Game-Theory-Based Distributed Power Splitting for Future Wireless Powered MTC Networks

This paper studies the emerging wireless power transfer for machine type communication (MTC) network, where one hybrid access point (AP) with constant power supply communicates with a set of users (i.e., wearable devices and sensors) without power supply. The information and energy are transferred simultaneously in downlink direction. For MTC networks, most devices only receive several bits control data from AP in downlink transmission. So it is possible to utilize part of the received power to execute energy harvesting provided that the transmission reliability is guaranteed. Since we assume that all devices are without power supply or battery, the power of uplink transmission is entirely from energy harvesting. After converting electromagnetic wave to electricity, the devices are able to transmit their measured and collected data in uplink. Based on these considerations, a non-cooperative game model is formulated and a utility function involving both downlink decoding signal to noise ratio (SNR) and uplink throughput is established. The existence of Nash equilibrium (NE) in the formulated game model is proved. The uniqueness of NE is discussed and the expected NE is selected based on fairness equilibrium selection mechanism. The optimal splitting ratio within the feasible set, which maximizes the utility function, is obtained by an iterative function derived from this utility function. The numerical results show that in addition to ensuring the downlink decoding SNR and maximizing uplink throughput of an individual device, our proposed algorithm outperforms the conventional algorithm in terms of system performance.