一种新的低/零自相关边峰训练序列设计方法

A Novel Design Method for Training Sequence with Low-Lobe Values

PN序列因其优良的自相关性而被广泛应用.然而,PN序列因其自相关函数存在众多边峰、序列长度必须为2m-1(m∈Z)等而影响了其实用价值.本文提出了通过巴克码/正交巴克码/超巴克码与PN序列组合构造一种新的命名为低/零自相关(LZA)训练序列的设计方法.仿真证明:该法产生的LZA训练序列具有长度可以任意确定、除零时延处外该序列均有低/零自相关值、产生时所需储存容量和电路单元小、保密性高等优点.与众所周知的PN序列相比,LZA序列弥补了训练序列的不足,具有较高的理论意义和实用价值.

Pseudo noise （PN） sequence has been apphed widely because of its optimum property of the autocorrelation function. However, it has been seriously hmited in practice by many side peaks and regular sequence length. In this paper, a novel method to design a new PN sequence named low or zero autocorrelation （LZA） sequence through Barker code or quaternary Barker code or super Barker code and PN sequence is pmpesed. Simulations show that LZA sequence has four merits when compared with the conventional PN sequences. First, the length of LZA sequence is optional. Second, the autocorrelation values of the LZA sequence are small or zero except at the point of time delay of zero. Third, the memory cell and circuit cell to construct the LZA sequence are very small even though the length of LZA sequence is as long as what you expect. Finally, the LZA sequence has great capabihty of security. Obviously, compared with the famous PN sequence, LZA sequence has supphed a gap of PN sequence with better performance, thus it will be useful in practice and theory.