Blind Estimation Of Pseudo-Random Sequence In Long-Code Direct Spread Spectrum Signals

The problem of pseudo-random (PN) sequence estimation in DSSS signals in non-cooperative communication systems has been widely studied by scholars from all over the world. The complex structure of long and short codes direct sequence spread spectrum (LSC-DSSS) signals which use a pair of PN sequences serving as spreading code and scrambling code makes the PN sequence estimation in non-cooperative communication systems very difficult, and it is quite necessary to conduct the relative research. This dissertation mainly studies the PN sequences estimation and blind despreading methods of LSC-DSSS signals in non-cooperative communication systems.First of all, a detailed analysis of current researches on PN sequence estimation of DSSS signals in non-cooperative communication system is given, and the basic principles and characteristics of DSSS communication systems are discussed. The signal models of short code direct sequence spread spectrum (SC-DSSS) signals, the long code direct sequence spread spectrum (LC-DSSS) signals that use a long PN sequence as spreading code and the LSC-DSSS signals that use a pair of PN sequences are given. The commonly used PN sequences as well as their generative principles and their properties are also briefly introduced.Secondly, the scrambling code estimation method of LSC-DSSS signals based on triple-correlation peaks matching and the blind despreading method are studied. With the properties of triple correlation functions of m-sequences, the existence of common peaks between the triple correlation functions of m-sequences with different periods is proved and the maximum likelihood estimation of triple correlation function of LSC-DSSS signal is deduced. The long scrambling code estimation methods of periodic LSC-DSSS signals and non-periodic LSC-DSSS signals based on triple-correlation peaks matching are put forward. After the long scrambling code is estimated, a descrambling method based on subsection cross-correlation is proposed by using the shift-and-add property and segmenting signal ingeniously. The primitive polynomial of the spreading code is estimated by using matrix's oblique elementary method and the information codes synchronization is achieved by utilizing information code synchronization method based on covariance norm. Finally, the blind despreading method of LSC-DSSS signal is obtained. The simulation results show the proposed methods have well performanceThirdly, the PN sequences estimation method of LSC-DSSS signals based on the common peaks is investigated. Starting from analyzing the relationship between the cyclotomic co-sets and the peaks of triple correlation functions of m-sequences, the corresponding relationship between the common peaks of triple correlation function of the LSC-DSSS signal and the primitive polynomials of m-sequences is proved. The PN sequences estimation methods of periodic LSC-DSSS signals and non-periodic LSC-DSSS signals are proposed respectively using the different characteristics of the signals. In addition, the detection probability and false alarm probability of common peaks of triple correlation functions of LSC-DSSS signals are deduced theoretically. The simulation results prove the validity of the proposed theories and deduction, and show the better performance of the proposed methods.Finally, the blind despreading method of wideband code division multiple access (WCDMA) signals is studied. By utilizing the characteristics of triple correlation functions of m sequences and the triple correlation functions of Gold sequences, the Gold sequence estimation method of WCDMA signal based on third order correlation statistics is put forward. Through the successful estimation of Gold sequence, the descrambling method of WCDMA signal based on the slices of forth-order cumulants is proposed by applying the second-order correlation properties of Orthogonal Variable Spreading Factor (OVSF). The information codes synchronization and OVSF code estimation are also achieved to realize the blind despeading of WCDMA signals finally. The simulation results show that this method can effectively achieve the identification of Gold sequence and blind despreading of WCDMA signals.