NOMA-Relevant Clustering and Resource Allocation for Proportional Fair Uplink Communications

Mylene Pischella (1) , Didier Le Ruyet (1)

1 CEDRIC - LAETITIA - CEDRIC. Traitement du signal et architectures électroniques

This letter focuses on clustering and resource block (RB) allocation in multi-carrier uplink networks using power-domain non orthogonal multiple access (PD-NOMA). The optimization objective is time-based proportional fairness. PD-NOMA is applied per frame and RB only if the achieved individual data rates are larger with PD-NOMA than with time division multiple access. The clustering algorithm recur-sively uses maximum weight matching to build clusters per RB. Given the clusters, RB allocation is then optimal. The proposed algorithm provides a good trade-off between reduced overall network latency and rate improvement compared to two reference algorithms.

Mots clés: optimal matching; resource allocation; proportional fairness; NOMA

Domaines: Théorie de l'information [cs.IT]; Traitement du signal et de l'image [eess.SP]

Mylene Pischella, Didier Le Ruyet. NOMA-Relevant Clustering and Resource Allocation for Proportional Fair Uplink Communications. IEEE Wireless Communications Letters, 2019, 8 (3), pp.873-876. ⟨10.1109/LWC.2019.2897796⟩. ⟨hal-02446283⟩