The Research On The DSSS Parameters Estimation And Sequence Reconstruction In Non-cooperative Communication System

Because of the fast development in modern communication technology, research in military Communications Counter-Measures field has been maken a great process. As a signal with large bandwidth, spread spectrum signals have low probability of being intercepted and captured, strong discrimination ability, so the Spread Spectrum technology are widely used in military and civil.Non-cooperate communication is a very important part of communication and is used in many occasions, where it is required to monitor and manage the communication signals in space and exactly receive the signal information content with little prior knowledge. Because the DSSS signal power is much lower than noise power and PN code has random characteristic, the ordinary algorithm can't estimate exactly. How to estimate the signal exactly even intercept its information in non-cooperative communication systems is a great challenge with a strong background of application.This dissertation studied these problems based on the statistical characteristic of communication signals. A blind estimate algorithm is proposed, it based on cyclic spectrum theory and matrix transform. This algorithm can distill carrier frequency, and rate of the PN code, eliminate the effect of noise using the second order cyclic stationary characteristic without a prior knowledge. Then demodulate the DSSS signal and get the base band signal. Using matrix transform, two matrix functions are proposed, matrix of estimate function and matrix of synchronization function. The first is used to estimate PN code period. The second is used to synchronize PN code and confirm the start bit. After that, reconstruct the PN code and intercept the signal information content. Finally, the algorithm complexity is given, the system structure is ameliorated and the performance is improved by using the influence between synchronous capacity and reliability. But the calculation will increase and the real-time performance will reduce. The Matlab simulation results show that this algorithm can distill parameter and reconstruct the PN code correctly on SNR below -18dB. And it verifies the validity of our theoretical analysis.