Throughput Analysis and User Barring Design for Uplink NOMA-Enabled Random Access

Abstract

Being able to accommodate multiple simultaneous transmissions on a single channel, non-orthogonal multiple access (NOMA) appears as an attractive solution to support massive machine type communication (mMTC) that faces a massive number of devices competing to access the limited number of shared radio resources. In this paper, we first analytically study the throughput performance of NOMA-based random access (RA), namely NOMA-RA. We show that while increasing the number of power levels in NOMA-RA leads to a further gain in maximum throughput, the growth of throughput gain is slower than linear. This is due to the higher-power dominance characteristic in power-domain NOMA known in the literature. We explicitly quantify the throughput gain for the very first time in this paper. With our analytical model, we verify the performance advantage of NOMA-RA scheme by comparing with the baseline multi-channel slotted ALOHA (MS-ALOHA), with and without capture effect. Despite the higher-power dominance effect, the maximum throughput of NOMA-RA with four power levels achieves over three times that of the MS-ALOHA. However, our analytical results also reveal the sensitivity of load on the throughput of NOMA-RA. To cope with the potential bursty traffic in mMTC scenarios, we propose adaptive load regulation through a practical user barring algorithm. By estimating the current load based on the observable channel feedback, the algorithm adaptively controls user access to maintain the optimal loading of channels to achieve maximum throughput. When the proposed user barring algorithm is applied, simulations demonstrate that the instantaneous throughput of NOMA-RA always remains close to the maximum throughput confirming the effectiveness of our load regulation.Index Terms-Massive machine type communication (mMTC), NOMA, random access, user barring.

On Asymmetric Game for NOMA-ALOHA under Fading

“…It is worthy to note that the approaches [4] [11] [21] do not need to consider the SINR in ( 6), as it is assumed that users know the CSI. If the CSI is known at a user, power control can be performed so that a required SINR for successful decoding can be achieved if no collision happens.…”

Reinforcement Learning for NOMA-ALOHA Under Fading

“…It is worthy to note that the approaches [7] [16] [29] do not need to consider the SINR in ( 6), as it is assumed that users know the CSI. If the CSI is known at a user, power control can be performed so that a required SINR for successful decoding can be achieved if no collision happens.…”

Uplink non-orthogonal multiple access in heterogeneous networks: A review of recent advances and open research challenges

“…Besides that, throughput is also a key metric whose maximization is an essential parameter for improved UL NOMA performance in HetNets. 52…”

“…Besides that, throughput is also a key metric whose maximization is an essential parameter for improved UL NOMA performance in HetNets. 52 The different metrics used in the shortlisted studies for evaluating the performance of NOMA-enabled HetNets are summarized in Table 4.…”