基于稀疏化LSSVM的物联网轻量级入侵检测方法

为适应物联网感知层节点计算能力弱、能量有限和存储空间不足等特点,提出基于稀疏化最小二乘支持向量机的物联网轻量级入侵检测方法,以最小二乘支持向量机作为分类器,通过改进的K均值数据稀疏和自适应剪枝的模型稀疏方法,使模型更好适应物联网苛刻的资源环境.实验测试结果表明:入侵检测模型的F1值达到0.9268,模型大小减少到81....

A lightweight intrusion detection method for the internet of things based on sparse LSSVM

In order to adapt to the weak computing capability, restricted energy and insufficient storage space of the sensing nodes in the pereception layer of internet of things (IoT), we propose a lightweight intrusion detection method based on sparse least square support vector machine (LSSVM), LSSVM is used as a classifier. This is helpful to better adapt to the harsh resource environment of IoT by the improved K-mean data sparsity and model sparseness of adaptive pruning. The results of experiments prove that the F1 value of the intrusion detection model reaches 0.926 8, and the size of the model is reduced to 81.3 KB. The proposed method can better adapt to the IoT scenarios and satisfy its demand in information security.