Abstract:

A method based on propagators for scrambler blind recognition is proposed. Firstly, the randomized-sequence is reshaped to a matrix according to the length of the scrambler. The scrambler which contains error codes could be estimated according to the propagator method. Secondly, the feedback polynomial of the scrambler is estimated based on the maximum likelihood method. Finally the correct scrambler is generated according to the polynomial and the synchronization is achieved according to the correlation between the scrambler and the recovered sequence. Validity of the algorithm is verified by the simulation results, and the computational load and error tolerance of the algorithm are analyzed. Case studies are presented to illustrate that the computational load of the new method is less than the eigenanalysis method.