AmBC-Enabled WBAN Toward Ultralow Power Intelligent Health Monitoring: A DDPG-Based Sensor Coordination Framework in Symbiotic Radio Networks

Abstract

The Internet of Medical Things (IoMT) and wireless body area networks (WBANs) provide a promising pattern for wearable health monitoring to support telemedicine. However, the high energy consumption of wireless communications limits the convenience and lifetime of the wearable sensors. To realize ultralow-power communications in WBANs, ambient backscatter communication (AmBC) has been considered as a potential technology. In this article, we propose an ultralow power AmBC-enabled WBAN framework in a symbiotic radio network (SRN), where the WBAN coordinator can also operate as a general user under the Wi-Fi access point (AP) to share spectrum and channel state information (CSI). By considering the unique characteristics of the WBAN channels, different data rate requirements and energy consumptions of different wearable AmBC sensors, and the transmit power allocation at multiple antennas of the Wi-Fi AP, we formulate the joint rate control and energy management (JRCEM) optimization problem. To solve the optimization problem and coordinate different types of wearable AmBC sensors to support comprehensive and personalized health monitoring, we propose a deep deterministic policy gradient (DDPG)-based intelligent power splitting coefficient adjustment (IPSCA) algorithm and design a piecewise reward function for the algorithm. The proposed algorithm does not require prior knowledge of the channel statistics distribution parameters and enables long-term optimization. Extensive numerical simulation results demonstrate that our proposed algorithm can intelligently control the data rates of different wearable AmBC sensors to reduce delay, with flexibility, adaptability, and sustainability in different scenarios.