SECURITY MEASUREMENT IN LTE/LTE-A NETWORK BASED ON zS-LR FEATURE SELECTION TECHNIQUE AND UM-tGAN ATTACK DETECTION TECHNIQUE

Abstract

Regarding raised data speeds, minimized access latency, robust resilience to multipath fading, enhanced spectral efficacy, and seamless integration, the LTE network performs great with other non-LTE (Long Term Evolution) wireless technologies. Nevertheless, there is an urgent need to execute extensive network Security Measurements (SMs) since the LTE/LTE-A (LTE-Advanced) network is hugely exposed to several security hazards. Grounded on z-Score-centric Lasso Regression (zS-LR) Feature Selection (FS) and Unified Momentum-based Tuned Generative Adversarial Network (GAN) (UM-tGAN) attack detection methodologies, the work has introduced precise SMs in LTE networks for performing accurate security measures. The effectual relevant data are gathered by the proposed system along with guarantees the quick as well as precise security attacks detection; in addition, it offers the base for network SM. By performing pre-processing like Missing Value (MV) handling, categorical features handling, scaling using MAD-PT, and imbalanced dataset handling that is succeeded by related FS employing the zS-LR technique, the gathered data are transmitted into structured data. In the end, by deploying the UM-tGAN method, the dimensional reduced features are trained and tested. Owing to heterogeneity, the proposed work is able to handle several attacks along with avoids adversarial-centric attacks. When weighed against the prevailing methodology, the proposed work acquires enhanced accuracy, precision, and sensitivity and avoids misclassification of attack by attaining a lower false prediction value.