Energy-Efficient Scheduling and Resource Allocation for Power-limited Cognitive IoT Devices

Energy-efficient scheduling and resource allocation strategies help reduce interference and extend the lifetime of power-limited Internet of Things (IoT) devices. This paper focuses on improving the transmission efficiency and working time of power-limited data acquisition equipment, e.g., low-power consumption IoT sensors. In particular, the cognitive device tunes its transmission time and power rationally to avoid interference and recharges itself by conducting energy harvesting. Inspired by the concept of the age of information, we coin the concept of the value of update (VoU) and use it to guide devices to upload data in a timely manner and optimize the key parameters through a deep deterministic policy gradient (DDPG) neural network to maximize the long-term VoU. Finally, extensive simulations are conducted to demonstrate the effectiveness and robustness of the proposed scheme.