Farhadi, A;
Olfat, A;
Masouros, C;
(2025)
Joint Beamforming and Resource Allocation for STAR-IRS-Aided SCMA ISAC Systems Using Meta Deep Reinforcement Learning.
IEEE Transactions on Wireless Communications
10.1109/TWC.2025.3580347.
(In press).
Preview |
Text
Joint_Beamforming_and_Resource_Allocation_for_STAR-IRS-Aided_SCMA_ISAC_Systems_Using_Meta_Deep_Reinforcement_Learning.pdf - Accepted Version Download (3MB) | Preview |
Abstract
In response to growing demands for both sensing and communication performance from future applications of wireless networks, simultaneous transmitting and reflecting intelligent reflecting surface (STAR-IRS), sparse code multiple access (SCMA), and integrated sensing and communication (ISAC) have been introduced. In this paper, we investigate the joint beamforming and resource allocation optimization problem (OP) for STAR-IRS-Aided SCMA for an ISAC System by considering ISAC, SCMA, beamforming, quality of service (QoS), and STAR-IRS constraints in the THz band. The investigated OP is a type of NP-hard OP that cannot be solved by conventional methods. Moreover, the possible solution must have the ability to adapt to different situations. Then to solve this problem efficiently, we employ a new meta deep reinforcement learning (MDRL). The simulation results demonstrate the effectiveness of the proposed solution for the resource allocation problem across different system model and MDRL parameters, as well as multiple access methods. Compared to non-orthogonal multiple access (NOMA) and orthogonal frequency-division multiple access (OFDMA), the proposed system model improves energy efficiency (EE) by 15% and 20%, respectively. The suggested MDRL approach achieves a 123% enhancement over classic deep deterministic policy gradient (DDPG). Additionally, a scenario involving a coupled phase-shift model was conducted for the proposed system model. To investigate the effect of channel state information (CSI) estimation on our proposed system model, we considered a simulation with both imperfect and perfect CSI. Additionally, when solving the optimization problem using a convex optimization approach, MDRL shows a 10% improvement.
| Type: | Article |
|---|---|
| Title: | Joint Beamforming and Resource Allocation for STAR-IRS-Aided SCMA ISAC Systems Using Meta Deep Reinforcement Learning |
| Open access status: | An open access version is available from UCL Discovery |
| DOI: | 10.1109/TWC.2025.3580347 |
| Publisher version: | https://ieeexplore.ieee.org/xpl/RecentIssue.jsp?pu... |
| Language: | English |
| Additional information: | This version is the author accepted manuscript. For information on re-use, please refer to the publisher's terms and conditions. |
| UCL classification: | UCL UCL > Provost and Vice Provost Offices > UCL BEAMS UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Electronic and Electrical Eng |
| URI: | https://discovery.ucl.ac.uk/id/eprint/10211011 |
Archive Staff Only
 |
View Item |
