No abstract

In today's wireless communication technology, spectrum occupancy is one of the major challenge. To perform all the task in wireless communication intelligently, Cognitive Radio (CR) is used. With the help of machine learning techniques, performance of CR will increase. In this paper, implementation of spectrum sensing (SS) in Cognitive Radio Network (CRN) is presented. To check the availability of spectrum, the supervised Machine Learning (ML) and conventional spectrum sensing method is used. To classify signal and noise, the Artificial Neural Network (ANN) classifier is used. The classifier's result shows better result than conventional method's result.

In communication system spectrum has a crucial role and wireless technologies are increasing rapidly it is required to make efficient use of spectrum to satisfy the spectrum scarcity problem. Using spectrum efficiently can be done by cognitive radio because of its ability to sense surrounding environment. Cognitive radio sense unoccupied spectrum by detecting the primary users' presence or absence in the spectrum. Using more than one cognitive radio in the detection process will increase the efficiency of spectrum usage and prevent interference between signals. Using machine learning techniques for the implementation of intelligent cognitive radio increase efficiency and detection performance and detect signal at low SNR condition. Fuzzy logic machine learning technique is implemented which is based on fuzzy membership functions and fusion centre. Energy detection is used to classify signal and noise at each cognitive radio after that each cognitive radio output information convert into membership function and apply fuzzy rules such as algebraic sum, the algebraic product give a final decision about the signal presence or absence. Simulation results show that the proposed system gives much better results compared to the conventional energy detection system and improves the performance of the system.

No abstract

No abstract