Reliable and prompt identification of active users is critical for enabling random access in massive machine-to-machine (M2M) type networks which typically operate within stringent access delay and energy constraints. Typically, M2M networks are comprised of sensors, actuators, etc. monitoring critical information which leads to a sparse and sporadic data traffic since only a small subset of devices are active at any given time instant. In order to properly allocate resources and to decode data in M2M systems, the base station (BS) needs to accurately identify the active devices as misdetections can lead to critical data loss. In our work, we focus on the active device identification problem in M2M systems with stringent requirements in terms of latency, reliability, energy efficiency, and security. We use information theoretic tools to characterize the fundamental limits on the number of channel-uses required to identify the active users. We also propose several practical active device identification strategies and study their performance gap in comparison to the theoretically established benchmark.

follow this research