改进的离散混合蛙跳算法压缩感知信号重构及应用  

Improvement of discrete shuffled frog-leaping algorithm and application in compressed sensing reconstruction

作　　者：刘洲洲[1,2] 王福豹[2] 

机构地区：[1]西安航空学院电子工程学院,西安710077 [2]西北工业大学电子信息学院,西安710072

出　　处：《吉林大学学报（工学版）》2016年第4期1261-1268,共8页Journal of Jilin University:Engineering and Technology Edition

基　　金：国家自然科学基金项目(61103242;61401499);陕西省教育厅科研计划项目(16JK1395)

摘　　要：针对工程离散优化问题特点,定义了具有普遍意义的青蛙编码方式,设计了编码位调换更新机制,提出了自适应权重因子和双模子族群策略。在此基础上,将改进的离散混合蛙跳算法（Discrete shuffled frog leaping algorithm,DSFLA）应用于压缩感知重构算法中,将未知重构信号理解为青蛙编码方式,利用DSFLA算法全局寻优能力得到次最优信号重构信息,从而实现了稀疏度未知情况下的信号重构。最后对典型TSP（Travelling salesman problem）问题算例和WSNs多目标定位问题进行仿真,仿真结果表明：改进的DSFLA具有更强的复杂问题求解能力,基于改进DSFLA压缩感知重构算法的WSNs目标定位精度优于传统信号重构算法,且抗噪能力达到25~45dB。An improvement of Discrete Shuffled Frog-leaping Algorithm（DSFLA）is proposed.Fist,according to the characteristics of discrete optimization problems,the frog coding of universal significance is defined,which is important for DSFLA in solving discrete optimization effectively.Then,the update mechanism based on ＂swapping of coded bits＂for DSFLA is designed,and an adaptive weighting factor and sub-ethic dual strategy are presented.Finally,the improved DSFLA is applied in compressed sensing reconstruction algorithm,in which the unknown reconstructed signal encoding is taken as the frog code.Typical TSP problems and multiple target localization in WSN are simulated.Simulation results show that the improved DSFLA has prominent ability to solve complex problems,and the perception accuracy of WSNs target reconstruction based on improved DSFLA CS reconstruction algorithm is better than that of the traditional signal reconstruction algorithm,and the anti noise capacity reaches 25~45dB.

关 键 词：计算机应用 无线传感器网络 离散混合蛙跳算法 压缩感知重构算法 多目标定位 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]