Lamport Certificateless Signcryption Deep Neural Networks for Data Aggregation Security in WSN

Confidentiality and data integrity are essential paradigms in aggregation owing to the various cyberattacks wireless sensor networks (WSNs). This study proposes a novel technique named Lamport certificateless signcryption-based shift-invariant connectionist artificial deep neural (LCS-SICADNN) by using develop security model. model utilises input layer with several nodes, four hidden layers overcome different attacks (data injection, compromised node, Sybil black hole attacks) output analyse given input. The one-time signcryption involving three processes (key generation, unsigncryption) is adopted achieve secure transmission between sender receiver. Firstly, one-way function executed generate public private keys for each node WSN. Secondly, digital signature generation encryption both performed. which handles encryption, forwards aggregator node. Then, receiver verifies sender’s key during decryption. Thus, can be guaranteed. Finally, authorised aggregates much higher confidentiality privacy. Performance analysis conducted simulating proposed LCS-SICADNN conventional models. Results of comparison indicate that improve throughput lesser delay, packet drop overhead compared methods.