Performance Analysis of Prediction Based Spectrum Sensing for Cognitive Radio Networks

Cognitive Radio (CR) was anticipated as a solution to the severe concern in wireless communication i.e. scarcity of accessible spectrum. Spectrum sensing is considered to be the most substantial part in CR. Due to the trade-off between spectrum sensing and throughput of the CR network, the licensed...

Full description

Saved in:

Bibliographic Details
Published in:	2019 2nd International Conference on Intelligent Communication and Computational Techniques (ICCT) pp. 271 - 274
Main Authors:	DasMahapatra, Suddhendu, Patnaik, Sharanya, Sharan, Shivendra Nath, Gupta, Mitali
Format:	Conference Proceeding
Language:	English
Published:	IEEE 01-09-2019
Subjects:	cognitive radio neural network prediction spectrum sensing
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	Cognitive Radio (CR) was anticipated as a solution to the severe concern in wireless communication i.e. scarcity of accessible spectrum. Spectrum sensing is considered to be the most substantial part in CR. Due to the trade-off between spectrum sensing and throughput of the CR network, the licensed users' transmission activity prediction is considered as a potential alternative to spectrum sensing. The present work considers a neural network based multilayer perceptron model to predict the availability of licensed spectrum. Performance of this model is evaluated in different traffic load. Stand-alone prediction model and prediction before sensing model, both are analysed with respect to sensing parameters, false alarm and misdetection probability. MATLAB software is used for simulation.
DOI:	10.1109/ICCT46177.2019.8969028