PCMA系统双同步序列联合盲估计

Joint Blind Estimation of Dual Synchronization Sequence in PCMA System

成对载波多址(Paired Carrier Multiple Access,PCMA)使用同一频率传输通信双方的调制波形,是一种重要的卫星通信多址方式。对PCMA信号进行盲分离是无线电频谱监测的重要内容,高精度的调制参数信息是盲分离的必要基础,同步序列辅助的调制参数估计方法具有精度高、鲁棒性好等特点。然而在PCMA信号的盲接收中,往往无法预知同步序列,且两路同步序列估计需要在无法完成分离解调的条件下得到,因此难度较大。本文针对幅相调制(Amplitude-Phase Modulation,APM)类PCMA信号,利用通信系统数据帧的结构特性,采用子空间分解思想,重点分析了不同条件下的信号相关矩阵奇异值分解(Singular Value Decomposition,SVD)结果,推导了特征值、特征向量与两路信号同步波形间的近似数学关系;提出“先整体后局部”的算法流程,首先对信号按帧长分段做整体SVD分解得到对同步波形及其位置的粗估计,而后对同步波形位置的原始PCMA信号做局部SVD分解,进一步提高估计性能,解决了PCMA信号盲接收中双同步序列估计的难题。仿真结果表明,本文算法可广泛适用于APM类PCMA信号的同步波形和同步序列盲估计问题,在低信噪比条件下算法性能优于最大似然的估计方法,随着信噪比的提升和使用数据量的增加,估计性能不断提高,能够得到对PCMA信号双同步序列的高精度估计,为数据辅助的参数估计方法应用于工程实践提供了先决条件,为PCMA信号调制参数估计和后续的盲分离技术研究提供重要基础。

Paired-carrier multiple-access(PCMA)is an important satellite-communication multiple-access method that uses the same frequency to transmit the modulation waveforms of both communication parties.The blind separation of the PCMA signals is an important part of radio spectrum monitoring,and high-precision modulation parameter information is needed for this blind separation,while the method of modulation-parameter estimation aided by the synchronous sequence has high accuracy and good robustness.However,in the blind reception of PCMA signals,it is often impossible to predict the synchronization sequence in advance,and estimations of the two synchronous sequences must be obtained under the condition that separation and demodulation cannot be completed,making it even more difficult.This study focused on PCMA signals using amplitude-phase modulation(APM).The singular value decomposition(SVD)results of the signal correlation matrix under different conditions were analyzed by exploiting the structural characteristics of the communication system’s data frames and the idea of subspace decomposition,and the approximate mathematical relationships between the eigenvalues,eigenvectors,and synchronous waveforms of the two signals were derived.A“whole first and then local”algorithm flow was proposed.First,overall SVD decomposition was performed on the signal segmented by frame length to obtain a rough estimation of the synchronous waveform and its position.Then,local SVD decomposition was performed on the original PCMA signal at the synchronous waveform position to further improve the estimation performance.This made it possible to solve the problem of dual synchronous sequence estimation in the blind reception of PCMA signals.Simulation results showed that the scheme could be widely applied to the blind estimation of a synchronization waveform and synchronization sequence of APM-modulated PCMA signals.Under a low signal-to-noise ratio,this method outperformed the maximum likelihood estimation method.By improving the signal-to-noise ratio and increasing the volume of data,the estimation performance could be further improved,enabling the high-precision estimation of PCMA-signal dual synchronous sequences.This would provide a prerequisite for the application of data-aided parameter estimation methods in engineering practice,and provide an important foundation for PCMA signal modulation parameter estimation and subsequent blind separation technology research.