Study On Single-channel Signal Separation And Detection Of Satellite-based AIS

The satellite-based AIS system is a new type of ship navigation system which has become a general system of the marine traffic.The AIS system has attracted more and more attention not only in the aspect of avoiding ship collision and automatic recognition,but also in improving the safety and efficiency of navigation.Based on the technical bottlenecks AIS system,the main work of this paper is as follows:(1)A single-channel signal separation algorithm based on the improved EMD is proposed.It adopts the improved EMD algorithm to decompose the observed signal into a series of IMF components of different frequency;Then select appropriate IMF components by caculating KL divergence values;Finally,the FastICA algorithm can be applied for multi-channel signal separation.The method overcomes the end effect of EMD algorithm.The results show that the proposed algorithm can achieve single-channel separation with some mixed signal,and the the larger frequency offset is,the better the performance is.(2)An algorithm based on redundant discrete wavelet transform is proposed.The observed signal can be decomposed into a series of detail signals and approximate signals after the redundant discrete wavelet transform.Then use ICA algorithm to separate the decomposed signals.The length of approximate signals and detail signals obtained by this method is as the same as the original signal.So that more information can be applied for single-channel signal separation.Theoretical analysis and simulation results show that the proposed algorithm achieves the separation of multi-channel mixed signals,and the separation effect is hardly affected by time delay.(3)A new signal detection algorithm based on channel and phase real-time tracking is proposed for satellite-based AIS system.The algorithm combines real-time tracking of the channel and phase to realize the joint estimation of parameters and sequences.The simulation results show that the algorithm is better than the differential detection,Viterbi decoding and and per-survivor processing algorithm,and the frequency offset ability is stronger.