A BIPMU-based network security situation assessment method for wireless network

Abstract

With the rapid development of information technology, wireless network has been accepted by the public and widely used in the daily life and work. Especially with the comprehensive deployment of 5G network technology, the application scope of wireless network is continuously increasing, which can greatly reduce the IT infrastructure overhead and human resource investment. Although the tremendous advantages, wireless network inevitably suffers from some severe security challenges. For example, the network data is characterized by large volumes, variety and high dimensions, which would greatly affect the efficiency and accuracy of network security situation assessment (NSSA). To solve this problem, we first design a novel model based on parsimonious memory unit (PMU), namely, bidirectional parsimonious memory unit (BIPMU). Compared with PMU, BIPMU can not only learn and characterize data through its time series relationship, but also comprehensively and effectively manage the potential connection of long-term and short-term dependence on time series data. Subsequently, we adopt BIPMU to design a novel NSSA method to evaluate the real-time security situation of wireless network. Finally, we develop a prototype implementation of our proposed NSSA method and provide the performance evaluation. The experimental results demonstrate that compared with the previous NSSA methods, our proposed NSSA method is much more attractive in efficiency and accuracy.