SVM based Throughput Maximization in Cognitive Radio
Adidela Rajya Lakshmi, Jada Shalini Singh · 2022 6th International Conference on Intelligent Computing and Control Systems (ICICCS) · 2022
in wireless networks, spectrum management has become a major issue. Utilizing the spectrum to its full potential among the many users, on the other hand, is a difficult task. During the preceding Wireless communication has seen a substantial surge in applications during the past two decades. Fixed spectrum band allocation schemes, on the other hand, are incapable of managing several requirements at the same time, which is a critical requirement of future wireless applications. Furthermore, in wireless networks, license users (PU’s) are only connected on occasion, resulting in spectrum underutilization. There are some flaws with the existing project, such as the fact that it is only relevant to a single system. Channel allocation is not optimistic, data transmission delays and overlaps occur, network function virtualization, and mobile edge computing are all issues that the proposed project would address. The contributed works successfully address the aforementioned limitation by utilizing SVM and energy sensing. Local spectrum sensing approaches, such as energy detection, employing standalone Cognitive Radio devices, on the other hand, may come to the wrong conclusion about the presence of a primary transmitter for a variety of reasons (SVM, PD, and PF). Cooperative sensing takes advantage of the spatial variance of SU’s to lessen the uncertainty produced by those components, then draws a single global conclusion about the existence or absence of the PU. SVM is a supervised machine learning approach with two phases: training and testing. In the training phase, the probability of detection is calculated by counting the number of successful detections among a secondary user’s total trials, allowing for maximum throughput.