Survey on Modulation Classification Methods for Cognitive Radio in Wireless Sensor Network

Abstract

The survey on modulation categorization for cognitive radio applications is the subject of this paper. More spectrum must now be assigned since more people are using mobile devices. Due to the limited availability of spectrum resources, it is essential to correctly detect main users' channel activity before allocating any RF spectrum to secondary users (SU). Spectrum sensing is used to do this job. The kind of modulation signal that is being sent across the channel and that has been received at the receiver side must be correctly identified. In this instance, determining the kind of modulation signal is greatly aided by modulation categorization. The complexity of the traditional techniques in modulation categorization is higher, but machine learning and deep learning provide more effective categorization. This article provides a detailed overview of the modulation categorization algorithms currently in use. From the survey, it can be shown that ML/DL-based strategies perform more effectively than traditional ones.