A Collaborative-Enhanced Sand Cat Swarm Optimization for Network Intrusion Detection

Due to the problems of redundant features and high dimensionality in collaborative intrusion detection, the direct use of the Sand Cat Swarm Optimization (SCSO) algorithm performs poorly and is prone to fall into local optimality. Thus, this paper proposes a Collaborative-Enhanced Sand Cat Optimization (CESCSO) algorithm to solve these problems for the purpose of promoting collaborative intrusion detection. Firstly, a better balance between the two phases of exploration and exploitation is achieved by introducing a non-linear transformation factor, which lays the foundation for the global search for the optimal solution. Secondly, the golden sine strategy is utilized to prevent individuals from falling into local optimality and improve the exploration ability of the SCSO algorithm. Thirdly, escaping behaviors are added to improve exploration efficiency and accuracy. Finally, a random opposition-based learning strategy is introduced at the end of each iteration to mutate the population and select the global optimal solution. The proposed CESCSO algorithm is then tested and compared with the original SCSO algorithm and other classical algorithms in six benchmark functions. The results show that the proposed CESCSO algorithm is highly competitive in terms of convergence speed and accuracy, and can significantly reduce the number of data dimensions and effectively improve the network intrusion detection performance.