Enabling transmission status detection in grant-free power domain non-orthogonal multiple access for massive Internet of Things

Grant-free access has been identified as a potential solution to support massive machine-type communications by avoiding the high signaling overhead and latency issues of existing grant-based channel access mechanisms. Motivated by this, the work in this article proposes a simple, yet efficient, strategy to enable power domain non-orthogonal multiple access (PD-NOMA) aided grant-free communication, where devices can transmit their data in an arrive-and-go manner by using randomly chosen power levels. The focus here is on collision detection, which is of prime importance in grant-free systems. In the considered grant-free PD-NOMA scheme, collisions occur when multiple users randomly choose the same resource block and power level for data transmission, which makes it difficult for the base station (BS) to distinguish between their data. Accordingly, the proposed technique aims at early collision detection by introducing special activity indicator symbols (AIS), which are transmitted by each active device before, or at the start of, the actual data packet. The BS then exploits these known AIS to identify successful transmissions and collisions before processing actual data, which avoids unnecessary data transmission by users and/or receiver processing in case of collisions, and prepares the receiver for appropriate data recovery otherwise. It is demonstrated that the proposed technique provides a low complexity and high performance solution to enable transmission status detection in grant-free PD-NOMA.