Performance Evaluation of Generalized Buffer-State-Based Relay Selection in NOMA-Aided Downlink

Abstract

In this paper, we propose a novel buffer-state-based (BSB) non-orthogonal multiple access (NOMA)-aided downlink scheme for a two-hop cooperative network supporting multiple relay and destination nodes. More specifically, in the broadcast phase, the source node broadcasts information packets to multiple selected relay nodes. In the relaying phase, a selected relay node transmits packets to multiple destination nodes based on the NOMA principle. Here, the use of buffer at the relay nodes allows us to achieve a flexible relay selection. Furthermore, with the aid of buffer-state-based relay selection, the detrimental effects of empty-and full-buffer states are avoided. The theoretical outage probability and average packet delay are derived for our proposed scheme, based on the Markov-chain model. Our simulation results demonstrate that the proposed scheme outperforms the existing max-link and BSB relay selection benchmarks.INDEX TERMS Broadcast, buffer, Markov chain, non-orthogonal multiple access (NOMA), outage probability, packet delay, successive interference cancellation, theoretical bound.

Hybrid NOMA/OMA Broadcasting-and-Buffer-State-Based Relay Selection

“…In this section, we derive the theoretical bounds of the throughput, the outage probability, the average packet delay, and the diversity order for the proposed scheme. Our theoretical analysis is based on a Markov chain model, similar to conventional studies [25,27,29]. For example, for the scenario of two relay nodes with a buffer size of L = 2, we have 19 states, as shown in Table 3 of [25].…”

“…Moreover, in [25][26][27][28][29][30], broadcasting an information packet from a source node to multiple relay nodes was introduced in the context of the aforementioned buffer-aided cooperative communications; this allows relay nodes to share a common information packet, hence increasing the degree of freedom in buffer states. As a result, the source-to-destination (SD) communication delay was kept at the minimum while achieving the high reliability of buffer-aided relay selection.…”

“…Such broadcasting-and buffer-state-based (BBSB) relay selection schemes were developed in the context of decode-and-forward (DF) [25] and amplify-and-forward (AF) relaying scenarios [26]. Furthermore, the cooperative beamforming mode was incorporated into BBSB relay selection in [27], while BBSB relay selection was introduced to physical-layer security [28], NOMA [29], and cognitive radio networks [30]. Note that, the conventional BBSB schemes of [25][26][27][28] does not rely on NOMA.…”

“…Note that, the conventional BBSB schemes of [25][26][27][28] does not rely on NOMA. Also, while in [29], the DL-NOMA concept was incorporated into the relaying phase of the buffer-aided relay selection for the multi-user downlink, cooperative beamforing and the NOMA mode in the source transmission phase were not exploited.…”

Generalized Buffer-State-Based Relay Selection in Cooperative Cognitive Radio Networks

“…These two schemes are capable of attaining a lower average packet delay than the original MMRS and max-link counterparts, owing to the beneficial effects of packet sharing among the relay nodes. Furthermore, the concept of BSB relay selection was proposed in [9], which was then introduced into the beamforming-assisted G-ML scheme [12], the fullduplex scheme [24], the NOMA-based downlink [25], and the secure relaying scheme [13]. Since one of the challenges in the buffer-aided relaying protocols is an increased packet delay [3], the relay selection scheme in [26] was proposed for achieving a significant reduction of the packet delay, compared to the max-link scheme [6] by introducing a priority on the relay selection criterion.…”