Optimal degree distribution with minimal stopping sets in coded random access for massive IoT communications

The future networks are predicted to serve massive number of devices with small packet loss rate (PLR) and fast processing. In this paper, we propose an access scheme that is capable of serving communication with minimum PLR due to the use of degree distribution having minimal stopping sets. This paper develops coded random access with low error floor caused by the appearance of stopping sets. We derive an optimal degree distribution using optimization by minimizing stopping set subject to small degree to keep low computational complexity. We evaluate the performance of the degree distribution using a visual tool, called Extrinsic Information Transfer (EXIT) chart, to validate the achievable throughput. In this paper, we also validate the theoretical probability of the stopping sets with a series of computer simulations. We found that the theoretical probability of stopping set is reliable, which is helpful for optimization of degree distribution having minimal stopping set. These results are expected to be useful for future massive IoT communications serving many devices with asymptotically zero failure for critical applications.