PI: Yury Polyanskiy, Department of Electrical Engineering & Computer Science, MIT
Abstract: The key challenge for the next generation radio-access networks is the massive numbers of infrequently communicating sensors (machine-type communication, MTC, or Internet-of-Things). The present solutions (LTE 4G etc) are inadequate because of the reliance on centralized resource allocation, orthogonalizing the access from different users. For MTC this is not acceptable as it incurs a significant control-layer overhead and latency. Thus, a new communication solution is required. Commercially, there is a strong demand for such a solution in both the unlicensed spectrum (so called low-power wide-area networks (LP-WANs)) and the licensed spectrum (5G). The novelties of this problem compared to traditional communication theory are as follows: (a) presence of a gigantic number of idle (inactive) users; (b) a large number (hundreds) of active users; (c) small amount of data payload in each user's packet; (d) focus on high energy-efficiency (low energy-per-bit).
This project will (1) develop theoretical understanding of the fundamental limits of random-access channels, and (2) suggest practically implementable solutions.