基于联合稀疏的阵列信号LOFAR谱重构  

Array-based LOFAR spectrum reconstruction via joint sparsity

作　　者：陆晨翔 曾向阳[1] 王强 LU Chenxiang;ZENG Xiangyang;WANG Qiang(School of Navigation,Northwestern Polytechnical University,Xi’an 710072,China;The 715th Research Institute of China Shipbuilding Industry Corporation,Hangzhou 310023,China)

机构地区：[1]西北工业大学航海学院,西安710072 [2]中国船舶重工集团公司第七一五研究所,杭州310023

出　　处：《自动化与仪器仪表》2023年第7期1-4,共4页Automation & Instrumentation

基　　金：国家自然科学基金资助项目(52271351)。

摘　　要：LOFAR谱广泛应用于被动声纳的阵列信号分析系统,针对阵列信号采用基于稀疏贝叶斯概率模型的联合稀疏方法,可同时实现目标高分辨DOA估计和目标信号LOFAR谱重构。该方法通过对宽带信号不同频率对应的角度域进行对联合稀疏约束,在不牺牲角度分辨能力的前提下,对不同频率获得一致的空域响应,实现恒定束宽效果。基于阵列实测数据的实验表明,该方法相比常规波束形成方法,能够提高水下目标信号LOFAR谱重构质量,并能有效抑制干扰混叠。LOFAR spectrum is widely used in passive sonar signal processing system.For hydrophone array signal,sparse Bayesian learning framework with joint sparsity priors can be utilized to perform high-resolution DOA estimation and LOFAR spectrum reconstruction at the same time.The proposed method can constrain the angle domain over different frequencies of wideband received signal via joint sparsity and obtain frequency-invariant spatial response for the purpose of attaining beamwidth constancy at no cost of reducing angular resolution.Experiments conduct on real measured array data shows that the proposed method,compared to conventional beamformer,can improve the reconstruction quality of LOFAR spectrum of underwater target signal and suppress interference effectively.

关 键 词：水声信号处理 LOFAR谱 结构化稀疏 恒定束宽 阵列信号处理 

分 类 号：TP29[自动化与计算机技术—检测技术与自动化装置] TN911.72[自动化与计算机技术—控制科学与工程]