Deep Q-learning based sparse code multiple access for ultra reliable low latency communication in industrial wireless networks

Sparse code multiple access (SCMA) is a technology that allows for extremely low latency and high reliability in modern wireless communication networks. Moreover, due to the sparse layout of its codebooks, SCMA competence for multiple access techniques leads the way for futuristic ultra reliable low latency communications (URLLC), which aims for high reliability with low latency. Therefore, a deep Q learning-based SCMA, called Q-SCMA, is proposed, in which codebooks are adaptively constructed to minimize bit error rate (BER), maximizing throughput within the constraints of reliability and latency, i.e., supporting URLLC. Therefore, the rewards for Q-SCMA are formulated such that, while imposing a latency constraint, they also provide excellent reliability. The performance of the proposed approach is evaluated in terms of BER, loading factor, level of interference, bit error probability, throughput, latency, and computational complexity. It is also analyzed and compared with contemporary approaches, conventional SCMA (C-SCMA) and deep neural network SCMA (D-SCMA), for average white Gaussian noise as well Rayleigh fading channel, and for industrial wireless network environment modified Rician fading channel is considered. Simulation outcomes show considerable performance disparities between the proposed approach and D-SCMA along with C-SCMA. This underlines the requirement for using a deep Q learning model as a performance indicator for developing URLLC supporting SCMA for industrial wireless networks.