Abstract:In this paper, we conduct a theoretical analysis of how to integrate reconfigurable intelligent surfaces (RIS) with cooperative non-orthogonal multiple access (NOMA), considering URLLC. We consider a downlink two-user cooperative NOMA system employing short-packet communications, where the two users are denoted by the central user (CU) and the cell-edge user (CEU), respectively, and an RIS is deployed to enhance signal quality. Specifically, compared to CEU, CU lies nearer from BS and enjoys the higher channel gains. Closed-form expressions for the CU's average block error rate (BLER) are derived. Furthermore, we evaluate the CEU's BLER performance utilizing selective combining (SC) and derive a tight lower bound under maximum ratio combining (MRC). Simulation results are provided to our analyses and demonstrate that the RIS-assisted system significantly outperforms its counterpart without RIS in terms of BLER. Notably, MRC achieves a squared multiple of the diversity gain of the SC, leading to more reliable performance, especially for the CEU. Furthermore, by dividing the RIS into two zones, each dedicated to a specific user, the average BLER can be further reduced, particularly for the CEU.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is: how to enhance the performance of the cooperative non - orthogonal multiple access (NOMA) ultra - reliable low - latency communication (URLLC) system through the reconfigurable intelligent surface (RIS), especially for the short - packet communication scenario. Specifically, the paper focuses on the following points:
1. **System Model**:
- The paper considers a downlink two - user cooperative NOMA system, where the two users are the central user (CU) and the cell - edge user (CEU) respectively. The CU is closer to the base station (BS) and thus enjoys a higher channel gain.
- The RIS is deployed to enhance the signal quality, especially in the short - packet communication scenario.
2. **Performance Evaluation**:
- The closed - form expression of the average block error rate (BLER) of the CU is derived.
- The BLER performance of the CEU under selective combining (SC) is evaluated, and a tight lower bound under maximum - ratio combining (MRC) is derived.
3. **Optimization Strategy**:
- It is proposed to divide the RIS into two regions, each region dedicated to serving one user, in order to further reduce the average BLER, especially for the significant performance improvement of the CEU.
4. **Theoretical Analysis and Simulation Verification**:
- Theoretical analysis shows that under high signal - to - noise ratio (SNR) conditions, the diversity gain of MRC is the square of that of SC, thus providing more reliable performance, especially for the CEU with poor receiving conditions.
- The accuracy of the theoretical analysis is verified through extensive simulations. The results show that the RIS - assisted cooperative NOMA system is significantly superior to the system without RIS in terms of BLER performance, especially when the SNR is high and MRC is used.
### Summary of Key Contributions
- **System Modeling**: A downlink cooperative NOMA URLLC system assisted by RIS, including a base station and two users (CU and CEU), is studied, and the closed - form expression of the average BLER of the CU is derived.
- **Performance Evaluation**: The derivation of the average BLER of the CEU under selective combining (SC) is provided, and a tight lower bound under maximum - ratio combining (MRC) is obtained.
- **Diversity Gain Analysis**: The diversity gain of the system under high SNR conditions is analyzed, and it is found that the diversity gain of MRC is the square of that of SC, indicating that MRC provides more reliable performance, especially beneficial for the CEU.
- **Simulation Verification**: The accuracy of the theoretical analysis is verified through simulation. The results show that the RIS - assisted cooperative NOMA system is significantly superior to the system without RIS in terms of BLER performance, especially when the SNR is high and MRC is used.
These research results provide valuable insights for the design and optimization of next - generation wireless networks, especially in applications requiring ultra - high reliability and low latency.