A Stochastic Game Approach for Collaborative Beamforming in SDN-Based Energy Harvesting Wireless Sensor Networks

Collaborative beamforming (CB) has recently emerged as a promising technique for transmission range extension and energy consumption reduction in wireless sensor networks (WSNs). However, due to the constrained energy and limited data processing capabilities of sensor nodes, the performance optimization of CB mainlobe and sidelobe control (SC) encounters challenges in the practical deployment. To address these challenges, we present an architecture of software-defined energy harvesting WSN (SD-EHWSN) for CB communications. Specifically, we first design the mechanism of CB communications based on the software-defined network (SDN) architecture to reduce the communication and computational overhead of sensor nodes. Then, we consider solar energy-harvesting system to achieve long-term operation of WSN and utilize a stationary Markov (SM) chain to model the arrival process of solar energy. Based on the stochastic nature of solar energy, a stochastic game model is developed to formulate the problem of CB optimization in SD-EHWSN, and the existence proof of Nash equilibria is provided. Based on the analytical results, we propose a reinforcement learning algorithm to maximize the long-term signal-to-noise ratio (SNR) performance with SC and prove the convergence of the algorithm. Simulation results are presented to validate the efficiency of the proposed scheme for CB communications in SD-EHWSN.