• 首页
• HFSS教程
• ADS教程
• CST教程
• 射频工程师培训
• 微波仪器操作培训
• 天线设计培训
• 射频专业基础课

Ansoft Designer / Ansys Designer 在线帮助文档：

System Simulator >
System Component Models >
WCDMA Rake Receiver >
   Matched Filter (MATCHFILTER)       

Matched Filter (MATCHFILTER)

Limits

Notes

1. The matched filter model can be used to resolve the received multi-path signal and despread symbols on each path in DS/CDMA systems. The block diagram of the matched filter is shown in Fig.1. If Method is set to Pilot Assisted {0}, the received signal is despread using the esti­mated multi-path delays (i.e., the output of the Multi-Path Search). If Method is set to Perfect {1}, the received signal is despread using the given parameter values D₀ ~ D_L-1.

Fig. 1 Block diagram of matched filter

2. Slot Structure
The slot structure is shown in Fig.2. Each slot comprises N_p pilot symbols and N_d data sym­bols. N₀ is the number of the first pilot symbol, i.e, the N₀ ~ (N₀ + N_p -1) symbols are pilots.

Fig.2 Slot structure

3. Sample Selection: Because of channel delay and shaping filter, the optimum sample position should be found to maximize the output power and minimize inter-symbol interference. The input signal is a discrete one sampled at the rate of 1/T_s in simulation. To determine the opti­mum sample position is to select an optimum sample from the S samples during a chip. The optimum sample position can be determined by comparing correlation values between the received signal starting at each sample point and the corresponding spreading code.

Assume complex vector x [ ] stores the received signal sampled at the rate of 1/T_s. Thus the correlation value between the received signal and the spreading code is given by


(1)

4. where G is the spreading factor and S is the number of samples per chip and the vector c [ ] stores the corresponding spreading code. The factor 1/2 is used to remove changes of symbol power caused by spreading and despreading. When the jth sample point of input signal hits the first chip of that symbol (which is spread by the spreading code c [ ]), the magnitude of the cor­relation value will be the greatest among these samples.

In this model, there are (D_max+1) correlation values, where D_max is the possible maximum path delay in terms of samples. Once the (D_max+1) correlation values are calculated, the sam­ple position where the magnitude of the correlation value is the greatest is selected. If this sam­ple is the jth sample, then j mod S becomes the optimum sample position (S_OPT..

Because all N_F pilot symbols in a slot are known for the matched filter in the receiver, their correlation values are added to increase processing gain. However, pilot symbols are not the same in a slot, these correlation values must be divided by the corresponding pilot symbol val­ues before the addition. So (D_max+1) values are produced. If the jth one has the greatest mag­nitude, then the (J mod S) sample position for each chip is selected as the optimum sample position for the current slot.

In a practical situation, it is very difficult to search, slot by slot. This model allows us to use the block-average power delay profile (K>1) ^[3]. First, the instantaneous power delay profile is measured by using pilot symbols belonging to each slot and, then, average them over multiple slots.

5. Multi-Path Search: The transmitted signal arrives at the receiver via different paths and delays. To use more signal power, multi-path delays of the received signal of the desired user are determined and signals on each resolved paths are combined. Because it is difficult to deter­mine the relative multi-path delays at the precision of T_S, multi-path delays are determined at the precision of T_c, where D_c = T_c x S is chip duration. Let D_c be the maximum delay in terms of chips, i.e., D_c equals the largest integer which is not larger than D_c/S, can be expressed as

(2)

6. After the optimum sample position D_OPT has been determined by Sample Selection, the corre­lation values between the received signal at each possible delay, in terms of chips, are calcu­lated as follows:


(3)

7. As in Sample Selection, the correlation values of different pilot symbols on the same path are divided by the corresponding pilot symbol values and added. D_c + 1 results of additions are obtained. In these results, L with the greatest magnitude are selected, the corresponding paths that belong to L results are determined as L valid paths. The Multi-Path Search process can also use the block-average power delay profile (K>1) ^[3].

8. Despread: After multi-path delays are obtained by the Multi-Path Search, symbols on each path in current slot are despread. If Method is set to Perfect {1}, symbols on each path in cur­rent slot are despread using the given parameter values D₀~D_L-1, rather than the multi-path delays are obtained by the Multi-Path Search. The despreading process is the same as the pro­cess of calculating correlation value. Finally, L samples are outputted successively for each symbol.

Netlist Form

MATCHFILTER:NAME n1 n2 n3 n4 [METHOD=val] L=val G=val [S=val] [K=val] NP=val ND=val + [N0=val] [DMAX=val] [D0=val,…, D15=val] [RIN1=val] [RIN2=val] [RIN3=val] [ROUT=val]

Netlist Example

MATCHFILTER:1 1 2 3 4 METHOD = 0 L =2 G = 32 S = 4 DMAX = 20 NP = 4 ND = 16 D0 = 1 D1 = 9

References

1. J. G. Proakis, Digital Communications, McGraw-Hill, 2001.

2. A. J. Viterbi, CDMA: Principles of Spread Spectrum Communication, Wesley Publishing Company, 1995.

3. K. Higuchi, H. Andoh, et al, “Experimental evaluation of combined effect of coherent RAKE combining and SIR-based fast transmit power control for reverse link of DS-CDMA mobile radio,” IEEE Journal on Selected Areas in Communications, vol. 18, No.8, pp.1526-1535, Aug. 2000.

                HFSS视频教程                                      ADS视频教程                               CST视频教程                           Ansoft Designer 中文教程

      Copyright © 2006 - 2013   微波EDA网, All Rights Reserved    业务联系：mweda@163.com