2000.

12.

K Park, Preventive and Social Medicine, M/s Banarsidas Bhanot publishers, Jabalpur, 2002.

13.

Suranjan Saha, Mathematics and Statistics Centre’s C.A. Foundation course series 2002, New central book agency (P) Ltd, Calcutta (India).

I E T E

T E C H N I C A L

R E V I E W

96

Authors V Malleswara Rao

Email: <[email protected]> Address: Department of Electronics and Communication Engineering, GITAM College of Engineering, Visakhapatnam 530 045, India. *

*

*

B Prabhakara Rao

Address: Depepartment of Electronics and Communication Engineering, Jawaharlal Nehru Technological University College of Engineering, Kakinada 533 003, India. *

*

*

D M Potukuchi

Email: <[email protected]> Address: Depepartment of Physics, Jawaharlal Nehru Technological University College of Engineering, Kakinada 533 003, India.

Paper No 79-A; Copyright © 2008 by the IETE.

Vol. 25, No. 2, March-April’08

I E T E

T E C H N I C A L

R E V I E W

To keep yourself up-to-date in state-of-the-art, subscribe to the Journals of the IETE (Revised subscription rates from 01.01.2007) There have been unprecedented advances in science and technology in recent years. New inventions and innovations are being reported in quick succession. The industry is becoming highly research-oriented and competitive. Growth of knowledge is increasing at such a rapid rate that a professional has an uphill task keeping oneself up-to-date in the areas of one’s specialization. Towards this end, IETE caters to the needs of electronics, telecommunication and IT professionals by publishing highly informative journals. IETE assists a professional through dissemination of current knowledge thereby enabling him / her to contribute substantially towards the profession and match the changing requirements of society. The IETE journals are indexed by reputed national and international abstracting agencies. 2. Leading organizations within the country and abroad including R&D establishments, educational institutions, polytechnics and libraries subscribe to the IETE journals. The non-member readers are invited to subscribe to IETE Journals. The brief scope of journals is as under:

²

IETE Journal of Research (Bi-monthly)*: contains R&D type articles, original research and review papers covering state-of-the-art technologies (ISSN 0377-2063)

²

IETE Technical Review (Bi-monthly)*: contains informative and general interest articles, reviews of technologies/products/designs (ISSN 0256-4602)

²

IETE Journal of Education (Four Monthly)**: contains tutorial and engineering education / academic type articles of interest to diploma/under-graduate level students, research scholars and teachers (ISSN-0970-1664)

3.

Annual Subscription Rates (Effective from 1st January 2007) for IETE Journals inclusive of inland delivery charges by Regd Book Post. Subscription Period: JAN to DEC Payments are accepted through Local Cheques /Demand Drafts / Pay Orders payable at New Delhi, in favour of IETE, New Delhi Inland ( Rs. ) Subscription

Foreign (US $) Subscription+Air Surcharge

IETE Journal of Research & IETE Technical Review

750.00

170 + 80

² ²

Above both Journals for Student Members of IETE

500.00

—

² ²

IETE Journal of Research

500.00

100 + 40

IETE Technical Review

500.00

100 + 40

IETE Journal of Education

150.00

35

IETE Journal of Education for Corporate Members of IETE

100.00

17

NOTE: For IETE Corporate members one Journal (either IETE Journal of Research or IETE Technical Review as per option received) is free of charge. Second journal, if required, is at nominal annual subscription of Rs.65/-.

NOTES

²

²

Subscription period: Jan to Dec Subscription Number will be allotted and journals will be mailed after receipt of payment in advance. Renewal requests latest by December All journals will be despatched within the country by Registered Book-Post only. Non-receipt complaints become time-barred after two months of publication. The same queries will not be attended repeatedly. WE DONOT HAVE ANY POLICY OF REFUNDS / REPLACEMENTS / SENDING OF BACK ISSUES. Kindly check the periodicity before claiming non-receipt of any issue: 1 st issue of bimonthly* and Four Monthly** Journals to be received by subscriber by mid March and mid May respectively.

IMPORTANT NOTE: This information may be treated as invoice where needed. No separate invoice will be issued. Mail completed form (visit www.iete.org) with payment, addressed to: Dy Managing Editor, IETE, 2, Institutional Area, Lodi Road, New Delhi- 110003, INDIA, Ph +91(11) 43538800-899 Extn.-843, E-mail: [email protected] & [email protected]

Registered with the Registrar of Newspapers for India, Delhi under Regn. No. 42814/84

For Advance Information of Authors A New System for On-Line Submission and Review of Articles for IETE Journals Subject of on-line submission & review of articles for IETE Journals has been under consideration for a long time. Keeping the present trend of Internet, automation and computerization at different levels, it was considered prudent and necessary to move from manual hard copy submission and review system to on-line system. The proposed software is expected to help authors, Honorary Editors, Reviewers, Editors-in-Chief and IETE HQs to streamline submission and reviewing process for timely publication of journals. The procurement of software system has been approved by the Council of IETE and will be operationalized in the near future. Effective date for system will be publicised in IETE Journals and website. The salient features of the proposed system are as follows:a)

The software system will create the website and a radio button to choose the journal from drop-down list. All the three journals will appear in the list. The authors will choose one of the journal and upload the manuscript.

b)

Resource templates based on approved standard would be hosted on IETE website the software as guidelines to authors for submission of articles on line.

c)

The chairpersons of the editorial boards would be the Editors-in-Chief for the three journals, i.e. IETE Journal of Education (JE), IETE Journal of Research (JR) and IETE Technical Review (TR).

d)

Each journal will have area editors from across the world as empanelled by the respective editor-in-chief. Those panels would be periodically reviewed taking into consideration the performance of the panelists.

e)

The concerned Editor-in-Chief will carry out on-line preliminary review for further processing or out-right rejection with intimation to author. The IETE HQ will keep the record of entire submission, revision, rejection and acceptance details.

f)

After final acceptance by the Editor-in-Chief, an e-mail will be generated automatically to the Administrator, Publication Section at IETE HQs who will have the access to the word / pdf document of the final accepted manuscript. The administrator will download it and carry out all editorial changes before it is given to Press for printing of the journals.

g)

The printed journals will be hosted on IETE website for free downloading by readers / authors and will be available for six months on IETE website & thereafter in our archive. Any suggestions and feedback are most welcome. Please email to [email protected]

Printed and Published by Brig (Retd) V K Panday for the Institution of Electronics and Telecommunication Engineers, 2, Institutional Area, Lodi Road, New Delhi 110 003 (India) and printed at the Shivam Offset Press, A-12/1, Naraina Indl. Area, Phase I, New Delhi-110 028 (India). Dy Managing Editor : CDR A P Sharma, IN (Retd) • Copyright © 2008 by the IETE