HARQ-IR Aided Non-Orthogonal Multiple Access for URLLC: Tradeoff Between Transmission Reliability and Data Freshness

Many emerging artificial intelligence-driven applications rely on the availability of large-scale, stable, and fresh sensing data, underscoring the Ultra-Reliable Low-Latency Communications (URLLC). This paper proposes to amalgamate Non-Orthogonal Multiple Access (NOMA) and Hybrid Automatic Repeat reQuest with Incremental Redundancy (HARQ-IR) to accommodate reliable and real-time wireless services. The outage probability and the Average Age of Information (AAoI) are used to evaluate the transmission reliability and data freshness for HARQ-IR-NOMA schemes. To reveal physical insights as well as ease system designs, the asymptotic outage probability in the high Signal-to-Noise Ratio (SNR) regime is derived by developing a recursive dominant term approximation method. Moreover, the AAoI is deduced in terms of the outage probability by considering an M/G/1/1 queuing model. The asymptotic AAoI at high SNR is shown to be an increasing function of the diversity order-deficiency, which theoretically justifies the tradeoff between the transmission reliability and the data freshness. Furthermore, the AAoI is minimized by optimizing the transmission powers between users as well as HARQ rounds while maintaining the outage and total power constraints. The minimal AAoI is obtained by developing Geometric Programming (GP)-based and Deep Reinforcement Learning (DRL)-based methods. Finally, the numerical results are presented for verification.