Cooperative Interference Mitigation And Energy Efficiency Optimization For Cognitive Sensor Networks

Abstract:Cognitive Sensor Networks, through the introduction of cognitive radio to the wireless sensor networks, cognitive sensor nodes achieve the ability of self-seeking and sharing the idle spectrum resources by sensing and learning from the surrounding spectrum environment. These cognitive radio features enhances the efficiency of available spectrum for wireless sensor nodes. But with the expand in the scale of the sensor networks and the sharply increase in the number of sensor nodes, interference between sensor nodes has become a bad factors reducing the network performance of information transmission. This impact of interference can be reduced by introduce cooperative communication into cognitive sensor networks. However, the available energy of wireless sensor nodes are limited, and the introduction of cooperative communication will accelerate the exhaustion of limited energy. This paper aims at improving the performance of information transmission as well as the energy efficiency, and studies two key issues: cooperative interference mitigation and energy efficiency optimization in cognitive sensor networks.First, to improve the communication performance of cognitive sensor networks in the interference environment as well as take the selfish characteristics of cognitive sensor nodes into account, the cooperative communication is introduced into cognitive sensor networks, and a cooperative interference mitigation with spectrum sharing mechanism is established for cognitive sensor networks. The source node shares partly of its spectrum to the relay nodes, in exchange for the cooperation of relay nodes to improve communication performance; relay nodes regard the obtained spectrum as a payback for their power consumption to help the source node. The model of leader to follower non-cooperative game is used to investigate the process of the proposed mechanism, then the existence and uniqueness of the Nash equilibrium is proved, an repeated iterative algorithm is proposed to reach the Nash equilibrium point, which make reasonable utilities for both sides of the game.Furthermore, sensor nodes only have limited energy, and energy saving is one of the essential concerning for the design of cognitive sensor networks. Therefore, the energy efficiency function for the cognitive sensor nodes is formulated and the energy efficiency optimization of each node is modeled as an equal entities’ non-cooperative game problem by considering the selfish properties of the cognitive nodes. Due to the high computation complexity for the solution of the non-cooperative game, a Q-learning algorithm is introduced to find the optimal energy efficiency value, the value then is used to balance the power consumption for cooperative interference mitigation.Finally, an information transmission model for cognitive sensor networks is constructed using the MATLAB. The performances of communication capacity, power consumption and energy efficiency are verified to prove the efficiency of proposed cooperative interference mitigation and energy efficiency optimization scheme.