基于RSSI的笼内鸡只定位方法研究

Research on positioning method of chickens in cage based on RSSI

为快速准确对笼养环境下鸡只多目标定位,设计了1种基于接收信号强度(Received Signal Strength Indicator,RSSI)融合混合滤波算法的笼内鸡只自动定位方法。首先设计了RSSI值采集系统,由PC机(笔记本电脑)、FU-M6-A型读写器、极化相匹配的天线和脚踝标签、读写器软件构成。其次,合理地布置天线,建立了信号衰减测距模型,使用狄克逊高斯混合滤波算法将原始数据预处理,优化笼养环境下信号传播路径损耗系数。最后计算脚踝标签到发射天线之间的距离,将测距值代入极大似然估计法求得脚踝标签的位置坐标初值,以标签坐标和测距值作为未知量采用牛顿迭代法求得最优的标签位置坐标。结果表明:所设计定位方法的定位误差在5 cm以内的概率接近83%,和传统的节点—节点定位算法相比,该方法提高了鸡只在移动过程中的定位精度,为鸡只行为分析提供了1种方法。

In order to quickly and accurately locate the multiple targets of chickens in a cage environment,this paper designed an automatic location method for chickens in cages based on fusion of the received signal strength(RSSI)and hybrid filtering algorithm.First,the RSSI value acquisition system was designed,which was composed of a PC(laptop),a FU-M6-A reader,an antenna and ankle tags with matching polarization,and a demo of the reader.Secondly,the antennas were arranged reasonably.A signal attenuation ranging model was established,and the original data was preprocessed using the Dixon-Gaussian mixture filter algorithm to optimize the signal propagation path loss coefficient in the cage environment.Finally,the distance between the ankle tag and the transmitting antenna was calculated,whose value was operated with the maximum likelihood estimation method to obtain the initial value of the position coordinate of the ankle tags.The Newton iteration method was used to obtain the optimal label position coordinates setting the label coordinates and the ranging value as unknown.The results showed that less than 5 cm position error of the designed positioning method was close to 83%.Compared with the traditional node-node positioning algorithm,this method improved the positioning accuracy of moving chickens and provided a method for chicken behavior analysis.