遗传算法在子阵级波束形成技术中的应用研究

发布时间：2018-05-04 01:32

本文选题：子阵 + 波束形成　；参考：《哈尔滨工程大学》2014年硕士论文

【摘要】：在水声信号处理中,常用到波束形成技术,特别是多波束技术。一个基阵,如何在预期方向得到期望信号,是基阵系统形成波束和多波束的目的所在。一般可以从两个角度理解,对于发射,在某一方位集中能量,这样一来,同样发射功率的声纳系统,就会得到相对好的远距离效果。对于接收,能更好的确定目标。在接收方面,如果目标不止一个,并且可能目标特指某方向,多波束是很好的选择。对常规的单一波束,声纳的接收和发射过程只能利用到单一的覆盖范围。而多波束声纳系统预成多波束,同时覆盖一个较大的扇面。在水声领域一般采用多波束形成技术。论文研究内容如下:首先,论文讨论了阵列信号处理以及波束形成技术研究的背景和意义、国内外研究现状、波束形成分类、基础阵列模型等知识。对论文需要的波束形成方法以及加权方式等基础知识进行了阐述。其次,本文先后介绍了均匀不重叠阵、均匀重叠阵和不均匀不重叠阵的研究及仿真工作。在了解掌握子阵划分影响波束形成的前提下,关于子阵重叠度对子阵多波束形成性能的影响进行了讨论。然后,论文研究了基于子阵的数字多波束形成技术,完成基于子阵的分别加权多波束和FFT多波束形成研究的工作。在均匀线阵和均匀面阵情况下对分别加权多波束方法做了重点研究和仿真分析。针对研究仿真结果提出了面临的问题和阻碍。最后,针对前面遇到的问题,分析了利用遗传算法寻找阵列的模拟权和数字权的方法的优点。通过遗传算法优化得到每个权系数来压制主瓣内的栅瓣,提高多波束性能。由于传统遗传算法计算过程中易早熟、不稳定,研究了一种小生境遗传算法来改进传统遗传算法计算上易陷入局部循环的不足,改进波束性能方面这种方法应用到子阵多波束领域可以更具针对性。可以对每个波束产生的栅瓣进行抑制,并能克服计算上的早熟现象。小生境遗传算法的使用对优化子阵多波束形成性能提出了一种新的理念,是传统遗传算法的延伸和扩展。
[Abstract]:In underwater acoustic signal processing, it is often used in beamforming technology, especially the multi beam technology. A matrix, how to get the desired signal in the expected direction is the purpose of forming the beam and multi beam in the array system. In general, it can be understood from two angles, the energy is concentrated in a certain direction, so that the same transmit power sound. The nanar system will get a relatively good distance effect. For receiving, it can better determine the target. In receiving, if the target is more than one, and the possible target is in a certain direction, the multi beam is a good choice. For the conventional single beam, the reception and transmission of sonar can only be used in a single coverage range. Multiple beams are preformed and a large fan is covered at the same time. The multi beam forming technology is generally used in the field of underwater acoustics. The contents of this paper are as follows: first, the background and significance of the research on array signal processing and beamforming technology are discussed, the status of research at home and abroad, the classification of beamforming, the basic array model and so on. The basic knowledge of beam forming and weighting is described in this paper. Secondly, the research and Simulation of uniform nonoverlapping array, uniform overlapping array and nonuniform nonoverlapping array are introduced successively. On the premise of understanding and mastering the influence beam formation of subarray division, the performance of subarray multi beam forming performance on subarray overlap degree is discussed. The influence is discussed. Then, the paper studies the digital multi beam forming technology based on subarray, and completes the work of the study of the separate weighted multi beam and FFT multi beam forming based on the subarray. In the case of uniform linear array and uniform surface array, we do the key research and simulation analysis of the respectively weighted multi beam method. In the end, in view of the problems encountered, the advantages of using genetic algorithm to find the analog and digital weights of the array are analyzed. Each weight coefficient is optimized by genetic algorithm to suppress the gate flap in the main lobe and improve the performance of the multi beam. A niche genetic algorithm is studied to improve the shortcomings of the traditional genetic algorithm, which is easy to fall into the local cycle. In improving the beam performance, this method can be more targeted in the subarray multi beam field. It can suppress the gate flap generated by each beam and overcome the precocious phenomenon in the calculation. The use of the method puts forward a new idea to optimize the performance of sub array multi beam forming, which is the extension and extension of traditional genetic algorithm.

【学位授予单位】：哈尔滨工程大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP18;TB56

【参考文献】