Abstract:<p>Spectrum sensing ranks amongst cognitive radio's most critical and demanding activities. For radio resource networks to use idle licensed bandwidth, secondary users need to identify the primary user's signal precisely. Cognitive radio networks (CRNs) have been proposed to improve the efficiency of spectrum use to meet the increasing demand for wireless bandwidth; they allow channel communication between secondary (unlicensed) and primary (licensed) users without interruption. A secondary user in a CRN must at all times track the presence of a primary user's signal to prevent the primary user (PU) from attempting to intervene whether the statistical variance of the PU data traffic would boost the prediction outcome provided by multiple current teaching algorithms in estimating the PU's average OFF time and the need for the best adaptive learning to predict accurate PU activities. The objective of the work has been modeled to identify the PUs signal. First, the data has been created via quadrature phase shift keying modulation and quadrature amplitude modulation by convolutional neural network (CNN). A dataset with different orthogonal frequency-division (OFDM) multiple access signals is generated. Then cyclo-stationary functions are extracted using the accumulation process fast Fourier Transform. This research aims to use glow worm swarm optimization (GWSO) to automatically check for the optimal design of CNNs without any manual work being required. Also, k-Nearest Neighbors (kNN) replaces Softmax, as it outperforms conventional CNN architectures with noisy signals. A comparison of the proposed classifier with another machine learning classifier, namely support vector machine, Decision tree (DT), clearly shows that SVM and DT are outperformed by the improved convolution neural network-glow worm swarm optimization (ICNN-GWSO) classification. Results of simulation show that the improved scheme outperforms conventional spectrum sensing systems, both in fading and non-fading settings, where efficiency is measured using measures such as chance, the total likelihood of error, and the ability to optimize data transfer possibilities.</p>

Analysis of Spectrum Occupancy Using Machine Learning Algorithms

Spectrum Occupancy Prediction for Realistic Traffic Scenarios: Time Series versus Learning-Based Models

Ensemble Classification-Based Spectrum Sensing Using Support Vector Machine for CRN

Survey on Spectrum Allocation Algorithms for Cognitive Radio Networks

Machine Learning Techniques for Spectrum Sensing when Primary User Has Multiple Transmit Powers

Spectrum Mapping in Large-Scale Cognitive Radio Networks with Historical Spectrum Decision Results Learning

Spectrum Sensing with Small-Sized Data Sets in Cognitive Radio: Algorithms and Analysis.

Analysis of cognitive radio spectrum access with finite primary users and infinite secondary users

The application and comparison between machine learning algorithms in cooperative spectrum sensing

Artificial Neural Networks, Support Vector Machine And Energy Detection For Spectrum Sensing Based On Real Signals

An effective spectrum sensing in cognitive radio networks using improved convolution neural network by glow worm swarm algorithm

Som-Ga-Svm Detection Based Spectrum Sensing in Cognitive Radio

SVM-based Spectrum Mobility Prediction Scheme in Mobile Cognitive Radio Networks.

Discrimination of primary user emulation attack on cognitive radio networks using machine learning based spectrum sensing scheme

Machine-learning-based spectrum sensing enhancement for software-defined radio applications

A Fuzzy Support Vector Machine Algorithm for Cooperative Spectrum Sensing with Noise Uncertainty

A Machine Learning-Enabled Spectrum Sensing Method for OFDM Systems

Human Occupancy Detection via Passive Cognitive Radio

Spectrum Hole Prediction Based On Historical Data: A Neural Network Approach

Artificial neural network design for improved spectrum sensing in cognitive radio

Effective energy detection using machine learning techniques for cooperative sensing in cognitive radio networks