基于谱范数和CPAST算法的周期长码直扩信号PN序列估计

针对周期长码直接序列扩频(DSSS)信号的伪噪声(PN)序列盲估计问题,提出一种新的谱范数和约束投影逼近子空间跟踪(CAPST)算法相结合的PN序列估计算法。算法首先根据最大化分段协方差矩阵的谱范数法,估计出PN序列的同步点;然后利用CPAST算法,分别估计各子段的PN码序列;最后按照一定顺序拼接成完整的周期长码PN序列。算法具有运算量小、复杂度低、实时处理及硬件实现的优点。仿真结果表明,在较低的信噪比情况,该算法能够有效地估计出同步点和PN码序列。

Estimation of PN Sequence of Periodic Long Code DSSS Based on the Spectral Norm and CPAST Algorithm

In non-cooperative direct sequence spread spectrum (DSSS)communication systems, the estimation of pseudo-noise (PN) sequence of periodic long code is an important research subject.A efficient estimation approach of spread sequence is proposed based on combination spectral norm and CPAST algorithm.we first search the start position of a data symbol, maximizing spectral norm of a sub-segment cross-variance matrix, and then estimate PN sequence of every sub-segment with fast constrained projection approximation subspace tracking (CPAST) algorithm, and get complete PN sequence after sorting and splicing in certain order . It has the advantage of lower complexity, less storage,real-time processing and easier realization in hardware and. The experimental simulation results show that the algorithm can retrieve the estimation of synchronization point and PN code sequence quickly and accurately in low SNR conditions.