A Double-Layer Indemnity Enhancement Using LSTM and HASH Function Technique for Intrusion Detection System

The Intrusion Detection System (IDS) is the most widely used network security mechanism for distinguishing between normal and malicious traffic activities. It aids in that it may identify unforeseen hazards traffic. Several techniques have been put forth by different researchers intrusion detection. However, because attacks increased dramatically, making difficult to execute precise detection rates quickly, demand effectively recognizing incursion growing. This research proposed an improved solution uses Long Short-Term Memory (LSTM) hash functions construct a revolutionary double-layer IoT Network Detection. presented framework utilizes standard well-known real-time IDS datasets such as KDDCUP99 UNSWNB-15. In framework, dataset was pre-processed, employed Shuffle Shepherd Optimization (SSO) algorithm tracking informative attributes from filtered database. Further, designed model LSTM classifying precisely. Finally, secure function SHA3-256 utilized countering attacks. intensive experimental assessment of approach with conventional algorithms emphasized efficiency terms accuracy, precision, recall, etc. analysis showed attained attack prediction accuracy 99.92% 99.91% UNSWNB-15, respectively.