采用压缩感知和GM（1,1)的无线传感器网络异常检测方法

刘洲洲 1; 2; 李士宁 2

doi:10.7652/xjtuxb201702007

您当前的位置：

首页 >

文章列表页 >

采用压缩感知和GM（1,1)的无线传感器网络异常检测方法

更新时间：2025-07-09

- 采用压缩感知和GM（1,1)的无线传感器网络异常检测方法
- An Anomaly Detection Method for Wireless Sensor Networks Based on Compressed Sensing and GM(1,1)
- 西安交通大学学报 2017年51卷第2期页码：40-46
- 作者机构：
  
  1. 西安航空学院电子工程学院,西安,710077
  2. 西北工业大学计算机学院,西安,710072
- 作者简介：
- 基金信息：
  
  : 2016-06-28。作者简介: 刘洲洲(1981—),男,博士,副教授。基金项目: 国家自然科学基金资助项目(61401499);陕西省教育厅科研计划项目资助(16JK1395)
- DOI：10.7652/xjtuxb201702007
  中图分类号： TP393
- 网络首发：2017-02-10，
  
  纸质出版：2017
- 稿件说明：
移动端阅览
刘洲洲 1, 2, 李士宁 2. 采用压缩感知和GM（1,1)的无线传感器网络异常检测方法[J]. 西安交通大学学报, 2017,51(2):40-46.

An Anomaly Detection Method for Wireless Sensor Networks Based on Compressed Sensing and GM(1,1)[J]. 2017, 51(2): 40-46.
刘洲洲 1, 2, 李士宁 2. 采用压缩感知和GM（1,1)的无线传感器网络异常检测方法[J]. 西安交通大学学报, 2017,51(2):40-46. DOI： 10.7652/xjtuxb201702007.

An Anomaly Detection Method for Wireless Sensor Networks Based on Compressed Sensing and GM(1,1)[J]. 2017, 51(2): 40-46. DOI： 10.7652/xjtuxb201702007.

摘要

针对当前无线传感器网络(WSNs)异常检测算法的检测准确率较低同时影响网络能耗均衡的问题

提出了一种基于改进压缩感知(CS)重构算法和智能优化GM(1

1)的WSNs异常检测方法。首先

通过建立双层异质WSNs异常检测模型

并采用压缩感知技术对上层观测节点收集到的下层检测节点温度测量数据进行处理

同时结合温度数据稀疏度未知特点

构造有效的稀疏矩阵和测量矩阵

并重新定义测量矩阵正交变换预处理策略

使得CS观测字典满足约束等距(RIP)条件; 其次

重新定义了离散蜘蛛编码方式

蜘蛛种群不断协同进化

以获得稀疏结果中非零元素的位置信息

利用最小二乘法得到非零元素的幅度信息

实现了对未知数量检测节点数据的精确重构。在此基础上可以由蜘蛛种群迭代进化得到优化后GM(1

1)的参数序列

通过检测参数序列的相关阈值来判定节点是否发生异常。实验仿真结果表明

与OMP-IGM等异常检测方法相比

该方法的异常检测准确率提高了约7%～33%

网络能耗降低了约18%～43%。

Abstract

A new anomaly detection scheme for wireless sensor networks(WSNs)based on an improved reconstruction method of compressed sensing(CS)and the intelligent optimizing GM(1

1)is proposed to improve the accuracy of existing anomaly detection algorithms and to reduce the network energy consumption. A double WSNs heterogeneous anomaly detection model is established

and the CS technology is used to process the upper observation nodes data collected from lower detection nodes. An effective sparse matrix and a measurement matrix are constructed by combining the anomaly detection model characteristics

then an orthogonal transformation pretreatment strategy is redefined for the measurement matrix such that the observation dictionary of CS satisfies the restricted isometry property(RIP). Since the data sparsity for CS is unknown

a new CS reconstruction algorithm based on discrete social spider optimization algorithm is proposed to realize the accurate reconstruction of the detection node data

and an improved GM(1

1)intelligent optimization scheme for anomaly detection is designed to achieve a reliable prediction of abnormal nodes in the network. The parameters of GM(1

1)are optimized through the iteration of the spider population

and abnormalities of nodes are determined by detecting the relevant thresholds of the parameter sequences. Experimental simulation results and comparisons with other anomaly detection algorithms show that the accuracy of the proposed scheme increases by about 7% to 33%

and the network energy consumption reduces by about 18% to 43%.

关键词

Keywords

references

穆天圆, 乔学工, 张敏. 基于Voronoi图的蜂群优化算法在WSN覆盖中的应用 [J]. 传感技术学报, 2015, 28(10): 1525-1530.

MU Tianyuan, QIAO Xuegong, ZHANG Min. Application of bee colony optimization algorithm based on Voronoi graph in WSN coverage [J]. Journal of Sensing Technology, 2015, 28(10): 1525-1530.

张波, 刘郁林, 王开, 等. 基于概率稀疏随机矩阵的压缩数据收集方法 [J]. 电子与信息学报, 2014, 36(6): 1478-1484.

ZHANG Bo, LIU Yulin, WANG Kai, et al. Compression of probabilistic sparse random based on the matrix data collection method [J]. Journal of Electronics and Information Technology, 2014, 36(6): 1478-1484.

肖政宏, 陈志刚, 李庆华. WSN中基于分布式机器学习的异常检测仿真研究 [J]. 系统仿真学报, 2011, 23(1): 121-127.

XIAO Zhenghong, CHEN Zhigang, LI Qinghua. WSN based distributed machine learning anomaly detection simulation [J]. Journal of System Simulation, 2011, 23(1): 121-127.

WANG Jin, TANG Shaojie, YIN Baocai, et al. Distributed compressive sampling for lifetime optimization in dense wireless sensor networks through intelligent compressive sensing [C]∥IEEE International Conference on Computer Communications. Piscataway, NJ, USA: IEEE, 2012: 603-611.

DONOHO D. Compressed sensing [J]. IEEE Transactions on Information Theory, 2006, 52(4): 1289-1306.

李洪成, 吴晓平, 严博. 面向MANET异常检测的分布式遗传k-means研究 [J]. 通信学报, 2015, 36(11): 167-173.

LI Hongcheng, WU Xiaoping, YAN Bo. Distributed genetic k-means for MANET anomaly detection [J]. Journal of Communication, 2015, 36(11): 167-173.

吴迪, 王奎民, 赵玉新, 等. 分段正则化正交匹配追踪算法 [J]. 光学精密工程, 2014, 22(5): 1395-1402.

WU Di, WANG Kuimin, ZHAO Yuxin, et al. Piecewise regularized orthogonal matching pursuit algorithm [J]. Optics and Precision Engineering, 2014, 22(5): 1395-1402.

焦李成, 杨淑媛, 刘芳, 等. 压缩感知回顾与展望 [J]. 电子学报, 2011, 39(7): 1651-1662.

JIAO Licheng, YANG Shuyuan, LIU Fang, et al. Review and prospect of compressed sensing [J]. Journal of Electronics, 2011, 39(7): 1651-1662.

CUEVAS E, CIENFUEGOS M, ZALDIVA D, et al. A swarm optimization algorithm inspired in the behavior of the social spider [J]. Expert System with Applications, 2013, 40(16): 6374-6384.

江艺羡, 张岐山. 近似非齐次无偏GM(1, 1)模型的递推解法及应用 [J]. 控制与决策, 2015, 30(12): 2199-2204.

JIANG Yixian, ZHANG Qishan. Approximate nonhomogeneous unbiased GM(1, 1)recursive solution and application of model [J]. Control and Decision, 2015, 30(12): 2199-2204.

CHENGC T, TSEC K, LAU F C M. A delay-aware data collection network structure for wireless sensor networks [J]. IEEE Sensors Journal, 2011, 11(3): 699-710.

王艳娇, 李晓杰, 肖婧. 基于动态学习策略的群集蜘蛛优化算法 [J]. 控制与决策, 2015, 30(9): 1575-1582.

WANG Yanjiao, LI Xiaojie, XIAO Jing. The cluster optimization algorithm based on spider dynamic learning strategy [J]. Control and Decision, 2015, 30(9): 1575-1582.

赵秀兰, 李克清. 弱稀疏性下的无线传感器网络事件检测算法 [J]. 计算机应用与软件, 2014, 31(3): 104-107.

ZHAO Xiulan, LI Keqing. An event detection algorithm for wireless sensor networks with weak sparsity [J]. Computer Applications and Software, 2014, 31(3): 104-107.

李鹏, 王建新, 曹建农. 无线传感器网络中基于压缩感知和GM(1, 1)的异常检测方案 [J]. 电子与信息学报, 2015, 37(7): 1586-1590.

LI Peng WANG Jianxin, CAO Jiannong. Anomaly detection scheme based on compressed sensing and GM(1, 1)in wireless sensor networks [J]. Journal of Electronics and Information, 2015, 37(7): 1586-1590.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

无线传感器网络节点协作的节能路由传输

采用时域测量矩阵的压缩感知稀疏信道估计方法

无线传感器网络可分负载调度算法

无线传感器网络分区能耗均衡的非均匀分簇算法

一种基于接收信号强度指示的改进型定位算法