基于重采样的无线信道信源数目估计算法研究

发布时间：2018-07-08 09:23

本文选题：信源数目估计 + 重采样理论　；参考：《天津大学》2014年硕士论文

【摘要】：无线信道信源数目估计是信道测量和建模的重要工作之一,是再现真实传播场景和改善通信质量的关键。对信道的分析有赖于多维参数的精确估计,而信源数目是影响参数提取算法性能的重要指标。由于无线信道具有时变特征,在工程应用中阵列通道接收信号相关性变化大,引起传统估计方法的性能恶化甚至失效。近年来,统计理论逐渐成为阵列信号处理领域新的有力工具,使得提升信源数目估计准确性和提高参数提取速度成为可能。本文首先回顾了信源数目估计问题国内外的发展历程和研究现状,然后以无线信号传播模型为切入点,引出信源数目估计对参数提取算法的重要影响。接着分别对白噪声和色噪声场景下的8种常用信源数目估计方法进行算法介绍和理论分析。在分析各种算法的优缺点后,本文以盖氏圆准则为理论基础,以重采样为统计手段,提出了一种能够同时适用于白噪声和色噪声场景的信源数目估计方法,并推导出了采样比例和子样本集个数的最优值。进而利用该方法在参数提取仿真实验中的成功应用,验证了其可行性、准确性和高效性。本文创新点在于将传统信源数目估计方法与重采样统计理论相结合,提出在未知噪声、低快拍、低采样率下的高精度信源数目估计方法。同时,结合重采样信源数目估计的高精度,完成了一套从信号入射到信道参数(时延、到达角、离开角、多普勒频移、复振幅)输出的完整实现过程,对信道建模和测量工作具有参考作用。
[Abstract]:The estimation of the number of wireless channel sources is one of the most important tasks in channel measurement and modeling. It is also the key to reproduce the real transmission scene and improve the communication quality. The analysis of the channel depends on the accurate estimation of the multidimensional parameters, and the number of sources is an important index affecting the performance of the parameter extraction algorithm. Due to the time-varying characteristics of wireless channels, the correlation of received signals in array channels varies greatly in engineering applications, which results in the deterioration or even failure of the performance of traditional estimation methods. In recent years, statistical theory has gradually become a new powerful tool in array signal processing, which makes it possible to improve the accuracy of source number estimation and improve the speed of parameter extraction. This paper first reviews the development and research status of the number of sources estimation problem at home and abroad, and then, with the wireless signal propagation model as the breakthrough point, leads to the important influence of the number of sources estimation on the parameter extraction algorithm. Then the algorithms and theoretical analysis of eight common methods for estimating the number of information sources in white noise and color noise scene are introduced and analyzed respectively. After analyzing the merits and demerits of various algorithms, this paper presents a method for estimating the number of sources which can be used in both white and color noise scenes based on the Ghakelian circle criterion and resampling as a statistical means. The optimal values of the sampling ratio and the number of subsamples are derived. Furthermore, the method is successfully applied in the simulation experiment of parameter extraction, and its feasibility, accuracy and efficiency are verified. The innovation of this paper is to combine the traditional source number estimation method with the resampling statistical theory, and propose a high precision source number estimation method under unknown noise, low rapid-beat and low sampling rate. At the same time, combined with the high accuracy of resampling source number estimation, a complete set of outputs from signal incident to channel parameters (time delay, angle of arrival, departure angle, Doppler frequency shift, complex amplitude) are completed. It can be used as a reference for channel modeling and measurement.
【学位授予单位】：天津大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN92

【相似文献】