A Meta Heuristic Inverse Ant Colony Optimization Based ADA Boost Classsifer (Io-HLBAB) For Cogntive Networks To Aid In Channel State Estimation

Abstract

Increasing scarcity in radio frequency spectrum has been on the rising trend off late with increasing state of art utilities demanding higher end of radio frequency spectrum. This has become more pronounced especially with the advent of 5G networks which are characterized by high data transfer rates and large bandwidth. Cognitive radio networks prove to be a promising solution to this increasing demand for radio spectrum by utilizing a set of intelligent mechanisms to allocate frequency band to the users based on a series of sensing and decision making. An opportunistic spectrum allocation policy through an optimized machine learning model with inbuilt classifier has been proposed in this research paper to improve the secondary user reallocation time in case of returning primary user thereby reducing the overall waiting time and loss of data. A hybrid Ada boost classifier based machine learning method has been proposed and implemented in this paper to arrive at a decision regarding primary user activity in received signal strength. An inverse ant colony optimization model (I-ACO) has been integrated into the machine learning model to detect availability of spectrum holes for secondary users in a condition of returning PUs. The proposed inverse optimized (IO-HBALB) has been compared against state of art and recent techniques and improved resource allocation with reduced collision analysis has been observed and recorded

Keywords: Cognitive radio networks, machine learning model, ant colony optimization, channel state estimation, spectrum hole detection.