Priority-Based Dynamic IoT-Downlink Communications with Blind Transmission-Mode Recognition

Abstract

In downlink Internet of things (IoT) applications, certain nodes can be given a higher transmission priority than others, and hence might be allocated more transmission resources. Consequently, some nodes may face a resource-starvation problem. To overcome such a predicament, this article proposes an opportunistic non-orthogonal multiple access (NOMA) scheme to enable low priority (LP) nodes (LPNs) to share the transmission resources of high priority (HP) nodes (HPNs) while prioritizing the quality of service (QoS) requirements of the HPNs. In the proposed scheme, the control center (CC) may decide to multiplex the data of one or more LPNs with the HPN data using NOMA, which can improve the network throughput substantially. Such a decision is based on the channel conditions and QoS requirements for the HPNs and LPNs. Hence, the proposed system may operate in one of the following four modes, HPN orthogonal multiple access (OMA), LPN-OMA, (HPN, LPN)-NOMA, and the OFF mode where all nodes are muted. To improve the network efficiency and reduce overhead, the network is considered uncoordinated where the receiving nodes do not receive side information about the selected transmission-mode by the CC, leaving it to the receiving nodes to blindly recognize the transmission-mode. The network complexity and throughput trade-off is considered for fixed and adaptive power assignments. The results show that the proposed system can provide substantial throughput improvement over conventional NOMA transmission. For the example of a pair of HPN and LPN, the maximum throughput gain with opti-mal power allocation is about 0.85 bits/symbol, which implies that the throughput relative improvement is about 9 folds. The article also discuss the challenges and potential solutions of the proposed framework.