Blind classification of the short-code and the long-code direct sequence spread spectrum signals

Abstract

Direct sequence spread spectrum (DSSS) signals are now widely used for communications. According to the relationship between spreading factor and code length, the DSSS signals can be divided into two types: the short-code (SC) and the long-code (LC) DSSS. For the cooperative receivers, the above type information is known in advance. However, under the non-cooperative contexts, such information becomes one of the unknown parameters of interest. To extract this information, a blind algorithm that is based on the second-order statistics of the matrix norms of the signal correlation matrices is proposed in this study. A set of correlation matrices is constructed from the received signal samples following which the matrix norm of each correlation matrix is computed and normalised by the maximum. Then, by comparing the variance of the normalised matrix norms with a preset threshold, the type of the received DSSS signal can be identified. Simulation results verify the capability of the proposed method in various scenarios, such as multipath fading channel and multiple access interference.