基于自然选择的线性递减权重PSO与Taylor算法的TDOA协同定位算法研究

TDOA co-location method based on natural selection linear decreasing weight PSO and Taylor algorithm

针对Taylor算法进行TDOA定位时,其初始估计位置的误差易导致Taylor算法不收敛和定位精度差的问题,提出一种基于自然选择的线性递减权重粒子群优化(W-SPSO)与Taylor算法协同定位的方法。该方法先通过W-SPSO算法得到一个初始估计位置(x,y),再通过Taylor算法在(x,y)处进行迭代运算得到最终定位结果。不同噪声情况下的仿真结果显示:W-SPSO与Taylor算法协同定位方法对MS坐标估计值的均方差(RMSE)小于标准PSO(粒子群优化)、SelPSO(基于自然选择的粒子群优化算法)、W-SPSO、Taylor以及Chan五种算法的RMSE。因此,所提出的定位方法在保留了SelPSO算法求解精度和收敛性的基础上,同时提高了全局搜索能力,使其具有更高的定位精度和收敛性。

When Taylor algorithm was used for TDOA positioning, the initial estimated position error easily led to the Taylor algorithm does not converge and the shortcomings of low positioning accuracy. To solve this problem, this paper proposed a co- located positioning method, which was based on natural selection linear decreasing weight particle swarm optimization algo- rithm （W-SPSO） and Taylor algorithm. This method used W-SPSO to get an initial estimated position （x, y）, and then got the final position by iterative calculation of Taylor algorithm at the （x, y）. Simulation results in different noise shows that the RMSE（the mean square error of the estimates coordinates ） of the co-location method is less than that of standard PSO（parti- cle optimization algorithm） ,SelPSO （particle optimization algorithm based on natural selection） ,W-SPSO,Taylor and Chan al- gorithms. Therefore, the co-location method retains solution accuracy and convergence of SelPSO algorithm, while improving the global search capability, the co-location method has higher positioning accuracy and convergence.