极低信噪比无线通信信号随机共振检测理论与方法研究

For deep-space and deep-ocean submarines wireless communication systems, the received signal-to-noise ratio (SNR) becomes increasingly lower as the distance of space exploration and diving depth constantly increases. In order to well perform the task of communication, some novel and effective signal detection methods are required under very low SNR region. In this proposal, the theory of stochastic resonance (SR) from the field of nonlinear kinetic systems is proposed to be applied for wireless communication signal detection under very low SNR scenarios, by virtue of its unique advantage in weak signal detection. The detection performance is improved by the cooperation among signal, noise and the nonlinear system. The relationship model between the parameters of nonlinear system and the performance of signal detection under very low SNR is first established. Then, the principle and mechanism of SR aided signal detection is explained in terms of SNR, mutual information and the correlation between input and output waveforms. Based this mechanism, the SR processing method of communication signal with very low SNR is further proposed. By feeding the received signals with very low SNR into a nonlinear system with specific parameters, the characteristics in time and frequency domains and the statistical distribution of noise are changed. As a result, the desired communication signal can be enhanced and the in-band noise can be suppressed at the same time. Moreover, the concerned communication signal is then demodulated by a nonlinear system, which can effectively distinguish signal and noise and estimate the signal parameters. With these manipulatoins, desired detection performance can be achieved. Finally, half hardware-in-loop simulations will be applied to verify the proposed SR aided communication signal detection under typical scenarios of deep-space and deep-ocean submarines wireless communication systems.

随着探测距离与下潜深度的增加，深空和远洋水下潜器通信系统接收信噪比（SNR）越来越低，需要研究高效的极低SNR无线通信信号检测方法。本项目利用非线性动力学系统的随机共振（SR）理论在微弱信号检测中的独特优势，研究极低SNR通信信号SR检测的理论和方法，利用信号、噪声和非线性系统间的协同提高检测性能。项目首先建立非线性系统参数和噪声与通信信号检测性能的关系模型，从SNR、波形相关性和互信息的角度，揭示通信信号SR检测的机理。以此为基础，项目提出极低SNR通信信号SR预处理方法，将接收信号通过特别设计的参数自适应非线性系统，利用SR改变噪声的时间、频率特性及其统计分布，在增强有用信号的同时抑制带内噪声；进一步，项目通过可以有效区分信号和噪声、估计信号参数的非线性方法，实现对极低SNR通信信号的判决解调。项目还将针对深空和远洋水下潜器通信的典型场景，通过半实物仿真对提出的理论和方法进行评估验证。

项目围绕极低信噪比（signal-to-noise ratio，SNR）无线通信信号随机共振（stochastic resonance，SR）增强检测和解调理论、方法和技术开展研究，主要研究内容包括以下四个方面：极低SNR无线通信信号SR检测机理研究；极低SNR无线通信信号SR预处理方法研究；经SR预处理的无线通信信号的非线性解调算法研究；基于半实物仿真平台的无线通信信号SR检测性能评估。.项目完成了上述四个方面的研究内容，取得了如下四方面的研究成果：.第一，针对宽带高斯噪声下的极低SNR信号高效检测问题，项目提出了基于双稳态SR系统的信号增强处理方法和架构，以最大化输出SNR为目标，给出了噪声、频率自适应的双稳态系统参数设计方法。.第二，针对非高斯脉冲噪声下带限无线通信信号的可靠接收问题，项目提出了基于超阈值量化阵列广义随机共振的信号增强处理方法，以最大化处理后SNR和波形相关性为优化目标，构建了优化处理模型，基于变分和凸优化理论，给出了优化处理方法的一般形式，和特殊处理方法（如clipper）的优化参数。.第三，针对非高斯脉冲噪声下线性和连续相位调制信号非线性处理和优化接收问题，项目提出了新型的有记忆（Myriad滤波）和优化的无记忆（单符号）预处理方法，给出了噪声参数估计、同步性能限分析、优化同步序列设计与同步算法、相干和非相干解调算法。.第四，面向对潜长波通信（典型的非高斯脉冲噪声应用场景），开发了实验室实验系统，对项目研究的关键算法进行了测试评估，验证了其有效性，相比于传统处理方法，最大可获得3dB的增益。.项目的研究成果可用于增强噪声下的微弱无线通信信号检测，尤其是长波通信等非高斯脉冲噪声的场景，提高信号接收的可靠性。项目申请发明专利10项（含授权1项），发表论文17篇,录用1篇，项目研究成果应用于电磁频谱监测项目，获省部级科技进步一等奖，成果应用于某探索研究项目，通过专家组验收。