基于贝叶斯压缩感知多目标定位算法

发布时间：2018-06-07 10:14

本文选题：多目标定位 + 贝叶斯压缩感知　；参考：《哈尔滨工程大学学报》2014年10期

【摘要】：针对室内多目标基于无线信号强度定位中的数据采集和精确度问题,引入基于贝叶斯压缩感知和拉普拉斯先验模型算法,从而满足在达到所需定位精确度的同时降低网络系统开销。所提出的方法是基于接收信号强度来感知位置变化,各移动设备上利用随机投影对接收到的信号强度进行压缩并传输,在采集中心通过基于拉普拉斯先验的贝叶斯压缩感知重构算法并结合最大似然函数法和迭代逼近法计算出各移动设备的位置。仿真结果表明了利用贝叶斯压缩感知重构算法实现室内多个移动设备的定位具有较高精确度,与orthogonal matching pursuit(OMP)重构算法相比较其定位精度至少提高了52.2%,与basis pursuit(BP)重构算法相比较至少提高了13.7%。
[Abstract]:Aiming at the problem of data acquisition and accuracy in indoor multi-target location based on wireless signal intensity, the Bayesian compression sensing and Laplace priori model algorithms are introduced. In order to achieve the desired positioning accuracy while reducing the network overhead. The proposed method is based on the received signal strength to perceive the change of position, each mobile device uses random projection to compress and transmit the received signal strength. In the acquisition center, the location of each mobile device is calculated by means of Bayesian compression perception reconstruction algorithm based on Laplacian priori, combined with the maximum likelihood function method and iterative approximation method. The simulation results show that the localization of indoor mobile devices using Bayesian compression perception reconstruction algorithm has a high accuracy. Compared with the orthogonal matching pursuit reconstruction algorithm, the positioning accuracy is at least 52.2% higher than that of the basis pursuit reconstruction algorithm, and at least 13.7% higher than that of the basis pursuit reconstruction algorithm.
【作者单位】：浙江工业大学信息学院;浙江工业大学艺术学院;
【基金】：浙江省自然科学基金资助项目(LY13F010011)
【分类号】：TN92;TP18

【参考文献】