Multi-User Communication

Transcription

1 Multi-User Communication Lecture 10 WCDMA Overview slide 1 Objective Introduce WCDMA from a systems perspective, but with a focus on lower layers (FDD mode) WCDMA Release 99 giving you, hopefully a technology context to which you can apply e.g. the theory on multi-user communication a system context from which you can explore recent advances on WCDMA (HSxPA) and its evolution (LTE) slide 2 1

2 Outline WCDMA introduction UMTS and 3GPP specifications UTRAN architecture Basic radio resource management Physical layer channels and procedures Short on TDD mode MUD in WCDMA uplink (gain potential) References Acronyms slide 3 WCDMA UE Node B orthogonal codes Soft/Softer Handover non-orthogonal codes (Code) Power DATA Bit rate Chip rate Channelisation code Scrampling code slide 4 Chip rate UE 4 UE 3 UE 2 UE 1 Available resources: Spreading Codes (OVSF) and Transmission Power Time 2

3 WCDMA Coverage and Capacity Dækning er begrænset af uplink db forbedring af dækningsområde Cell range (km) Maximum path loss (db) Uplink (144 kbps / 125 mw terminal) Downlink 10W Typisk maks. tab Downlink Downlink 20W 20W slide Kapacitet er begrænset af downlink Cell load (kbps) 3GPP Specifications slide 6 3

4 UMTS releases Standardized by 3rd Generation Partnership Project (3GPP), see [North America: 3GPP2] Release 99 12/99 Release 4 03/01 v3.0.0 v3.1.0 v3.2.0 v3.3.0 v3.4.0 v4.0.0 v4.1.0 v4.2.0 etc. etc. Release 5 06/02 v5.0.0 v5.1.0 etc. Release 6 06/05 UMTS Long Term Evolution slide 7 Corrections New Functions v6.0.0 etc. UMTS used for designating 3rd generation systems (ITU: IMT-2000) 3GPP specs Main rule for 3GPP specifications ( XX.INN XX: series specification I: (0) applies to both 3G and GSM (GPRS/EDGE) (1,2) applies to 3G only GSM means GERAN 3GPP RAN while 3G means a 3GPP UTRAN RAN Examples TS (v ), Physical channels and mapping of transport channels onto physical channels (FDD), release 6, Technical Specification Group Radio Access Network, July 2004 TS (v ), Spreading and Modulation (FDD), Technical Specification Group Radio Access Network, December 2003 TS (v ), Base Station (BS) radio transmission and reception (FDD), Technical Specification Group Radio Access Network, December 2004 TS (v ), Multiplexing and channel coding (FDD), Technical Specification Group Radio Access Network, December 2004 TR (v ), Beamforming enhancements (release 6), Technical Specification Group Radio Access Network, March 2004 TR (v ), Multiple input multiple output in UTRA, Technical Specification Group Radio Access Network, August 2004 TR (v ), Tx diversity solutions for multiple antennas, Technical Specification Group Radio Access Network, February 2004 slide 8 4

5 3GPP Series slide 9 UTRAN Architecture UMTS Terrestrial Radio Access Network slide 10 5

6 PLMN Architecture Public Land Mobile Network Radio-specific part Uu/Um Iu UE/MS UTRAN/GERAN CN From Release 5 GSM and UMTS have the same interface to the radio specific part of the network slide 11 PLMN Architecture Radio-specific part The geographical area covered by a PLMN is partitioned into MSC serving areas; a location area is a subset of a single MSC serving area. Typically, there is one (logically speaking) HLR in an operators PLMN. USIM Uu/Um Node B/ BTS Node B/ BTS Iub/Abis RNC/ BSC Iu HLR/AuC Iur Node B/ ME PS BTS RNC/ Internet, SGSN GGSN Node B/ BSC X25, etc. BTS UTRAN/ UE/MS GERAN CN External Network Public Land Mobile Network MSC/ VLR PLMN GMSC PLMN, PSTN, ISDN, etc. CS slide 12 6

7 BSS (RAN/GERAN) CS MGW Nb MSC-Serv./VLR G/E/Nc Circuit-switched core network MSC Mc BTS Abis A CS MGW MSC-Serv./VLR Nc BSC GMSC-Serv. Um BTS Gb Iu Cs D C Nb Mc SIM-ME CS MGW SIM ME HSS/AuC CS Domain USIM Cu MS Node B Iu bis UTRAN RNC Gs Gr Cx Gc IMS Dom ain (Release 5) Mb/Gi Uu Node B Gn Iur RNC Iu PS SGSN GGSN PS Domain Packet-switched core network SGSN Access Network Domain Core Network Domain User Equipment Domain slide 13 Infrastructure Domain NRT Packet Switched Data Retransmission, sequence numbering, flow control, multiplexing, etc. Protocol stack of a NRT packet switched session in UMTS Release 99 slide 14 7

8 Basic RRM Radio Resource Management slide 15 RRM Overview PC PC LC I u b AC LC HC PS RM PC UE Node B RNC Uu Iub RRM in UMTS Release 99 AC Admission Control; PS Packet Scheduler; LC Load Control; RM Resource Manager; HC Handover control; PC Power Control slide 16 8

9 Power Control Slow Outer Loop Power Control (OLPC) at rate Hz Fast Closed Loop Power Control (CLPC) at rate 1500 Hz UE Node B Node B adjusts the power to keep the SIR at the SIR target RNC RNC adjusts the SIR target in the Node B for the fast CLPC in response to link quality In uplink to keep the? received signal level the same for all users (near-far effect) In downlink to increase the reception quality of stationary users and users at the cell edge To increase spectral efficiency slide 17 Uplink Fast PC UE 1 and UE 2 are transmitting at the same frequency => equalizing received powers at Node B is critical to avoid near-far problems Closed loop power control: Node B commands UE to increase or to decrease its transmission power at a rate of 1.5 khz (±1 db steps) Closed loop power control follows also the fast fading pattern at low and medium speeds (< 50 km/h) slide 18 UE 1 L 1 PC commands L 2 Node B UE 2 Fast PC algorithm in Node B: If E b /N 0 <E b /N 0,target, send "power-up" command. Else If E b /N 0 >E b /N 0,target, send "power-down" command. 9

10 Outer Loop PC General outer loop algorithm Decrease Eb/N0 target Yes Estimated quality better than required? No Increase Eb/N0 target 7 Example adjustments of E b /N 0 target for AMR speech service, BLER target 1% If error in frame, increase E b /N 0 target by 0.5 db If no errors, decrease E b /N 0 target with such a rate that BLER = 1% on average minute period slide 19 Softer Handover Softer handover UE is connected to two sectors of one base station Softer handover probability 5-15 % UL/DL Basically same Rake combining as for multipath and antenna diversity (Node B and UE) Uplink combing from two sectors in Node B Rake receiver (maximal ratio combining) Sector 1 Sector 2 RNC slide 20 10

11 Soft Handover Soft handover UE is connected to two base stations Soft handover probability is % Required to avoid near-far effects Extra transmission over Iub More baseband processing needed (both base stations) DL Maximal ratio combining in UE in the same way as with softer handover or multipath diversity UL Frame selection combining in RNC RNC Uplink combing from two base stations in RNC (selection combining) slide 21 Soft Handover Execution (1/2) Active Set (AS) cells have the knowledge of service used by UE RNC informs the new cell (to be added to AS) about the needed connection, forwarding the following: Coding schemes, number of parallel code channels, the different transport channel configuration parameters in use by UL and DL UE ID and uplink scrambling code The relative timing information of the new cell with respect to the existing connection (as measured by the UE at its location). Based on this, the new Node B can determine what should be the timing of the transmission initiated with respect to the timing of the common channels (CPICH) of the new cell MS is informed about the channelisation codes to be used in transmission and relative timing information through existing connection slide 22 11

12 BS A Soft Handover Execution (2/2) BS B Handover command and T offset T offset BS B channel information Measure T offset UTRAN Transmision channel and T offset PCCCH frame PDCH/PCCH frame The relative timing information, which needs to be made available at the new cell is indicated in the above figure It makes transmissions capable to be combined in the Rake receiver from timing point of view slide 23 Fast Power Control in Soft Handover Power drifting BS 1 Reliability check RNC: Power drifting control Both Node Bs Detect downlink PC command from mobile Adjust downlink transmission power BS 2 UE: Check reliability of uplink PC command Adjust uplink transmission power Base stations detect independently the power control command from mobile to control downlink transmission power Independent power control commands are sent from Node Bs to UE to control uplink transmission power slide 24 12

13 Resource Management Bit rate Chip rate Chip rate DATA Channelisation code Scrampling code Code Allocation and Code Tree Management Uplink Downlink Spreading Separate bearer services Separate users/ bearer services Scrambling Separate users Separate cells All physical channels are spread with individual spreading codes, C m (n) and subsequently by the scrambling code, C FSCR Resource Manager generates DL spreading codes. The code layer, m and the code number, n designates each and every code in the layered orthogonal code sequences. slide 25 Code Types Downlink OVSF channelisation (or spreading) codes (SF 4-512) Scrambling codes long scrambling code (Gold code with 18 degree polynomial), but using only one frame (38400 chips) complex valued code is formed by time delayed version of the same code limited to 512 possible codes divided into 64 code groups Uplink OVSF channelisation (or spreading) codes (VSF 4 256) Scrambling codes short and long codes long scrambling code (Gold code with 25 degree polynomial), but using only chips» complex valued code is formed by time delayed version of the same code short 256 chips extended S(2) code family» complex valued code is formed by combining two codes millions of scrambling codes slide 26 13

14 Resource Manager Code Allocation Code Allocation Algorithm chooses the proper spreading code depending on the transport format combination type. Layer 0 C 0(0)=(1) Layer 1 C 1(0)=(1,1) C 1(1)=(1,-1) C 2 (0)=(1,1,1,1) C 2(1)=(1,1,-1,-1) C 2 (2)=(1,-1,1,-1) C 2(3)=(1,-1,-1,1) Layer 2 Layer 3 C 3(0)=( ) C 3 (1)=( ) C 3(2)=( ) C 3 (3)=( ) C 3(4)=( ) C 3 (5)=( ) C 3(6)=( ) C 3(7)=( ) The codes are layered from 0 to 11 according to the code type (~SF) Only layers 2 to 8 are available for DL and 2 to 7 for UL slide 27 RM Examples Examples: Ordinary DL speech 30 kbps channel (AMR kbps & control part with 1/3 channel coding - code type 7 (128 chips/symbol) C 2 (1) code layer = 2; code number = 1 code = kbps channel - code type 5 (32 chips/symbol) C 4 (5) code layer = 4; code number = 5 code = The Resource Manager maintains code tree orthogonality If a code C m (n) is in use, all the codes that are below it in the same branch and the codes that are above it in the same branch to the root are made unavailable slide 28 14

15 Physical Layer Channels and Procedures slide 29 Channel Types MAC selects appropriate bit rate according to the instantaneous source bit rate. Medium Access Control (MAC), Layer 2 Logical channels Transport channels Physical Layer, Layer 1 Physical channels slide 30 Physical layer supports variable bit rates up to 2 Mbps 15

16 WCDMA Channels BCH Broadcast how and with what characteristics FACH Forward Access PCH Paging Transport Channels RACH Random Access DCH Dedicated CPCH Common Packet DSCH Downlink Shared PCCPCH Primary Common Control SCH Synchronisation SCCPCH Secondary Common Control Physical Channels CPICH Common Pilot slide 31 AICH Acquisition Indication PRACH PICH Paging Indication DPDCH DPCCH CSICH CPCH Status Indication PCPCH PDSCH CD/CA-ICH Collision Detect/Avoidance Mobile Transport Channels (1/2) Dedicated channel DCH: data + signaling to one user Broadcast channel BCH: Cell and system info Forward access channel FACH: Data + signaling for one or more users within one cell Paging channel PCH: For mobile terminated calls Downlink shared channel DSCH: Packet data channel. Time multiplexed by several users. Node B DSCH optional for network Random access channel RACH: Data + signaling from one user Common packet channel CPCH: Extension of RACH for longer data packets slide 32 CPCH optional for network 16

17 Transport Channels (2/2) Due to direct support of variable bit rate and service multiplexing in UTRA/FDD there is only one dedicated transport channel (DCH). DCH contains user data and control information from higher layers. There exist a total of six common transport channels in UTRA/FDD: Broadcast channel (BCH): General information of UTRA network or the current cell (e.g. random access codes, access slots). BCH is sent at low data rate (single TF) and high power to reach all users in intended coverage area. Forward access channel (FACH): Downlink transmission of control information to UE's in current cell. Slow power control and low data rates. Paging channel (PCH): Downlink paging information (e.g. call initiation). Random access channel (RACH): Uplink control information (e.g. UE requests to set up connection/initiate call). Single frame only. Optional uplink common packet channel (CPCH): Extended RACH for sending data over multiple frames. Optional downlink shared channel (DSCH): Somewhat similar to RACH but can be shared by multiple users to increase data throughput. slide 33 RABs and TrChs Services in UMTS are classified according to their QoS requirements into one of 4 service classes The service classes are characterised by certain bearer attributes provided by the UMTS Radio Access Bearer Each Radio Access Bearer (RAB) is transmitted on a specific Transport Channel (TrCh) In a multi-service environment (with different QoS requirements) transmission is done on a combination of TrChs which are transmitted on the same physical channel(s) slide 34 17

18 TrCh Details (1/2) Transport Format (TF) group of parameters describing the transmission "mode" on a specific TrCh during a TTI (TTI size is part of the TF) TF Set (TFS) corresponds to a group of TFs applying to one specific TrCh TF Combination Set (TFCS) the product of TF Sets of all the TrChs forming the combination TF Indicator (TFCI) Each TF Combination (TFC) of the TFCS is indexed with the TFC Index (TFCI) at the physical layer slide 35 TrCh Details (2/2) TFI4 64 TFI4 TrCh 1 TrCh 2 TFI4 TFCI 5 64 TFCI 2 Example with radio bearer for user data and signalling TFI3 16 TFI2 32 TFI3 TFI1 16 TFCI 4 32 TFCI 1 TFI3 8 8 TFI1 0 0 TFI0 0 TFI0 TFI0 0 TFI0 TFCI 3 0 TFCI 0 Example bit rates for NRT Peak bit rate in bearer parameters is requested from PS Scheduled bit rate TFS for NRT RB includes all intermediate rates TFS subset for TFCS construction TFCS (SL & NRT RB) TFCI TFI TrC TFI TrC 0 H1 0 0H TFCS Construction by cartesian product slide 36 18

19 Dedicated Channel (DCH/DPCH) Speech and data services slide 37 Characteristics for UL and DL Dedicated channel (DPCH) consists of two physical channels: DPCCH keeps physical layer connection running reliably DPDCH carriers user bits with variable bit rate Possible to have a power offset between the two channels Target Content Bit rate Dedicated physical control channel (DPCCH) Keep physical layer connection running (1) Reference symbol: Channel and SIR estimation (2) Power control signaling (3) TFCI: bit rate information Constant bit rate for reliable detection Dedicated physical data channel (DPDCH) Carry user data and higher layer control data (1) User data (2) Higher layer signaling (RRC) Variable bit rate. Bit rate indicated with TFCI on DPCCH. slide 38 19

20 Solution Multiplexing of DPCCH and DPDCH Uplink I/Q code multiplexing Downlink Time multiplexing Target Continuous transmission reduce audible interference (1) Only one code needed saves orthogonal codes (2) Support for blind rate detection The code consumption is not an issue in uplink since the number of codes is very large The discontinuous transmission is not an issue in downlink since common channels (10-20% of BTS max power) are transmitted all the time Blind rate detection (no TFCI bits) is easier for the mobile when the channel bit rate remains constant in time multiplexed solution Variable rate transmission for data can be implemented by discontinuous transmission (DTX) on a slot interval (DL) and frame (UL) basis, symbol repetition where frame is always full, or variable spreading factor (UL). slide 39 Variable Rate in Uplink Continuous mobile transmission regardless of the bit rate (also during service DTX) Reduced audible interference to other equipment (nothing to do with normal interference, does not affect the spectral efficiency) Services can still have DTX, like silence period in speech. During that time no DPDCH transmitted but still continuous DPCCH Fast power control keeps received power of DPCCH constant Higher bit rate Service in DTX (e.g. silence in speech) Low bit rate DPDCH 10 ms frame 10 ms frame 10 ms frame DPCCH slide 40 20

21 Code(Power) Pairedwith Channel Structure (DPCH) UL SF = DPDCH 10 ms Code (Power) DPDCH DPCCH SF = 256 Paired with DL DPCH (DPCCH/DPDCH) TTI Transmission Time Interval (TTI) TCFI, (DL) TPC, PILOT TFCI (DL), TPC, PILOT TCFI, TFCI (UL) (UL), TPC, (PILOT) TPC, (PILOT) Carries the Dedicated (DCH) transport channel SF = (4-256) DPDCH DPCCH Dedicated Physical Data Channel Dedicated Physical Control Channel slide 41 TFCI TPC Transport Format Combination Indicator Transmitter Power Control Uplink DPDCH/DPCCH DPDCH Data N data bits T slot = 2560 chips, N data = 10*2 k bits (k=0..6) DPCCH Pilot N pilot bits TFCI N TFCI bits FBI N FBI bits TPC N TPC bits Fixed SF256 T slot = 2560 chips, 10 bits Let s the receiver know what is coming which TrCh s are active Slot #0 Slot #1 Slot #i Slot #14 in frame! 1 radio frame: T f = 10 ms Variable SF from 4 to 256 on a frame-by-frame basis are supported in the uplink slide 42 21

22 Uplink Processing Super frame 720 ms Frame 1 Frame 22 Frame ms (2) Detect PC command and adjust DL tx power Slot 1 Slot 2 (1) Channel estimate Slot 15 + SIR estimate for PC for adjusting UL tx power DPCCH Pilot TFCI TPC DPDCH Data (4) Interleaving (TTI) : Detect data Slot ms = 2/3 ms (3) Detect TFCI (10 ms frame) slide 43 Uplink TX (I) CRC encoding: Cyclic redundancy check (CRC) attachment is done to enable error detection at the receiver. The CRC indicator length can be set to 0/8/12/16/24 bits depending on the desired error detection accuracy. Encoder block size adjustment: Transport block concatenation is used for smaller amounts of data in order to reduce the overhead of tail bits and to increase the block size to improve the channel encoding performance. On the other hand, code blocks segmentation is done to avoid excessively large block sizes. Raw bits CRC encoding Encoder block size adjustment slide 44 22

23 Uplink TX (II) Channel encoding is done in order to improve the bit or frame error rate (BER/FER) performance of the link. Variable coding is supported (from no coding to high coding). For the relatively low data rates (similar to second generation systems), convolutional encoding (½ and 1/3 rate) is used for simplified detection and good performance. The highest data rates uses 1/3-rate Turbo encoding for best coding gain. Radio frame equalization is done by either concatenating transport blocks together or by segmenting blocks such that data is divided into equal-sized blocks when they do not fit a single 10 ms frame. Channel encoding Radio frame equalization slide 45 Uplink TX (III) Inter-frame interleaving is done whenever the delay-budget (for the current QoS) allows for more than 10 ms (1 frame) of delay. The interleaving length may be 20/40/80 ms. Interleaving reduces correlation between adjacent chips and thus improves detection (basic assumption for efficient channel decoding). Radio frame segmentation is padding the input bit sequence in order to ensure that the output can be segmented in an integer number of data segments of same size (subclause in TR25.212). The frame segmentation is only performed in the uplink since in the downlink, the rate matching output block length is always an integer multiple of the desired number of data segments. Inter-frame interleaving Radio frame segmentation slide 46 23

24 Uplink TX (IV) Rate matching ensures that the frames are filled up with data. To do this, either by puncturing or by repetition. Repetition is usually preferred for the uplink. The rate matching is dynamically updated on a frame-to-frame basis. The rate matching algorithm is detailed in TR Multiplexing: Finally, all the active transport channels are multiplexed and a 10 ms intraframe interleaving is conducted. After the interleaving, the data is mapped onto the physical channels. Rate matching Transport channel multiplexing + intra-frame interleaving + physical layer mapping slide 47 DPCCH/DPDCH# Uplink TX (V) block diagram Depending on which data rate is desired, each user can simultaneously have 6 DPDCH channels (data) and one DPCCH channel (control information). Spreading Scaling Example 3 x DPDCH configuration DPDCH1 β d Spreading DPDCH3 Scaling β d Σ Complex Scrambling Re{} RRC cos(ωt) Spreading DPDCH2 Spreading DPCCH Scaling β d Scaling β c Σ j Rotation Im{} RRC sin(ωt) S(t) Dual-channel QPSK modulation (BPSK modulation + I/Q code multiplexing) slide 48 24

25 Physical Layer Rates (Uplink) Slot Format #i Channel Bit Rate Channel Symbol SF Bits/ Bits/ N data (kbps) Rate (ksps) Frame Slot A single code at SF 4 allows 960 kbps which turns into a user data rate of 480 kbps with ½ rate coding; 6 parallel DPDCHs at ½ rate coding leads to a maximum user data rate in excess of 2 Mbps. Beneficial to stick to a single DPDCH for as long as possible to reduce Peak to Average Ratio (PAR). slide 49 Downlink DPDCH/DPCCH TPC N TPC bits TFCI N TFCI bits T slot = 2560 chips, 10*2 k bits (k=0..7) Let s the receiver know what is coming which TrCh s are active Slot #0 Slot #1 Slot #i Slot #14in frame! One radio frame, T f = 10 ms DPDCH Data2 N data2 bits DPDCH DPCCH DPCCH Data1 N data1 bits Pilot N pilot bits Constant SFs from 4 to 512 are supported in the downlink (some restrictions for SF 512). The SF for the highest transmission data rate determines the channelisation code reserved from the code tree. slide 50 25

26 Downlink Processing Super frame 720 ms (1) Channel estimate + SIR estimate for PC for adjusting DL tx power Can use CPICH Frame 1 Frame 22 Frame ms (2) Detect PC command and adjust UL tx power Slot 1 Slot 2 Slot Data TPC TFCI Data Pilot DPDCH DPCCH DPDCH Slot ms = 2/3 ms (3) Detect TFCI (10 ms frame DPCCH (4) Interleaving (TTI) : Detect data slide 51 Downlink transmitter The downlink uses time multiplexing between data and control information. This is possible since there are multiple users and there are always general control channels being transmitted from the BS (e.g. SCH). On the uplink, multiplexing like this would cause audible interference during discontinuous transmission. Spreading Even bit Odd bit All channels (except SCH) Spreading Rotation j Other channels (users)... Complex Scrambling Σ SCH Re{} Im{} RRC cos(ωt) sin(ωt) RRC S(t) slide 52 26

27 Physical Layer Rates (Downlink) Symbol_rate =Chip_rate/SF Bit_rate =Symbol_rate*2 DPCCH overhead User_bit_rate =Channel_bit_rate/2 Spreading factor Channel symbol rate (kbps) Channel bit rate (kbps) DPDCH channel bit rate range (kbps) Maximum user data rate with ½- rate coding (approx.) kbps kbps kbps kbps kbps kbps kbps kbps 4, with 3 parallel codes Mbps Half rate speech Full rate speech 144 kbps 384 kbps 2 Mbps slide 53 Example Downlink Speech + Signalling 40 ms Signaling 3.4 kbps with 40 ms interleaving Speech 12.2 kbps Speech 12.2 kbps Radio frame Radio frame Radio frame Radio frame 10 ms AMR 12.2 kbps 81 class A bits 12 CRC 8 tail class B bits 8 tail + 60 class C bits 8 tail DPCH 3.4 kbps 136 data bits RLC+ MAC 12 bits 24 CRC slide 54 27

28 Channel coding AMR Class A 1/3 rate conv AMR Class B 1/3 rate conv AMR Class C 1/2 rate conv DPCH SF=256 SF=128 1/3 rate conv Downlink L1 bit rates Spreading factor Bits per frame 240 bits 510 bits Example Channel coding Bits per 40 ms 960 bits 2040 bits Bits per 20 ms Bits per 40 ms AMR 12.2 kbps 772 bits 1544 bits DPCH 3.4 kbps bits Rate matching Transport channel multiplexing AMR 12.2 AMR12.2+DPCH 1544 bits 2060 bits Most suitable spreading factors and required rate matching AMR 12.2 AMR12.2+DPCH SF=128 SF=128 32% repetition 1% puncturing slide 55 Speech, full rate 128 channels Number of codes with spreading factor of 128 (AMR 12.2 kbps *(128 4)/128 Common channel overhead and 10.2 kbps) /1.2 Soft handover overhead = 103 channels Speech, half rate 2*103 channels Spreading factor of 256 (AMR 7.95 kbps) Downlink Capacity = 206 channels Packet data 3.84e6 Chip rate *(128 4)/128 Common channel overhead /1.2 Soft handover overhead *2 QPSK modulation *0.9 DPCCH overhead /3 1/3 rate channel coding /(1 0.3) 30% puncturing = 2.65 Mbps Note: usually interference limits the capacity before the number of orthogonal codes 4 channels with SF=128 for common channels assumed 20% soft handover overhead assumed Result 103 speech channels or 2.65 Mbps data with one scrambling code slide 56 28

29 Basic Procedures Common channels and synchronisation slide 57 Additional Downlink Physical Channels These channels do not carry transport channels but are needed for network operation SCH, CPICH, AICH, PICH Synchronization channel SCH For the mobile to synchronize to the cell. Common pilot channel CPICH Acquisition indicator channel AICH Paging indicator channel PICH slide 58 For the mobile synchronization, channel estimation, and for the neighbor cell measurements Response to RACH preamble For indicating to the mobile that there is paging on PCH 29

30 Common channels (I) Common pilot channel (CPICH): Purpose of run-time synchronization between the BS and UE's located in the cell. CPICH unmodulated, scrambled by cellspecific primary scrambling code (SF=256). Used for initial synchronization, channel estimation and measurements for handover and cell selection. With multiple BS antennas (antenna diversity), CPICH's from each BS antenna are separated by simple modulation patterns. slide 59 CPICH CPICH is unmodulated signal under the cell specific scrambling code Other cell measurements Channel estimation CPICH is transmitted continuously and it takes typically 5-15% of the base station max power (IS-95 typically 20-25%, narrowband => relatively higher overhead) CPICH is used for downlink channel estimation in the mobile for coherent combining of multipath components slide 60 30

31 Common channels (II) Synchronization channel (SCH): Purpose of initial synchronization between the BS and UE's located in the cell. SCH is used for cell search. It consists of primary and secondary synchronization channels. The primary channel uses a 256-chip spreading sequence which is identical for every cell (global). Secondary channels use sequences individual to each group of cells and which identify one out of 64 possible scrambling code groups. Once the UE has found the secondary SCH it has obtained both frame and slot synchronization. (determined by the sequence used on the secondary SCH channel) slide 61 SCH 256-chip sequence the same in every cell Primary SCH chip sequence modulated, identifies the code group of the cell Secondary SCH chips slide =2304 chips 31

32 Cell Search 512 scrambling codes in downlink are divided into 64 groups to speed up the cell search, each group contains 8 codes (8 x 64 = 512) Step 1 Which channel is used Primary SCH Which part of synchronization is obtained (1) Chip synchronization (2) Symbol synchronization (3) Slot synchronization Step 2 Secondary SCH (1) Code group (which of 64) (2) Frame synchronization Step 3 Pilot channel CPICH Exact scrambling code (which of 8) Note: SCH is not under the cell specific scrambling code because it must be received before knowing the scrambling code As a consequence, SCH is non-orthogonal to other channels All other downlink channels are under the scrambling code slide 63 Synchronization ML approach: Correlate with the PN sequence at all delays within the uncertainty region, and then determine the delay τ. Thus, the mean synchronization time equals KLT, where T is the correlation time and K is the number of correlations per chip interval. Serial search: Correlate with the PN sequence at one delay and determine if the output is above the noise+mai floor. If the output is below the the noise+mai floor, then move the correlator to the next delay. Here the mean synchronization time is less than KLT. slide 64 32

33 Serial Acquisition Scheme slide 65 Probability of Detection and False Alarm slide 66 33

34 Dual-Dwell Serial Search slide 67 Tracking of PN- Sequences After coarse synchronization is obtained within +/- one chip, a more accurate synchronization is initiated (tracking). Tracking of the received PN-sequence is performed separately for each RAKE finger. Tracking is performed continuously during the transmission since there is a time-varying drift between the received and locally generated PN-sequence. The time-drift is mainly caused by two factors: Movement of mobile unit. At a speed of 100km/h, the time drift is on the order of 100nsec/sec. Oscillator drift between Tx and Rx. slide 68 34

35 Early-Late Gate Tracking The early-late gate algorithm aims at maximising the autocorrelation between the received and the locally generated PNsequence. The tracking algorithm is a simple gradient search algorithm The two power estimates can be obtained from the pilot signal/symbol. slide 69 Tracking Uncertainty Deterministic uncertainty due to filtering slide 70 35

36 TDD Mode In brief! slide 71 General Characteristics Combined TDMA/CDMA (TDD) multiple access Allows operation in unpaired band Requires synchronization between base stations to avoid uplink/downlink interference Allows for assymmetric uplink/downlink capacity Discontinuous transmission leads to power disadvantage cell range reduction Has the advantage of a reciprocal channel used for (open loop) uplink power control slide 72 36

37 UE WCDMA TDD Node B Single-user detection orthogonal codes non-orthogonal codes Multi-user detection (Code) Power UE 4 UE 3 UE 2 UE 1 UE 2 UE 1 frame n frame n+1 slide 73 Available resources: Spreading Codes (OVSF) and Slots Up to 16 users code multiplexed per slot Time Generalized TDD Frame Data symbols (976 chips) Data symbols (976 chips) Guard (96 chips) Data symbols (976 chips) Midamble (512 chips) Data symbols (976 chips) Guard (96 chips) 2560 chips TS 0 TS ms # allocated codes (SF=16) 1 Number of allocated time slots kbps 48.8 kbps 158 kbps Midamble (training sequence) for joint channel estimation kbps 195 kbps 390 kbps 781 kbps 1.26 Mbps 2.54 Mbps Burst Type I slide 74 37

38 MultiUser Detection Interference Cancellation slide 75 MUD analysis If we define the Interference Cancellation receiver efficiency, β, as the ratio between the equivalent intra-cell interference after and before interference cancellation [Hämäläinen], then the required (matched filter) SINR (E b /N o ) of user j (per antenna) can be expressed as γ W Pj j = Rj ( Pown Pj )( 1 β ) + Pother + Pnoise where W is the chip rate, R j the selected data rate for transmission, P j the total receiver power (per antenna), P own the total received own-cell power (per antenna), P other the total received other-cell power (per antenna), and P noise is the background noise power (per antenna). A practical IC implementation (with acceptable complexity) can achieve an efficiency of 30%, whereas about optimum for multi-stage IC achieves 70% efficiency slide 76 38

39 The gain from IC can be approximated as: i G i η UL IC Gain i is the other-to-own interference ratio β is the efficiency of the IC receiver η UL is the uplink fractional load The cell throughput gain from IC decreases with i since IC is only effective towards intra-cell interference. Also, the impact of the IC efficiency β on the cell throughput gain is scaled by the uplink fractional load η UL. β from [ C. Rosa, Enhanced plink Packet Access in WCDMA, Ph.D. dissertation, AAU, December 2004] η UL slide 77 Cell throughput gain [%] β = 0.3 Cell throughput gain [%] β = η UL i = 0.0 i = 0.2 i = 0.4 i = 0.6 i = 0.8 i = 1.0 i = 0.0 i = 0.2 i = 0.4 i = 0.6 i = 0.8 i = % 54 % 27% 100% 3 5 Noise Rise and fractional load Wideband received power based RRM η UL PN = 1 = I total Noise rise -1 Noise rise Throughput based RRM Measure total wideband received power Itotal η UL = N Calculate sum of the bit rates in a cell ( 1+ i) L = ( 1+ i) j j= 1 j= 1 N 1+ ( E / N ) b 1 W R 0 j j υ j slide 78 39

40 References and Acronyms slide 79 References H. Holma and A. Toskala, WCDMA for UMTS Radio Access for Third Generation Mobile Communications, John Wiley & Sons, 3rd edition, 2004 (HSDPA chapter!) T.E. Kolding et al., High Speed Downlink Packet Access: WCDMA Evolution, IEEE Vehicular Technology Society (VTS) News, vol. 50, no. 1, pp. 4-10, February 2003 S. Hämäläinen, H. Holma, and A. Toskala, Capacity Evaluation of a Cellular CDMA Uplink with Multiuser Detection, International Symposium on Spread Spectrum Techniques and Applications, vol. 1, pp , September 1996 C. Rosa, T.B. Sørensen, J. Wigard, and P.E. Mogensen, Interference Cancellation and 4-Branch Antenna Diversity for WCDMA Uplink Packet Access, Proceedings of VTC Spring 2005, Stockholm, Sweden, May-June 2005 B. Vejlgaard, Data Receiver for the Universal Mobile Telecommunications System (UMTS), Ph.D dissertation, AAU, 2000 slide 80 40

41 Acronyms 3GPP 3rd Generation Partnership Project AC Admission Control AuC Authentication Centre BSS Base Station Subsystem BTS Base Transceiver Station CDMA Code Division Multiple Access CN Core Network CS Circuit Switched DL Downlink (broadcast) EUTRA Evolved UMTS Terrestrial Radio Access FDD Frequency Division Duplexing FDMA Frequency Division Multiple Access GERAN GSM Evolved Radio Access Network GGSN Gateway GPRS Support Node GPRS General Packet Radio Service GSM Global System for Mobile communications HC Handover Control HLR Home Location Register HSS Home Subscriber Services HSxPA High Speed Downlink/Uplink Packet Access IMS Internet Multimedia Subsystem IMT International Mobile Telephony (ITU-2000) ITU International Telecommunications Union LTE Long Term Evolution LC Load Control ME Mobile Equipment MS Mobile Station slide 81 Acronyms (cont.) MSC Mobile Switching Centre PLMN Public Land Mobile Network PS Packet Switched QoS Quality of Service PC Power Control PS Packet Scheduler RM Resource Manager RNC Radio Network Controller RNS Radio Network Subsystem RRM Radio Resource Management RTT Round Trip Time SF Spreading Factor SGSN Serving GPRS Support Node SHO Soft Handover SIP Session Initiation Protocol SS7 Signalling System 7 TDD Time Division Duplexing TDMA Time Division Multiple Access TMSI Temporary Mobile Subscriber Identity UE User Equipment UL Uplink (multiple access) UMTS Universal Mobile Telecommunications System USIM UE Subscriber Identification Module UTRAN UMTS Terrestrial Radio Access Network VLR Visitor Location Register VSF Variable Spreading Factor WCDMA Wideband Code Division Multiple Access slide 82 41