WCDMA Basics Chapter 2 OBJECTIVES:

Transcription

1 WCDMA Basics Chapter 2 This chapter is designed to give the students a brief review of the WCDMA basics of the WCDMA Experimental System. This is meant as a review only as the WCDMA basics have already been covered in the WCDMA Overview course, LZU , which is a prerequisite to this course. OBJECTIVES: Upon completion of this chapter the student will be able to: Have access to reference information on the multiple access technologies TDMA, FDMA and CDMA Have access to reference information on spreading and coding in the WCDMA Experimental System Have access to reference information on DS-CDMA aspects such as:- power control, cell breathing, RAKE receiver and handover

2 WCDMA Experimental System Survey Intent ionally B l a n k EN/LZT R2A

3 2 WCDMA Basics 2 WCDMA Basics Table of Contents Topic Page MULTIPLE ACCESS TECHNOLOGIES OVERVIEW...17 TRANSMISSION FDMA TDMA CDMA SPREADING...22 SPREADING PRINCIPLES DS-CDMA ASPECTS...29 POWER CONTROL CELL BREATHING RAKE RECEIVER HANDOVER (HARD, SOFT, SOFTER) SUMMARY DESCRIPTION OF THE WCDMA EXPERIMENTAL SYSTEM ACCESS TECHNIQUE...33 EN/LZT R2A i

4 WCDMA Experimental System Survey Intent ionally B l a n k ii EN/LZT R2A

5 2 WCDMA Basics MULTIPLE ACCESS TECHNOLOGIES OVERVIEW TRANSMISSION As an introduction to the radio access technology used in the WCDMA Experimental System, that is DS-CDMA (Direct Sequence Code Division Multiple Access), a brief overview of transmission and the following radio access technologies will be given: Transmission Frequency Division Multiple Access (FDMA) Time Division Multiple Access (TDMA) Code Division Multiple Access (CDMA) in general and Direct Sequence CDMA (DS-CDMA) in particular Before introducing the multiple access methods as listed above, transmission will be discussed briefly. Modern digital mobile communication systems are full-duplex capable. Bi-directional communication, that is (quasi-) simultaneous transmission and receiving is possible. Transmission can be in the Uplink (UL) direction and the Downlink (DL) direction. The UL is the transmission direction from the Mobile Station (MS) to the Base Transceiver Station (BTS). The DL designates the transmission direction from the BTS to the MS. In the United States this is usually called the reverse and forward direction. There are two different duplex transmission principles, i.e. Frequency Division Duplex (FDD) and Time Division Duplex (TDD). In FDD, the UL and DL transmission takes place in different frequency bands which are separated from each other by a duplex distance. In TDD, the UL and DL transmission are implemented in the same frequency band and this band is divided into Time Slots (TSs). The UL and DL are separated by time and occur in different TSs. In the following discussion on the access methods we will only discuss the FDD duplex technique. EN/LZT R2A 17

6 WCDMA Experimental System Survey FDMA Frequency Division Multiple Access (FDMA) is very common in the first generation of mobile communication systems, that is analogue systems. The available spectrum in FDMA is divided into physical channels of equal bandwidth. One physical channel is allocated per subscriber. The physical channel required for the transmission of one subscriber s information uses FDD. In pure FDMA systems different speech/data/signalling transmissions can take place (per subscriber) at the same time on different frequencies. The physical channel allocated to the subscriber is used during the entire duration of the call and is unavailable for use by another subscriber during that time. The physical channel is released at the end of the call where it is then made available for the next subscriber who needs to make a call. In summary, in FDMA narrow bandwidth is used as is continuous transmission and reception. There is orthogonality in frequency within the cell, that is different users use different frequencies in the cell. See Figure 2-1. Frequency Division Multiple Access (FDMA) Orthogonal in frequency within cell Narrow bandwidth per carrier Continuous transmission and reception No synchronisation in time t Power Figure 2-1 Frequency Division Multiple Access (FDMA) f 18 EN/LZT R2A

7 2 WCDMA Basics TDMA Note: When discussing Time Division Multiple Access (TDMA) in this section, it is the access technique that is discussed and not the standard. TDMA is used in some digital systems and is normally used in combination with the FDMA method described above, for example GSM and TDMA/D-AMPS) which uses FDD. In TDMA, the available spectrum is divided in time into Time Slots (TSs). The subscriber is allocated a TS and only that TS can be used during the time that is assigned to that subscriber. A physical channel in TDMA is defined as one TS and the subscriber has cyclical access to it. The subscriber information (speech/ data/ signalling) is divided up and transmitted, bit by bit, via the assigned TS. The high frequency transmission of each TS is called a burst. A TS is typically in the order of a millisecond. TDMA requires strict timing of the burst transmission in order to avoid overlapping of adjacent TSs. The time delay caused by the transmission of bursts (which have the finite speed of light) is a problem in cellular systems with large cells. A very precise synchronization between the MS and the BTS is required. Timing Advance information and Guard Periods between adjacent time slots prevent interference of bursts of adjacent TSs. In summary, in TDMA there is syncronisation, increased bandwidth and increased peak power. The transmission and reception is discontinuous. There is also orthogonality in time within the cell, that is different users use the spectrum at different times. See Figure 2-2. Time Division Multiple Access (TDMA) Orthogonal in time within cell Increased bandwidth per carrier Discontinuous transmission and reception Increased peak power Synchronisation in time t Power Figure 2-2 Time Division Multiple Access (TDMA) f EN/LZT R2A 19

8 WCDMA Experimental System Survey CDMA Code Division Multiple Access (CDMA) is used in some digital systems whereby many subscribers use the same frequency at the same time. As all subscribers are using the same frequency at the same time within the cell, there needs to be a way for each subscriber to find the information that is destined for him/her. In Direct Sequence CDMA (DS-CDMA), uniquely identifying information for a particular subscriber is achieved by spreading the information. By spreading the information we mean we multiply the information with short and long codes. Spreading and short and long codes will be explained in more detail in the section on spreading below. The term WCDMA used in the WCDMA Experimental System means that CDMA is used as an access method. CDMA is also known as a Spread Spectrum Technology. Spread Spectrum Technologies have been used since the 1940s, so it is not a new technology. In the 1940s it was mainly used by the military so that the radio communication between different military groups could not be monitored or jammed by outsiders. More information on the usage of this technology by the military can be found in Appendix A. CDMA can be further divided into Time Hopping CDMA (TH- CDMA), Frequency Hopping CDMA (FH-CDMA) and Direct Sequence CDMA DS-CDMA. The main difference between these CDMA methods is in the modulation techniques used to generate the spread-spectrum signal. Modulation will be discussed in more detail under the section QPSK below. The difference between TH-CDMA and FH-CDMA is that in TH-CDMA the information-bearing signal is not transmitted in a continuous manner, but rather in short bursts and the time of the short bursts is decided by the spreading codes. In FH- CDMA, depending on the spreading codes used, the carrier frequency at which the information-bearing signal is transmitted is changed rapidly. TH-DCMA and FH-CDMA are not used in the WCDMA Experimental System. As only DS-CDMA is used the Experimental System, only this technology and aspects relating to this technology will be discussed further in the following sections. 20 EN/LZT R2A

9 2 WCDMA Basics Direct Sequence Code Division Multiple Access (DS-CDMA) Separates users through different codes Large bandwidth Continuous transmission and reception t Power Figure 2-3 Direct Sequence Code Division Multiple Access (DS -CDMA) As transmission and multiple access techniques have been briefly discussed, an explanation of spreading (giving the advantages and disadvantages) in the WCDMA Experimental System is given in the following section. f EN/LZT R2A 21

10 WCDMA Experimental System Survey SPREADING By using the spread spectrum technology in the WCDMA Experimental System, the advantages of using such as system are gained. In summary some of the many advantages are as follows: The wideband transmission has the advantage of being less sensitive to frequency selective interference and fading. The power density of the spectrum is decreased several times and the transfer of information is still possible even below background noise. Radio Network planning is much easier compared to networks using FDMA or TDMA as all the available frequencies can be used in all cells. CDMA is very spectrum efficient due to the possibility of using each carrier in each cell. There is no fixed capacity limit (number of users at the same time). The main limit is the increase in the level of background noise from other subscribers and this reduces the quality of service. The possibility of having soft handover in the WCDMA Experimental System is an advantage. Soft handover is explained in more detail in the section on Handover. Some of the disadvantages associated with WCDMA are: The power levels of all MS transmissions received at the BTS must be equal if the bit rates are equal and therefore fast power control is necessary. This will be discussed in further detail in the DS-CDMA concepts section. As MSs in soft handover mode require the resources of more than one cell, the system capacity is reduced. SPREADING PRINCIPLES Spreading in the WCDMA Experimental System involves the use of short and long codes. In this section the following will be discussed: Spreading code groups Modulation Spreading with short and long codes 22 EN/LZT R2A

11 2 WCDMA Basics Spreading Code Groups Spreading codes can be divided into two groups as follows: Orthogonal Codes - Orthogonal codes are typically used for channel separation because there is a minimum of interference between different users if they are using an orthogonal code. An example of orthogonal codes are the short codes called Walsh Codes. More information on Walsh coding is given below. Pseudo Noise (PN) Codes - These are generated by a shift register with a feedback arrangement. In Direct Sequence systems PN codes are generated using linear or shift registers. Codes can be divided into a number of PN code families such as Gold codes, Gold like codes, S-Kasami codes, L-Kasami codes, VL-Kasami codes and so on. The long codes used in the WCDMA Experimental System are the long codes of Gold type. More information on Gold coding is given below. Spreading of the subscriber information is achieved by multiplying the transmitted information with short and long codes. On the DL, synchronisation is easy (short codes are only orthogonal if synchronised). On the UL it is not sufficiently easy from an implementation point of view, to synchronise, so different long codes are required. Modulation More information on short and long codes is given after the following section on modulation. Quadrature Phase Shift Keying (QPSK) is the modulation method used in the WCDMA Experimental System. However there are several modulation techniques, of which Binary Phase Shift Keying (BSPK) is another example. As only QPSK is used in the WCDMA Experimental System, that is the modulation scheme which will be discussed in this section. The simplest method of producing QPSK is when two phase modulators are operated in parallel as shown in Figure 2-4. EN/LZT R2A 23

12 WCDMA Experimental System Survey X(t) +/- 180 Phase modulator Y X Oscillator f c f c Phase divider X=1 270 sin ω ct cos ω ct X=0 90 Σ S(t) = X(t)sin ( ω c +/- 180) + Y(t)cos ( ω c +/- 180) Y(t) Y=1 180 Y=0 0 +/- 180 Phase modulator Figure 2-4 Block diagram of a QPSK modulator X= Y=1 Y= X=1 The effect of phase shifts of a multiple of 90 can be achieved by modulating every second bit with cos ω c t and the next with sin ω c t. In this way there are two streams of digital data, X(t) and Y(t) which are modulated in parallel and the sum of the results produces a multiple of 90. See the vector diagram in Figure 2-4 above. The result of the modulation is that the bit data rate is double the modulation rate. The two bits of each modulation state can be in any of the four phases, i.e. 45, 135, 225 and 315. This is why this type of modulation is referred to as Quadrature Phase Shift Keying. In the WCDMA Experimental System spreading is done after splitting the bits of information into two branches. The diagram below shows QPSK modulation of data. The spreading is the multiplication of the information by short and long codes and this can be done simultaneously, or by the short code first and then the long code. More details on spreading are given after Figure 2-5 below. 24 EN/LZT R2A

13 2 WCDMA Basics User Information 1 Spreading x with short and long codes Modulation Mchips/s Bits 1 with short and long codes x Figure 2-5 Diagram showing information bits, spreading of the information and QPSK modulation in the WCDMA Experimental System Spreading with Short and Long Codes In advance of outlining the process of spreading, some basic terms will be reviewed as follows: A bit of information is a 1 or a 0 or a -1, +1 The bit rate is the rate at which the subscriber is transmitting the bits of information i.e. the 1s and the 0s. Subscribers can be transmitting at variable bit rates on the same frequency at the same time in the WCDMA Experimental System, but ultimately the information is transmitted over the air at a constant chip rate. A symbol of information in the WCDMA Experimental System is two bits of information, as QPSK modulation is used. The symbol rate is thus half the bit rate. The Spreading Factor (SF) is the ratio between the chip rate and the symbol rate. This is equal to the spreading gain (i.e. the protection against interference) and it is achieved by spreading the spectrum. The combination of a chip rate of Mchips/s and the modulation scheme (which is QPSK) give a bandwidth of about 5 MHz. EN/LZT R2A 25

14 WCDMA Experimental System Survey When information from various subscribers is sent over the air this information can be sent at different bit rates depending on what service the subscriber is using at the time, that is speech/video conferencing and so on. 1. If subscriber A transmits data at a symbol rate of 128 ksps the Spreading Factor (SF) becomes If subscriber B is transmitting data at a bit rate of 64 ksps the SF becomes 64. The following is valid for QPSK modulation: Symbol Rate* Bit Rate Multiply by SF Processing Gain (db) Chip Rate 256 ksps 512 kbps Mchips/s 128 ksps 256 kbps Mchips/s 64 ksps 128 kbps Mchips/s 32 ksps 64 kbps Mchips/s 16 ksps 32 kbps Mchips/s ksps =Kilosymbols per second kbps=kilobits per second Mchips/s=Megachips per second Note: In the example above 16 ksps is the lowest rate shown as this is the lowest symbol rate chosen for use in the WCDMA Experimental System. *Symbol Rate is 2 bits of information. 26 EN/LZT R2A

15 2 WCDMA Basics Short and Long Codes Long codes are allocated in UL and DL from Gold sequences of different lengths. All BTSs and all MSs have unique long codes. The short codes are allocated from layered orthogonal Walsh codes. Each code is used to distinguish an individual channel. An allocation method is used that does not block codes in the code tree unnecessarily. Depending on the SF used a certain Walsh code will be picked from the Walsh tree as seen in Figure 2-6 below. c 1 = {1} c 2,1 = {1 1} c 2,2 = {1-1} c 4,1 = { } c 4,2 = { } c 4,3 = { } c 4,4 = { } SF = 1 SF = 2 SF = 4 Figure 2-6 Diagram showing Spreading Factor 1, 2 and 4 and how the Walsh tree is built. The primary function of the long codes in the WCDMA Experimental System is to distinguish between all BTSs and all MSs. Good long codes should have low out-of phase autocorrelation peaks to maximize the probability of correct synchronization. The long codes should also have low crosscorrelation peaks in order to minimize the interference between different BTSs and different MSs. The short codes are used for channelization, i.e. to separate different logical channels that are transmitted using the same long code. These short codes are mutually orthogonal which means that it is theoretically possible to separate the logical channels at the receiver. Unfortunately, due to multipath propagation it is not possible to do this fully. Figure 2-7 shows the generation of PN sequences (long codes). EN/LZT R2A 27

16 WCDMA Experimental System Survey Maximum-length (ML) and Pseudo-noise (pn) shift-register sequences n A shift register with n stages can assume (2-1) states (excluding the "0" state) A shift register with a suitable feedback passes all states and generates a periodic ML sequence with a period of 2 n -1. Example: n=4 Modulus addition: 0 + 0=0, 0 + 1=1, 1 + 0=1, 1 + 1=0, Modulus-2 + addition Binary p-n sequence Binary sequence Period= M=4-1=15 Bipolar sequence Characteristics a) Bipolar ML sequence ("1" -1, "0" +1) has an autocorrelation function in the form of a pulse train (see Fig. B1-2). This corresponds to an almost white spectrum. b) The sequence contains a random train of "0" and "1"s. (e.g. Number of "1"s minus number of "0"s =1) (a) and b) explain the term "pseudo-noise sequence".) c) Modulus-2 addition of two ML sequences of the same classproduces another ML sequence of the same class. (Of the same class means from the same shift registerarrangement but with different starting states.) Figure 2-7 Generating binary maximum length PN sequences (long codes) Long and short codes are combined in the WCDMA Experimental System and are thereafter used to spread the user data. See Figure 2-5. The rate of both the long and short code generators is Mchips/s while the rate of the data varies between 16 ksps and 256 ksps. The short code generator is restarted after each data symbol (i.e. the period of the short code is always adapted to the user data rate) but the period of the long code generator is longer. For example on traffic channels DL, the long code is repeated every radio frame (i.e. every 10 ms). 28 EN/LZT R2A

17 2 WCDMA Basics DS-CDMA ASPECTS POWER CONTROL As modulation (QPSK), and spreading (with short and long codes) have been explained, the following basic aspects of Direct Sequence Code Division Multiple Access (DS-CDMA) are discussed below: Power Control Cell Breathing RAKE Receiver Handover (Hard, Soft, and Softer) Power control is the most important element in DS-CDMA. Because many users access and use the same frequency and bandwidth at the same time, there is a high possibility of interference between the users. In the case where there is no power control, if an MS is close to the BTS the signal could be stronger from that MS than from the MS which is furthest from the BTS. This is known as the near-far problem. In order to maintain good capacity levels in the network, the signals received by the BTS, no matter where the MSs are transmitting from (that is near or far) should be of equal power assuming that all MSs are transmitting at the same user bit rate. There are three types of power control: Open Loop power control Closed Loop power control and Outer/Quality Loop power control Open Loop power control is used to compensate for near-far problems and shadow fading effects. Closed Loop power control is used to compensate for multipath fading, near far problems and shadow effects. Outer/Quality Loop power control is used to set the target quality level for the Closed Loop power control. As already mentioned, in CDMA many subscribers use the same frequency channel at the same time, but there is however a threshold level which shouldn t be exceeded. That is the number EN/LZT R2A 29

18 WCDMA Experimental System Survey CELL BREATHING RAKE RECEIVER of subscribers that can be added should not exceed a certain level. If this level is exceeded then other subscribers will be affected by excessive interference. The threshold level is decided and set by the operator. If the threshold level is very high then the service quality offered to the subscriber will be lower. What is required is a sufficiently high threshold level to make efficient use of the network, but also a level which will provide good service quality to the subscribers In comparison to a traditional TDMA system the coverage of (W)CDMA depends on the traffic load in the cells. The more traffic, the more interference and the shorter must the distance be, between the BTS and the MS. In a system where the traffic load changes this will cause the cells to grow and shrink with time. This effect is often referred to as cell breathing. In the DL all connections on a certain carrier share the same power amplifier. If at one moment the load is low, a particular MS will have the possibility to connect to the BTS even if it is very far away from it. On the other hand, if the traffic load is high, the MS will not be able to connect unless it is close to the BTS. This effect makes it somewhat difficult to use the term coverage for the DL. The plain receiver sensitivity depends on the required Carrier to Noise (C/N) ratio. However, the received Carrier (C) power must be large enough to combat both Noise (I) and Interference (I), that is the C/(N+I) must exceed the receiver threshold. In order to get an accurate coverage prediction in a busy system, a margin accounting for the noise rise on the UL is needed, as the interference increases with system load. The purpose of the RAKE receiver is to rake together multiple replicas of the same signal received from the MS in order to receive a stronger, combined signal. When the signal is sent from the MS it hits off different objects and these objects reflect the signal towards the receiving antenna at the BTS. At the antenna, the same signal arrives from the MS with different delays due to the differences in path distance as shown in the Figure 2-8 below. The RAKE needs to find out what the delays for the strongest replicas of the signal are. 30 EN/LZT R2A

19 2 WCDMA Basics h(t) Delay Power Spectrum τ 1 τ 2 τ 3 τ 4 τ 5 Figure 2-8 Multipath delays in the RAKE Receiver The delays are then used to adjust the code sequence as seen in the Figure 2-9. Summation of all RAKE finger contributors Threshold Detector Amplification Code Sequence Generator τ 1 τ 2 Multiplication symbol Weighting symbol. Depending on the combining method, this tap has different values τ 3 Phase compensation of local oscillator Figure 2-9 RAKE Receiver with three fingers HANDOVER (HARD, SOFT, SOFTER) The main reasons for handover are connectivity and mode changes. Due to the movements of the MS, radio links and connections need to be changed from one or several sectors or BTSs to another or several others, without dropping the call. Regarding mode changes, a change of a connection from a common channel to a dedicated channel and vice versa is required. In soft handover an MS is connected to two or more BTSs at the same time. In softer handover an MS is connected to two or more sectors belonging to the same BTS. EN/LZT R2A 31

20 WCDMA Experimental System Survey The purpose of soft and softer handover is to make use of macrodiversity. Macrodiversity is a powerful method of combating radio fading, and the soft/softer handover makes it possible to take advantage of macrodiversity especially during mobility. The purpose of hard handover is the relocation of DL shortcodes (non-interrupted hard handover) or to realize a new active set of logical sectors (re-synchronization hard handover). The main reason for re-synchronization hard handover is to perform an interfrequency handover. Hard handover is for the relocation of DL short codes and for interfrequency handover. Soft handover is for intrafrequency handover and softer handover is for intra frequency, intersector handover. 32 EN/LZT R2A

21 2 WCDMA Basics SUMMARY DESCRIPTION OF THE WCDMA EXPERIMENTAL SYSTEM ACCESS TECHNIQUE A summary of the WCDMA Experimental System access technique is given below: A cell is a geographical area served by one BTS. The cell is further divided into sectors. A sector is a geographical area served by one sector antenna. The sector antenna is used for transmission and reception. In the WCDMA Experimental System the cell can be divided into a maximum of six sectors. The spreading codes used are both short and long. A cell has a cell specific long code. Sectors in the cell are separated by using the long codes. Each physical channel in a specific sector has its own short code. The carrier symbol rate in the WCDMA Experimental System is 16, 32, 64, 128 or 256 ksymbols per second. One modulated symbol in the WCDMA Experimental System is 2 bits of information. The modulation method used is QPSK. The main reasons for handover are connectivity and mode changes. The access method used in the Experimental System is DS-CDMA with a chiprate of Mchips/sec. EN/LZT R2A 33

22 WCDMA Experimental System Survey Intent ionally B l a n k 34 EN/LZT R2A