Elements Of Rf Link Design

Elements of Communication Link Design

S.Pal Ph.D S.N.Prassad

Draft Digital & Communication Area. ISRO Satellite Centre. Bangalore.

2005.

2

CONTENTS PAGE 1.0 2.0 2.1 2.2 2.3 2.4 2.5 2.6 2.7

Preface. 8 General. 9 Power Spectral Density. 9 Spectral Capacity 9 Transmission Area 9 TABLE 2.1:International Radio Frequency Designated Bands 4-11. TABLE 2.2:The old and new Listing of Frequency Bands. 10 TABLE 2.3: Most commonly used International System. 10 TABLE 2.4: Frequency Bands for Public and Military

Communications. 2.8 TABLE 2.5: comparison of Low-side and High side Tuning for AM broadcast band with fIF=455KHz. 2.9 TABLE 2.6 Power Spectral density of some common wireless broadcast and communication systems. 2.10 Satellite Slant Range Power. 2.10.1 Range or Wireless Distance Measurement. 2.11 Flux Density. 2.12 Spatial Capacity.

11

APPENDIX-A

16

A-1 A-2 A-3 A-4

17 18 19

A-5 A-6 A-7 A-8

Some Common Conversion Factors: Electromagnetic Frequency Spectrum. Frequency-Wavelength Conversion Spiral Nomogram. Satellite Elevation angle, Range and Loss due to Range; Spiral Nomogram. Power Flux Density Limitations. Power Flux Density Limitation Space to Earth. Maximum EIRP Limits. Low Energy Density & High Energy Density Systems 3

11 12 12 13 15 15

20 21

22 23 24

9

A-9 A-10 A-11 A-12 A-13 A-14 A-15 A-16 A-17 A-18 3.0 3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9 3.10 3.11 3.12

Fundamentals of Range Intersection Methods. 25 Fundamentals of multiple Range Intersection Methods. 26 Global Positioning System (GPS) with satellite navigation. 27 GPS Receiver Function. 28 Digital Signal Processing for GPS Receivers. 29 Precise Positioning Service (PPS) error budget. 30 Standard Positioning Service (SPS) error budget. 31 Differential GPS error budget. 32 GPS for civil aviation. 33 Key requirements for military grade GPS receivers. 34 LINK. 35 EIRP: Effective Isotropic Radiated Power. 35 Free-Space Loss /Spreading or Path Loss. 35 Elements of Link Budget. 36 TABLE 3-1: Steps in Ground to satellite Link Calculation. 37 TABLE 3-2:Uplink Link Parameters in tabular form. 38 C . 38 N C System . 39 N0 Link Budget for Digital satellites. 39 Table 3-3: Uplink and Downlink values of Eb/N0 for PE = 10-6 39 Typical example of a link. 40 G/T of different categories of INTELSAT Ground Stations. 42 Propagation above 10 GHz. 43

APPENDIX-B

44

B-1 B-2 B-3 B-4 B-5 B-6 B-7

45 46 47 48 49 50

Point to multipoint service. Typical antenna pattern in one plane. Electrical field polarization. Radiation pattern-intensity pattern. Frequency reuse by orthogonal polarization. Some typical antenna gain and bandwidths. Antenna gain wrt diameter.

4

51

B-8 B-9 B-10 B-11 B-12 B-13 B-14 4.0 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 4.10 4.11 4.12 4.13 4,14 4.15 4.15.1 4.16 4.17 4.18 4.19 4.20 4,21 4.22 4.23 4.24 4.24.1

Frequency reuse by Spatial Separation. Frequency reuse by orthogonal polarization. Carrier to Noise ratio in a link. Digital link performance. Spiral Nomogram for total link. Digital Television Performance. Television total quality measurement. NOISE. White Noise. Colored Noise. Noise bandwidth. Matched Filter. Thermal noise threshold. Intermodulation Noise. S Signal to Noise Ratio   . N Eb/N0 Table 4-1: Uplink and Downlink values of Eb/N0 required for PE = 10-6 Noise Figure. Noise Temperature. Thermal Noise aspects of Low-Noise Systems. C N0 C Calculation of . N0 S/N of a typical FM system. FM/FM Configuration. G/T. G/T Example. Intelsat Standard Ground Stations. Signal Fading. Diversity Reception. Types of Diversity combiners. FM threshold. FM threshold Extension. Noise in Analog modulated systems. S/N Ratio.

5

53 54 55 56 57 58

61

52

61 61 61 61

61 62 62 62 63 64 64 64 66 66 67

67 68 69 70 71 71 73 75 75 76 76

4.24.2 4.24.3 4.24.4 4.24.5 4.3 4.4 4.4.1 4.4.2

4.5.1 4.5.2 4.6

Base band System: Coherent System. Signal to noise ratios: DSB Systems. Signal to noise ratios: SSB System. Signal to noise ratios: AM system. Noise in Angle Modulation Systems. Pre emphasis & De-emphasis. 79 Pre Emphasis. 79 De-Emphasis. 80 Threshold Extension.

4.5 FMFB. Phase Locked Loop (PLL). PLL as a demodulator.

C-10 C-11 C-12 C-13 C-14 C-15 C-16 C-17 C-18

81 81 81 82

APPENDIX-C C-1 C-2 C-3 C-4 C-5 C-6 C-7 C-8 C-9

76 76 77 77 78

83

Angle Modulation with sinusoidal message signal. 84 Illustration of spike noise with sinusoidal modulation. 85 Spiral Noise figure to Noise temperature chart. 86 Tropospheric Noise Temperature. 87 Cosmic & Galatic noise temperatures. 88 Carrier/Total IMP power ratio for n carriers. 89 TABLE 4.1 Summary of noise performance characteristics, 90 0 Attenuation due to rainfall at 18 C. 91 S/N ratio improvement in Diversity system for various orders of diversity. 92 Rainfall attenuation at 24 GHz. 93 Spectra of angle modulated signal. 94 Amplitude spectrum of a FM signal wrt β. 95 FM performance with sinusoidal modulation. 96 Comparison of PM & FM modulator output. 97 FM power density spectra at discriminator output and at LPF output. 98 S/N and C/N of FM system. 99 FM output vs. input SNR due to non-sinusoidal modulating signal. 100 S/N ratio to probability of BER. 101

6

5.0 5.1 5.2 5.3 5.4 5.5 5.6 5.7

MODULATION. 102 Why Modulation. 102 Types of Modulation. 102 Types of Analog Modulation. 102 Types of Digital Modulation. 103 Amplitude Modulation. 103 Features of Amplitude Modulation/ Double Side band AM. 103 Single Side Band (SSB) AM. 104 5.8 Vestigial Side band (VSB) AM. 104 5.9 S/N ratio Improvement after demodulation. 104 5.10 S/N ratio Improvement AM System. 104 5.11 Frequency Modulation. 105 5.12 Phase Modulation. 106 5.13 Demodulation of narrow band FM and PM signals. 107 5.14 Local carrier generation methods. 108 5.15 Digital Modulation. 109 5.16 Types of Digital Modulation. 111 5.17 Amplitude Shift Keying. 112 5.18 Frequency Shift Keying. 112 5.19 FSK signal waveform. 113 5.20 Phase Shift Keying. 113 5.21 Comparison of Binary Digital Modulation Schemes. 114

7

5.22 5.23 5.24 5.25 5.26 5.27 5.28 5.29

Systems.

5.30 5.31 5.32 5.33 5.34 5.35 5.36

Differential Phase Shift Keying (DPSK). 116 Optimum Receiver. 117 TABLE 5-1: Typical Prrobability of Error For a PE of 10-4. 118 M-ary Data Communication. 118 Qadrature Multiplexing. 119 OQPSK or Staggered QPSK system. 121 MSK. 122 TABLE 5-2:Comparison of different Digital modulation 123 QPSK & MSK Waveforms. 123 Implementation Techniques of MSK Modulator. 124 Spectral Efficiency. 126 Spectral Features of Digital Modulation Systems. 126 Eb forPE = 10 −6 TABLE 5-3: N0

127 Bandwidth Efficiency on a main lobe spectrum of M-ary systems. 127 Bandwidth efficiency on a main lobe spectrum of M-ary Systems. 128

APPENDIX-D D-1 D-2 D-3 D-4

129

Waveform of ASK, PSK, FSK modulations. An example of digital modulation schemes. Bit error probability vs. Eb/No for M-ary PSK and 16QASK. Bit error probability Pe for various digital systems.

8

130 131 132 133

D-5 D-6 D-7 D-7 6.0 6.1 6.2 6.3 6.3.1 6.3.2 6.3.3 6.3.4 6.4 6.4.1 6.4.2 6.4.3

6.4.4 6.5 6.5.1 6.5.2 6.5.3 6.5.4

Bit error probability vs. Eb/No for coherent M-ary FSK. Digital link performance. Waveform for different line codes. Power Spectral Density of different line codes. RF Radiation hazards. Introduction. RF Radiation. Sources of Radiation. Broadcasting. 140 Communications. Radar. 141 RF Equipments or Machines. Effects of RF Radiation. Thermal Effects. Non-thermal Effects.

140 142 143 141 144 TABLE6.2 146 146 147 148 148

Non thermal Effects RF Safety standards. Leakage Standard. Exposure Standards TABLE- 6.3 149

TABL E- 6.4 151 TABL E- 6.5

6.5.5

6.6 7.0 8.0

134 135 136 137 138 138 138 139

152 153 154 155

Radiation Safety Management. Abbreviations. References. XXX---XXX

9

1.0 PREFACE: During estimation of a communication link for satellite lot of issues have to be considered. In this brief report an effort has been made to include as many related topics as possible, which a communication engineer normally faces during day-to-day working. It puts forward a brief reference to move along with design or analysis. It is recommended to consult more specialized books or literature for any detailed analysis of a topic. On these topics, related formulae, curves, nomograms and some examples have been also provided. S.N.Prasad

10

2.0 GENERAL: 2.1 Power Spectral Density: PSD = P (Power)/B (Bandwidth) = W/Hz. 2.2 Spectral Capacity: For narrow band systems the spectral capacity is given as: Maxm data rate (bps)/ Bandwidth (Hz) : bps/ Hz. 2,3 Spatial Capacity: The normally used term equivalent to spectral capacity in wide band systems is Spatial Capacity. It is defined as maximum data rate of a system over an area in which it is transmitted. The transmission area is calculated from the circular area assuming a transmitter in the centre. However, as a rule of thumb is to use the square of the maximum transmission distance. Maxm data rate (bps)/ Transmission Area (m2); bps/ m2. For narrow band systems, Spectral Capacity: bps/ Hz = Maxm data rate (bps)/ (Bandwidth) 2.3 Transmission Area (m2) = π (Transmission distance) 2 2.4 TABLE 2.1 International Radio Frequency Designated Bands 4-11 Band no Symbols Frequency range Metric subdivision 4 VLF 3-30 KHz Myriametric waves 5 LF 30-300 KHz Kilometric waves 6 MF 300-3000 KHz Hectometric waves 7 HF 3-30MHz Decametric waves 8 VHF 30-300MHz Metric waves 9 UHF 300-3000MHz Decimetric waves 10 SHF 3-30GHz Centrimetric waves 11 EHF 30-300GHz Millimetric waves The microwave frequencies apply 300 MHz and above.

11

2.5

2.6

TABLE: 2.2 The old and new listing of Frequency Bands. Old Frequency New Frequency band letter (GHz) NATO band (GHz) L 1-2 D 1-2 S 2-4 E 2-3 C 4-8 F 3-4 X 9-12.4 G 4-6 J/Ku 12.4-18 H 6-8 K 18-26.5 I 8-10 Q/Ka 26.5-40 J 10-20 U 33-60 K 20-40 V 50-70 L 40-60 E, O 60-90 M 60-100 W 75-110 T 110-170 TABLE 2-3: Most commonly used Symbol K M G T M µ N P

International Name Kilo Mega Giga Terra milli micro nano pico

12

System. Factor 103 106 109 1012 10-3 10-6 10-9 10-12

2.7

TABLE 2.4 Frequency Bands for Public and Military Communications Use Frequency Omega navigation 10-14 KHz World-wide submarine communication 30 KHz Loran C navigation 100 KHz Standard AM broadcast 540-1600 KHz FM broadcast 88-108 MHz Television: channels 2-4 54-72 MHz Channels 5-6 76-88 MHz Channels 7-13 174-216 MHz Channels 14-83 420-890 MHz

2.8 TABLE 2.5: Comparison of Low-side and High side tuning for AM fIF=455KHz. Lower frequency Upper frequency Standard AM 540 KHz 1600 KHz broadcast band LO frequency (540 –455) KHz (1600-455) KHz for low-side = 85 KHz = 1145 KHz tuning LO frequency (540 +455) KHz (1600+455) KHz for high-side = 995 KHz = 2055 KHz tuning

13

broadcast band with Tuning range of Local Oscillator ----13.7 to 1

2.07 to 1

2.9 TABLE 2.6 Power Spectral Density of some wireless broadcast and communication systems: Systems

Transmission Bandwidth Power Power(W) (Hz) Spectral Density (W/MHz) 50KW 75KHz 666.6 100 KW 6MHz 16.7 10mW 8.33 KHz 1.2 1mW 7.5 GHz 0.013

Radio Television 2G Cellular Ultra Wideband

Classificatio

Narrowband Narrowband Narrowband Ultra wideband

2.10 Satellite Slant Range:

Earth station

γ

Earth radius Re

R=Range

β

Orbit radius = r

C Fig 2.12

Satellite Position

α

Satellite slant Range

Slant range is given as: R = r cos α − Re2 − r 2 Sin 2α r = Re + h 0 Here α + (90 + γ + β ) = 180 : Re = Earth’s radius, R=slant range:

h= Satellite height from Earth: C= centre of earth. Example: Range =

( R + H ) 2 − R 2 cos 2 θ

− RSinθ

R= Radius of earth 6380 Km; H=Height of the satellite=800Km θ = Angle of elevation= 150

14

Range =

( 6380 + 800) 2 − ( 6380) 2 cos 2 150

− 6380 Sin15 0 = 3684.4 − 1651.27 Km = 2034 Km

2.10.1 Range or Wireless Distance Measurement: Most wireless distance measuring systems operate either by resolving two-ways phase delays of a modulated electromagnetic carrier mono-pulse signal between the object and the reference transmitter or by measuring the two-way propagation time of a coded electromagnetic pulse between these points. For pulse system the distance or range is computed as: t −t d =c m d 2 Where c = Velocity of propagation of wave (m/sec), t m is the measured round trip time delay (secs), and td is the internal system time delays (secs), it should be known precisely, the resultant range d is in meters. For a ranging signal, multi-path propagation is a major cause of errors in determining the position. Errors due to that are difficult to ascertain. Due to various uncertainties exact modeling of signal path (medium) is not possible, but fairly good approximation is possible. The ranging data obtained is used for satellite orbit prediction and its orbit maintainance. Positioning by microwave systems is accomplished by determining the coordinates of intersection of two or more measured ranges from different known ground stations. This is called trilateration method. The ranges of a radio source from different stations are measured. The intersection of the circular ranges (Figs A-9 & A-10) gives the desired range of the radio source. Since each range contains errors, all the circles will not intersect at the same point. In case of three observed ranges (Fig A-10), three different coordinates result. The final position is derived by adjustment of these redundant coordinates, usually by least square minimizing techniques. The range accuracy is obtained by computing the residual errors v1, v2, and v3 for each position. These are the corrections to be added to each range value obtained. Using multiple ranging can minimize positional uncertainties. Using multiple ranging can minimize positional uncertainties. Many times tones are transmitted to satellite and round trip delay of the tones are measured. This delay is used to compute the range. It time delay or the phase change of the transmitted and received tone provides range. A set of coherently generated tones is used for obtaining successive

15

range accuracy. A lower frequency tone is for estimating an unambiguous range value. The global positioning system or GPS (Fig A-11) has rapidly become the standard surveying and navigation mode, replacing microwave and other types of ranging systems. It provides the best positional accuracy, does not require time consuming calibration and is very user friendly providing all weather, 24 hrs, world wide, 3-dimensional satellite-based positioning system. There are 24 satellites at 10000 km height orbiting in 6 planes. It provides: • The precise positioning service (PPS) developed for military and other authorized users and provides an accuracy of 5-10 meters in absolute positioning mode. It uses precise P code. • The standard positioning service (SPS), is available for civilian use and provides an accuracy of 10-20 meters in absolute positioning mode. It uses coarse acquisition C/A code. The measurement is made by phase comparison techniques: • Carrier-phase tracking. • Code-phase tracking. Either the satellite’s carrier frequency phase or the phase of the digital code modulated on the carrier is tracked for resolving the distance between the satellite and the receiver. The GPS satellites broadcast at two carrier frequencies: L 1 at 1575.42 MHz (λ=19 cm) & L2 at 1227.60 MHz (λ=24 cm). The coarse acquisition code (C/A, 300m wavelength) and the precision code (P, 30m wavelength) modulate the carrier. C/A and P codes are both present on L 1 frequency while only P code is present on L2 carrier. In addition to that a 50 bps satellite navigation message containing the ephemeris and health status of each satellite is also transmitted. For fixing a position minimum 4 satellites should be visible at ground station to provide coordinates and timing information. The main sources of errors in range measurements are: Tropospheric, Ionospheric and Multipath errors. The Differential GPS (DGPS) technique is used to improve accuracy in range measurement from a few meters to a few millimeters. 2.11 Power Flux Density: PFD at receiving station is expressed in terms of ratio (in dB) of RF power wrt 1W/m2 per 4KHz. Power received by parabolic antenna P = EIRP-L+G

16

Where L= Path loss = 92.45+20log10R+20log10f 4πAη a G = Receiving antenna gain= G = λ2 Where A=Antenna area in sq cm. ηa=antenna efficiency, λ=Wave length in cm. D=antenna dia in meters, F=frequency in GHz. πD 2 2 cm 2 For circular antenna A = πr = 4 2  πDf  Hence G =   η a wrt isotropic antenna.  c  Or, G = ( 20 log10 π − 20 log10 c ) + 20 log10 D + 20 log10 f + 20 log10 η a (dBi) 20log10π, 20log10c, D& f can be combined = a constant ka. G = k a + 20 log10 D + 20 log10 f + 10 log10 η a  B  PFD = EIRP − L + G − 10 log10  t  dBW/m2per BCCIR  BCCIR  BCCIR= 4KHz: Bt = Transponder Bandwidth.  PT GT  w / m2 In general PFD =  2  4 π R   Power Spectral Density: PSD = P (Power)/B (Bandwidth) = W/Hz. 2.12 Spectral Capacity: Maxm data rate (bps)/ Transmission Area (m2); bps/ m2 For narrow band systems Spectral Capacity: bps/ Hz = Maxm data rate (bps)/ (Bandwidth) Transmission Area (m2) = π (Transmission distance)2

17

APPENDIX-A

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

3.0 LINK: Before designing a radio link it is essential that link estimates be made. This would help in understanding about the uplink and downlink power level requirements, sizing of ground and onboard antenna systems, configuring the type of ground and onboard systems for desired data quality. 3.1 EIRP: Effective Isotropic Radiated Power; If Transmitter power = Pt Antenna Gain = Gt Ground station Losses = Lg EIRPdBW = Pt + Gt − Lg When EIRP is expressed as EIRPdBW then all power notations should be in dBW. The factors affecting the onboard EIRP are the available power, number of carriers in transponder, the antenna type and size, and the required flux density on ground. A limit on EIRP also comes from the tolerable interference limit at other ground stations formulated by CCIR standard. 3.2 Free-Space Loss /Spreading or Path Loss: The attenuation undergone by an EM wave in transit between a transmitter and a receiver in a communication system is called Path loss. Free Space

Antenna

Antenna

R TX

RX

If the distance between the receive (RX) and transmit (TX) antenna is d, the frequency of transmission is f MHz, then the free space loss PL is emperically given as: PL dB = 32.44 + 20logR + 20logf: Where R is in Km, f is in MHz. PL dB = 36.58 + 20logR + 20logf: Where R is in miles, f is in MHz. PL dB = 37.8 + 20logR + 20logf: Where R is in nautical miles, f is in MHz.

37

• For double the distance LdB is increases by 6 dB. • For half the distance LdB is decreased by 6 dB. Path loss may be due the following effects such as: • Free space loss • Refraction. • Reflection. • Diffraction. • Clutter. • Antenna aperture-medium coupling loss. • Absorption. 3,3 Elements of Link Budget: A break up of various elements to be taken into consideration for link estimate are shown below. Ground losses Pt

EIRP Gt

Power received at the input of onboard antenna Free space loss (PL)

Onboard losses Gr

Power in dBW

Prec

Distance in Kms Fig: 3.3 Typical Satellite link

38

3.4 TABLE 3.1: Steps in Ground to satellite Link Calculation: Items

Symbols fu

Unit MHz/GHz

Typical values 6GHz

Pt Lb

Watts/dBW dB

Lg

dB

12 dBW 3-4 dB, depends upon requirement 2-3 dB

Gt

dBi

45-50 dB

EIRPu

Watts/dBW

EIRPu = Pt - Lb- Lg + Gt

1)Free Space loss: It dpends upon: satellite altitude, elevation angle subtended at ground station by satellite, Slant range of satellite. 2) Loss due to polarization mismatch

PL

dB

Ppol

dB

3) Antenna contour in which ground station exists. 4) Rain and all other losses

Lgu

dB

(4πR/λ) 2 λ= Signal wavelength R= Distance of satellite from ground station. 0.1 for matched polarization Around 2dB

LAu

dB

A) Uplink Frequency B) Uplink EIRP 1) Ground transmitter Power 2) Input back off: for multi carrier operation 3) Ground Station Losses eg; connectors, cables/wave guides, filters, Rotary joints etc. 4) Ground Station antenna gain: Function of: size, Beam width, Efficiency (η =60%), Directivity. 5) EIRP

C) Losses

dB

Up to C band = 1dB maxm. At 18 GHz: 8-10dB PL + Ppol + Lgu + LAu

Gr

dBi

25-30

T

Deg K

T = Tant + Ttot + Trec

Gr / T

dB/K

Total Losses Ptot

D) Satellite G/T Satellite antenna gain: Size (0.6m), Efficiency (η =60%), 1) Satellite system noise temperature: Antenna noise temp Tant, All losses from antenna to receiver input Ttot 2) Receiver noise temperature Trec Satellite G/T

39

3.5 TABLE 3.2: Uplink Parameters in tabular form: Ground transmitter power Pt Ground station losses Lg Ground station antenna gain Gt Ground station EIRP Pt- Lg+ Gt 2 PL  4πR  Free Space Loss:    λ  Signal received at the input of EIRP- PL onboard antenna Onboard antenna gain Gr Onboard losses Lo Power received at the input of C = Prec = EIRP+ Gr- Lo onboard receiver Onboard receiver Bandwidth B Receiver Noise Temperature Tsys Noise Power N = KTB C C/N available at the input of = EIRP+ Gr- Lo – (KTB) N detector in dB 3.6

C C : can be also expressed as: N N

C/N = EIRP + Gr + PL – K TSYS – B; all expressed in dB. Or, C/N = EIRP + (Gr/TSYS) – PL – K - all other losses. C/N0 = EIRP +(Gr/TSYS) – PL - all other losses. The threshold requirement of C/N varies for different types of detector as well as for different types of signal modulation. A threshold of 10 dB is recommended for FM receivers while an AM detectors theoretically do not have any threshold. The other losses may include: • Polarization loss. • Antenna pointing loss. • Antenna pattern off-contour loss. • Gaseous absorption loss. • Excess attenuation due to rainfall if any. 40

3.7

System

C : N0

C C we must know for uplink, N0 N0 C C C for downlink, for intermodulation (in case multicarrier operation), N0 N0 N0 For estimating over all link of a system

for the system is given as: (C/N0) sys = 1/1/(C/N0)-1u+ (C/N0)-1d+ (C/N0)-1i The subscripts: sys, u, d, and i refer to link uplink, downlink, and intermodulation, respectively. To reduce IM products to an acceptable limits a “back off” is employed to the input of power amplifier. A C/IM ratio of 20 dB or better is commonly recommended value. 3.8

Link Budget for Digital satellites: Steps involved are to calculate: - Uplink C/N0 - Downlink C/N0 - Overall C/N0 - Calculate Eb/N0 for the modulation type. - Subtract appropriate value of modulation implementation loss. - Find out BER from the available final Eb/N0 value for the station. Eb/N0 = C/N0 – 10log(Bit rate) 3.9 Table 3.3: Uplink and Downlink values of Eb/N0 required for PE = 10-6 (Eb/N0) U dB 50.0 30.0 20.0 15.0 14.0 13.55 12.0 11.0 10.6 10.56

(Eb/N0) D dB 10.54 10.59 11.06 12.47 13.14 13.55 15.98 20.52 29.17 33.92

3.10 Typical example of a satellite link: 41

An earth station has a G/T of +25 dB/K. up and downlink frequencies are 6.0 & 4 GHz respectively. The bit rate is 45 Mbps and satellite EIRP is +34 dBW; satellite G/T is +0.5 dB/k. The uplink and downlink free space losses are 199.6 dB & 196.2 dB. The required Eb/N0 is 12dB that includes modulation implementation loss. The earth station EIRP is +67dBW. Uplink Budget: Earth Station EIRP Pointing Loss Polarization Loss Satellite Pointing Loss Free Space Loss Power received by satellite G/T of satellite Total carrier power received C K C N0 Downlink Budget: Satellite EIRP Pointing Loss Polarization Loss Satellite Pointing Loss Free Space Loss Power received by earth station G/T of earth station Total carrier power received C K C N0

+67 dBW 0.5 dB 0.5 dB 0.5 dB 199.6 dB -134 dBW +0.5 dB/K -133.6 dBW -(-228.6) dBW/Hz 95 dB/Hz

+34 dBW 0.5 dB 0.5 dB 0.5 dB 196.2 dB -163.7 dBW +25 dB/K -138.7 dBW -(-228.6) dBWHz 89.9 dBHz

Using formula: 89.9 dB = 0.977x109: 95 dB = 3.16 x109 1  C    = 1 1 1 + + N  0  SYS (C / N ) (C / N ) D (C / N ) IM U

42

1   = 1 1 = 88.73 dB.  SYS 0.977x10 9 + 3.16 x10 9  C  C  – 10log(Bit rate) Applying equation: avl =  N0  N 0  SYS  C   N0

= 88.73-10log(45x104) C avl = 12.19 dB. N0 C required = 12.0 dB. N0

Margin = 0.19 dB.

Typical G/T for ISTRAC ground stations: X band: 32-34 dB/K S band: 19-22 dB/K VHF band: -13 to -15 dB/K

43

3.11 G/T of different categories of INTELSAT ground stations: Standards Frequency bands G/T A 6/4 GHz G/T≥35.0 20Log(f/4) dB/K Previous: 3.7-4.2 GHz (UL) 5.925-6.425 GHz (DL) Extended: 3.625-4.2 GHz (UL) 5.850-6.425 GHz (DL) B 6/4 GHz G/T≥31.7+ 20Log(f/4) dB/K C 14/12 and 14/11 GHz G/T≥37.0+ 20Log(f/11.2) dB/K For degraded weather conditions: G/T≥37.0+ 20Log(f/11.2) + XdB dB/K; where X=Excess attenuation of downlink degradation D There are two versions of Standard D station D1 ----------------------G/T≥22.7.7+ 20Log(f/4) dB/K D2 ----------------------G/T≥31.7+ 20Log(f/4) dB/K E 14/12 and 14/11 GHz There are three versions of Standard E station. E1 ----------------------G/T≥25.0(but<29.0) + 20Log(f/11) dB/K E2 ----------------------G/T≥29.0(but<34.0) + 20Log(f/11) dB/K E3 ----------------------G/T≥34.0 + 20Log(f/11) dB/K For degraded weather E1 ----------------------conditions: E2 ----------------------G/T≥25.0 + 20Log(f/11) + X dB/K E3 ----------------------G/T≥29.0 + 20Log(f/11) + X dB/K G/T≥34.0 + 20Log(f/11) + X dB/K

44

3.12

Propagation above 10 GHz: The propagation of radio waves above 10 GHz not only involves freespace loss but several other factors are also to be considered as below: 1) The contribution of in homogeneities in the atmosphere. 2) The gaseous contribution of the homogeneous atmosphere due to resonant and non-resonant polarization mechanism. 3) Contribution due to rain, fog, mist, and haze (dust, smoke, and salt particles on air). Propagation through atmosphere gets affected due to several molecular resonance such as water vapor at 22 and 183 GHz, Oxygen with lines around 60 GHz. Other gasses like N2O2, SO2, O3, NO2, AND NH3, also display resonance but do not have much effect on propagation of radio waves. The major offender is precipitation. It can exceed that of all other sources of attenuation in atmosphere above 10 GHz. The total transmission loss is given as: PL dB = 32.44 + 20logR + 20logf: Where R is in Km, f is in MHz. It can be re written as below: PL dB = 32.44 + 20logR + 20logf + a + b + c + d + e Where: a = Excess attenuation in dB due to water vapor. b = Excess attenuation in dB due to mist and fog. c = Excess attenuation in dB due to O2 . d = Sum of absorption losses in dB due to other gasses. e = Excess attenuation in dB due to rainfall.

45

APPENDIX-B

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

4.0

Noise: Noise has very drastic effects on signal transmission. The quality of signal suffers unpredictably in presence of noise. Sometimes making it impossible to receive. There are various types of noise such as below. 4.1

White noise: White noise has constant power spectrum density at all frequencies. Thermal noise derived from resistive networks approximates white noise up to 1013Hz. Other noise sources can also be considered as white noise if the bandwidth of the noise is much larger than the bandwidth of the network or to the system to which it is applied. 4.2

Colored noise: Noise characterized by a power density, which is either band limited or non-uniform, is called a colored noise. A white noise passed through a network of finite bandwidth (filter) will come out as colored noise. 4.3

Noise bandwidth: The noise bandwidth of a network is the bandwidth of an equivalent ideal rectangular pass band filter, having same mid band gain as that of an actual filter, and producing the same output noise power for white noise applied at the input. 4.4

Matched filter: A digital signal may comprise of a data stream of ‘zeros’, ‘ones’ and transmission associated noise. During reception it is very essential to determine if, at an instant of time, ‘zeros’, ‘ones’ or just noise is present. For a given input signal it is conceivable that some filter may exist that enhances the signal as much as possible while reducing noise as much as possible until the ratio of peak signal power to average noise power is maximum. Such a filter is called a “matched” filter. Its transfer function is matched to the signal spectrum. The optimum Eb/N0 ratio for the signal is achieved at some selected instant of time. 4.5

Thermal noise threshold: The received signal must be higher in magnitude than the receiver noise floor or noise threshold (Pn). 63

( Pn ) = KTB; Where

( Pn ) dB

K = Boltzmann Constant = -228.6 dBW/Hz T = Absolute temperature of thermal noise. B = Rx Bandwidth in Hz. = −228.6dBW + 10 log T + 10 log B

For T = 170 or 290 K, the above equation reduces to ( Pn ) dB = −204dBW + NFdB + 10 log BHz 4.6 Intermodulation Noise: Due to Intermodulation products (IMP) generated in a system Intermodulation noise is produced. If two signals with frequencies f1 & f2 are passed through a nonlinear device or medium IMP’s would be generated. These frequencies may be produced from the harmonics of the signals or the main signals themselves. The product results due to beating of two signals in nonlinear environment, Second Order Products 2f1 , 2f2 , f1± f2 Third Order Products 2f1 ± f2 ; 2f2±f1 Fourth Order Products 2f1 ± 2f2 ; 3f1±f2 In multichannel radios the IMP’s produced resembles white noise. The Intermodulation noise may result due to : • Improper level settings driving a device to saturation. • Improper alignment allowing a device to function in nonlinear region. • Nonlinear envelope delay. S 4.7 Signal to Noise Ratio   . N The S/N ratio is expressed in decibels. It is the amount by which a signal level exeeds the noise level in the system. Expressed in dB it can be written as: S   = ( Signallevel ) dB − ( Noiselevel ) Db  N  dB  Eb    :  N0  For digital signal transmission instead of S/N, Eb/N0 is considered. As the signal energy is contained in bit duration the Eb/N0 is defined as the received signal energy per bit per hertz of thermal noise. 4.8

64

 Eb  C C   = = − 10 log(bitrate)  N 0  KTR N 0  Eb   N   can be also written as:  0

 Eb  S  1     N  =  0  N 0R Where R = Bit rate, C=Received carrier power. 1 Data rate R = . Tb Expressed in dB:  Eb    = C dBW − 10 log( bitrate ) − ( − 228.6dBW ) − 10 log Te N  0 Where Te = effective noise temperature of the receiving system. The value of Eb/N0 varies depending upon the modulation scheme, type of detector used, the coding done on the transmitted signal, and on the BER requirement.  Eb  S •  N   depends only upon N 0 and R and is unaffected by any system  0 design choices, such as modulation and coding.

4.9 Table 4.1: Uplink and Downlink values of Eb/N0 required for PE = 10-6 (Eb/N0) U dB (Eb/N0) D dB 50.0 10.54 30.0 10.59 20.0 11.06 15.0 12.47 14.0 13.14 13.55 13.55 12.0 15.98 11.0 20.52 10.6 29.17 10.56 33.92

65

4.10 Noise Figure: Noise Figure (NF) is a measure of the noise produced by a practical network compared to an ideal noiseless network.

S     N in

S     N out

G

(S N ) NF = (S N )

in

out

If Sout /Sin = G; Gain of the system, then N out NF = . KTBG (NF)dB = 10log NF NF expressed in ratio is called Noise Factor. 4.11 Noise Temperature: The noise temperature of two port device e.g. a receiver is the thermal noise that device adds to the system. If the device is connected to a noise free source then its effective input noise temperature is given as: P Te = ne GKdf Where G = Gain of the system. df = specified band of frequency Pne = Available noise power of the device. The noise temperature (Te) and NF are related as below: T NF = 1 + e : T0 Where T0 = equivalent room temperature or 290 K: Te = T0 ( NF − 1)

( NF ) dB

T   = 10 log1 − e   290 

4.12 Thermal Noise aspects of Low-Noise Systems: Downlink signals of the order of –154 to –188 dBW are commonly encountered in space communication. The aim is to achieve sufficiently high

66

S/N or Eb/N0 (in case of digital communication) at the demodulator output. This can be usually done by : a) Increasing System gain, normally antenna gain and LNA gain. b) Reducing system noise. The total effective noise temperature Tsys of a receiving system is conventionally referred to the input of the receiver. It is given as: TSYS = Tant + Trec Where Tant = Effective noise temperature of antenna subsystem. Trec = Effective input noise temperature of receiving system. The ohmic loss components from the antenna feed to the receiver input also generate noise. Other loss producing elements are transmission lines, directional couplers, circulators, isolators, waveguide & switches. The loss factor of these components are inverse of gain ie; The respective noise temperatures would be 1 La = ga Te The equivalent noise figure or noise factor n f = La + T0 Te = T ( La − 1) NFdB = 10 log10 n f And the noise figure on dB At room temperature (290 K) : the noise factor equals the loss of the system ie; n f = La C 4.13 N0 Transmission line Antenna LNA Reference point of measurement

Rx

Fig: 4.13 C/N0 Estimation

C/N0 is a very important element in communication link. By knowing C/N0 one can estimate about the quality of demodulated signal or it can be estimated whether a signal can demodulated properly or not. C/N0 is carrier to noise density, where N0 = noise density in 1Hz & C is the received signal power. 67

It can be deduced that: N 0 = KT = 228.6dBW + 10 log TSYS C dBW = Prec + Gant − linelosses( Ll ) K= -228.6 dBW is the theoretical value of noise level in dBW for a perfect receiver (noise factor of 1) at absolute zero in 1 Hz bandwidth. C/N0 of a link can be also expressed as :   C G  = ( EIRP ) dB − ( Freespaceloss ) dB − ( otherlosses ) dB +  N0  TdB / k − K  Where K is the Boltzmann’s constant. The “other losses” may include: • Polarization loss. • Antenna pointing loss. • Antenna pattern off-contour loss. • Gaseous absorption loss. • Excess attenuation due to rainfall if any. C 4.14 Calculation of : N0 C The calculation of final at the receiver input is essential for N0 estimation of proper functioning of over all link. This must include, as a  C   C  C C C  ,  , minimum, for uplink  for downlink  for N0 N0 N0  N 0 U  N0 D  C   . The basic intermodulation (in case of multi carrier operation)  N  0  IM equation is given as:  C   N0

1   = 1 1 1  SYS (C / N ) + (C / N ) + (C / N ) U D IM

 C    should meet the minimum requirement for the receiving system to  N 0  SYS work. Normally it should be greater than 10 dB. In order to bring down the  C   , due to IM products to an acceptable level so as not to degrade   N 0  SYS poor C/IM ratio, the total uplink power must be “backed off” or reduced. A C/IM ratio of 20 dB or above is an acceptable value.

68

4.15 S/N of a Typical FM System: C/N is a vital parameter in any modulation scheme. For demodulation certain minimum C/N value has to be met to produce acceptable S/N value in demodulated signal. S/N in general can be given as: S C = + I + P +W N N Where: I = Modulation improvement. P = Improvement due to pre-emphasis / de-emphasis techniques. W = Noise weighting factor. 4.15.1 FM/FM Configuration: S   in FM/FM system is given as below. N  ∆FTT  B  S C  + 10 log IF  + P + W  = + 20log  N N  f ch   Bch  Where BIF = IF Bandwidth ∆FTT = Rms test tone deviation.

fch = highest base band frequency. Bch = voice channel BW (3.1 KHz) P = top VP channel emphasis improvement factor. W = psophometric weighting improvement Factor (2.5 dB). Once the voice channel S/N has been calculated, the noise in the voice channel may be calculated by relation:

 90 − ( S / N )    10  pWpO Noise = log-1 

For a video channel:  S  C  ∆f   BIF    =  + W + CF + 20log3   + 10 log  2 Bv   Nv  N  fm   S   = p-p luminance signal-noise ratio. Where  N  v ∆f = p-p composite deviation of the video. f m = Highest base band frequency. Bv = Video noise BW (for NTSC it is 4.2 MHz) 69

B IF = IF noise BW. W = weighting factor. CF = rms to p-p luminance signal conversion factor (6dB) 4.16 G/T: G/T is called the figure of merit of a receiving system. It gives “feel” of receiving systems ability to receive low-level signals effectively. It is a very important element of link budget analysis. G = GdB − 10 log TSYS T TSYS = Tant + Trec Where G= Antenna gain in dB. Tsys= Total Receiving System noise temperature. Tant = Antenna noise temperature. Trec = Receiver noise temperature. The system noise basically consists of: a) Cosmic background noise. b) Galatic noise. c) Noise temperature due to precipitation in the path. d) Solar noise (either in main lobe or side lobes). e) Presence of earth (typically) 290K in side lobes. f) Contribution from nearby objects like buildings, radomes, etc. g) Temperature of blockage items in antenna subsystems such as booms, feeds, absorbers etc. Basically there are two main noise contributors: sky noise and noise from ohmic losses. Sky noise is the main noise source entering through antenna’s main and side lobes. It is largely due to extraterrestrial sources and thermal radiation from atmosphere and earth. Cosmic noise is extraterrestrial radiation coming from all directions. The Sun is another extremely strong source of noise capable of complete communication breakdown. The atmosphere affects external noise in two ways. It attenuates noise passing through it, and generates noise because of energy of its constituents. Ground radiation is thermal in nature. Tsky arise with frequency, elevation angle and water vapor concentration. Tant =

(l a − 1) + TSKY la

70

Where LA = Sum of all losses due to waveguide, filters, directional couplers etc, up to reference plane. The numeric value of LA denoted as la is expressed as: −1  L A  l a = log10    10  4.17 G/T Example: A satellite downlink operates at 21.5 GHz. To calculate the G/T of a terminal operating with the satellite, the reference plane is taken at the input to the LNA. The antenna has a 3ft aperture displaying 44 dB gross gains. There is 2ft of waveguide with 0.2 dB/ft of loss. There is a feed loss of 0.1dB, a band pass filter has 0.4 dB insertion loss, and a radome has a loss of 1.0 dB. The LNA has a noise figure of 5.0 dB and a gain of 30 dB. The LNA connects directly to a down converter/IF amplifier combination with a single side band noise figure of 13 dB. Net gain of the antenna to the reference plane: Gnet = (44-1.0-0.1-0.4-0.4) dB = 42.1 dB. The sky noise for 100 elevation (from graph) for 21.5 GHz is 145K. The sum of all losses to reference plane (including radome loss) will be: La = (1.0 + 0.1 + 0.4 + 0.4) = 1.9 dB The numeric value of La denoted as la = log-1 (1.9/10) = 1.55 Tant = (la – 1) 290 + Tsky  / la = (1.55 – 1) 290 + 145  / 1.55 = 196.45K To calculate Trec, first convert all noise figures to its equivalent noise temperatures using formula T ( NF ) dB = 10 log(1 − e ) 290 Equivalent noise temperatures of 5 dB noise figure & down-converter /IF stage noise figure of 13dB are 627K & 5496K respectively. Hence Trec = 627 + (5496/1000) = 632.5K. Tsys = Tant + Trec = (196.45 + 632.5) dB = 828.95K. G/T = GdB – 10logTsys = 42.1 dB –10log (828.95) = 12.91 dB/K G/T

for X band 32-34 dB/K S band 19-22 dB/K VHF-13 to -15 dB/K

71

4.18 Intelsat Standard Ground Stations: • Standard A Frequency band: 6/4 GHz. G/T≥ 35.0 + 20log(f/4) dB/K; Where f is the receive frequency in GHz. Extension of 6/4 GHz band: Previous: 3,700-4,200 GHz downlink. 5,925-6,425 GHz uplink. Extended: 3,625-4,2000 GHz downlink. 5,850-6,425 GHz uplink. • Standard B Frequency band: 6/4 GHz. G/T≥ 31.5 + 20log(f/4) dB/K; Where f is the receive frequency in GHz. • Standard C Frequency band: 14/12 GHz and 14/11 GHz. G/T≥ 37.0 + 20log(f/11.2) dB/K; Where f is the receive frequency in GHz, for clear sky. For degraded weather condition: G/T≥ 37.0 + 20log(f/11.2) +X dB dB/K; Where f is the receive frequency in GHz. • Standard D Frequency band: 6/4 GHz. Standard D station has two versions D-1 & D-2 G/T≥ 22.0 + 20log(f/4) dB/K; for D-1 Frequency band: 6/4 GHz. G/T≥ 31.7 + 20log(f/4) dB/K; for D-2 Where f is the receive frequency in GHz. • Standard E Frequency band: 14/12 GHz and 14/11 GHz. Standard E station has three versions: E-1, E-2, E-3. For clear sky condition: G/T≥ 25.0 (but<29.0) + 20log(f/11) dB/K; for E-1 G/T≥ 29.0 (but<34.0) + 20log(f/11) dB/K; for E-2 G/T≥ 34.0 + 20log(f/11) dB/K; for E-1; for E-3 For degraded sky condition: G/T≥ 25.0 + 20log(f/11) + X dB/K; for E-1 G/T≥ 29.0 + 20log(f/11) + X dB/K; for E-2

72

G/T≥ 34.0 + 20log(f/11) + X dB/K; for E-3; Where f is the receive frequency in GHz. • Standard F; Frequency band: 6/4 GHz Standard F station has three versions: F-1, F-2, F-3. G/T≥ 22.7 + 20log(f/4) dB/K; for F-1 G/T≥ 7.0 + 20log(f/4) dB/K; for F-2 G/T≥ 29.0 + 20log(f/4) dB/K; for F-3 Where f is the receive frequency in GHz. 4.19 Signal Fading: A transmitted signal not only experiences free space loss due to distance, but if the path gets longer signal fading also takes place. It depends upon on intervening media conditions like terrain roughness, climate, reflectivity, rainfall and path distance for frequencies above 10 GHz. Some times signal fading could be 30 dB or more at these frequencies. Multi path fading: This is most commonly encountered fading. It stems out from the interference between a direct wave and a reflected wave. The reflected wave could be from the ground, or from atmospheric sheet or layer. This type of fading is observed during windless, and foggy nights, when temperature inversion near the ground occurs. Reflections also take place from a body of water, salt beds, or flat desert between transmitting and receiving antennas. Signal fading is mainly due to: a) Intrusion of the earth’s surface or atmosphere layers into propagation path (earth bulge or diffraction fading). b) Antenna decoupling due to variation of the refractive index gradient. c) Partial reflection from elevated layers between terminal antenna elevations. d) Due to “Duct” formation containing only one of the antenna terminals. e) Precipitation along the signal propagation path. “Multipath fading increases with frequency, but much faster with path length, following a relation f.d3.5 where f is carrier frequency and d is the distance.” 4.20 Diversity Reception: Diversity reception reduces fade margin requirement, it tends to reduce depth of fades. Assume a random fading (Rayleigh distribution) with a 35dB fade margin requirement without diversity. Using frequency diversity with 2% frequency separation, only a 23.5dB fade margin would be required to maintain same propagation reliability.

73

Diversity reception is used in many radio systems. A special form of diversity is used on satellite communication link operating above 10GHz. Diversity reception: • Reduces depth of fades on combined output. • If one diversity path is lost other path remains in operation. • Depending upon the combiner used the S/N ratio of the combined output improves over S/N ratio of any single signal path. Diversity reception technique is based on the fact that radio signal arriving through two separate paths may not suffer same amount of signal degradation. The separation may be in: • Frequency. • Space (including angle of arrival and polarization). • Time (a time delay of two identical signals on parallel paths). • Path (signals received at two geographically different paths).

Receiver at f1 Combined output

Base band Combiner Receiver at f2 Fig 4.20a Frequency Diversity

Receiver at f

Combined output

Base band Combiner

Vertical Separation > 100λ Receiver at f

Fig 4.20b Space Diversity

The frequently used diversity in radio link is the frequency and space. The frequency diversity is based on the fact that fading is different for different frequencies. A 2-5% frequency separation is good enough for this 74

purpose. This scheme requires two transmitters and two receivers. For space diversity signal is received at two different locations (several wave lengths apart) at the same frequency. No additional assignment is required. Sufficient output is always available from one of the locations. Two antennas at different heights may also provide space diversity, to a certain degree. The antenna separation required for space diversity is given as: ( 3λR ) S= L Where S = Antenna separation (m). R = Effective earth radius (m). λ = Wave length (m). L = Path length (m). The aim of space diversity is that the reflected wave travels half wavelength more than the normal path. 4.21 Types of Diversity combiners: • Selection combiner. • Equal gain combiner. • Maximal ratio combiner. The selection combiner uses one receiver at a time. The highest S/N channel is connected to the output. The output S/N ratio is equal to the input S/N ratio from the receiver selected at that time. The equal gain combiner adds the diversity receiver outputs, and the output S/N ratio of the combiner is: S0 S + S2 = 1 ; Where N = Receiver noise. N0 2N The Maximal ratio combiner uses relative gain change between the output signals in use. Let us assume the stronger signal has unity output and the weaker signal has an output proportional to gain G. It can be shown that G S1 = , the signal gain is adjusted to be proportional to the ratio of the input S2 signals. In that case: 2

2

2

 S0  S  S    =  1  +  2  N N  N0  The basic assumption involved for the latter two combiners are: • All receivers have equal gain. • Signals add linearly; noise adds on rms basis. • Noise is random. • All receivers have equal noise output N. 75

• The output (from the combiner) signal-to-noise ratio S0/N0 is constant. Performance wise Maximal ratio combiner tops the list while Equal gain combiner & Selection combiner occupy 2nd & 3rd positions respectively. Maximal ratio combiner gives an improvement of 1.5 to 3dB improvement over selection combiner. The Diversity combining has two categories: • Pre-detection. • Post-detection. Pre-detection combining takes place at IF level i.e. at RF level. Input Signal-1 normal

IF AMP

Mixer

70 MHz to FM Det

LO Phase Det Balanced mod

IF Hybrid Combiner

LO

Input Signal-2 Diversity

IF AMP

Mixer

FM Discri

Fig 4.21a

Base band Amp

Input Signal-1 normal

Pre-detection Combiner Combined base Band output

Amp

Noise & pilot filter

Detector

Pilot filter

Noise amp Fig 4.21b

Post-detection Combiner

Amp Input Signal-2 Diversity

Pilot alarm Relay

Identical Receiver

Pilot alarm Relay 76

In the second type, combining takes place at base band level after detection. Combining at IF level is tricky because a strict phase control is very much essential. 4.22 FM threshold: Saturation FM Capture Effect (S/N) dB

FM Improvement Threshold

Noise (C/N) in dB

Any detector requires Fig: a minimum 4.22 FM value Thresold of C/N to produce an acceptable signal of a particular S/N; below this minimum value of C/N the quality of output would degrade drastically. This minimum value of C/N is called threshold. The threshold corresponds to a sudden rise in effective output noise power. In FM radios at lower C/N “snaps” or “clicks” are heard due to the noise spikes produced in the demodulator. Further decrease in C/N “clicks” merge into a sort of crackling or sputtering sound and the desired signal is undetectable. Above the threshold value with increase in C/N, there would be significant improvement in S/N at the output of the detector. At C/N of 12 dB or above the S/N improvement is very significant. This FM phenomenon is also called as FM capture effect. FM threshold is given as: FM threshold in dBW = KTB FM threshold Improvement dBW = KTB +10 dB = -204 dBW + NFdB +10log(BW) +10dB 4.23 FM threshold Extension: Two widely used techniques for threshold extension are: a) Frequency-compressive feedback loop & b) Phase locked loop

77

4.24 Noise in Analog Modulated Systems: 4.24.1 Signal to noise ratios: 4.24.2 Base band Systems: Coherent system. Message signal m (t) Bandwidth W

Low Pass Filter

∑

Noise

-B`

-W

0

Yd (t)

Signal Noise

+W

B

f

Fig-24a Base band Signal at the input of filter Signal Noise

-W

0

+W

f

Fig-24b Base band to Signal output filter PTthe Assuming the signal power be at and the of noise added to the double1 sided power spectral density N 0 over the signal bandwidth B that exceeds 2 the signal bandwidth W, as shown in Fig-a. The signal to noise ratio at the output of BPF is: P SNR = T N 0W

The filter enhances the SNR by a factor of B/W. 4.24.3 Signal to noise ratios: DSB Systems: The coherent DSB demodulator is preceded by a pre detection IF filter of minimum BW of 2W. A post detection low pass filter removes the double frequency and all unwanted frequencies produced during Xr (t)= xc (t)+n (t)

Low Pass Filter

Pre detection IF filter

2cos(ωct+θ) Fig 4.24.3 DSB Demodulator 78

Yd (t)

multiplication of the received and locally generated signals. The ratio of post and pre detection SNR or detection gain is given as:

( SNR ) post det ection ( SNR ) Pr e det ection

=2

This is because coherent demodulator suppresses the Quadrature component of noise. It is apparent that there is 3 dB improvement in SNR after demodulation. But this advantage is off set as pre detection filter bandwidth must be 2W to pass a DSB signal, there by doubling the noise content. So here effective SNR improvement does not take place after DSB signal demodulation. 4.24.4 Signal to noise ratio: SSB System: In case of SSB signal as carrier is suppressed coherent demodulation is only possible and: ( SNR ) post det ection =1 ( SNR ) Pr e det ection The SSB system lacks the 3-dB detection gain of DSB system. However, the pre detection noise power of the SSB system is 3 dB less than that of the DSB system. Hence coherent demodulation of both DSB and SSB result in performance of base band. 4.24.5 Signal to noise ratios AM system: The required transmission bandwidth of AM is 2W. The detection gain is given as: ( SNR ) post det ection 2m 2 = ( SNR ) Pr e det ection 1 + m 2 Where m = percentage of modulation, m=1 for 100% modulation. Efficiency (η) of AM transmission is defined as the ratio of side band power to total power in transmitted signal. m2 η= 1+ m2

79

( SNR ) post det ection

Hence

( SNR ) Pr e det ection

= 2η

If the efficiency of AM were 100%, then AM would have the same post detection S/N ratio as the ideal DSB or SSB systems. But the efficiency of AM is typically much less than 100% therefore the post detection S/N ratio is correspondingly lower. For low ( SNR) input to envelop detector, the analysis becomes complex. The principle component of demodulated output contains the Rayleigh-distributed noise envelope and it is not proportional to the signal. This has significantly much degrading effect than the additive noise. At low ( SNR) input a threshold effect sets in and for slight decrease in input SNR to the demodulator the output SNR falls drastically. For high ( SNR) input the envelope detection is as good as coherent detection. As for coherent demodulation a demodulation carrier generated within receiver multiplies the modulated carrier. This demodulation carrier must be phase coherent with the received modulated carrier. Thus the demodulation carrier must be derived from the received carrier. 4.3 Noise in Angle Modulation Systems: Angle modulated systems are non-linear modulation systems. There are great differences between linear and non-linear modulation systems when noise effects are considered. There are significant differences between FM and PM systems. FM offers great improvement over both linear and PM system in noisy environments.

Xr (t)

Pre detection Filter

Discriminator

Fig 4.3

Post detection Filter

Yd (t)

Angle Demodulation system

The pre detection filter bandwidth BT is determined by Carson’s rule. In other words BT = 2( β + 1) f m where β = modulation index for sinusoidal signal, and fm is the frequency of the modulating signal. For arbitrary signal m (t), a general expression for bandwidth results where the deviation ratio D is defined as: f D= d W Where fd = peak frequency deviation

80

W = bandwidth of m (t). A general expression for bandwidth B = 2( D + 1)W The output of pre detection filter consists of two components: a) Signal alone and b) Signal and noise. For small ( SNR) input the only term containing the signal is multiplied by the noise. Thus a threshold effect is observed. If the ( SNR) input > than the Rayleigh-distributed noise envelope of the FM demodulator then the effect of noise is reduced if the transmitted signal amplitude is increased. Thus the output noise is affected by the transmitted signal strength even for abovethreshold operations. FM System: ( So / No ) = 3β 2 ( Ci / N i ) For high (Ci/Ni) ratio using frequency discriminator. • 3 dB improvement expected for a PLL demodulator. • Around 2-2.5 dB improvement is expected for a FMFB demodulator is used. 3 For large β: ( S o / N o ) = 3( β / f m ) ( C i / N i ) 4.4 Pre emphasis & De-emphasis: The white noise at the input of receiver produces a hyperbolic noise spectrum at the demodulator output. The demodulator output S/N decreases with increase in base band frequency. For desired constant S/N at the demodulator output the base band signal components need to be boosted. At the receiver end these boosted signal components should be de-emphasized or reduced to reproduce the original signal and to avoid signal distortion. A high pass filter pre emphasizes these high frequency components at the transmission end, and at the receiving side a low pass filter de-emphasizes these enhanced signal components to restore the signal in its original form. 4.4.1 Pre Emphasis: r C

H (jω) R

ω1 Fig a Pre emphasis filter 81 Fig: 4.4.1 Pre emphasis

ω2 logω

Fig b Frequency Response

1 1 : ω2 = : τ = rC = 75µ sec s Signal between f1 and f2 are pre rC RC emphasized at transmitter end so that higher deviation for high frequency components can be achieved.

ω1 =

4.4.2 De-Emphasis: H (jω)

r R

ω1

ω2 logω -6dB/octave

Fig a De emphasis filter

Fig b Frequency Response

Fig: 4.4.2a De emphasis The receiver receives the transmitted pre emphasized signal. To restore the received signal to original condition the frequency components between f1 and f2 are de-emphasized.

Data

HPF

FM Mod

Data TX

FM Rx

LPF

Preemphasis emphasis &isDe The improvements dueFig to4.4.2b pre/de asemphasis below: schemes FM system: • For τ =75 µsecs, fm = 15 KHz, the typical improvement in S/N is around 13 dB. • For τ =75 µsecs, fm = 21 KHz, the typical improvement in S/N is around 16 dB. AM system: • For τ =75 µsecs, fm = 15 & 21 KHz, the typical improvement in S/N is around 6 & 8 dB respectively. • For τ =75 µsecs, fm = 5 KHz, the typical improvement in S/N is around 3 dB.

4.5 Threshold Extension: 82

The Frequency-Compressive feedback loop (FMFB) and the phaselocked loop (PLL) have widespread application in threshold extension in highnoise environments. 4.5.1 FMFB: FM Signal

Discriminator

BPF

Demodulated Output VCO Fig 4.5.1

Frequency-compressive feed back demodulator

For larger value of KDKV the phase deviation can be made small, thereby reducing significantly the bandwidth of signal at the discriminator input. It is even possible to compress the bandwidth of a wide band FM signal to that of a narrowband FM signal at the input of the discriminator. The bandwidth B p of the BPF is smaller than the transmission bandwidth BT due to bandwidth compression. Thus the rate of spikes due to noise is reduced and threshold extended. Above threshold the post detection SNR for FMFB system is same as that of an ordinary discriminator. The threshold extension in FMFB is less than obtained using a PLL. 4.5.2 Phase Locked Loop (PLL): Phase Detector FM Signal

∑

Loop Filter

FM Demodulated Output

VCO Fig 4.5.2 PLL demodulator

A PLL can be also used as FM or PM demodulator. The phase detector (PD) compares the phases of the incoming signal and the locally generated signal. A voltage proportional to the phase difference is produced which after filtering is fed to the VCO as control voltage. A VCO is essentially a frequency modulator, the frequency deviation of the output dθ/dt is proportional to the input signal. The phase error signal drives the PLL into lock. As the VCO input signal is proportional to the frequency deviation of the PLL input, the integral

83

of this signal is proportional to the phase deviation of the PLL input. The noise accompanying the signal has in-phase and out of phase noise components. The PLL correlates with the in-phase component of noise while effect of the out of phase noise gets cancelled, thereby extending the threshold by maximum by 3dB. The parameters of importance in a Phase Locked Loop are: i. Order of the PLL ii. Loop Bandwidth iii. Damping Factor iv. Natural Frequency v. Lock-in, Pull-in and Hold-in vi. Sweep Rate 4.6

PLL as a demodulator: PM Demodulated Output PD

∑

Loop Filter

FM Demodulated Output

VCO

AM/FM/PM modulated Signal input

900

LPF

QA D

Fig 4.6

AM Demodulated Output

A PLL as AM/FM/PM Demodulator

84

APPENDIX-C

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

102

103

5.0 Modulation: 5.1 Why Modulation!!! Modulation is the process of encoding information from a message source in a manner suitable for transmission. It generally involves translating a base band signal (called the source) to a band pass signal at frequencies that are very high when compared to the base band frequency. The band pass signal is called the modulated signal and the base band message signal is called the modulating signal. Modulation may be done by varying the amplitude, phase, or frequency of a high frequency carrier in accordance with the amplitude of the message signal Need For Modulation: Modulation is extremely necessary in communication system due to the following reason: • Practical antenna length: In order to transmit a wave fully, the length of the transmitting antenna should be equal to the wavelength of the wave. Velocity 3 x10 8 Wavelengthλ = = meters Frequency FrequencyinHz The practical antenna length to transmit audio frequencies become so big in size that it becomes impossible to be constructed practically. For this reason, it is impracticable to radiate audio signal directly into space. Therefore a very high frequency signal is chosen over which the low frequency signal or message is superimposed (modulated) and then transmitted. At this carrier frequency an antenna can be easily constructed. • Channelization of different signals: Different signals at different sub carriers can modulate a single carrier and are transmitted. 5.2 Types of Modulation: There are mainly two types of modulation they are • Analog modulation. • Digital Modulation 5.3 Types of Analog Modulation: • Amplitude Modulation • Frequency Modulation • Phase Modulation 5.4 Types of Digital Modulation: Digital Modulation is variant of analog modulation but the modulating signal is digital in nature. The different types are:

104

• Amplitude Shift Keying. • Frequency Shift Keying. • Phase Shift Keying. • Differential Phase Shift Keying. • M-ary Phase Shift Keying. • M-ary Frequency Shift Keying. • Minimum Shift Keying. • Spread Spectrum Modulation. 5.5 Amplitude Modulation: In amplitude modulation, only the instantaneous amplitude of the carrier wave is varied in accordance with the intensity of the signal. However, the frequency of the carrier remains constant. es

e

ec

Signal

t A.M

Carrier

Wave

Fig 5.5 Principle Of Amplitude Modulation

5.6 Features of Amplitude Modulation: • The instantaneous amplitude of the carrier wave changes according to the intensity of the modulating signal. • Two side bands about the carrier at the signal frequency are produced. • AM is susceptible to channel noise. Carrier Power

Upper side band Power

Lower side band Power

PAM = P-fcm+ Pf;

fc

+fm

Fig 5.6 AM modulated spectrum Where PAM = power in AM signal, Pc = power in carrier. Pf = power in side bands.

105

Frequency

η = (Am / A0)2/1 + (Am / A0)2  for Am = A0 i.e., (for square wave modulating signal) 100% modulation, η = 50%, in that case side band power equals carrier power. For 100% modulation (sine wave modulating signal) i.e., p-p amplitude of 2 Am, Pf = Am2/2. Modulation index βAM = Am / A0 ηAM = (m) 2/2+(m) 2 = 1/3 = 33.3% for m=1. 5.7 Double Side band AM: It is suppressed carrier transmission. ηDSB = 100% as there is no carrier transmission hence all power remains in the side bands. But the suppressed carrier transmission and reception both are complex and expensive. 5.8 Single Side band AM: Both side bands of DSB signal carry same information. Hence one is redundant, as full information can transmitted by using only one side band. DSB-AM bandwidth = 2(SSB-AM) bandwidth. Here also ηSSB = 100%. 5.9 Vestigial Side band (VSB) AM: Stringent sharp cut-off filters are required for generation of SSB signals, which is quite difficult to realize in practice. In Vestigial Side band AM instead of cutting off one side band completely a part of other side band is also transmitted. So a more realistic filtering is used for generating as well as reception of a VSB signal. ηVSB = 100%. 5.10 Amplitude Modulation:

(Si/Ni)

Demodulator (So/No)

If the input and output signal to noise ratios of the demodulator be (Si/Ni) & (S0/N0) respectively then for small signal AM system, the S/N ratio improvement after demodulation is given as: (S0/N0)/(Si/Ni) = 2/3 For large carrier AM system the performance of diode detector is as good as that of a synchronous detector.

106

5.11 Frequency Modulation: In FM the instantaneous frequency of carrier wave is changed in accordance with the intensity of the signal. The carrier frequency increases or decreases with the increase or decrease of the amplitude of the modulating signal. A sinusoid is said to be frequency modulated if instantaneous frequency is a linear function of the information signal. es A

F

B

e

C E

t

G

t

c

D

Carrier

Signal

e m

t

Frequency Modulation

Fig 5.11 FM wave

Thus, FM waveform is given as: SFM= A cos ω 0 t + θ 0 + k PM ∫ (t )dt

[

]

Where ω0 & kPM are positive constants and θ0 is an arbitrary phase angle. A FM system can produce infinite side bands depending on the modulation index β. ∆ω Modulation index β = fm Where ∆ω = peak frequency deviation and f m = modulating frequency. Many frequencies are generated which are not present in the signal. Because of this fact the FM waveform may have much larger BW than the signal. The FM spectrum is analyzed using Bessel functions where all the frequency components can be analyzed. At many values of β the carrier component (coefficient J0 of Bessel function) becomes zero and the whole power is distributed over the side bands. If β ≤ 1radian it is called narrow band FM (NBFM), where the bandwidth is restricted while for β ≥ 1radian wide band FM (WBFM) is generated with larger bandwidth. The NBFM modulated spectrum is like that of AM system with two side bands. While

107

for WBFM the BW depends upon the value of β. Normally the number of significant side bands n = β + 1 . Frequency modulation has better noise immunity that compared to amplitude modulation. Since signals are represented as frequency variations rather than amplitude variations, FM signals are less susceptible to atmospheric and impulse noise, which tend to cause rapid fluctuations in the amplitude of the received signal. Also message amplitude variations do not carry information in FM, so burst noise does not affect FM system performance as much as AM systems, provided that the FM received signal is above the FM threshold. In a FM system, it is possible to tradeoff bandwidth occupancy for improved noise performance. Unlike AM, in a FM system, the modulation index, and hence bandwidth occupancy, can be varied to get greater signal-to-noise performance. It can be shown that, under certain conditions, the FM signal-to-noise ratio improves 6dB for each doubling of bandwidth occupancy. This ability of a FM system to trade bandwidth for SNR is perhaps the most important reason for its superiority over AM. However, AM signals are able to occupy less bandwidth as compared to FM signals, since the transmission system is linear. In modern AM systems, susceptibility to fading has been dramatically improved through the use of in-band pilot tones, which are transmitted along with the standard AM signals. The modern AM receiver is able to monitor the pilot tone and rapidly adjust the receiver gain to compensate for the amplitude fluctuations. 5.12 Phase Modulation: The instantaneous phase of the carrier wave is varied linearly in accordance with the intensity of the signal it is called phase modulation. A sinusoid is said to be phase modulated if instantaneous angle is a linear function of the information signal. Thus, PM wave form is given as: SPM= A cos[ω 0 t + θ 0 + k PM f (t )] In PM like FM the numbers of side bands depend upon the phase Phase Frequency modulation index. The small phase deviation m(t) restricts the BW resulting in m(t) Modulator narrowModulator band PM (NBPM). In PM signal the carrier frequency PM remains FM constant. Hence normally a satellite down link signal is phase modulated so (a) (c) that error free Doppler frequency extraction is possible. m(t)

∫

Phase Modulator

m(t)

FM

(b)

d dt

Frequency Modulator (d)

Fig (a) & (b) Generators of FM Fig 5.12

108

Generation of FM and PM signals

Fig © & (d) Generators of PM

PM

5-13 Demodulation of narrow band FM and PM signals:

m(t)

Amplifier

∫

m(t)

Σ

900

900

SNBPM(t)

SNBFM(t)

Carrier

Fig 5.13a

Carrier

Narrow band PM modulation

SNBPM

Fig 5.13b Narrow band FM modulation

BPF

LPF m(t)

− Sin(ω 0 t + θ 0 ) Fig 5.13c NBPM Demodulation

SNBFM

Σ

BPF

LPF

− Sin(ω 0 t + θ 0 ) Fig 5.13d

109Demodulation NBFM

d

dt

m(t)

PM Demodulated Output PD Loop Filter

FM Demodulated Output

VCO

AM/FM/PM modulated Signal input

900 Fig 5.13e A PLL as AM/FM/PM Demodulator

LPF

QA D

AM Demodulated Output

5.14 Local carrier generation methods: SDSB(t)

Narrow band Filter

Amplifier Local carrier

Fig 5.14a Local carrier generation for signals with pilot carrier Balanced Mod

Local carrier

LPF Cos (2πfct)

LPF

VCO

Product

900 SDSB(t)

Sin (2πfct)

LPF 110 Balanced Mod

Fig 5.14b Costa’s Loop

() 4 QPSK Signal

Narrow band Filter

LPF

VCO

÷4

Generated Reference Carrier

Fig 5.14c Reference Carrier generation Loop for QPSK Receiver

5.15 Digital Modulation: Modern mobile communication systems use digital modulation techniques. The advancements in VLSI and DSP technology have made digital modulation more cost effective than analog transmission systems. Digital modulation offers many advantages over analog modulation. Some advantages include • Greater noise immunity. • Robustness to channel impairments. • Easier multiplexing of various forms of information (e.g., voice, data and video). • Improved spectral efficiency, because digital signals are more robust against channel impairments.

111

Greater security. Has a digital error-control code, which detect and/or correct transmission errors. It supports complex signal conditioning and processing techniques such as source coding, encryption, and equalization to improve the performance of the overall communication link. New multipurpose programmable digital signal processors have made it possible to implement digital modulators and demodulators completely in software. Instead of having a particular modem design permanently frozen as hardware, embedded software implementations now allow alterations and improvements without having to redesign or replace the modem. Digital signals such as PCM, DM, ADM etc, must some times be transmitted over a channel requiring carrier modulation. It would be impractical to attempt direct transmission of a time multiplexed PCM signal to a satellite. This is because of the impractical size of antenna required to transmit such signal. The fundamental methods existing for carrier modulation are: amplitude, phase, and frequency. • •

Source

ADC #

Encoder

Modulator To channel

# ADC absent if source is digital Carrier Fig 5.15a A Digital Transmitter

Demodulator

Detector

Decoder

From channel Ref carrier (coh system)

DAC *

User

* DAC absent if user needs digital output Fig 5.15b

A Digital Receiver

The demodulation could be coherent or non-coherent. If a local reference in phase with the transmitted carrier is available for demodulation, it is called Coherent demodulation; otherwise it is called Non-coherent. Likewise, if a periodic reference signal is available at the receiver that is in synchronism with the transmitted sequence of digital signal (clock), the system is referred as Synchronous otherwise Asynchronous.

112

5.16 Types of Digital Modulation: A desirable digital modulation scheme should provides low bit error rates at low received signal-to-noise ratios, performs well in multi-path and fading conditions, occupies a minimum of bandwidth, and is simple and cost effective to implement. Depending on the demands of the particular application, trade-offs are made while selecting a digital modulation scheme. Some schemes are better in terms of the bit error rate performance, while others are better in terms of bandwidth efficiency. The performance of a modulation scheme is also measured in terms of its power efficiency and bandwidth efficiency. In communication systems power and available transmission bandwidth are the two precious items. Most systems can be either bandwidth or power limited. In bandwidth-limited systems, spectrally efficient modulation techniques can be used to save bandwidth at the expense of power; in power-limited systems, power efficient modulation techniques can be used to save power at the expense of bandwidth. In both power-and-bandwidth limited systems, error-correcting codes can be used to save power or to improve error performance at the expense of bandwidth. Recently, trelliscoded modulation (TCM) have been used to improve the error performance of bandwidth-limited channels without increase in bandwidth. • Power efficiency is the ability of a modulation technique to preserve the fidelity of the digital message at low power levels • Bandwidth efficiency is the ability of a modulation scheme to accommodate data within a limited bandwidth. In general, increasing the data rate implies decreasing the pulse width of a digital symbol, which increases the bandwidth of the signal. Bandwidth efficiency reflects how efficiently the allocated bandwidth is utilized and is defined as the ratio of the data rate throughput per Hertz in a given bandwidth. If R is the data rate in bits per second, and B is the bandwidth occupied by the modulated RF signal, then bandwidth efficiency ηB is expressed as: R η B = bps / Hz B A greater value of ηB will transmit more data in a given spectrum allocation. The upper limit on ηB is limited by channel capacity. In a digital communication system, there is often a trade off between bandwidth efficiency and power efficiency. Adding error control coding to a

113

message increase the bandwidth occupancy (and this, in turn, reduces the bandwidth efficiency), but at the same time reduces the required received power for a particular bit error rate, and hence trades bandwidth efficiency for power efficiency. On the other hand, higher-level modulation schemes (M-ary keying) decrease bandwidth occupancy but increase the required power, and hence trade power efficiency for bandwidth efficiency. Digital modulation techniques may be classified as coherent and noncoherent based on the kind of receiver used. ASK, FSK and PSK- amplitude, frequency and phase shift keying respectively are the three basic forms of digital modulation corresponding to AM, FM, and PM, for analog modulation. A special form of coherent binary FSK is Minimum Shift Keying (MSK). 5.17 Amplitude Shift Keying: The binary ASK waveform carrier is present (on) or absent (off), depending upon whether signal is 1 or a 0.

1

0

1

1

0

1

0

0

Thus an idealized ASK signal can be represented as f(t) =

{

A Sin ωct;

0

0
; elsewhere

The signal energy E is given by E = ∫0 A2 Sin2 ωct dt = A2T/2 5.18 Frequency Shift Keying: A FSK signal is given as: s1 (t ) = A cos ω c t

s 2 (t ) = A cos( ω c + ∆ω ) t

For 0 ≤ t ≤ T A FSK signal can be viewed as frequency of a carrier wave switched to two values f1 & f2 depending upon the binary code’s 0 or 1 status. 5.19 FSK signal waveform: f1

f2

f1

f2

114

An idealized FSK waveform can also be considered as composed of two ASK waveforms of different carrier frequencies. This is shown in the figure below. FSK Signal

ASK Signal

ASK Signal The average energy per bit is given by: E = ∫ A2 Sin2 mω0t dt = A2T/2 5.20 Phase Shift Keying: In phase shift keying the carrier phase of a signal is switched between two or more values in response to the PCM code. For binary PCM, an 180º phase shift is convenient because it simplifies the modulator design and hence is often used. This particular choice is commonly known as phase reversal keying (PRK).

1800 phase transition

The PSK waveform as shown in the above figure can be represented as: f1 (t) = A Sin ωct; f2 (t) = -A Sin ωct

115

5.21 Comparison of Binary Digital Modulation Schemes: ASK: • The transmitters for ASK systems are very easy to build and have an advantage in that there is no power transmitted when there is no data being sent (if OOK is used). • Receivers for non-coherent ASK systems are easy to build. • The difference in performance between coherent and non-coherent detection is slight compared to the increase in complexity required so that coherent detection of OOK is generally not used. • A disadvantage of ASK is that the decision threshold in the receiver must be adjusted with changes in received signal levels. These adjustments are normally made with an automatic gain control. FSK: • Systems, in contrast to ASK operates symmetrically about a zero decision threshold level regardless of carrier signal strength. • Receiver complexity depends primarily upon whether a coherent or noncoherent modulation method is used. • Non-coherent FSK is relatively easy to implement and is a popular choice for low-to-medium data-transmission rates such as Teletype. • FSK transmissions intended for non-coherent demodulation require more bandwidth for a given bit rate than either ASK or PSK. Bandwidths of FSK transmission intended for coherent demodulation can be made as small as desired by controlling ∆f, but with S/N penalty. • Bandwidths of FSK transmissions intended for coherent demodulation are typically equal to or slightly greater than those required for ASK or PSK. PSK: • Superior to both ASK and FSK systems in that they require less transmitted power for a given error probability. However, synchronous detection is required and carrier recovery systems are more difficult (and therefore more expensive) to build. • DPSK systems are often a good compromise that sacrifices some error performance but permits a more economical receiver. The measure of system performance for digital data communication is the probability of error PE. Synchronous detection in a white Gaussian noise background requires a correlation or a matched filter detection to have minimum PE (maximum Eb/N0) for a fixed S/N. Demodulation of digital signals

116

can be performed coherently or non-coherently. A coherent demodulator requires generation of a phase coherent reference signal in the receiver. The non-coherent system is inferior to the coherent one but it is mainly used due to the simplicity of receiving system, and where ever the resulting PE requirement is not stringent. Demodulation: An ASK system information is in amplitude ot the signal’s presence or absence. Therefore the demodulation scheme is similar to the AM system. The bandwidth of the BPF should be 1/T to pass the signal without distortion.

Received Signal Plus Noise

BPF

Threshold Voltage

Envelope Detector

Decision

Fig 5.21a Non-coherent ASK Receiver

BPF ωc

Envelope Detector

Received Signal Plus Noise

∑ BPF ωc+∆ω

Threshold Voltage

Envelope Detector

Fig 5.21.b Non-coherent FSK Receiver

For FSK signal the transmitted signals are: s1 (t ) = A cos( ω c t + θ ) for 0 ≤ t ≤ T s 2 (t ) = A cos[ ( ω c + ∆ω ) t + θ ] for 0 ≤ t ≤ T Where ∆ω is sufficiently large such that s1 (t )ands 2 (t ) occupy different spectral region. A FSK receiver is two ASK receivers in parellel. A FSK receiver is basically a FM receiver where a BPF and an envelope detector forms a discrimnator.

117

Decision

5.22 Differential Phase Shift Keying (DPSK). Message Equivalent gate

One bit Delay

Level shift

±A Cos

±1

A Cos Fig 5.22a

Received Signal

A DPSK Modulator

Threshold Voltage

One bit Delay

Decision

Fig 5.22b Demodulation of DPSK

A possible implementation of a differentially coherent demodulator for DPSK is shown above. The received signal + noise is correlated bit by bit with one bit delayed version of signal + noise. The output of the correlator is compared with a threshold set at zero, a decission being made in favor of a 1 or a 0, depending on whether the correlator output is positive or negative. The above receiver is a sub-optimum receiver.

5.23 Optimum Receiver:

Received Signal Plus Noise

Cos ωct

Decision Logic

Sin ωct

118 Fig 5.23 Coherent Demodulation of DPSK signal

Data

The optimum receiver is a Coherent DPSK receiver. It provides about 2dB improvement in PE over sub-optimum receiver. For large S/N ratio this degradation is less than 1 dB. But only significant disadvantage is that errors tend to occur in group of two because the DPSK receiver makes decision on basis of signal received in two sucessive bit intervals during correlation in the encoding process. As a portion of the received noisy signal is used for correlation in receiver, the determination of overall PE becomes some what complcated. A truly non-coherent PSK system does not exist, since it would be imposible to convey information in phase of a carrier of completely random phase. For PSK system the transmitted signals are once again 1 or 0. S1 (t ) = A cos( 2πf c t ) for1 S 2 (t ) = A cos( 2πf c t + π ) for 0

Where fcis large compared to 1/Tb. Assuming 0 or 1 ocurring with equal probability. The average probability of error PE is given as:  E b (1 − ρ )  1 , PE = erfc  2 2 N 0   Where ρ= Correlation coefficient of signals S2(t) & S1(t). As S2(t) & S1(t) are antipodal, it means S2(t)=-S1(t): In that case ρ= -1 in expression of PE reduces to  Eb  1  PE = erfc  2 N 0   In case of orthogonal signals: ρ= 0  Eb  1  PE = erfc  2 2 N 0   ρ= -1 gives the minimum value of PE possible as S2(t) & S1(t) vary.

119

The choice of type of digital data system depends upon cost and complexity of system. In signal fading or intervening high randomly varying noise media, coherent scheme is preferable. But where ever it is not possible to generate a coherent reference, non-coherent system is used. 5.24 TABLE 5.1: Typical Prrobability of error For a PE of 10-4 Modulation Required SNR Non-coherent ASK 12.5 dB Coherent ASK 11.5 dB Non-coherent FSK 12.5 dB Coherent FSK 11.5 dB DPSK 9 dB Coherent PRK 8.5 dB (QPSK) 5.25 M-ary Data Communication: In binary digital communication systems one of the two possible signals can be transmitted during each signalling interval. In M-ary system, one of M possible signals may be transmitted during each Ts second sigalling interval (symbol). Binary data transmission is special case of M-ary data transmission, where M=2. The rate at which M-ary symbols are transmitted is called baud rate in bauds. For M=2, one state represents 0 while the other represents a 1. In M=4 systems such as QPSK, each level or state represents two information bits or coded symbols. For 8-ary FSK or PSK systems, 3bits are transmitted for each transition or change of state. Of cource for M-ary systems some form of coding or combining is required prior to modulation and decoding after demodulationto recover the original bit stream.

5.26 Qadrature Multiplexing: Two different messages can be sent through the same channel by means of quadrature multiplexing. In quadrature multiplex system, the messages m1(t) and m2(t) are used to double sideband modulate two carrier signals, which are in phase quadrature. The optimum Eb/No performance achievable with BPSK led to a search for mechanisms to improve the bandwidth efficiency of PSK schemes without any loss of performance. It was found that since Cos2πfct and Sin2πfct (where fc

120

is the carrier frequency) are coherently orthogonal signals, two binary bit streams modulating the two carrier signals in quadrature can be demodulated separately. In analog communication, this idea has been used for a long time to multiplex two signals on the same carrier, so as to occupy the same bandwidth, e.g., the two-chrominance signals in color sub carrier. Such a modulation scheme, increasing the bandwidth efficiency of binary PSK by two, is known as QPSK

m1(t)

∑

Oscillator Binary data m (t)

Serial to Parallel Converter

QPSK Output

900

m2 (t)

Fig 5.26 a QPSK Transmitter

The modulated spectrum of QPSK is typical Sin2X/X2 nature. The main lobe of QPSK spectrum contains 90% of the signal energy. Still, the small power outside the main lobe is a source of trouble when QPSK is to be used for multi-channel communication on adjacent carriers. The wide spectrum of QPSK is due to the character of the base band signal. This signal consists of abrupt changes and abrupt changes give rise to spectral components of high frequencies. In short, the base band spectral range is very large and multiplication by a carrier translates the spectral pattern without changing its form. This problem is overcome by passing the base band signal through a premodulation LPF to suppress the side lobes as far as possible. But such filtering will causes intersymbol interference.

∫ T0 dt

Cos (2πfct) Oscillator QPSK Signal

Parallel to Serial Converter

900 Sin (2πfct) 121

∫ T0 dt

Fig 5.26 b

QPSK Receiver

Data Output

In the receiving system signals have to be generated in receiver to track the phase of the QPSK signal. This can be done using four phase Costas loop or by quadrupler loop circuit. A commonly used circuit for reference carrier generation is ahown below. () 4 QPSK Signal

Narrow band Filter

LPF

VCO

÷4

Reference Carrier

Fig 26 c Ref Carrier generation Loop for QPSK Receiver

The reference carrier generator for a PSK signal requires a squaring circuit and a ÷2 network instead of the quadrupler and the ÷4 circuit in above diagram for QPSK reception. Comparing average symbol-to-noise-spectraldensity ratio, QPSK is approximately 3dB worse than PSK. But twice the data is transmitted in QPSK system compared to PSK. Hence for a given Eb/N0 the probability of error is same for QPSK & PSK systems. 5.27 OQPSK or Staggered QPSK system: Since the possibility of instantaneous phase transitions of 1800 is eliminated in OQPSK the band limited signal has a much smaller envelope fluctuation than QPSK. Consequently, absence of fast phase transitions of 1800 means that the undesired high frequency components originally removed by the band limiting filter will not be regenerated. In QPSK system the quadrature data streams can switch sign simultateously, due to that the data-bearing phase of the modulated signal can change by 1800. This sharp phase change can have an undesirable effect in

122

terms of envelope deviation when the modulated signal is filtered. To avoid this, the quadrature–channel data signals of QPSK system can be offset by half the signalling interval relative to each other.The resulting modulation scheme is referred to as offset QPSK or OQPSK system. With this staggering the phase change due to modulation of the transmitted carrier is 900. For an ideal system the error probability of OQPSK is identical to that of QPSK. But one limitaion of OQPSK system is that the data streams must have same symbol intervals, whereas for QPSK they need not. OQPSK also known as staggered QPSK (SQPSK) has relatively low spectral side lobes after undergoing non-linear amplification. The difference in time alignment in the base band components doesn't change power spectral density of modulated signal and hence in linear channels both QPSK and OQPSK spectra have the same shape. However the two modulated signals respond differently when they undergo band limiting and then non-linear amplification. Because of the coincident alignment of the two base band components in QPSK modulation, an instantaneous phase transition of 180° can take place; hence the band limited QPSK signal exhibits 100% envelope fluctuation. Since the possibility of instantaneous phase transitions of 180° is eliminated in OQPSK the band limited signal has a much smaller envelope fluctuation than QPSK. Consequently, absence of fast phase transitions 1800 means that non-linear amplifiers will not regenerate undesired high frequency components originally removed by the band limiting filter. Spectral advantage of OQPSK stems mainly from the fact that OQPSK avoids the late phase transitions of 180° associated with the QPSK format. This suggests that further suppression of spectral spreading in band limited non-linear applications can be obtained if the OQPSK signal can be modified to avoid step phase transitions altogether. Hence, "Minimum Shift Keying" or MSK is considered as an effective alternative. To minimize the effects of adjacent channel interference, it is desirable that power radiated into the adjacent channel be 60-80 dB below that in the desired channel. Hence narrow main lobe and fast roll-off of sidelobes are desirable. Majority of communication systems use class-C power amplifiers. To prevent re-growth of spectral side-lobes due to nonlinear amplification it is essential that the input signal should have constant envelope. In Continuous Phase Modulation (CPM) class of frequency modulation techniques the carrier phase varies continuously, which produce constant amplitude envelope and excellent spectral characteristics. MSK signals have

123

all the attributes for mobile radio except for a compact power density spectrum. This is achieved by filtering the modulating signal by a low pass filter. A GMSK uses a pre-modulation filter with a sharper pre-modulation low pass filter (Gaussian bell shaped filter) which produces much more sharper main lobe compared to normal MSK signal. A GMSK spectrum is difficult to realize practically. 5.28 MSK: In MSK the base band waveform, that multiplies the quadrature carrier, is much "smoother" than the abrupt rectangular waveform of QPSK. While the spectrum of MSK has a main center lobe which is 1.5 times as wide as the main lobe making filtering much easier. The waveform of MSK exhibits phase continuity, that is, there are no abrupt phase changes as in QPSK. As a result the intersymbol interference caused by non-linear amplifiers is avoided. The staggering which is optional in QPSK is essential in MSK Minimum Shift Keying which is also called continuous-Phase Frequency-Shift Keying (CPFSK) with modulation index 0.5 or Fast Frequency-Shift Keying (FFSK), is one of the most effective digital modulation techniques for digital transmission via satellite or mobile channels. The difference in the rates of fall off of these spectra can be explained on the basis of the smoothness of the pulse shape p (t). The smoother the pulse shape, the faster is the drop of spectral tail to zero. Thus, MSK, having a smoother pulse, has lower side lobes than QPSK and OQPSK. It turns out that in MSK 99% of the signal power is contained in a bandwidth of about 1.2f b while in QPSK the corresponding BW is about 8fb. This indicates that in relatively wide-band satellite links (where, for e.g., filtering is not used after the non-linearities), MSK may be spectrally more efficient than QPSK or OQPSK. However, the MSK spectrum has a wider main lobe than QPSK or OQPSk. In MSK the bandwidth required to accommodate this lobe is 1.5fb while it is only 1fb in QPSK. Since the MSK spectrum has a wider main lobe, this suggests that in narrow-band satellite links, MSK may not be a preferred method. 5.29 TABLE 5.2: Comparison of different Digital modulation systems: Modulationon Scheme ASK

Signalling Waveform

Bandwidth

S1 (t) = A Cosω0t S0 (t) = 0

Double of band signal

FSK

S1 (t) = A Cosω1t

Depends

base

on

124

Probability Of Error

Frequency Spectrum

Pe = erfc [√Eb/N0] where Eb: Energy per bit N0/2: Noise Spectral Density Pe = 1/2 erfc [√Eb/2N0] for

Power spectral density is centered at center frequency Two line spectrum

BPSK

S2 (t) = A Cosω2t Where ω1 and ω2 are two different frequencies. S1 (t) = A Cosω0t S2 (t)=Acos (ω0t+π)

QPSK

S1 (t)=ai (t) Cos2πfct + aq (t) Sin2πfct

MSK

S1 (t)=ai (t) Cos2πfct Cos (πt/2T)+ aq (t) Sin2πfct Sin (πt/2T)

seperation of both the frequencies

coherent detection. Pe = 1/2 erfc [-Eb/2N0] For non coherent detection.

occur for different frequencies.

BW=1.25 times the bit rate (practical). Theoretically BW = bit rate. BW = 0.5 times bitrate (theoretical) BW = 0.6 times bit rate (practical) Bandwidth is 1.5 times more than that of QPSK

Pe = 1/2 erfc √[Eb/No]

Spectrum: Sin2X/X2 without spectral line at carrier frequency

For each Pe = 1/2 erfc √[Eb/No]

Spectrum: Sin2X/X2 and no discrete spectral line at carrier frequency

Pe = 1/2 [erfc (√γb)1/4 erfc2 (√γb)]

Spectrum Sin6X/X6

5.30 QPSK & MSK Waveforms:

QPSK and OQPSK

MSK Power In dB

MSK has some excellent special properties that make it an attractive fT alternative when other channel constraints require bandwidth efficiencies Fig 5.30 the QPSKcontinuous & MSK Spectrum below 1bits/s/HZ. For example, phase nature of MSK makes it highly desirable for high power transmitters driving highly reactive loads. Since intersymbol switching occurs when the instantaneous amplitude of p (t) is zero, the finite rise and fall times and data asymmetry inevitably present in practical situations have a minimal effect on the MSK performance. In addition, MSK has simple demodulation and synchronization circuits.

125

two

5.31 Implementation Techniques of MSK Modulator: The most straightforward way of implementing a MSK modulator is to apply the data stream to a voltage controlled oscillator. The output of the VCO is then a frequency modulated MSK signal.

Data In Fig 5.31 a MSK Modulator

MSK Signal

VCO

The main disadvantage of the above modulator is the instability of the VCO. This modulator is thus not suitable for coherent demodulation. A second disadvantage is that it is difficult to maintain center frequency at allowable value under restriction of maintenance of linearity and sensitivity for the required FM modulation. Modulation and demodulation can be accomplished in 2 equivalent fashions viz. Serial and parallel. The serial technique is advantageous for high data rate implementations. In parallel modulation, the serial data stream is thought of as being 'demultiplexed' into its even and odd bits to produce two bit streams, ai (t) and aq (t) staggered half symbol and then used to biphase the input signals. A serial modulator structure for serial MSK is as in the figures below. X(t)

MSK signal

MSK

Outpu Fig 5.31 b A MSK Modulator

ACos (2πf1t)

Band Pass Matched Filter

Band Pass Conversion Filter

Z(t)

X(t)

Acos (2πf1t+θ)

Low Pass Filter

Fig 5.31 b

YD(t)

A MSK Demodulator

It is seen to consist of a biphase shift keyed modulator with carrier 1 frequency T Hz, and a band pass conversion filter. The serial demodulator 4

126

structure for MSK is essentially the reverse of the serial MSK modulator. It consists of a band pass conversion filter followed by a coherent demodulator and low pass filter which elliminates double frequency components at the mixer output. The major difference between the serial modulator and demodulator is the matched filter, which has a transfer function proportional to the square root of the power spectrum of the MSK signal. Serial MSK modulation and demodulation have the advantage that all operations are performed serially, and therefore offer significant implementation tradeoffs at high data rates. The precise synchronization and balancing required for the quadrature signals of the parallel structures are no longer present. The critical system components are the biphase modulator, the bandpass conversion and matched filters, and the coherent demodulator. Even data

Oscillator

Data

Serial to parallel converter

1 F0 = T 4

Carrier Oscillator

Σ 900

900 Odd data

Fig 5.32c Block Diagram of Quadrature MSK Modulator

5.32 Spectral efficiency: In any communication system, the two primary communication resources are the transmitted power and channel bandwidth. A general system design objective would be to use these two resources as efficiently as possible. In many communication channels, one of the resources may be more precious than the other and hence most channels can be classified primarily as powerlimited or band-limited. (The voice grade telephone circuit, with approximately 3KHz bandwidth, is a typical band-limited channel, whereas space communication links are typically power-limited). In power-limited channels, coding schemes would be generally used to save power at the expense of bandwidth, whereas in band-limited channels "spectrally efficient modulation" techniques would be used to save bandwidth. 127

The primary objective of spectrally efficient modulation is to maximize the bandwidth efficiency, defined as the ratio of data rate to channel bandwidth (in units of bits/s/Hz). A secondary objective of such modulation schemes may be to achieve this bandwidth efficiency at a prescribed average bit error rate with minimum expenditure of signal power. Some channels may have other restrictions and limitations, which may force other constraints on the modulation techniques. For example, communication systems using certain types of non-linear channels call for an additional feature, namely a constant envelope, which makes the modulation impervious to such impairments. This is needed because a memoryless non-linearity produces extraneous sidebands when passing a signal with amplitude fluctuations. Such sidebands introduce out-of-band interference with other communication systems. 5.33 Spectral Features of Digital Modulation Systems: ASK: • Transmitters are easy to design. No power is transmitted when there is no data being sent. FSK: • Symmetrical about a zero division threshold. Requires more bandwidth for a given bit rate than either PSK/ASK For same Eb/No, FSK has 3dB advantage over ASK on a peak power Requirement (but average power requirement is same). Frequencies are so selected that inter channel interference is less. BPSK: • DSB-SC Modulation Superior to ASK and FSK by 3dB in average power requirement for a given probability of error. QPSK: • Bit error probability in BPSK and QPSK is equal but not symbol error probability MSK: • Constant envelope. • Narrow bandwidth and sharp cut-off of LPF. • Compact spectrum. • Coherent detection capability. Out of band radiation is a serious constraint.

128

5.34 TABLE 5.3: Eb forPE = 10 −6 N0 M 2 4 8 16

Non-coherent 14.2 dB 11.3 dB 9.7 dB 8.7 dB

Coherent 13.6 dB 10.8 dB 9.3 dB 8.2 dB

If one considers the required BW to be that required to pass the main lobe of the signal spectrum (null to null), then the BW efficiency of various Mary schemes be as:

5.35 TABLE 5.4 Bandwidth efficiency on a main lobe spectrum of M-ary systems. M-ary scheme BW Efficiency (bits/s/Hz) PSK, QASK 0.5log2M log 2 M FSK M +1

5.36 TABLE 5.5 Bandwidth efficiency on a main lobe spectrum of M-ary systems. M PSK, QASK or QAM FSK 2 0.5 0.33 4 1.0 0.4 8 1.5 0.33 16 2.0 0.24 32 2.5 0.15 Note: BW efficiency of M-ary PSK goes up with increasing M, while for FSK it goes down.

129

Efficiency (η) of M-ary signaling: R log 2 M 1 = = η of M-ary signaling is given by W WTs WTb

( b s ) / Hz

Where W=BW in Hz, Tb=bit duration, R= bit rate. The smaller the WTb product the more efficient the digital communication system. GSM system uses GMSK modulation where WTb =0.3 Hz/(b/s), where W= 3dB bandwidth.

130

APPENDIX-D

131

132

133

134

135

136

137

138

139

6.0 RF Radiation Hazards-A Review 6.1 Introduction: The transmission of intelligence or radiation of a signal is by Electromagnetic (EM) waves. These EM waves are also referred as “radio waves” and the frequencies at which they are transmitted are called “radio frequencies”. The EM waves propagated in free space have electric and magnetic fields perpendicular to each other and to the direction of propagation. They are called transverse electromagnetic (TEM) waves. The radiation could be a desired one or an undesired interference. The example of desired radiation could be a radio/TV/distress signal/point to point telecommunication signal/or a satellite communication. An undesired radiation could be an interfering signal to a wanted communication or could be a leakage signal originating from an electronic instrument in laboratories or a high power transmitter. The undesired/unwanted signals can disturb the wanted communication channel. A radiation whether wanted or unwanted above certain specified power level and at certain frequencies can affect the health of human beings. Some times during development of simple low power circuits in laboratories can also produce unwanted radiation of unexpected high magnitude which goes undetected. It may seriously affect the health of a working person. These days the use of mobile phones has gone up very much. Efforts are on to find out the effects of RF signals emanating from mobiles on human health but till now no concrete evidence of radiation from mobile phones have been established. The absorption of RF energy affects a human being directly by the way of heating the limbs. Shock and burns result from contact with conductive objects. Its athermal effect is affecting the tissues without heating. Some people wearing heart pacemakers, insulin pumps, passive metallic plates may also get affected by radiation. 6.2 RF Radiation: The RF radiation around us in day-to-day life affects our life very much. A required amount of controlled radiation exposure is used, as medicine while an undesired uncontrolled radiation is injurious or could be even fatal for human being. As the harming radiation energy cannot be measured or quantified in numbers it remains a dark area. The safety limits have not been quantified in many areas. The safety management deals with well being of

140

people. Humans respond to many stimuli as part of the normal process of living but a biological effect due to unwanted radiation may lead to a health hazard. As it is not possible to find out the effects due to unwanted radiation experiment on human beings, the effects of RF radiation have been studied and documented on small animals like rats, rabbits, bacteria, yeast cell etc. But the extrapolation of these results straight away to human beings is not possible in many cases because of the physical sizes, producing different resonant frequencies, different thermal properties due to different thermoregulatory systems. Then it also difficult to extend research to very wide radio frequency spectrum of 10kHz to say 300GHz, to low or high levels of field, employing different modulation techniques etc. The situation complicates further when certain effects take place only in certain frequency bands, power levels, and modulation scheme. This suggests a host of combination of situations. Moreover there is seldom equipment that can produce many simulation signatures for test purposes. 6.3 Sources of radiation: The radio transmission has become the integral part of our day-to-day life. Stimulated by the needs of the world wars, radio transmission has become an established technology, which is taken for granted, and which, among other things, provides for the broadcasting to our homes of entertainment, news and information of every kinds. The uses of TV, radio, mobile & cordless phones, domestic satellite dishs bringing quasi-optical nature of microwaves are very common sight today. The terrestrial and satellite broadcasting and communications, and the enormous number of mobile phones now in use, homes, work, and recreational places are irradiated by vast number of electromagnetic signals. Many of these radiations are at low level while those emanating from transmitters or some other equipment are high. Some radiations are unintentional, resulting from the leakage from devices or equipments, due to inadequate shielding or improper metal cases. Apart from any effects of leakage on people, it also causes interference to other equipments. Transmitting unwanted noise like signal can be used to obstruct or jam signals from enemy country causes the intentional interference. This technique was widely used during cold war era. The transmitted signal could be radar continuous wave signal or a modulated wave like radio, TV broadcasts, mobile phones etc. The modulating signal could be a voice, a sine wave, a pulse, a noise or a mixture of different signals.

141

The chief sources of RF radiation are: • Broadcasting. • Communications. • Radar Systems. • Machines. 6.3.1 Broadcasting: Broadcast transmitters in medium-wave/short-wave/Television and microwave relay stations normally use quite high power. They radiate RF power from a few kilowatts to a few megawatts. Medium frequency (MF) transmitters are usually used for national broadcasting while the high frequency (HF) bands equipments are used for long distance or overseas services. BBC, Voice of America or German overseas services cover throughout world. They use very high RF power in megawatt range. These transmitters would be normally large in size. If proper dummy loads or effective RF shielding are not employed then they could be a big source of RF leakage. For HF broadcasting the antenna system are supported by a lot of masts and towers. As more than one frequency is normally used additional antenna systems are required. The unused antenna systems can become “live” due to parasitic energisation from the working antenna. They become a good source of unwanted radiators. Large antenna sites make job of ensuring the radiation safety issues much more difficult. 6.3.2 Communications: There are infinite variety of communication equipments ranging from mobile phones, MF/HF/VHF/UHF/Microwave ground communication systems, air to ship / satellite communications in VHF/UHF/Microwave bands. They use variety of antenna systems at different power levels. To optimize communication some times a single scheme uses diversity techniques where more than one set of antenna systems are used for different polarizing signals. Various elements of RF communication systems affect the health of a common man directly or indirectly. Some of them are listed below: a) Air traffic control (ATC) communication. ATC communication is normally in VHF And UHF bands. They transmit in range of about 50 watts. Several transmitters are operated simultaneously to secure wide coverage. They are quite organized networks keeping the health of workers around. But the radar used in ATC sites may illuminate parts the

142

towers, and radiation could be uncontrolled. This could be injurious to inhabitants near by. b) Microwave transmitting towers. Microwave transmission is very common for civilian or military communication. These communications could be either land based line of sight or tropospheric in nature. For tropospheric communication the transmitted signal from one station is deflected from the tropospheric layers and received at other desired station. It is often used to communicate between oilrigs and base stations on ground. Typical RF power could be any where between 5W to a few KW in 4-6 GHz band. The microwave communication towers for civilian telecommunication networks also operate in 4-6 GHz band in a few KW range of RF power. The antennas are relatively near to ground therefore safety clearance is an important factor. The tropospheric communication uses fixed or mobile antenna. The safety aspects in both the cases would be different. For portable antenna due their changing positions sometimes they may come very near to human population. In that case the safety aspects become very important. c) Satellite Communication Systems. Presently most of the satellites operate in C, X, Ku, Ka bands of frequencies. The satellites are in LEO (Low Earth Orbit). MEO (Medium Earth Orbit), HEO (High Earth Orbit) or Geostationary Orbits, depending upon the service requirements. The ground stations are used to transmit signals to the satellite and to receive signals from them. The normal antenna used is parabolic dishes of varying diameters. For Geostationary satellites the ground station antenna are fixed in the direction of the satellite at particular azimuth and elevation. But for orbiting satellites the ground station antenna must track the movement of satellite in space from low to high elevation angles. The normal RF power transmitted from a ground station could range from tens of watts to a few kilowatts. The scanning ground station antenna while uplinking to a satellite may have bad effect on the human beings in near by houses. The effect would be more if the undesired side lobes of the antenna are more. For “receive only” terminals this problem is minimal as the signal strength received from a satellite on ground would have very insignificant effect on human health. 6.3.3 Radar: Radars are very widely used element in civil aviation, civil and defense (“identification of friend or foe” IFF, military jamming radars), meteorology,

143

police departments. These radars could be fixed land based, ship borne, air borne (aircrafts, satellites) or mobile land based (police). They could be covered by RF transparent domes or without domes. The radar antennas are scanning one with a typical rate of 6-15rpm. Surveillance radar is normally high power and is placed in isolated places. They are less injurious to human health compared to the space and land borne ones. The powers radiated by radars differ depending upon their mode of operation. The typical mean power and effective peak powers radiated by modern radars of planar array type are around 4kW and 54kW respectively. All these could be potential health hazards if the systems are not used properly or left unserviced for long period of time. 6.3.4 RF Equipments or Machines: In scientific development laboratories or at electronic systems manufacturing industries electronic equipments are needed to develop a new system, test the developed systems, or to calibrate an existing system. Many food-processing industries use RF drying machines to process their products. Other RF processing machines are used for welding, vulcanizing, heat-sealing, plastic processing, brazing, soldering, forging etc. RF induction machines are used for container sealing. Many of these machines if not properly RF shielded may radiate appreciable power levels, which may harm a worker, near by. Some of the induction machines for sealing oil containers are rated at 2-3kW at around 30 MHz. Some of the RF applicators used in heavy industries generate in range of 100kW. They are as good as a large radio transmitter. An engineer or a scientist developing a new electronic circuit in laboratory uses a signal generator or measuring instruments like Oscilloscope or a Spectrum analyzer. These equipments generate a host of RF signals of various levels. If these levels were not in safe level then they would be harmful. Apart from that, often while aligning or tuning a circuit very high levels of unwanted oscillations are produced. This unpredicted signal burst very near to a worker may harm him/her without knowing him/her. The effect of which may be felt in short or long course of time depending upon the damage caused. Use of short wave therapy to human body is known since early twentieth century. The RF energy (typical 27MHz) penetrates in human tissues and induces heating in joints. Many physiotherapists employ pulsed RF energy for treating the muscular systems. One use of microwave energy is to induce localized hyperthermia to destroy malignant tumors. Here the basic safety concerns are:

144

• Exposure of the patient. • Exposure attending physiotherapists. • Exposure of other people near the patient. The patient should not be exposed to stray unwanted radiation during treatment. In absence of any radiation measuring instrument it may not be possible to ascertain as to how much dose has been received by him. It may cause serious damage to the patient. A reasonable information about the equipments used is essential to avoid health hazards. 6.4 Effects of RF radiation: The known effects of RF radiation on human beings documented are: 6.4.1 Thermal effects. The most demonstrable effect of RF radiation above 100kHz on human body is the thermal effect. High percentage of human body is made of water and water molecules. On RF exposure the tissues having significant water contents get influenced by impinging electromagnetic field and heat is generated (just like what happens in microwave oven). This heating effect is different at different power levels and frequency bands. The amount of heat produced also depends upon the amount of heat absorbed and the thermo-regulatory system of an individual. On excessive exposure the thermo-regulatory system gives way and give rise to hyperthermia or heat exhaustion. If the cell temperature reaches around 430C irreversible damage takes place. A rise in 2.20C is often taken as the limit of endurance for clinical trials. The quantity of absorption of energy in tissues is expressed in watts per unit mass (Wkg-1) of tissues; this is called as Specific Absorption Rate (SAR). At low frequency (tens of kilohertz) energy absorption is low. It is reported that absorption is maximum at human resonance, which is typically 30-80 MHz depending upon height for adults, above this it reduces monotonously. Practical SAR measurement for human is not possible, only estimation by computer simulation on dummy models has been possible. The RF energy through human tissues may be absorbed, reflected or pass, depending upon body structure and tissue interface. As per Gandhi and Riazi, the depth of penetration of energy depends upon its frequency. With higher frequency the penetration of energy decreases. At microwave frequencies deposition of energy is confined to surface layers of skin. For diathermy treatments the energy frequency is controlled for heating at desired depth in body. A radiation at microwave frequencies may damage the skin of a human being. This risk increases with increase in RF frequency. So the radiation exposure should be quite small for microwave frequencies signals compared to that at lower frequency signals.

145

Some persons may have high local SAR (hot spot) in their body. These hot spots are produced due to complex mixture of tissues. They may produce nonuniform localized distribution of heating in the body damaging the local cells. The absorption of a radiation is maximum at resonance frequency of a human body. The resonance frequency of a human being has been found to be related with height of an erect person. As per an empirical relation arrived at, the resonance occurs at when height corresponds to approximately 0.36 to 0.4 wavelengths (λ). For a standard man of height 1.75m; 1.75 λ= = 4.37 meters . 0.4 300 Re sonantfrequency = MHz = 69 MHz . 4.37 Small children appear to resonate at higher frequencies and tall adults at lower frequencies. Similarly it has been observed that if a person is effectively earthed due to bare feet or conductive shoe, the resonance occurs at approximately half the above frequency. 6.4.2 TABLE: 6.1 Effective resonant frequency in polarized field; h=0.4λ Height of a person (m) Resonance (MHz) 0.5 240 0.75 160 1.0 120 1.25 96 1.5 80 1.75 69 Gandhi[13] suggests that human being absorb energy 4.2 times greater than what is expected considering physical cross-section of the body. The most susceptible organs from heat transfer point of view are the eyes and male testes. These organs do not have means of dissipating heat due to that reversible or some times irreversible damages to may take place. The production of cataracts due to the thermal effects on animals has been well documented. The frequencies likely to cause cataract in human beings have been found to lie between 1-10 GHz. Here too the depth of RF radiation penetration in the eye tissues depends of upon its frequency. Some reported work claim that microwave pulsed radiation at low levels could affect susceptible part of eye. They suggest that at millimetric wavelengths the

146

power absorption of human eyes could be of the order of 15-25mW for an incident power of 100Wm-2 after 30-60 minutes of exposure. Many reported works suggest that eye exposure to RF radiation be treated with caution especially with high power pulsed radiation. Exposure limits and level should be controlled. Unnecessary exposure to eye, for example by holding the head close to open RF power amplifier or any electronic circuit, during alignment is avoided. The same precaution should be taken while removing a waveguide flange without switching off the source. Experiments with anaesthetized rats and mice have shown that exposure of SAR of 8-10 Wkg-1, and 30 Wkg-1 depletes male germ cells respectively. Conscious rats and mice exposed to 9 Wkg-1 and 20 Wkg-1 respectively. As anaesthetized animals could not regulate their testicular temperature hence the radiation effect was more. Some times temporary decrease in fertility have been also reported. But detailed studies on human beings are not available on organized basis hence it is not possible to conclude for sure whether radiation hazard affects productivity of a person. Many experiments results are available to find the effect of RF radiation on human auditory system. Human auditory system responds to frequencies as low as 200 MHz to 3GHz. The reported volunteers when exposed to RF radiation heard buzzing, hissing, and clicking sounds depending upon the modulation characteristics. But persons responded in different way to the same problems. Many times contradictory responses were received from different people. Detailed survey on various persons engaged in various fields of occupation revealed no conclusive evidence of auditory defects produced by radiation hazards. The currents induced in human body especially in limbs (limb current) have been observed as of some concern from a radiation frequency up to 100MHz. It has been established that currents in legs of an adult in RF field produce large SAR wherever conductive cross-sectional area (like ankle) is small. The current density produced will be quite large in those narrow areas. Gandhi et al, have shown by measurements that the induced currents are highest when a person is erect and barefoot. The leg currents are proportional to frequency and to the square of height of a person. They found that the currents peaked around 40MHz, the resonant frequency of a standard man. The current when wearing shoes is about 0.8 times the bare foot current. Due to reduction in impedance to ground with frequency the current increases with frequency.

147

RF shock or even burn may be experienced if a person touches a passive conductive object in the electric field. These objects could be fences, scrap metal, any equipment stored in open, etc. Burn may take place if the current density (mA/cm-2) is quite high due to small contact area. Chatterjee et al, studied this effect on several males and females in frequency range of 10kHz to 3MHz. They experimented on threshold currents for perception and pain. As burn results from the current density and contact area. If the contact area is large then currents even more than the safe limit may not harm a person. Touch burns have been reported at 60mA current with a contact area of 0.2cm-2. 6.4.3 TABLE: 6.2 Range of threshold currents (ICNIRP*) Threshold current (mA) Frequency 50/60 Hz 1kHz 100kHz Touch Perception 0.2-0.4 0.4-0.8 25-40 Pain on finger contacts 0.9-1.8 1.6-3.3 33-55 Painful Shock 8-16 12-24 112-224 Severe shock/ 12-23 21-41 160-320 Breathing difficulty * ICNIRP-International Commission for Non-ionizing Radiation Protection. Pulsed RF transmission is very common. Radar is an example. It is considered that pulsed radiation may affect nervous systems. A large SAR is produced during the pulse period. 6.4.4

Non-thermal effects: Till now no damaging non-thermal effects of RF radiation have been reported. Still with exceedingly wide use of RF energy in day-to-day life it is imperative that all aspects be examined thoroughly. Some people feel that the exposure to RF radiation may lead to growth of tumors in human body. Divergent results were reported from different laboratories leading no concrete conclusion. On this issue a leading authority on non-ionizing radiation of UK the National Radiological Protection Board (NRPB) could not find good evidence that electromagnetic radiations with frequencies less than about 100 kHz were carcinogenic. This clears electrical and electronics home appliances. If a person wearing heart pacemaker is exposed to a radiation it is possible that the radiation may affect the working of pacemaker. Similarly many

148

passive devices like metal plates, rods and other fixings fitted in a person may resonate with radiated energy and harm him. With higher frequencies some laboratories have suggested that they may act like tumor promoters at certain power and frequency levels. The increasing wide spread use of cell phones all over the world has brought back the fear of possibility of radiation borne carcinogenic development in users. Arguments and counter arguments were put forward on the issue, leading to a stale met. A study funded by the European union was constituted “to show conclusively that EM radiation emitted by mobile phones and power lines could affect human cells at energy levels generally considered harmless.” The four-year REFLEX project involving 12 groups of seven European countries carried out supposedly identical experiments. The results were compared. The conclusion as per project leader Franz Adlkofer of Verum Foundation in Munich, Germany was: “Electromagnetic radiation of low and high frequencies is able to generate a genotoxic effect on certain but not all types of cells and is also able to change the function of certain genes, activating them and deactivating them.” But many scientists including Michael Repacholi of WHO do not agree with results and feel that the experiments conducted by various groups were not completely standardized hence the results are not conclusive. More study is required to come to a conclusion. Ground breaking research to understand the effects of mm waves on skin is being carried out at Cranfield University at Oxford shire, UK. Dr. Clive Alabaster of Radar Systems Group at the university leads the study team. Sponsored by Anritsu the program has arrived at some preliminary results. Using the safety benchmark set by the NRPB of 10 mW/cm2, Dr. Alabaster calculated the temperature rise of skin exposed to this level of mm wave radiation for 30 seconds. He found that this exposure could heat up the skin surface by 0.20C. This heating would be hardly noticed by a human body. He wants to reconfirm the result and seeks to extend the study to variety of skin samples. 6.5 RF Safety standards: In controlling human exposure to RF radiation the safety standards deals to whether the potential hazard relates to: • Leakage or unwanted radiation from RF source. • Intended or wanted radiation from a transmitter, antenna, machine or applicator. The governing bodies for these standards in different countries are:

149

• •

•

•

•

The American National Standard Association (ANSI). International Commission on Non-ionising Radiation Protection (ICNIRP). Is has link with WHO, International Radiation Protection Association (IRPA). It is non-governmental and non-political organization. The UK National Radiology Protection Board (NRPB). NRPB is statutory adviser to the Health and Safety Commission on ionization and nonionization radiation for United Kingdom. The USA Federal Communications Commission (FCC). It is responsible for assessing RF emissions and provides technical information from time to time via Internet. The Commission of the European Communities (EC). It maintains draft for various requirements for on ionization and non- ionization radiation and also machine safety directives.

6.5.1 Leakage Standard: Leakage Standards are set for maximum expected safe radiation level at a defined distance from a radiating source in specified condition of use. For specific gadgets like microwave ovens the safe radiation levels are specified at safe working distance. National safety standard is specified for leakage from radio or TV broadcast transmitters. Some times in-house local standards are also applied in isolated cases for different gadgets. Leakage limits are often defined and carried out by legal provisions of a country. This is monitored closely at the time of testing of a gadget. 6.5.2 Exposure Standards: It is limit for human exposure to RF radiation. Beyond this limit RF exposure could be dangerous to human beings. Basically the leakage and exposure mean same thing. The leakage is specified at a distance between from a radiation source. A leakage is a criterion for pass-fail standard for an instrument, while in an intentional radiation (like a transmitting station) there is no requirement for reduction in radiation level as the power level is required for normal functioning of the system. But these levels may injurious to a worker or people near by. Hence there is a continuing requirement for active safety management. Many times these power levels get enhanced due to reflection from metal objects or from a resonant antenna. Some standardization bodies recommend different limits for “occupational” and “public” use. People engaged in RF system development work

150

are termed as “occupational” while people not engaged in RF system development or not knowledgeable on the subject have been characterized as “public”. The radiation limits prescribed for public is normally less than the 1 occupational category. Typically the power density limits for “public” is of 5 that of “occupational”. For estimating the radiation doses the product of power density and 1 th of an hour (6 exposure time is taken into consideration. Since 1960, 10 minutes) averaging has been established. Considering an example, a continuous limit for power density of 50Wm-2; in watt-hour or joule units can be calculated as: 50Wm-2 x 0.1h = 5Whm-2 Or 50Wm-2 x 360 secs = 18000Jm-2; (since 1W=1Js-1) For peak power radiations used in radars or pulsed equipments, adverse effects on human body have not been reported due to its very short period of occurrence. Different RF radiation safety documents have tabulated in different ways. ICNIRP has tabulate under headings of “occupational” and “public”. The uses “controlled” and “uncontrolled” areas while the NRPB use the classification as “adults only” and “children present”. The following safety limits in TABLES A-B for “occupational” and “public” have been reproduced from NRPB, IEEE, & ICNIR standard documents. 6.5.3 TABLE: 6.3 “Occupational” limits

Power Flux Density: NRPB-Adult Frequency Wm-2 10-60 MHz 10 60-137 MHz 2700 f2 137MHz to 50 1.1 GHz 41f2 1.1GHz to •

ANSI/IEEE- Controlled Frequency Wm-2 100-300 MHz 10 300 MHz-3 GHz f/30 3-15 GHz 100 15-300 GHz

100

1.5 GHz

1,5 GHz to

100

300 GHz

f= GHz

f= MHz

151

ICNIRP-Occupational Frequency Wm-2 10-400 MHz 10 400-2000 MHz f/40 2-300 GHz 50 f= MHz

•

Electric Field Strength: NRPB-Adult Frequency Vm-1 600KHz-10 MHz 600/f1 10-60 MHz 60 60 MHz to 1000f2 137 MHz 137 137MHz to

ANSI/IEEE- Controlled Frequency Vm-1 0.1-3 MHz 614 3-30 MHz 1842/f 30-100 MHz 61.4 100-300 MHz

ICNIRP-Occupational Frequency Vm-1 0.065-1 MHz 610 1-10 MHz 610/f 10-400 MHz 61

61.4

f= MHz

1.1 GHz

1,1 GHz to

1251 f2

1.55 GHz

f1= MHz; f2= GHz; •

Magnetic Field NRPB-Adult Frequency 535KHz-10.6 MHz 10.6 -60 MHz 60 MHz to 137 MHz

Strength:

137MHz to 1.1 GHz 1,1 GHz to 1.55 GHz

f= MHz

Am 1.8/f21

ANSI/IEEE- Controlled Frequency Am-1 0.1-3 MHz 16.3/f

ICNIRP-Occupational Frequency Am-1 0.065-1 MHz 1.6/f

0.16 2.7f2

3-30 MHz 30-100 MHz

16.3/f 16.3/f

1-10 MHz 10-400 MHz

1.6/f 0.16

0.36

100-300 MHz

.1630

400-2000 MHz

0.008 f

2-300 GHz

0.36

-1

.33f2 f= MHz

1.55GHz to 0.52 300GHz f1= MHz; f2= GHz;

f= MHz

152

6.5.4 TABLE: 6.4 “Public”limits

• Power Flux Density: NRPB-Child present Frequency Wm-2 12-200 MHz 6.6 200-400 MHz 165 f2 400MHz to 26 800 MHz 41f2 800MHz to

ANSI/IEEE- Controlled Frequency Wm-2 100-300 MHz 2 300 MHz-3 GHz f/150 3-15 GHz f/150 15-300 GHz

100

ICNIRP-Public Frequency 10-400 MHz 400-2000 MHz 2-300 GHz

Wm-2 2 f/200 10

f= MHz

1.55 GHz 1,5 GHz to

100

300 GHz f= GHz •

f= MHz

Electric Field Strength: NRPB-Child present Frequency Vm-1 600KHz-12 MHz 600/f1 12-200 MHz 50

ANSI/IEEE- Controlled Frequency Vm-1 0.1-3 MHz 614 1.34-3 MHz 823.8/f

ICNIRP-Public Frequency Vm-1 0.15-1 MHz 87 1-10 MHz 87

f 200 MHz to 400 MHz

250f2

3-30 MHz

823.8/f

10-400 MHz

28

400-2000MHz

1.375 f

2-300 GHz

400MHz to 800 MHz 800 MHz to 1.55 GHz

100

30-100 MHz

27.4

125 f2

1,55-300ghZ 194 f1= MHz; f2= GHz;

100-300 MHz 27.4 f= MHz

153

61 f= MHz

•

Magnetic Field Strength: NRPB-Child present Frequency Am-1 535KHz-12 MHz 1.8/f21 12-200 MHz 0.13 200-400 MHz 0.66f2 0.26 400-800 MHz 800MHz to

ANSI/IEEE- Controlled Frequency Am-1 0.1-34 MHz 16.3/f 1.34-30 MHz 16.3/f 30-100 MHz 158.3/f1.668 100-300 MHz 0.0729

0.33f2

1.55GHz

ICNIRP-Public Frequency Am-1 0.15-1 MHz 0.73/f 1-10 MHz 0.73/f 10-400 MHz 0.073 400-2000 MHz 0.003 2-300 GHz

f

0.16

f= MHz

1.55GHz to 0.52 300GHz f1= MHz; f2= GHz;

f= MHz

For Amateur Radio users NRPB and FCC both have issued guidelines. The FCC guidelines for power thresholds have been shown in TABLE-6.3C below. 6.5.5 TABLE:6.5 FCC power thresholds for Amateur Radio Stations: Wave length band Power Limits 160,80,75, and 40m 500W 30m 425 20m 225 17m 125 15m 100 12m 75 10m 50 VHF all bands 50 70cm 70 33cm 150 23cm 200 13cm 250 SHF all bands 250 EHF all bands 250 Exceeding these power limits requires a station evaluation. The people around an amateur radio station fall into category of “public”.

154

6.6 Radiation Safety Management: The radiation safety management is a very important aspect in day-today life for the occupational or public around a working or even a recreational place. Some advanced countries have the management system in place. But many countries are yet to evolve a norm. An international standard is absent on this issue. This is mainly due to lack of consensus in views on the existing standards. The objectives of the safety management should include: • Identification of potential radiation hazards in occupational and public domains. • Methods to reduce them &easurements of the levels if possible. • Controlling of the hazardous levels. • Education or training of the technical and non-technical personnel about the hazards. • Development of the safety measures. • Periodic calibration and checking of RF as well the radiation measuring equipments. • Use of anechoic chambers with proper RF absorbers while calibrating high gain antenna. • Use of RF shielded chambers while aligning a transmitter. • Proper RF shielding is provided while developing an RF system in laboratories. • Proper EMI filtering is provided at required points during development of RF systems to avoid unwanted leakage. • Maintaining a record of various measurements made, safety precautions initiated and their implementation, and action taken if any. Safety audit should be maintained seriously. • A record of all incidents occurred be maintained. • Keeping of a copy of international safety document in laboratories or in work place. • A specific area is ear marked in laboratory for developing RF systems and it should be indicated by the way of a notice. Pasted on wall. • Companies developing RF equipments of their own design or manufacturing a customer’s designed system be made aware of the hazard safety levels and ways to keep the levels within limits. Similarly the customers should be made aware of radiation characteristics and safe use of the gadgets he/she is going to handle.

155

7.0 Abbreviations: AM Am-1 ANSI BBC CISPR CW DC FCC FM EIRP EMC ICNIRP IEEE ITU NRPB Pfd P.r.f RF RFI Rms SAR Vm-1 WHO Wm-2

Amplitude Modulation. Magnetic Field Strength (ampere per meter). American National Standards Institution. British Broadcasting Corporation. International Special Committee for Electro technical Standardization. Continuous Wave (un modulated wave). Direct Current. USA Federal Communication Committee. Frequency Modulation. Effective Isotropic Radiated Power. Electromagnetic Compatibility. International Commission for Non-Ionising Radiation Protection. Institution of Electrical and Electronic Engineers (USA). International Telecommunication Union. Jm-2 Energy-Joule per square meter. National Radiological Protection Board. Power Flux Density (Wm-2). Pulse Repetition Rate (Hz). Radio Frequency. Radio Frequency Interference. Root Mean Square value. Specific Absorption Rate (Wkg-1) Electric Field Strength (volt per meter) World Health Organization. Power Flux Density.

156

8.0 References : 1. Adair, E.R., Thermo physical Effects of Electromagnetic Radiation; IEEE Engineering in Medicine and Biology Magazine, March 1987. 2. Johnson, C.C and Guy, A, W., Non-ionizing Electromagnetic Wave Effects in Biological Materials and Systems; Proc. IEEE, Vol.60, June 1972. 3. Gandhi, O.P. and Riazi, A., absorption of Millimeter Waves by Human Beings and its Biological Implications; IEEE Trans. On Microwave Theory & Technology, Vol. MTT No.2, February 1986. 4. Gandhi, O.P, State of Knowledge for Electromagnetic Absorbed Dose in Man and Animals; Proc. IEEE, Vol, 68 No, 1 January 1980. 5. Gandhi, O.P, Advances in Dosimetry of RF radiation and their past and projected impact on safety standards; IEEE Proc IMTC Sandiago USA, April 1988. 6. Roger L. Freeman, Telecommunications Transmission Handbook Fourth Edition, John Wiley & Sons, INC, 1998. 7. N. Couch II, Digital & Analog Communication Systems, Macmillan Publishing Co, USA, 1990. 8. Peyton Z. Peebles, Jr. Communication Systems Principles. Addison-Wesley Publishing Company 1976. 9. Ghavami, Michel, Kohno, Ultra Wideband Signals & Systems in Communication Engineering. 10. Roland Kichen, RF and Microwave Radiation Safety Handbook, Newnes, Oxford, 2001. 11. Roland Kichen, RF and Microwave Radiation Safety Handbook, ButterworthHeinemann Ltd, 1993, 12 .Flinpowsky & Muehldorf, Space Communication Systems. 13 R.E.Ziemer, W.H. Trante, Principles of Communications Systems, Modulation, and Noise, 3rd Edition, Jaico Publishing House,1993 14. Walter Morgan & Gordan, Communications Satellite Handbook, John Willy &Sons, 1989. 15. T.D.Gibson, The Mobile Communications Handbook, CRC Press LLC, 2nd Edition, 1999. XXX---XXX

157