An Artificial Intelligence Approach for Cognitive Spectrum Management Jamal Elhachimi, Zouhair Guennoun Laboratory of Electronics and Telecommunications Mohammadia School of Engineers (EMI) Mohamed V – Agdal University(UM5A) Rabat, Morocco Jamal_elhachimi@hotmail.com, zouhair@emi.ac.ma Abstract The dramatically progress of wireless technologies in the last years have augmented the demand for channel resources in a limited spectrum with inextensible character. However, some studies have shown that most of the allocated frequency bands are unused or underutilized. Cognitive radio (CR) has emerged as an efficient technology to exploit the unused available spectrum resources in an opportunistic manner. To take advantage of this concept, we analyzed the criteria for this problem, we determined the common objectives to solve it, and we developed an intelligent technique for spectrum assignment in cognitive radio. Such technique uses an artificial neural network (ANN) for improving wireless communication for cognitive radio mobile terminals. We also discuss the open issues and the prospects for future research in this area. Key word: cognitive radio networks. Wireless Networks. Channel assignment. Neural network. 1. Introduction The enormous growth in demand for wireless communication systems and the area of user needs for new services during the last few decades leads to the growth of number and the popularity of smart mobile devices. Unfortunately, the limited character of available spectrum presents a crucial obstacle to the continued growth of wireless systems and services. A great effort is made in this context and several algorithms were investigated to search for channel assignment to achieve network robustness and maintain minimum network interference [1]. Furthermore, many studies have shown that more than 90% of the allocated spectrum is unused or underutilized [2]. Consequently, new opportunistic approaches for a distributed management and optimal radio access must be developed. Cognitive radio (CR) presents one of the most efficient technologies to remedy the dynamic spectrum management. Cognitive radio was conceived to operate across different spectrum bands and heterogeneous radio access technologies (RAT) [3]. It is worthy to recall that the main idea of this technology is to have a radio terminal (called secondary user - SU), that can sense and access the spectrum when the spectrum is unoccupied by the primary user (PU) [4]. When a PU requests to access its own spectrum, the SUs using the same spectrum opportunistically should switch to other unoccupied spectra to protect the transmission of the PU and continue their own data delivery [1]. However, severe throughput degradation may occur in this situation [1]. Cognitive radio aim consists to improve the spectrum allocation, and investigate different methods and protocols such as game theory to exploit the unused spectrum. CR nodes possess the necessary qualities to make a considerable progress in the reliability of wireless networks [5], [1]. To perform this task, radio terminal should acquire more intelligence through the cognitive cycle and new algorithms should be developed to detect gaps as cyclo-stationary [6] and allows the CR terminal to operate in a smart way in such environment, scanning a set of frequency bands in the purpose of exploiting them [7]. In this paper, we propose an artificial neural network approach for channel assignment in CR to satisfy the network robustness constraints and resist any single channel interruption by PUs. To this end, we formulate the problem as a constrained optimization problem, in which the objective is to maximize the use of each available frequency for secondary users and minimize the maximum interferences under all possible PUs’ appearances. In the next sections, we give an overview of the state of the art for channel assignment in cognitive radio, and then we present the suitable formulation, and explain the proposed approach. We terminate with discussion of the results and a conclusion. 2. Related Work The current challenges in radio networks are many folds: to ensure an efficient and full use of radio frequency resources and multimedia applications. To get connected, at best, anywhere, any-time and with any network. To customize the more power-full features stimulated by the increasing consumers’ demand. To find solutions for mobile business. And, finally to tend toward several access technologies whose assignment is local and continuously and independently updated, Jamal Elhachimi et al, Int.J.Computer Technology & Applications,Vol 5 (3),1219-1225 IJCTA | May-June 2014 Available online@www.ijcta.com 1219 ISSN:2229-6093