CN1549473B - Beam forming method adapted to wide band CDMA system - Google Patents

Abstract

A beamforming method applicable to a wideband code division multiple access system, comprising: performing spatial processing on an array signal, the array signal being a time-delay-aligned baseband signal X, comprising: for each frame, at the first time During the pilot bit period of the slot, the respread scrambling symbol of the known uplink channel pilot symbol is used as the reference signal r, and the cross-correlation operation is performed with the baseband signal X to obtain the cross-correlation matrix r _xr =E[Xr ^* ] , calculate the sub-optimal weight until all the pilot bits of the first time slot end; during the period of the information symbol bit and the pilot bits of other time slots, the sub-optimal weight is used as the initial value, respectively with The respread scrambling signal of despreading and descrambling information symbol bits and the respread scrambling signal of known pilot bits are used as reference signals, and iterated according to the minimum mean square error criterion to find the adaptive suboptimal weight W= r _xr ; use the adaptive sub-optimal weight W to perform beamforming to obtain Y=W×X. The invention simplifies the system structure and greatly reduces the technical difficulty and calculation amount.

Description

A Beamforming Method Applicable to Wideband Code Division Multiple Access System

technical field

The invention relates to a wideband code division multiple access system, in particular to an adaptive beam forming method of an intelligent antenna base station.

Background technique

Wideband Code Division Multiple Access (WCDMA) is the third generation of communication technology. Compared with the previous two generations of communication technology, it has better security, higher transmission rate and other performances, so it has become the trend of communication system development. Since the CDMA system is an interference-limited system, the existence of multiple access interference will seriously affect the system performance, so how to effectively suppress multiple access interference (MAI) is the key to the research of the third generation mobile communication. Taking advantage of the differences in the signal structure, space and time propagation characteristics of the useful signal and the interference signal, there are mainly two MAIN suppression technologies proposed at present: intelligent wireless and multi-user detection. And smart antenna technology is the key technology of WCDMA system, it helps to fight against interference, increase system capacity and increase cell coverage area. There are currently two implementations of smart antennas: adaptive and pre-multi-beam. Although the realization of adaptive antennas is much more complicated than that of pre-multi-beam antennas in theory and engineering practice, adaptive antennas will be better performance. In the adaptive mode, the corresponding antenna weight or beam shape can be adjusted arbitrarily according to a certain space-time two-dimensional adaptive algorithm, so as to match the current transmission environment to the greatest possible extent, and the corresponding smart antenna receiving beam can be in any direction In this way, the maximum pointing of the beam can be aimed at the desired user, and the null of the beam can be aimed at the interference user, so as to obtain higher processing gain. In the pre-multiple beams, each weight can only be selected from several pre-specified groups of values, and the working mode of the smart antenna at a certain moment can only be selected from several pre-designed beams, not arbitrary directions, so only It is possible to partially match the current transmission environment, which is not optimal from a theoretical point of view.

Therefore, the core technology in the smart antenna system is the adaptive beamforming technology. The improvement degree of the smart antenna system to the communication system mainly depends on the performance of the adaptive beamforming method. At present, various adaptive beamforming methods have been proposed. Such as the constant modulus algorithm, but this algorithm is a blind algorithm, which does not use the pilot bits specified in the WCDMA protocol, which is a waste of resources; there are other beamforming algorithms, such as code filtering and cascade training methods, in The operation process involves multiplication and inversion of large matrices, which is too complex and computationally intensive to be easily realized by hardware. At present, there is no technical solution for fundamentally solving the above problems.

Contents of the invention

The present invention is just to overcome the problems existing in the above-mentioned prior art, and proposes an adaptive beamforming method suitable for wideband code division multiple access systems, so as to reduce the amount of calculation, reduce the complexity of calculation, and improve the processing accuracy at the same time and speed.

In order to achieve the purpose of the invention, the present invention provides a beamforming method suitable for wideband code division multiple access systems, which includes the following steps:

Perform spatial processing on the array signal, where the array signal is a delay-aligned baseband signal X, including: for each frame,

计算出次最优权值，直到第一个时隙的所有导频位结束；During the pilot bit period of the first time slot, the respread scrambling symbol of the known uplink channel pilot symbol is used as the reference signal r, and the cross-correlation operation is performed with the baseband signal X to obtain the cross-correlation matrix

Calculate the suboptimal weight until the end of all pilot bits in the first time slot;

During the period of the information symbol bit and the pilot bit of other time slots, with the suboptimal weight as the initial value, the respread scrambled signal of the despread and descrambled information symbol bit and the respread scrambled signal of the known pilot bit are respectively Spread the scrambled signal as a reference signal, iterate according to the minimum mean square error criterion, and find the adaptive suboptimal weight performing beamforming with the adaptive sub-optimal weight W to obtain Y=W×X;

Among them, iterate according to the minimum mean square error criterion to find the adaptive suboptimal weight Using the adaptive suboptimal weight W to perform beamforming to obtain Y=W×X includes:

Input the weight value at the end of the pilot bit of the first time slot to the multiplier, and multiply it with the baseband signal X to obtain the beamforming signal of each array element, and pass the beamforming signal of each array element through The adder synthesizes one signal, and then input this signal to the calculation iteration error module of the respread iteration module and the descrambling module of the time domain processing module, wherein, the signal input to the calculation iteration error module of the respread iteration module is used as the calculation The subtracted vector during the error; the signal descrambling input to the descrambling module of the time domain processing module is divided into I road data and Q road data, and the Q road data enters the despreading module; Utilize channel estimation and compensation module to despread The final data is processed, and then judged, and the obtained sign bits are input to the respread iteration module;

When the current moment is the period of the information symbol bit, the symbol bit obtained after the decision is respread and scrambled and then input to the calculation iteration error module of the respread iterative module, and subtracted from the previously input adder signal to obtain the error signal , input to the calculation new weight module of the airspace beamforming module;

When the current moment is the pilot bit period of other time slots, the respread scrambling signal of the known uplink channel pilot bit is used as an input of the calculation iteration error module of the respread iterative module, which is compared with the previous input adder signal Subtract the error signal to get the error signal, which is input to the calculation new weight module of the airspace beamforming module;

Judging whether the iterative error obtained by performing beamforming with the weight output by the calculation new weight module is within the convergence threshold; if not, return to the step of performing spatial processing on the array signal; if so, save the weight as the The adaptive suboptimal weight W is used for beamforming.

The positive effect that the present invention can bring is that adopting the method provided by the present invention to realize adaptive beamforming not only simplifies the system structure, greatly reduces the technical difficulty and calculation amount, shortens the research and development period, but also enables wideband code division multiple access system The performance of smart antenna base station processing is greatly improved, which is convenient for engineering realization.

The present invention will be described in detail with reference to the accompanying drawings in conjunction with the embodiments, so as to have a deeper understanding of the purpose, technical features and positive effects of the present invention.

Description of drawings

The overall flowchart of the present invention of Fig. 1;

Fig. 2 is a structural diagram of the multi-objective array algorithm of the pilot bit assisted despreading and respreading based on the minimum mean square error criterion of the present invention.

Detailed ways

进行波束形成Y＝W×X，Y是时域处理模块22和重扩迭代模块23的输入，步骤102；对波束形成的信号进行解扰、解扩，在信息符号位和其它时隙的导频位期间，以次最优权值为初值，分别以解扰解扩的信息符号位的重扩加扰信号和已知导频位的重扩加扰信号为参考信号，进行LMS(最小均方算法)迭代，求出权值，步骤103；判断最小均方误差是否在收敛门限之内，步骤104；如果不满足，则重复执行步骤102、步骤103；如果误差满足要求，则保存该权值，用来波束形成该时隙的信号数据，步骤105；用收敛权值波束形成后，解扰解扩上行信道的I路数据，完成信息的接收，统计并输出信噪比，步骤106；最后流程结束，步骤107。As shown in Figure 1, the flow process starts, step 101; first, during the pilot bit period of the first time slot, use the respread scrambling symbol of the known uplink channel pilot symbol as a reference signal, and the time delay alignment The array receiving vector X of the current information bit performs a cross-correlation operation to find the suboptimal weight value. During the pilot bit period of the first time slot of each frame, the signal after the pilot signal is spread and scrambled is used as the reference signal r, find the cross-correlation matrix of the pilot signal X received by the array and the reference signal r The approximate solution obtained according to the minimum mean square error criterion is used as the adaptive suboptimal weight And the weight calculated by the above two steps

Carry out beamforming Y=W×X, Y is the input of time domain processing module 22 and re-spreading iterative module 23, step 102; Descrambling and despreading are carried out to the signal of beamforming, in the guide of information symbol bit and other time slots During the frequency period, the suboptimal weight is used as the initial value, and the respread scrambling signal of the descrambled and despread information symbol bits and the respread scrambling signal of the known pilot bit are used as reference signals respectively, and the LMS (minimum mean square algorithm) iterations to find the weight, step 103; judge whether the minimum mean square error is within the convergence threshold, step 104; if not satisfied, then repeat step 102, step 103; if the error meets the requirements, then save the The weight is used to beamform the signal data of the time slot, step 105; after beamforming with the convergent weight, descramble and despread the I-way data of the uplink channel, complete the reception of information, count and output the signal-to-noise ratio, step 106 ; Finally, the process ends, step 107.

As shown in Figure 2, it is a schematic diagram of the realization of the pilot bit assisted despreading respreading multi-target array adaptive algorithm based on the minimum mean square error criterion provided by the present invention. This figure shows the beam for a single path of a single user For multi-path and multi-user, the single path of a single user contained in it can be superimposed. The specific description of this figure is as follows: the signal received by the array is converted into a baseband signal through the radio frequency front end and the baseband, and adaptive processing is performed at the baseband. Due to the influence of the spatial channel, the baseband signal will have performance changes such as delay extension and angle extension. First, it is assumed that the array received signal is a delay-aligned baseband signal 214. Next, we will perform algorithm processing on 214. During the pilot period of the first time slot, the signal 234 respread and scrambled by the known pilot bit is used as a reference signal, and the cross-correlation matrix is calculated together with the baseband signal 214, and the suboptimal weight value is calculated until the first All the pilot bits of the first time slot end; the weight at the end of the pilot bit of the first time slot is input to the multiplier 211, and the beamformed signal of each array element is synthesized into a signal by the adder 212, and then Decompose this signal into two-way signals, one way is input to the calculation iteration error module 233 of the respread iteration module 23, as the subtracted vector when calculating the error, the other way is input to the descrambling module 221 of the time domain processing module 22, descrambling The Q path data of the final data enters the despreading module 222, and the data unit of the despreading is a bit, and the channel estimation and compensation modules 223 and 224 can be used to process the despreading data to reduce the influence of the channel, and the data after the judgment 225 is The sign bits of 1, -1, ... are input to the respread iteration module 23; when the current moment is an information bit period, the sign bits input by the time domain processing module 22 are input after the respread module 231 and the scrambling module 232 To calculate the iterative error module 233, and subtract the previously input adder 212 signal to obtain an error signal, which is input to the calculation new weight module 213 of the spatial beamforming module 21; when the current moment is the pilot bit period of other time slots , the respread scrambling signal 235 of the known pilot bit is used as an input of the calculation iteration error module 233, and is subtracted from the previously input adder 212 signal to obtain an error signal, which is input to the calculation new weight of the spatial beamforming module 21 Value module 213. The signal input to the new weight calculation module 213 and the time-delay-aligned signal 214 are based on the minimum mean square error criterion, iterated, and the weight is obtained. If the iterative error of beamforming with this weight meets the requirements, then the beam The formed data is descrambled, despreaded, channel estimated and compensated, combined with RAKE, and output.

Claims (2)

1. A beamforming method applicable to wideband code division multiple access systems, said beamforming method, comprising the following steps: Perform spatial processing on the array signal, where the array signal is a delay-aligned baseband signal X, including: for each frame, During the pilot bit period of the first time slot, the respread scrambling symbol of the known uplink channel pilot symbol is used as the reference signal r, and the cross-correlation operation is performed with the baseband signal X to obtain the cross-correlation matrix

Calculate the suboptimal weight until the end of all pilot bits in the first time slot; During the period of the information symbol bit and the pilot bit of other time slots, with the suboptimal weight as the initial value, the respread scrambled signal of the despread and descrambled information symbol bit and the respread scrambled signal of the known pilot bit are respectively Spread the scrambled signal as a reference signal, iterate according to the minimum mean square error criterion, and find the adaptive suboptimal weight performing beamforming with the adaptive sub-optimal weight W to obtain Y=W×X; Among them, iterate according to the minimum mean square error criterion to find the adaptive suboptimal weight

Using the adaptive suboptimal weight W to perform beamforming to obtain Y=W×X includes: Input the weight value at the end of the pilot bit of the first time slot to the multiplier, and multiply it with the baseband signal X to obtain the beamforming signal of each array element, and pass the beamforming signal of each array element through The adder synthesizes one signal, and then input this signal to the calculation iteration error module of the respread iteration module and the descrambling module of the time domain processing module, wherein, the signal input to the calculation iteration error module of the respread iteration module is used as the calculation The subtracted vector during the error; the signal descrambling input to the descrambling module of the time domain processing module is divided into I road data and Q road data, and the Q road data enters the despreading module; Utilize channel estimation and compensation module to despread The final data is processed, and then judged, and the obtained sign bits are input to the respread iteration module; When the current moment is the period of the information symbol bit, the symbol bit obtained after the decision is respread and scrambled and then input to the calculation iteration error module of the respread iterative module, and subtracted from the previously input adder signal to obtain the error signal , input to the calculation new weight module of the airspace beamforming module; When the current moment is the pilot bit period of other time slots, the respread scrambling signal of the known uplink channel pilot bit is used as an input of the calculation iteration error module of the respread iterative module, which is compared with the previous input adder signal Subtract the error signal to get the error signal, which is input to the calculation new weight module of the airspace beamforming module; Judging whether the iterative error obtained by performing beamforming with the weight output by the calculation new weight module is within the convergence threshold; if not, return to the step of performing spatial processing on the array signal; if so, save the weight as the The adaptive suboptimal weight W is used for beamforming. 2. a kind of beamforming method that is applicable to wideband code division multiple access system as claimed in claim 1, it is characterized in that said array signal is carried out spatial domain processing is for single user single path, and multipath and multiuser is the superposition of the results of the single path of the single user.

Images