计算机科学与探索 ›› 2018, Vol. 12 ›› Issue (10): 1633-1644.DOI: 10.3778/j.issn.1673-9418.1801011

• 人工智能与模式识别 • 上一篇    下一篇

萤火虫优化和随机森林的WSN异常数据检测

许欧阳1,李光辉1,2,3+   

  1. 1. 江南大学 物联网工程学院,江苏 无锡 214122
    2. 江苏省无线传感网高技术研究重点实验室,南京 210003
    3. 物联网技术应用教育部工程技术研究中心,江苏 无锡 214122
  • 出版日期:2018-10-01 发布日期:2018-10-08

Anomaly Data Detection Using Glowworm Optimization and Random Forest in Wireless Sensor Networks

XU Ouyang1, LI Guanghui1,2,3+   

  1. 1. School of Internet of Things Engineering, Jiangnan University, Wuxi, Jiangsu 214122, China
    2. Jiangsu High Technology Research Key Laboratory for Wireless Sensor Networks, Nanjing 210003, China
    3. Research Center of IoT Technology Application Engineering (MOE), Wuxi, Jiangsu 214122, China
  • Online:2018-10-01 Published:2018-10-08

摘要: 异常数据检测在无线传感器网络(wireless sensor network,WSN)环境监测系统中发挥重要作用。针对传统的随机森林(random forest,RF)算法因冗余决策树导致异常数据检测效率不高的问题,根据选择性集成思想,提出了一种基于变异二进制萤火虫算法(mutation binary glowworm swarm optimization,MBGSO)以及自适应更新随机森林的WSN异常数据检测算法MBGSO-ARF。该算法使用改进的BGSO算法优化RF进行选择性集成以得到最优子集成分类器,并使得检测模型随数据流的变化而自适应更新,提高了检测准确性并节省了检测时间,对优化算法MBGSO和二进制粒子群算法(binary particle swarm optimization,BPSO)进行了实验对比。仿真实验结果表明:该优化算法优于BPSO算法,MBGSO-ARF算法在准确率上较其余算法都有提升,且集成模型大小得到了压缩。以上结果证明了MBGSO-ARF算法的有效性。

关键词: 异常检测, 随机森林, 无线传感器网络(WSN), 萤火虫算法, 选择性集成学习

Abstract: Data anomaly detection plays an important role in wireless sensor network (WSN) application systems for environmental monitoring. The redundant decision trees decrease the efficiency of anomaly detection algorithm based on traditional random forest (RF). To address this issue, a new anomaly detection algorithm (MBGSO-ARF) for WSNs using the theory of selective ensemble is proposed in this paper. MBGSO-ARF is based on mutation binary glowworm swarm optimization (MBGSO) and adaptive updating random forest (ARF) algorithm. MBGSO-ARF uses the improved BGSO algorithm to optimize the RF for selective ensemble to obtain the optimal sub-ensemble classifiers, which reduces the ensemble scale of the detection model, and can update the model adaptively with the evolution of data stream. Thus the accuracy of anomaly detection is improved and the time cost is reduced. An experimental comparison between MBGSO and binary particle swarm optimization (BPSO) is given. Simulation results show that the proposed algorithm is superior to BPSO algorithm, and MBGSO-ARF algorithm has higher accuracy than the other algorithms. And the size of the ensemble model is compressed. The above results demonstrate the effectiveness of MBGSO-ARF.

Key words: anomaly detection, random forest, wireless sensor network (WSN), glowworm swarm optimization, selective ensemble learning