DS-SS Signal Spread Spectrum Codes Blind Estimation Based On Computational Intelligence

It is usually that the receiver must know the spreading sequence used by the sender to despread with correlators, and then it can recover the transmitted data. Direct sequence spread spectrum communications as the main means of spread spectrum communication, has a good low-emission power spectral density of concealment, the secrecy capacity of pseudo-random coding and signal processing related to anti-jamming capability, that have brought new challenges to detect and identify the spread spectrum communication , spread spectrum communications signal monitoring and management. How to achieve the spread spectrum communication signal detection and identification is the focus of communications surveillance, and it is also a pressing research topic. Therefore, the spread spectrum signal parameter detection and pattern recognition is an important step on how to detect and identify a direct sequence spread spectrum communication signals.This paper focuses on direct sequence spread spectrum (DS-SS) signal spreading code blind estimation based on computational intelligence. The main work includes the following aspects:(1) It introduces several model spread spectrum signal briefly, and then notes that the focus is the DS-SS signals. It introduces applications of computational intelligence in signal processing briefly. In this paper, DS-SS signals study is mainly on the blind estimation of the unknown spreading codes.(2) It proposed an approach that can estimate the pseudo-noise (PN) sequence from low SNR DS / SS signal, which is based on the eigen-decomposition of signal correlation matrix. Further, using the PCA based on variable step-size neural network method to achieve the pseudo-code sequence blind estimation. This method based on the self-adaptive variable step-size learning algorithm, a larger step size in the initial stages, can make faster convergence over time, gradually reducing the step in order to ultimately achieve a better neural network steady-state convergence. The algorithm overcomes the traditional algorithm's inherent conflicts between convergence speed and steady-state error.(3) In the context of blind detection and estimation problems for weak DS signals, a method to recover the spread spectrum code of direct sequence spread spectrum signal is presented, whereas the receiver has no knowledge of the transmitter's spreading sequence. The method is based on back propagation (BP) neural network, whose input is the received signal, and the expected output is the same as the input signal. It is supervised to adjust the neural network according to the error back-propagation, and then it can blind estimate the spread spectrum code sequences according to the symbol function value of second layer weights when adjusting the network to reach convergence.(4) In view of blind detection and estimation problems for weak DS signals, a method to recover the spread spectrum code of direct sequence spread spectrum signal is presented, which based on radial basis function (RBF) neural network, whereas the receiver has no knowledge of the transmitter's spreading sequence.In this paper, we study the algorithm of DS-SS signals PN spreading sequence blind estimation, further more, methods of spread spectrum code blind estimation are verified by the simulation results. Theoretical analysis and computer simulation results are provided to show that the method can work well on lower SNR DS/SS for longer PN sequence. Therefore, this approach will provide a means for solving the DS-SS signal PN sequence blind estimation and it also pave the way for the management, surveillance and interference of DS communication, blind multi-user detection of DS-CDMA.