Experimental testing and analysis of regression algorithms for spectrum sensing in cognitive radio networks

The allocation policy has resulted in some bands becoming oversaturated and hindering the efficient utilization of these frequencies by unlicensed users. Cognitive radio can dynamically allocate these users across different bands. The utilization of spectrum prediction techniques can help improve the efficiency of cognitive radio networks (CRN) by lowering their energy consumption and reducing the number of users. There are various kinds of prediction techniques that are utilized in the spectrum, such as regression analysis, machine learning, and the Markov model. The paper presents a framework for analyzing the spectrum's presence and absence of a licensed user with the help of machine learning. It is evaluated by taking into account the different kinds of regression techniques and their performance on a set of data that's been created using a software-defined testbed. The performance of the proposed method has been compared with that of the literature. It shows that the accuracy has significantly improved.