Novel Soft Spread Spectrum with Unequal Probabilistic Spreading codes against Clustering-based Blind Despreading

Blind despreading is one of the methods to analyze soft spread spectrum. It mainly uses mathematical method or machine learning to analyze spreading codes. In this paper, a novel soft spread spectrum scheme based on random mapping with unequal probability is proposed. In this scheme, AES(Advanced Encryption Standard) algorithm is used to select spreading codes randomly, and Kasami sequence is used for spreading codes. Anti-blind despreading analysis on the scheme is carried out compared with the traditional scheme. Simulation results demonstrate that, for the traditional spread spectrum scheme, when the transmitted information reaches 1,720 bits and 18,800 bits respectively, the corresponding spreading codes can be recovered by using the clustering algorithm. In our scheme, when the transmitted information reaches 120,000 bits, the clustering algorithm still fails to converge effectively.