A clustering‐based user pairing scheme for downlink 3D non‐orthogonal multiple access‐based visible light communication systems
Tanuja Dogra,Manoranjan Rai Bharti
DOI: https://doi.org/10.1002/dac.5717
2024-01-29
International Journal of Communication Systems
Abstract:In this paper, we have proposed a user pairing scheme based on clustering for downlink 3D NOMA‐based VLC systems. In this user pairing scheme clustering is performed prior user pairing. The number of clusters are selected using K‐means algorithm and Elbow method, and the user allocation in a particular cluster is validated using Artificial Bee Colony (ABC) optimization. In order to achieve high throughput, inter cluster user pairing is performed by forming pairs of two or three users with maximum channel gain difference in between them. The simulation results (as shown in Figs. 1 and 2) show that, for given values of signal‐to‐noise ratio (SNR), the proposed user pairing scheme offers higher average user data rates (AUDR) as compared to the existing user pairing schemes for downlink NOMA‐based VLC systems. Summary Non‐orthogonal multiple access (NOMA) has proven to be a very effective multiple access scheme to be employed in visible light communication (VLC) systems. NOMA scheme is capable of enhancing the spectral efficiency, sum rate performance, and various other performance parameters in VLC systems as compared to traditional orthogonal multiple access schemes. However, it is not practical to apply the concept of NOMA to all the users jointly. Therefore, the concept of user pairing has been introduced in literature to implement NOMA effectively so as to reduce the decoding order of successive interference cancellation (SIC) mechanism and complexity. In this paper, we have proposed a user pairing scheme based on clustering for downlink 3D NOMA‐based VLC systems. In this user pairing scheme, clustering is performed prior user pairing. The number of clusters are selected using K‐means algorithm and Elbow method, and the user allocation in a particular cluster is validated using artificial bee colony (ABC) optimization. In order to achieve high throughput, inter‐cluster user pairing is performed by forming pairs of two or three users with maximum channel gain difference in between them. The simulation results show that, for given values of signal‐to‐noise ratio (SNR), the proposed user pairing scheme offers higher average user data rates (AUDRs) as compared to the existing user pairing schemes for downlink NOMA‐based VLC systems. The energy efficiency (EE) and the bit error rate (BER) performance of the proposed user pairing scheme have also been studied.
telecommunications,engineering, electrical & electronic