Energy-Efficient Cross-Layer Spectrum Sharing in CR Green IoT Networks

Internet-of-Things (IoT) aims to connect intelligent things to the Internet, enable thing-to-everything communication, and sustainability sensing and control. However, the exponential increase of connected IoT devices is resulting in tremendous demand on the limited spectrum. Cognitive radio (CR) technology is envisioned to be one of the main enabling technology for efficient IoT deployment. CR offers huge spectrum opportunities for sustainable green IoT devices by opportunistically utilizing the idle licensed bands. Since most green IoT devices are expected to be sensor nodes with limited energy sources, efficient communication protocols for CR-based IoT sensor networks are needed. This paper proposes an efficient cross-layer design consisting of joint optimization of the modulation order as a physical-layer parameter and the backoff probability as a MAC-layer parameter with the objective of minimizing the energy consumption in the CR-based green IoT networks subject to IoT delay guarantees, licensed primary radio (PR) channel availability and PR user activities. Simulation results reveal that the proposed cross-layer design has significantly reduced energy consumption while achieving the delay requirements in the network compared to the physical-layer-only and MAC-layer-only adaptation schemes.

I. Introduction

Green Internet-of-Things (IoT) is an emerging technology that promises to provide sustainable services at the edge of the network by enabling large deployment of automated services and applications [1], [2]. IoT is expanded and being used in enormous services and applications, including smart cities, connected mobility, industrial applications, health and fitness, etc. The growth in the connected IoT devices (mainly wireless sensors) to the Internet requires a huge demand on the limited spectrum [3]–[5]. Cognitive radio (CR) is a mature and promising technology that offers the needed spectrum for IoT devices [6]–[8]. CR technology allows CR IoT devices (unlicensed spectrum users) to dynamically allocate and share the underutilized licensed spectrum, which is owned by primary radio (PR) licensed networks. When sharing the spectrum with highly-dynamic legacy PR networks (PRNs), the forced-termination rate for CR-IoT transmissions can increase, causing sever degradation in network throughput and energy consumption.