QQ客服 官方微博 反馈留言

题名	干扰抑制阵列处理算法及相关技术的研究
作者	王昭辉
学位类别	博士
答辩日期	2009-05-27
授予单位	中国科学院声学研究所
授予地点	声学研究所
关键词	干扰抑制 阵列信号处理 规整化理论 稀疏重建技术 加挡圆阵
其他题名	Research on Critical Array Processing Algorithms and Related Techniques for Interference Suppression
学位专业	信号与信息处理
中文摘要	干扰问题在当前无线通信、雷达、声纳、射电天文等系统中普遍存在，并日益突出。本文主要对干扰抑制阵列信号处理关键技术进行探讨，并进一步考虑其在加挡圆阵上的应用。 传统自适应波束形成算法在各类误差源影响下，性能严重下降。线性约束最小均方误差方法作为一种稳健的自适应波束形成器，虽然可以降低阵列增益对指向性误差的敏感性，但由于其自适应过程中可用自由度数目的降低，干扰抑制性能并不理想。另外，该方法并不能改善阵元幅相误差和阵元位置扰动、以及空间谱估计误差存在时的系统性能。 为解决自由度降低问题，将线性约束转化为软约束，通过控制零点深度改善算法的干扰抑制性能，该方法相比于传统特征矢量约束方法具有更大的灵活性。考虑到算法性能对阵列扰动的敏感性，引入了三类规整化技术，即Tikohonov规整化、截断总体最小二乘规整化和规整化总体最小二乘理论，提高误差存在情况下的系统性能。仿真试验表明，此三类算法获得了比传统对角加载技术更高的阵列增益，其中规整化总体最小二乘算法具有最优性能。 在观测时间有限或信号和干扰高度统计相关场合，各类波束形成方法不能有效抑制干扰。为解决该问题，本文采用稀疏重建技术，根据信号源空间分布的物理稀疏特性，通过对波形估计施加 范数、组范数或混合范数约束，将估计问题转化为标准的稀疏重建问题。本算法在有限快拍数或者信号源高度相关的情况下，可精确重建信号源波形，并保持了较高的空间分辨率和良好的阵列处理增益。当信号和干扰在方位、时间和频段上相重叠时，将稀疏重建技术应用于时域干扰抑制领域，可有效进行干扰抑制。仿真试验和实际试验数据处理结果验证了上述算法的有效性。 最后，根据加挡圆阵数学模型，对本文各类阵列信号处理方法加以修正并应用到该阵列中。仿真试验对算法的有效性进行了验证。
英文摘要	Interference suppression become more and more imperative in current numerous systems: wireless communications, radar, sonar, radio astronomy, seismology, etc. Among various stratergies to combat interference, this article concentrates on array processing techniques. Conventional adaptive beamforming algorithm deteriorates rapidly with the increasing errors between norminal and real parameters. Although linearly constrained minimum variance method could reduce the sensitivity to directional errors, due to the reduced freedom avaliable in the adaptation process, capability of interference suppression is far from satisfying. Moreover, this method still could not maintain its good performance when array disturbance error or spatial spetrum estimation error exists. To reduce the negtive impacts caused by less freedom numbers, linear constraints were replaced by soft constraints, which controled the depth of null pionts by one scaler and provides much more flexibility compared with eign-vector contrained method. Considering the array disturbance error, three kinds of regularization techniques were employed, ie. Tikhonov regularization method, truncated total least square principle and regularized total least square principle, all of which led to higher array processing gain than the conventional diagonal loading techniques, and the third one presented the best results. Given the spasity of targets’ distribution in space, the Sparse Reconstruction techniques were introduced to achieve accurate waveform estimation with limited snapshots and highly correlated targets’ waveforms, and basis pusuit noising methods were employed to reduced the computational complexity. Additionlly, application of sparse reconstruction in time domain to achieve interference suppression is also quite promising. Simulation and experimental results validated the above conclusions. Finally, modification were made to the aformentioned array processing algorithms to find their applications in the uniform circular arrays, performance of which was verified by simulation results.
语种	中文
公开日期	2011-05-07
页码	102
内容类型	学位论文
源URL	[http://159.226.59.140/handle/311008/568]
专题	声学研究所_声学所博硕士学位论文_1981-2009博硕士学位论文
推荐引用方式 GB/T 7714	王昭辉. 干扰抑制阵列处理算法及相关技术的研究[D]. 声学研究所. 中国科学院声学研究所. 2009.

个性服务

查看访问统计

相关权益政策

暂无数据

收藏/分享

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。