Executive Summary : | The fifth generation (5G) of wireless communication systems brings a paradigm shift in wireless communication compared to the earlier generations such that in 5G and beyond, the end-users are autonomous machines rather than human-operated mobile devices. By such machine-type communications, the emerging Internet of Things (IoT) in 5G and beyond enables a wide variety of applications. Examples include smart homes, smart cities, and industrial manufacturing and control. These applications require the connectivity of a massive number of devices with short-packet transmissions. Furthermore, since a significant fraction of these devices operates on limited battery, energy-efficient communication is crucial. Hence, the emerging wireless systems call for a theoretical framework that can be used to analyze energy-efficient short-packet communication in a network with a massive number of nodes. The presence of a massive number of devices will result in a scenario where the number of devices may be comparable to the packet length that each device transmits. To model such networks, a channel was recently introduced, where the number of transmitters also grows as the packet length increases. In our recent work, we analyzed the asymptotic limits of energy efficiency of this network by taking the packet length large enough. While these limits can be used as benchmarks for packet length of 5000 bits, they may not be approachable for short packets consisting of 100 bits. One of the objectives of this project is to provide suitable benchmarks on energy efficiency in transmitting packet length consisting of several 100 bits. Furthermore, we plan to explore the benefits of feedback from the receiver to the transmitters. In a channel with one transmitter and a single receiver, the availability of feedback does not improve the fundamental limits from the ones under no feedback. However, it helps in approaching the limits using packets with short lengths. This motivates us to explore the benefits of different kinds of feedback on energy efficiency in massive networks. This project will also explore computationally-efficient methods to numerically evaluate the bounds on energy efficiency. |