Improved blind-spreading sequence estimation algorithm for direct sequence spread spectrum signals

Direct sequence spread spectrum (DSSS) signals are now widely used for communications. DSSS transmitters use a spreading sequence to modulate the baseband signal before transmission. A receiver which does not know the spreading sequence cannot demodulate the signal. Burel and Bouder introduced an eigenanalysis-based blind-spreading sequence estimation algorithm, which performs well even when the received signal is far below the noise level. However, this algorithm does not applied to the long-code DSSS signals. An improved blind-spreading sequence estimation algorithm is presented. This algorithm is based on segmentation. The received signal is divided into K collections of temporal windows, from which K covariance matrices can be computed. The authors prove that K short-time segments of the spreading waveform can be recovered from these matrices using the eigenanalysis technique. Then, the spreading sequence can be reconstructed by concatenating these short-time segments. Simulations show that the proposed algorithm can provide a good estimation for long- or short-code DSSS signals in non-cooperative context, even with low signal-to-noise ratio. Furthermore, for short-code DSSS signals, the computational cost of the proposed algorithm is much lower than that of the original algorithm.