Enabling Aloha-NOMA for Massive M2M Communication in IoT Networks

The Internet of things (IoT), which is the network of physical devices embedded with sensors, actuators, and connec- tivity, is being accelerated into the mainstream by the emergence of 5G wireless networking. This paper presents an uncoordinated non-orthogonal random access protocol, an enhancement to the recently introduced Aloha-NOMA protocol, which provides high throughput, while being matched to the low complexity requirements and the sporadic traffic pattern of IoT devices. Under ideal conditions it has been shown that Aloha-NOMA, using power-domain orthogonality, can significantly increase the throughput using SIC (Successive Interference Cancellation) to enable correct reception of multiple simultaneous transmitted signals. For this ideal performance, the enhanced Aloha-NOMA receiver adaptively learns the number of active devices (which is not known a priori) using a form of multi-hypothesis testing. For small numbers of simultaneous transmissions, it is shown that there can be substantial throughput gain of 6.9 dB relative to pure Aloha for 0.25 probability of transmission and up to 3 active transmitters.

💡 Research Summary

**
The paper addresses the challenge of supporting massive machine‑to‑machine (M2M) communication in the emerging 5G‑based Internet of Things (IoT) by enhancing the recently proposed Aloha‑NOMA random access protocol. Traditional Aloha suffers from low throughput due to frequent collisions, while Aloha‑NOMA improves performance by exploiting power‑domain non‑orthogonal multiple access (NOMA) and successive interference cancellation (SIC). However, the original scheme assumes a known number of simultaneously active devices and a fixed set of power levels (or SIC degree), which limits its applicability in heterogeneous IoT environments where the number of contending devices varies dynamically.

The authors propose two major innovations: (1) a multi‑hypothesis testing (MHT) module at the gateway that estimates the number of active devices in real time, and (2) a flexible frame structure that dynamically adjusts the SIC degree (i.e., the number of distinct power levels) based on the estimate. In the second phase of each frame, all devices transmit a common training sequence at the same power. The gateway computes the average received energy and applies a Neyman‑Pearson likelihood‑ratio test to decide among hypotheses (H_0, H_1, …, H_K), where (H_k) corresponds to exactly (k) active devices. By setting a target false‑alarm probability, the detection threshold is analytically derived as a function of the signal‑to‑noise ratio (SNR). Once the estimate (\hat K) is obtained, the gateway selects an SIC degree (m) such that (m \ge \hat K). If (m < \hat K), the frame is aborted and restarted; otherwise, the gateway broadcasts the chosen SIC degree and each device randomly selects one of the (m) pre‑defined power levels. Distinct power selections enable the SIC receiver to decode the signals sequentially; collisions in power selection trigger a limited number of re‑selection attempts, after which a NACK and back‑off are issued.

The frame consists of five phases: (1) beacon transmission, (2) training sequence transmission, (3) SIC‑degree broadcast, (4) data transmission, and (5) ACK/NACK feedback. This structure avoids the rigid slot allocation of TDMA/FDMA and can accommodate a varying number of active users without redesigning the whole frame.

Simulation results consider a single‑antenna gateway and (M = 10) devices, with the number of active devices following a binomial distribution with transmission probability (p_T). Throughput is evaluated for SIC degrees (m = 2) and (m = 3). The results show that, for (p_T = 0.25), Aloha‑NOMA with three power levels achieves roughly five times the successful transmission count of pure Aloha, corresponding to a 6.9 dB gain. Increasing the number of power levels beyond five yields diminishing returns (throughput saturation). Moreover, allowing more re‑selection attempts improves throughput at the expense of higher latency, which the authors note but do not analyze in depth.

The paper concludes that the combination of real‑time active‑user detection and adaptive SIC configuration makes the enhanced Aloha‑NOMA protocol highly suitable for massive IoT deployments. It retains the low‑complexity, energy‑efficient nature of pure Aloha while delivering substantially higher spectral efficiency. The proposed method also respects the power constraints of typical IoT devices by limiting the number of power levels used in a given contention round. Overall, the work provides a practical, scalable MAC solution that bridges the gap between simple random access and the performance demands of future massive IoT networks.