WCDMA Uplink Channel Scramble Code Estimation And Blind Decoding

Wideband Code Division Multiple Access has become one of the third generation mobile communication technology standards due to its flexible and multi-service high-speed wireless air interface technology.WCDMA signal has the advantages of anti-jamming,anti-interception,strong concealment due to the fact that its power spectrum is usually submerged in the noise,which makes it become one of the key technologies in various communication environments like military communications,civil communications and business communications.In order to better capture useful signals in the communication confrontation of comple x environment,it is especially important to detect and identify subtle features like scrambling sequences and source information.WCDMA signal contains multiple pseudo-random sequences,where the scrambling code is a truncated Gold sequence,and its structure is extremely complex.Therefore,in the military communications,it is very difficult to estimate the information sequence and scrambling sequence of the intercepted WCDMA signal,and its research is practically significant.Based on this,this dissertation mainly studies the source information estimation and long-period scrambling code identification in the WCDMA signals under the condition of multi-user and unknown carrier frequency offset.This dissertation firstly introduces the status of Direct Sequence Spread Spectrum reconnaissance.Then,the WCDMA uplink spreading and scrambling mechanism are analyzed,which are reconnoitered and researched by an established WCDMA signal model.Finally,based on the research shortcoming that the spread spectrum sequence estimation and blind decoding algorithm of existing DSSS signals are basically carried out under the ideal circumstance of single user and demodulated carrier,this dissertation,under the complex condition of multi-user and unknown carrier frequency offset,firstly proposes a WCDMA signal blind decoding algorithm based on space-time domain and secondly proposes a long-period scrambling code estimation and blind source estimation algorithm for WCDMA signals under single channel.In the space-time domain environment,space-domain filtering of the array antenna is firstly used to separate the multi-user signals,and then the principal component analysis and the phase-locked loop technology is applied to estimate the source information of different users.The proposed method combines the spatial gain of the array antenna and the principal component analysis of time-domain projection gain,and uses the phase-locked loop to correct the influence of the residual frequency offset on the scrambling code estimation.The theoretical analysis and simulation results show that the proposed space-time domain method can improve the decoding performance of multi-user WCDMA signals by using array antenna gain in the case of multi-user and unknown carrier frequency offset.In the single-channel environment,the signal received by the receiver cannot be separated in the in the time,frequency domain and airspace domains,and only very little amount of information can be used.This dissertation adopts independent component analysis combined with phase locked loop to estimate the scramble sequence and signal data of different users.The proposed algorithm uses the frame repetition of scramble code and OVSF code to sample with equal frame intervals,and then a new blind source separation model with carrier modulation is constructed to operate the signal separation and scrambling code estimation.Analysis results show that the proposed method can effectively identify the scrambling sequence of multi-user WCDMA signals and blindly decode the signal under the condition of multi-user and unknown carrier frequency offset.The WCDMA signal reconnaissance algorithm in complex environment proposed by this dissertation not only solves the problem of long code DSSS signal reconnaissance which contains multiple pseudo-codes at the same time,but also improves the performance of decoding and pseudo-code blind estimation compared with the traditional singular value decomposition algorithms.