Fuzzy based clustering in CWPSN using machine learning model

Cognitive wireless power sensor network (CWPSN) technology, widely used in almost all fields, has addressed various issues. The researchers have addressed the problems in the lack of radio spectrum availability and enabled the allocation of dynamic spectrum access in specific fields. The main challenge has been to support the radio spectrum allocation using intelligent adaptive learning and decision-making techniques so that various requirements of 5G wireless networks can be encountered. Machine learning (ML) is one of the most promising artificial intelligence tools conceived to support cognitive wireless networks. This paper aims to provide energy optimization and enhance security to cognitive wireless power sensor networks using a novel protocol during resource allocation. In addition to the existing methods, a novel protocol, fuzzy cluster-based greedy algorithms for attack prediction and energy harvesting using a machine-language model based on neural network techniques have been introduced. The simulation has been done using MATLAB software tools which gives efficient results.