基于联合分步APIT定位算法的改进

Improved APIT Localization Algorithm Based on Joint Step

针对无线传感网络（wirelesssensornetwork，WSNs）APIT节点定位算法在锚节点低密度环境下的节点误判和节点失效等问题给出了改进，在APICT定位算法的基础上提出了联合分步定位算法UNION-APICT（unionapproximatepoint-in-circumcircleTest），该算法是结合连通性的测距技术，RSSI测距技术以及质心定位和APICT等技术，来联合解决对未知节点定位问题；通过仿真实验结果表明，改进后的UNION-APICT在APICT算法的基础之上平均定位误差减少了10％～25％，定位性能有了明显的提升；随着通信半径R和最大探测距离rmax的增加，定位误差也在逐渐减小，该算法较APIT和APICT定位算法在锚节点密度、节点覆盖率和定位精度上都有所提高。

The classic APIT （approximate Point--in--Triangulation test） algorithm has been utilized in node localization in wireless sen- sor networks for major applications, however, it is easily fall into nodes misjudgments and node failure as its accuracy can not be guaranteed in low node density scenarios. To solve this problem, this paper proposes an improved node localization algorithm UNION--APICT （union approximate Point--in--Circumcircle test） which focuses on improving the localization accuracy by combining the connected ranging method, the RSSI ranging technology, the centroid localization and APICT together, which can solve the unknown node localization problem effective- ly. The simulation experiment results demonstrate that the UNION--APICT can achieve the lower average location error as much as 10% to 25 % than APICT scheme. And when the communication radius R and the maximum detection range Rmax increasing, the node localization error decreasing accordingly. Compared with the APIT and APICT localization algorithms, the new UNION--APICT can get performance improvement significantly both in anchor nodes density, the coverage, and localization accuracy.