Channel prediction in cognitive radio network using hierarchical neural system for throughput maximization
Loading...
Date
2019-09
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
G.B. Pant University of Agriculture and Technology, Pantnagar - 263145 (Uttarakhand)
Abstract
Wireless Communication has shown exponential upsurge in demand of wireless devices. Due to ever increasing demand of wireless communication, there is spectrum scarcity which is caused by a fixed spectrum allocation policy. Studies on usage pattern reveal that most of the allocated spectrum experiences low utilization under fixed spectrum allocation regime and spectrum occupancy varies from milliseconds to hours. This motivated the concept of frequency reuse that would allow secondary networks to borrow unused radio spectrum from primary licensed networks (users). The core technology behind frequency reuse is Cognitive Radio for which one of the essential components is channel sensing, i.e., the wireless devices can sense the radio spectrum environment within their operating range to detect frequency bands that are not occupied by primary users. This thesis presents a study on maximizing the throughput for secondary user by predicting the primary free channel in advance to save time in sensing and maximizing transmission time in a frame. The study involves exhaustive exploration of various classifiers for predictive decision making and later development of hierarchal neural network model for predicting best possible channel for transmission in following frame. It starts with performance analysis of spectrum sensing techniques, energy detector and matched filter are discussed with simulation results. On the basis of these results, energy detector is found best technique which is further analysed in fading scenario and with mobile SU. There exists a sensing-throughput trade off, hence proposed model focuses on parallel processing for predicting the next channel while primary user continues to transmit in current frame. The model considers three type of radio channel condition scenario namely, constant fading, cyclostationary fading and fast fading. The performance of the system is found to be closer to upper bound for best channel throughput and results are discussed in comparison to lower bound statistics.
Description
Keywords
null