基于自适应遗传的E-CID定位算法研究

针对LTE系统NLOS（非视距）环境下基于传统遗传的E-CID（增强小区识别）定位算法过早收敛于某局部最优解而非全局最优，文章提出了一种改进的自适应遗传E-CID定位算法，该算法通过对LTE终端位置数据进行加权最小二乘估算，利用遗传算法进行非线性最优解全局搜索，自适应的改变交叉及变异概率，避免了传统遗传算法过早收敛于局部最优解缺点。仿真结果表明：自适应遗传法比传统遗传算法优势更明显，定位精度更准确。

E-CID location algorithm based on adaptive genetic algorithm

In view of the traditional genetic algorithm for E-CID localization algorithm based on the traditional genetic algorithm in NLOS environment, it converges to a local optimal solution rather than the global optimum. The paper proposes an improved adaptive genetic E-CID algorithm which uses weighted least squares estimation for LTE terminal location data, genetic algorithm for nonlinear global searching optimal solution, the change of the adaptive crossover and mutation probability. It avoids the premature convergence of traditional genetic algorithm in local optimal solution. The simulation results show that the adaptive genetic algorithm is better re than the traditional genetic algorithm and is more accurate in positioning.