基于欠定盲分离的跳频信号恢复与分选技术研究

发布时间：2018-04-20 04:05

本文选题：欠定盲源分离 + 跳频通信　；参考：《西安电子科技大学》2014年硕士论文

【摘要】：由于良好的抗干扰能力和低截获率,跳频信号在军事通信中获得了广泛应用。研究跳频信号的侦察技术,具有重要的现实意义。为了获得跳频信号的时域数据,本文在假设跳频信号参数已知的情况下,从应用欠定盲分离算法恢复每跳信号和跳频信号分选两部分出发,研究了欠定混合跳频信号的恢复问题。本文首先从解决跳频信号的稀疏性问题出发,针对时频域转换方式复杂度高且无法直接得到时域跳频信号数据的问题,提出了一种基于滤波的跳频信号欠定盲分离算法。该算法首先通过滤波分解跳频信号欠定盲分离问题,使得分解后的时域跳频信号满足稀疏性要求,然后直接应用欠定盲分离算法即可得到时域跳频信号的每跳数据。仿真实验表明,和将跳频信号转换到时频域的方式相比,该算法在保障恢复精度的同时,降低了算法复杂度,并可以直接得到时域跳频信号数据,在解决跳频信号的欠定盲分离时,更加高效。为了解决欠定盲分离带来的跳频信号每跳数据排列模糊性的问题,本文研究了同步正交跳频信号的分选技术。首先介绍了欠定盲分离中信号到达角信息的提取,然后引入码元速率用于分选,提出了基于到达角和码元速率的多参数融合的同步正交跳频信号的分选算法。仿真实验表明,多参数融合的分选算法在跳频信号分选方面表现良好,算法适用面更广,当信噪比在8 dB以上时,分选正确的概率可以达到0.9以上。
[Abstract]:Frequency hopping signal has been widely used in military communication because of its good anti-jamming ability and low interception rate. It is of great practical significance to study the reconnaissance technology of frequency hopping signal. In order to obtain time-domain data of frequency-hopping signals, in this paper, based on the assumption that the parameters of frequency-hopping signals are known, two parts of each hopping signal and frequency-hopping signal sorting are recovered by using under-determined blind separation algorithm. The problem of the recovery of underdetermined hybrid frequency hopping signals is studied. In order to solve the problem of frequency hopping (FH) signal sparsity, in this paper, a filter based undetermined blind separation algorithm for frequency-hopping signals is proposed to solve the problem of high complexity of time-frequency domain conversion and the inability to directly obtain time-domain frequency-hopping signal data. The algorithm firstly decomposes the frequency hopping signal under-fixed blind separation problem by filtering, which makes the decomposed time-domain frequency-hopping signal satisfy the sparse requirement, and then directly applies the under-determined blind separation algorithm to obtain the time-domain frequency-hopping signal per hop data. The simulation results show that the algorithm not only guarantees the recovery accuracy, but also reduces the complexity of the algorithm compared with the method of converting the frequency hopping signal to the time-domain frequency domain, and can directly obtain the time-domain frequency-hopping signal data. It is more efficient in solving the under-determined blind separation of frequency-hopping signals. In order to solve the problem of fuzzy arrangement of frequency hopping signals caused by under-determined blind separation, this paper studies the sorting technology of synchronous orthogonal frequency-hopping signals. This paper first introduces the extraction of the signal angle of arrival in underdetermined blind separation, then introduces the symbol rate for sorting, and proposes a multi-parameter fusion algorithm based on the angle of arrival and symbol rate for the separation of synchronous quadrature frequency-hopping signals. The simulation results show that the multi-parameter fusion algorithm performs well in frequency hopping signal sorting, and the algorithm has a wider range of applications. When the SNR is above 8 dB, the correct sorting probability can be more than 0.9.
【学位授予单位】：西安电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN914.41

【相似文献】