On integrating natural computing based optimization with channel assignment mining and decision making towards efficient spectrum reuse in cellular networks modelled through multi-agent system schemes

Abstract

Heuristic schemes and algorithms are successfully used for solving optimization problems. A great challenge in large-scale wireless networks is the channel allocation adaptability to dynamic network traffic conditions, which can be formulated as a discrete optimization problem. Several approaches such as genetic algorithms and multi-agent techniques have been applied so far in the literature for solving the resource management problem focusing mainly on network base-stations representation. A very promising intelligent approach known as ant colony optimization (ACO) which constitutes a special form of swarm intelligence has been used for solving routing problems. This approach has been introduced by the authors for improving the channel allocation in large-scale wireless networks, focusing on network procedures as the basic model component and not on network nodes as so far found in the literature. A comprehensive intelligent model architecture based on multi-agent systems technology and ACO for channel allocation in cellular networks is herein analysed and proposed. Moreover, the decision making for channel allocation is presented as well as the network-ant agent communication model. The new methodology for integrating ACO schemes and decision making regarding spectrum reuse in a multi-agent framework, through a novel agent negotiation approach, is the main contribution of this research. Finally, the simulation results show the large-scale wireless network performance improvement in terms of efficient resource management.