针对星载船舶自动识别系统(AIS)中无法避免的信号混叠问题,研究了突发混叠信号单通道分离中广泛应用的逐幸存路径处理(PSP)与最大似然接收解调,提出了一种基于相位跟踪的AIS 信号单通道分离算法。该算法通过最小化混叠信号状态转移表中的欧氏距离,解决了信道估计与跟踪中由频偏和调制系数估计误差导致的相位畸变问题。仿真表明,该算法相位跟踪性能在归一化频偏误差(相对符号速率)大于2×10-2时仍然能收敛,该算法的解调误码率与抗频偏的PSP 算法及线性联合检测算法相比,当误码率等于10-3时,同道干扰信道下具有2 dB ～4 dB 的信干比性能优势。

A single-channel separation algorithm using phase tracking is proposed in this paper to demodulate automatic identification system (AIS) signals, so as to enhance the abilities of the widely-used per-survivor processing ( PSP) and maximum likelihood detection ones in adaptation to co-channel interference (CCI) . The main contribution of this paper is to resolve the phase recovery problem in channel estimation and tracking for an unknown modulation index and frequency offset, coped with a time-varying phase shift by minimizing the Euclidean distance used in the state transferred trellis. Simulation results show that the proposed algorithm can work well even when the unitary frequency offset (compared to symbol rate) grows up to 2×10-2 . Compared with the PSP detection based on anti-frequency-offset and joint linearly complexity detection, the proposed algorithm gains an improvement of 2 dB ～4 dB in signal-to-interference ratio when the bit error rate is set to 10-3.

TN924