摘要
引入象群算法,对无线传感网络入侵检测技术予以优化。在传统算法的基础上加入了飞行及融合粒子策略,以提高象群算法的全局搜索能力和收敛速度。将象群算法与Spark分析相结合,以缩短分类时间、提高稳定性;同时,SVM函数相融合,以增强算法的精准度、降低误报率。象群算法优化模型效能评测结果表明:与传统算法相比,优化后的象群算法入侵检测结果精准度约提高5.548%,平均误报率约降低3.074%,整体稳定性和运行效率有所提高。
In this paper,the image group algorithm is introduced to optimize the infinite sensor network intrusion detection technology.In order to improve the global search ability and convergence speed of the image swarm optimization algorithm,the flight and fusion particle strategy is added to the algorithm.The combination of spark algorithm and image classification algorithm improves the stability and implementation time of image group.Meanwhile,the fusion of image group algorithm and SVM function is realized to enhance the accuracy of the algorithm and reduce the false alarm rate.The effectiveness evaluation results of the optimization model of image group algorithm show that compared with the traditional algorithm,the accuracy of the improved algorithm is improved by about 5.548%,the average false alarm rate is reduced by about 3.074%,and the overall stability and operation efficiency are improved.
作者
周跃
ZHOU Yue(Anhui Industry Polytechnic,Tongling Anhui 244000,China)
出处
《重庆科技学院学报（自然科学版）》
CAS
2021年第4期94-100,共7页
Journal of Chongqing University of Science and Technology：Natural Sciences Edition
基金
安徽省质量工程项目“周跃技术技能型大师工作室”(2019DSGZS53)。
关键词
EHO算法
无线传感网络
入侵检测
特征向量
参数优化
EHO algorithm
wireless sensor network
intrusion detection
feature vector
parameter optimization
