Radio Transmission Aspects Of Umts

Radio Transmission Aspects of UMTS Sergio BARBERIS [email protected] Tel +39 011 228 7309 Wireless Techniques and Methodologies

TELECOM ITALIA LAB

Radio transmission Aspects of UMTS OUTLINE

• Principles of spread spectrum communications (DS-CDMA, hybrid TD-CDMA) • UTRA physical layer specification • System aspects (power control, soft handover, capacity)

Principles of spread spectrum communications

ACCESS TECHNIQUES FOR MOBILE COMMUNICATIONS FDMA (TACS) P

F

TDMA (GSM, DECT)

T P

F CDMA (UMTS)

T P P - Power T - Time F - Frequency

F T

Spread spectrum systems

• •

Transmission systems where the bandwidth of the transmitted signal >> bandwidth of the information signal The bandwidth spreading is performed exploiting the properties of PN sequences (codes) that must be known at the receiving side

AWGN bandlimited channel C=B log2 (1 + S/N) How to obtain a desired bit rate R

[C]=bit/s

[B]=Hz

S/N=signal to noise ratio

Narrowband B and high S/N Wideband B and low S/N

General characteristics of spread spectrum systems

• • •

Robustness against jammers and fading Low interception probability Multiple access capability

DS-CDMA • Users sharing the same band are transmitted simultaneously on the same carrier • Users are distinguished each other by means of a “code” and, the mutual interference is reduced during the “decoding” process • In the DS-CDMA technique, the spreading is obtained multiplying the user signal by the signal associated to a “code” (a PN sequence or an orthogonal sequence)

DS-CDMA b(t) c(t)

BPSK MOD.

x(t) j(t)

BPSK z(t) DEMOD. c(t)

~ ~

fo b(t): information signal (Rate Rb) c(t): PN sequence (Rate Rc) Rc>>Rb The bandwidth spreading is obtained multiplying the information signal by the PN sequence

DS-CDMA (II) Gb(f)

b(t) 1

t f

-1

1/Tb

c(t) 1

Gc(f) t f

-1

1/Tc

c(t)b(t) 1

Gc*Gb

Spectrum of the product signal

t f

-1

1/Tc

DS-CDMA (III) The spread signal b(t)c(t) is then modulated, transmitted, interfered by a narrowband signal j(t) and demodulated at the receiveng end obtaining: z(t)= b(t)c(t) + ~j(t) The information signal b(t) is recovered multiplying z(t) by c(t): ~ ~ 2 z(t)c(t)= b(t)c (t) + j(t)c(t) = b(t) + j(t)c(t) Spread interferer

DS-CDMA (IV) Gz(f) jammer Information signal

f Spectrum of z(t)c(t) Information signal Spread interferer

f After the low pass filtering, we detect only a fraction of the original interfering power (reduction factor=Rc/Rb) f

Coding and bandwidth spreading • The bandwidth spreading can also be obtained by means of repetition codes or error correcting codes. • Terminology: – Processing gain: ratio between chip rate and user net bit rate – Spreading factor: number of chips representing a user coded bit

CDMA system

f0

0

DATA

DATA

f0

WIDEBAND SPECTRUM

0

CORRELATOR DIGITAL FILTER

ENCODING & INTERLEAVING

PN SOURCE

DEINTERLEAVING & DECODING

PN SOURCE

CARRIER

f0

BACKGROUND NOISE

CARRIER

f0

EXTERNAL INTERFERENCE

f0

OTHER CELL INTERFERENCE

f0

OTHER USER INTERFERENCE

DATA

“Long codes” e “short codes”

• Short codes: they allow a better interference control but, a code management could be necessary • Long codes: no code management is required but, the mutual interference cannot easily be controlled.

Pseudo-Noise sequences out

0

0

1

• They are obtained by means of linear shift registers (with feedback defined by a characteristic polynomial) • A N-cell LSR can provide a PN sequence with period 2N-1 • PN sequences are suitable for multiple access because their autocorrelation function decrease sharply after time shifts of few chips (i.e., equal to zero everywhere but the origin)

WALSH CODES (I) It is a set of orthogonal codes generated by the rows of a Hadamard matrix The Hadamard matrix of order two is defined as: 1

H2 =

1 1

1 -1

Tc

t

1

Walsh functions

-1

Tc

t

WALSH CODES (II) The Hadamard matrix of order 2N (and then the Walsh codes corresponding to the matrix rows) is defined as H2N =

HN HN HN -HN

Walsh functions are perfectly orthogonal (i.e., no mutual interference); actually, after transmission over a multipath channel, orthogonality is lost

Walsh codes: examp. of channeliz. (I) w1 = [-1, +1, -1, +1] w2 = [-1, -1, +1, +1] w3 = [-1, +1, +1, -1]

We assume for example Rc = 4Rb i.e., SF = 4

Data stream to be transmitted d1 = [1, -1, 1] d2 = [1, 1, -1] d3 = [-1, 1, 1]

Walsh codes: examp. of channeliz. (II) d1 d2 d3

1 1 -1

-1 1 1

1 -1 1

w1 w2 w3

-1 -1 -1

1 -1 1

-1 1 1

1 1 -1

-1 -1 -1

1 -1 1

-1 1 1

1 1 -1

-1 -1 -1

1 -1 1

-1 1 1

1 1 -1

d1w1 d2w2 d3w3

-1 -1 1

1 -1 -1

-1 1 -1

1 1 1

1 -1 -1

-1 -1 1

1 1 1

-1 1 -1

-1 1 -1

1 1 1

-1 -1 1

1 -1 -1

r(t) = d1w1+ d2w2+ d3w3

-1

-1

-1

3

-1

-1

3

-1

-1

3

-1

-1

Tc

Tb

Walsh codes: examp. of channeliz. (III) To recover the information d1 multiply the composite signal r(t) by the code w1 and then we sum up (integrate) over the bit time w1 r(t)w1 D1=

ΣTb

-1

1

-1

1

-1

1

-1

1

3

1

4

1∑ -1 Tb

-1

-3

1

-1

1

-1

1

-1

1

3

1

-1

-4

D1>0 ⇒ transmitted d1= 1 D1<0 ⇒ transmitted d1=-1

4

⇒ d1=[1, -1, 1]

Time-variant multipath channel: example of impulse response Received signal

Transmitted signal

t=t1

t=t0

t=t0+a

t=t0+b

t=t1+τ11

t=t1+τ12

t=t2+τ21

t=t2

t=t3

t=t3+τ31

t=t3+τ32

Propagat. impairments due to multipath

τ1

f

f

t

t

α1ejφφ1 x

αnejφφn

τn

+

x

t

t

Receiver: Rake receiver

τi

αi e jϕ

τn

τn

despreading

αne jϕ

i

n

X

Channel estimat.

X

Σ

Phase recovery

τi

despreading

Propagation channel Spreading sequence

Phase recovery Channel estimat.

receiver

Σ

TD-CDMA Access Technique

TDD (TD-CDMA) technique

One Time Slot

Energy

WB-TDMA/CDMA

3.

84

M

Fr e

ch ip /s

qu en

cy

Codes

1

2

3

.

.

frame with 15 time slots

.

14

15

Time

TD-CDMA spreading codes • Within each 0.666 ms time slot, more channels can be allocated and separated each other by means of spreading codes • The codes can be allocated to different users or to a same user, according to the needs. • The number of codes in a time slot is not fixed but depends on the rate and spreading factor of each physical channel. • SFMAX = 16 • After spreading, data are scrambled with a cell specific scrambling sequence

Resource allocation

• In the TD-CDMA component a physical channel is identified by a combination of carrier, time slot and code. • Resources are allocated to cells by means of slow DCA: – slot clustering – each slot can be used in both uplink and downlink transmission, according to the needs

• resources are allocated to bearers by means of fast DCA: – high bit rate services can be provided allocating to a same user several codes in a same or different time slots.

Joint detection

• It reduces (ideally cancels completely) the mutual interference among signals • The receiver exploits the knowledge of all the spreading sequences used by the other users on the same slot/carrier and perform the simultaneous demodulation of all signals. • The output is a vector of information sequences (one for each user)

Joint detection basic principle

Mobile 1

CDMA code 1

Base station

Midamble channel 1

X

1

Traffic channel 1 input data

Radio channels

Mobile K

CDMA code K

X Traffic channel K input data

Midambles channels 1 - K

Channel estimation of K radio channels Estimated radio

1 2

K channels 1 - K

Traffic channel 1 output data

Joint Detection (JD) of K traffic channels

Midamble channel K

Traffic channel K output data

K

CDMA codes 1 - K

Joint Detection If e = (d ·c)*h + n is the received vector We have to estimate the transmitted vector d by means of the following equation: d’ = M ·e The matrix M has to be calculated so as to maximize performance and minimize complexity

UTRA: physical layer specification

Technical Specification Group responsible for the L1 specification • 3GPP/RAN-WG1 “Radio layer 1 specification” • Chairman: Antti TOSKALA (Nokia) E-mail “[email protected]” • Secretary: Shinobu IKEDA (ETSI) E-mail “[email protected]” • Meeting attendance: about 130 delegates • Documents available at the following address: “ftp://ftp.3gpp.org”

UTRA L1 spec. organisation (rel.’99) The technical spec. is organised in 11 documents: • A general overview of the specification (TS25.201) • Five specification documents on the FDD component (TS 25.211-TS 25.215) • Five specification documents on the TDD component (TS 25.221-TS 25.225) • Two technical reports (TR 25.833, TR 25.944)

Technical specification documents (I) • TS 25.201: Physical layer - General description It describes the content of the TS 25.2xx documents and provides an overview of the physical layer. • TS 25.211: Physical channels and mapping of transport channels onto physical channels (FDD); the correspondent document for the TDD component is TS 25.221 • TS 25.212: Multiplexing and channel coding (FDD); the correspondent document for the TDD component is TS 25.222 • TS 25.213: Spreading and modulation (FDD); the correspondent document for the TDD component is TS 25.223.

Technical specification documents (II) • TS 25.214: Physical layer procedures (FDD); the correspondent document for the TDD component is TS 25.224. • TS 25.215: Physical layer - Measurements (FDD); the correspondent document for the TDD component is TS 25.225.

Technical Reports • TR 25.833: Physical layer items not for inclusion in Release ‘99. • TR 25.944: Channel coding and multiplexing examples It is a document “strongly supported” by NTT DoCoMo containing several examples of channel coding and multiplexing for some typical transport channels. • A technical report is being produced for each Work Item to be included in Release 4 and 5. When the Work Item is completed the text from the TR is moved in the relevant TS by a CR procedure.

Release 4 • Same document structure as release 99 (the first digit of the version is 4) • Main new features with respect to release 99: – – – –

1.28 Mchip/s TDD option DSCH power control improvement in soft handover TDD Node B synchronisation UE positioning

Some Work/Study Items for Release 5 • • • •

High Speed Downlink Packet Access (HSDPA) Uplink Synchronous Transmission (USTS) Radio Link Performance enhancements Node B synchronisation for 1.28 Mchip/s TDD

Main parameters UTRA/FDD

UTRA/TDD

Access technique

WCDMA

Hybrid WCDMA+TDMA

Chip rate

3.84 Mcps (SF FDD:4-256, TDD 1-16)

Carrier spacing

4.4-5 MHz (200 kHz carrier raster)

Frame duration

10 ms

N. slot per frame BTS synchronization Modulation Coherent receiver Multi-rate

15 Not required

Not required (advisable) DL: QPSK DL: QPSK UL: Dual-channel QPSK UL: QPSK Uplink e downlink Variabile SF + Multi-code + Multi-slot (TDD only)

Main parameters: Wideband TDD vs Narrowband TDD UTRA/N-TDD Access technique

CDMA (Synch)+TDMA

UTRA/W-TDD Hybrid WCDMA+TDMA

1.28 Mcps (SF 1-16)

3.84 Mcps (SF 1-16)

Carrier spacing

1.6 MHz

4.4-5 MHz (200 kHz carrier raster)

Frame duration

10 ms (5 ms subframes)

10 ms

N. slot per frame

Subframe: 7+3 minislots

15

synchronization

Both uplink and downlink

Chip rate

DL: Not required (advisable) UL: not required

Coherent receiver

QPSK QPSK 8 PSK optional Uplink e downlink

Multi-rate

Variabile SF + Multi-code + Multi-slot

Modulation

Smart antennas

Strongly advised

Optional

Map. of transport channels onto physical channels T r a n s p o r t C h a n n e ls

P h y s ic a l C h a n n e l s

DCH

D e d ic a te d P h y s ic a l D a ta C h a n n e l ( D P D C H ) D e d ic a te d P h y s ic a l C o n tr o l C h a n n e l ( D P C C H )

RACH

P h y s ic a l R a n d o m A c c e s s C h a n n e l ( P R A C H )

CPCH

P h y s ic a l C o m m o n P a c k e t C h a n n e l ( P C P C H ) C o m m o n P ilo t C h a n n e l ( C P I C H )

BCH

P r im a r y C o m m o n C o n tr o l P h y s ic a l C h a n n e l ( P - C C P C H )

FACH

S e c o n d a r y C o m m o n C o n tr o l P h y s ic a l C h a n n e l ( S - C C P C H )

PCH S y n c h r o n is a tio n C h a n n e l ( S C H ) D SCH

P h y s ic a l D o w n li n k S h a r e d C h a n n e l ( P D S C H ) A c q u is it io n I n d ic a to r C h a n n e l ( A I C H ) A c c e s s P r e a m b le A c q u i s itio n I n d ic a to r C h a n n e l ( A P - A I C H ) P a g in g I n d ic a to r C h a n n e l ( P I C H ) C P C H S ta t u s I n d ic a to r C h a n n e l ( C S I C H ) C o llis io n - D e te c tio n /C h a n n e l - A s s i g n m e n t I n d ic a to r C h a n n e l (C D /C A -IC H )

Physical channels UL • Physical channels: – DPDCH (Dedicated Physical Data Channel): it is used to carry dedicated data generated at OSI layer 2 and above (user data or associated signalling). – DPCCH (Dedicated Physical Control Channel): it is used to carry control information generated at OSI layer 1; the information include pilot bits for channel estimation, Transmit Power Control bits (TPC), bits to indicate the bit rate (TFCI) and, in the UL only, Feedback Information bits required for transmission diversity (FBI) – DPDCH and DPCCH are transmitted on the I and Q branch respectively of a QPSK modulator; they are distinguished by means of different codes. – PRACH (Physical Random Access Channel): it is used to carry the RACH, the transport channel used by the mobile to access the system – PCPCH (Physical Common Packet Channel): it is used to carry the CPCH, the transport channel for packet transmission (contention access)

Frame structure for channels DPxCH (UL) 0.667 ms

k

Tslot = 2560 chips , 10x2 bits (k=0..6) Data Ndata bits

DPDCH

Pilot N pilot bits

DPCCH

TFCI N TFCI bits

FBI NFBI bits

TPC NTPC bits

Tslot = 2560 chips, 10 bits

Slot #0

Slot #1

Slot #i Tf = 10 ms

Slot #14

Frame duration: 10 ms each frame is split into 15 slot (0.667 ms) corresponding to one power control period. Spreading factor SF: 4 ≤ SF ≤ 256 DPDCH and DPCCH can be characterised by different values of SF

Spreading and modulation - up link Channelization codes (OVSF) cD Real

I

DPDCH

c’scramb

p(t)

I+jQ

cC DPCCH

cos(ω ωt)

Q

sin(ω ωt) Imag

∗j

p(t)

cD, cC : channelization codes c’ scramb: scrambling code (short or long) p(t): pulse-shaping filter (root raised cosine, roll-off 0.22)

DPDCH and DPCCH are separate by means of different codes. During a call (circuit switched), at least the DPCCH is always active A same code can be reused on the I and Q branch. The scrambling codes are complex sequences QPSK modulation is used

Physical and logical channels: downlink • Physical channels: – DPDCH and DPCCH are transmitted as in the uplink case but, here are time multiplexed – The Downlink Shared Channel (DSCH) is used to transmit packet traffic scheduled by the base station according to the traffic originated by the users. A dedicated channel used to carry the physical layer control information is always associated to DSCH. – The Broadcast CHannel BCH is a downlink transport channel that is used to broadcast system and cell specific information; the BCH is time multiplexed with the SCH (Synchronisation Channel), the channel which allows the mobile to acquire the synchronisation so as to demodulate the signal received from the base station. The resulting time multiplex is transmitted over the Primary CCPCH (Common Control Physical Control Channel). A code multiplexed common pilot (CPICH) is transmitted too, separated from the Primary CCPCH and transmitted on a separate code. – The secondary CCPCH is used to transmit the paging channel (PCH) and the Forward Access Channel (FACH). They can carry also short user packets.

Physical channels - down link DPDCH

DPCCH

DPDCH

TPC TFCI Data1 N data1bits N TPC bits N TFCI bits

Data2 N data2 bits

DPCCH N

Pilot pilotbits

k

Tslot = 0.666 ms (2560 chips), 10x2 bits (k=0..7)

Slot #0 Slot #1

Slot #i Tf = 10 ms

Slot #14

DPCCH and DPDCH are defined as the UL but, they are time multiplexed 4 ≤ SF ≤ 512

Spreading and modulation - down link cos(ω ωt)

I

DPDCH/DPCCH

S

p(t)

cch

P

sin(ω ωt)

c scramb

Q

p(t)

cch: channelization codes c’ scramb: scrambling code p(t): pulse-shaping filter (root raised cosine, roll-off 0.22)

A same channelisation codes is used on the I and Q branch. The scrambling code is a real sequence that is used on both the I and Q branch. QPSK modulation

Spreading codes

C4,1 = (1,1,1,1) C2,1 = (1,1) C4,2 = (1,1,-1,-1) C1,1 = (1) C4,3 = (1,-1,1,-1) C2,2 = (1,-1) C4,4 = (1,-1,-1,1)

SF = 1

SF = 2

SF = 4

Two kind of codes are used: -Orthogonal Variable Spreading Factor (OVSF) codes are used as channelisation codes; OVSF are defined by means of a binary tree -scrambling codes are used so as to guarantee good autocorrelation properties and in order to distinguish different cells (in the downlink) and to distinguish mobile users (in the uplink)

Uplink Variable Rate (No DTX) 10 ms

1-rate

1/2-rate 1/4-rate 0-rate

Variable rate

R=1

R = 1/2

: DPCCH (Pilot+TPC+TFCI) : DPDCH (Data)

R=0

R=0

R = 1/2

Downlink Variable Rate (DTX based) 0.666 ms 1-rate 1/2-rate 1/4-rate 0-rate

: DPCCH-part (Pilot+TPC+TFCI) : DPDCH-part (Data)

TrCH

TrCH

Add CRC per Tr. block

Add CRC per Tr. block

Channel coding

Channel coding

1st Interleaving

1st Interleaving

Rate-Matching

Multiplexing

2nd Interleaving

Mapping to Physical channels

Transport channel multiplexing

Packet Access

• There are three possible cases – Short and infrequent packets are transmitted on common control channels (FACH, RACH) – Big size packets or scheduled packets can be transmitted over a dedicated channel – In the downlink case it is possible to use a shared channel (DSCH) where the access of the different user packets is scheduled by the BS; DSCH is suitable for medium and large amount of data – On the uplink it is possible to use the CPCH (contention channel); CPCH is suitable for medium and large amount of data

Common Channel Packet Access Access request

User packet

Arbitrary time

Access request

User packet

Common Channel (RACH/FACH)

• • • •

No link maintenance when no packet to transmit Limited to small packets and medium data rates No fast power control No soft handover

Dedicated Channel Multi-Packet Transmission Scheduled packets Non-scheduled packet Access request

User packet

User packet

Access request

User packet

Dedicated Channel (DCH) Link maintenance (pilot, TPC, TFCI)

• Scheduled and non-scheduled packet access • Closed-loop power control and soft handover • Link released after time-out period has expired

DSCH (Downlink Shared Channel)

• The DSCH is a downlink channel,shared in time among all users (orthogonal code shared between users) • DSCH is used in parallel with a low bit rate dedicated channel • Closed loop power control allowed; no soft handover

Common Packet Channel (Uplink • the CPCH is an uplink channel: users can access CPCH by means of a contention mechanism • Channels used to optimise the radio resources in case of packet transmission • Fixed code per cell • closed loop power control allowed; no soft handover