Enhancing Cognitive Radio Network Design with New Energy Detection versus Pilot and Radio-Based Techniques

Abstract

This study aimed to enhance the energy efficiency (EE) and accuracy of the Cognitive Radio Network (CRN)
system design by using a unique energy detection approach, contrasting it with the conventional Pilot and
Radio Based Detection Technique. A model was developed and processed in Python, using a network dataset
for initial exploration, sourced from the UCI Machine Learning Repository. Statistically, with a confidence
interval of 95% and sample size of 140, the energy detection's precision was assessed. In evaluating spectrum
allocation, the conventional technique had a slightly higher accuracy. However, our proposed energy detection
method achieved an impressive 95.2713% accuracy. Surprisingly, it processed in just 4 seconds, half the time
taken by the conventional method. The results confirm the new method's superiority in energy efficiency.