Wake-up radio is a promising approach to mitigate the problem of idle listening, which incurs additional power consumption for the Internet of Things (IoT) wireless transmission. Radio frequency (RF) energy harvesting technique allows the wake-up radio to remain in a deep sleep and only become active after receiving an external RF signal to ‘wake-up’ the radio, thus eliminating necessary hardware and signal processing to perform idle listening, resulting in higher energy efficiency. This review paper focuses on cross-layer; physical and media access control (PHY and MAC) approaches on passive wake-up radio based on the previous works from the literature. First, an explanation of the circuit design and system architecture of the passive wake-up radios is presented. Afterward, the previous works on RF energy harvesting techniques and the existing passive wake-up radio hardware architectures available in the literature are surveyed and classified. An evaluation of the various MAC protocols utilized for the novel passive wake-up radio technologies is presented. Finally, the paper highlights the potential research opportunities and practical challenges related to the practical implementation of wake-up technology for future IoT applications.

Traditional methods for designing metasurfaces with the intended scattering response are frequently laborious and time-consuming. Selecting the structure’s shape, substrate type, and structure dimensions requires a great deal of design expertise. This paper proposes a time and resource-efficient dual-task multichannel machine learning (DTM2L) framework for the constrained inverse design of multiple metasurface structures. The proposed DTM2L framework is used for joint classification and regression for the prediction of the type and dimensions of the structure. Several examples to achieve a desired scattering response with the constraint on the frequency are provided to verify the performance of the proposed method.