传感器网络中阈值Nesterov加速梯度下降定位方法

Threshold Nesterov Accelerating Gradient Descent Location Method in Sensor Network

在传感器网络定位问题中,利用接收信号强度RSSI(Received Signal Strength Indication)的定位方法存在着接收信号传播不稳定,定位精度较低的问题。为解决该问题,提出了一种基于阈值Nesterov加速梯度下降NAGT(Nesterov Accelerated Gradient Descent with Threshold)的RSSI定位算法。算法引入Nesterov思想,不断更新寻优动量,以达到损失函数最小,从而求取对应的未知基站坐标,通过增设阈值,降低了算法陷入局部最优的概率。经仿真比较分析,NAGT方法相对于粒子群算法与随机梯度法,在定位精度与效率上有着较为明显的优势。

Among sensor network location issues,The location method based on Received Signal Strength Indication（RSSI）remains several problems including unstable signal propagation and low positioning accuracy. In order to solve this problem,combing RSSI and Nesterov Accelerated Gradient Descent with Threshold（NAGT）,a novel location algorithm is proposed. For the purpose of obtaining unknown base station coordinate,the algorithm introduces the idea of Nesterov to update the optimal momentum to minimize the loss function. By adding the threshold,the probability of converging to local optimum is also reduced. Through simulation experiments,NAGT has an obvious advantage over the particle swarm algorithm and the stochastic gradient method in location accuracy and efficiency.