TL;DR
This paper proposes a novel adaptive interference avoidance and mode selection scheme for Device-to-Device (D2D) enabled small cells in 5G Industrial Internet of Things (IIoT) networks. By intelligently managing D2D communication modes and optimizing resource allocation, the scheme significantly enhances system throughput and signal-to-interference-plus-noise ratio (SINR) while reducing interference.
What Problem Does This Paper Address?
The Industrial Internet of Things (IIoT) demands high-speed, reliable, and ubiquitous wireless connectivity, especially for mission-critical applications and devices at the cell edge. While small cells (SCs) and Device-to-Device (D2D) communication offer promising solutions for improving coverage and throughput, they introduce significant interference challenges. Managing this interference is crucial for efficient resource sharing (RS) and maintaining Quality of Service (QoS) for both cellular users (IUEs) and D2D users. Existing approaches often rely on fixed power levels or simplified mode selection, failing to adapt to the dynamic nature of IIoT environments and leading to suboptimal performance. This paper addresses the need for an adaptive mechanism that can dynamically manage interference and optimize D2D communication modes within SC networks to maximize overall system performance.
Methods
The proposed solution employs a two-phase adaptive interference avoidance (IA) and mode selection (MS) framework:
- Phase 1: Channel Reuse Partner Selection (CRPS) and Mode Selection (MS): This phase adaptively determines the optimal mode (Cellular Mode - CM, or Reuse Mode - RM) for each D2D pair. It utilizes the Channel Gain Ratio (CGR) between cellular and reuse modes to prioritize D2D pairs. For D2D pairs operating in RM, a Channel Gain Factor (CGF) is computed to select the best reuse partner, minimizing interference. The mode selection criteria are based on a threshold derived from the CGF, allowing for dynamic assignment to CM or RM.
- Phase 2: Power Optimization: This phase focuses on optimizing the transmit power for both D2D users and cellular IUEs. It employs a Lagrangian dual decomposition method to find the optimal distributed power allocation that maximizes the network sum rate while adhering to interference constraints. The problem is formulated as NP-hard and then decomposed into a convex optimization problem for efficient solving.
The overall architecture is a hierarchically centralized system where SC base stations (SCBSs) communicate with the central macro base station (MeNB) for resource management and interference control.
Datasets
Not discussed in the paper. The evaluation is based on simulations.
Key Insights and Ideas
- Adaptive Mode Selection is Crucial: The paper highlights that dynamically switching between Cellular Mode (CM) and Reuse Mode (RM) for D2D communication, based on real-time channel conditions, is key to optimizing performance and managing interference.
- Channel Gain Factor (CGF) for Reuse Partner Selection: Introducing the CGF as a metric to select the best D2D reuse partner is a novel approach that effectively minimizes interference and optimizes resource sharing.
- Hierarchical IA and MS: The proposed hierarchical approach, where SCBSs manage local IA and MS while coordinating with the MeNB, balances centralized control with distributed intelligence, reducing complexity and latency.
- Impact of CSI: The results strongly suggest that accurate Channel State Information (CSI) is paramount for D2D communication, especially in reuse mode. Schemes with poor CSI estimation suffer significantly.
- Trade-off between SINR and Interference: The study demonstrates that while increasing interference power generally reduces SINR, the proposed scheme effectively manages this trade-off, maintaining a higher SINR compared to other methods.
Key Results
The proposed adaptive IA and MS scheme demonstrated superior performance across various metrics compared to five other benchmark schemes (SINR-Aware, ARSA, QoS-Aware, RSS, and JMAPC):
- System SINR: The proposed scheme achieved the highest system SINR CDF, outperforming others by a significant margin (e.g., 15.46% better than SINR-aware, 32.46% better than JMAPC).
- System Throughput: It also yielded the highest system throughput CDF, showing improvements of 6.41% over SINR-aware and 25.39% over JMAPC.
- Interference Management: The scheme effectively managed interference, showing a more controlled response to increasing interference power, leading to better system SINR and throughput.
- Robustness to Distance: While performance degraded with increased D2D pair distance, the proposed scheme still maintained a competitive edge, especially compared to schemes that struggled with CSI.
Limitations and Future Work
- Distance Sensitivity: The paper acknowledges that the proposed scheme suffers slightly from increased distance between D2D pairs, leading to performance degradation in SINR and throughput.
- CSI Measurement Challenges: The MS scheme's inability to accurately measure CSI between D2D pairs and interfering links with SCBSs is noted as a limitation, particularly affecting performance in certain scenarios.
- Future Work: The authors suggest that further research could explore more advanced techniques to address the distance sensitivity and improve CSI measurement accuracy. The paper also implies that extending this framework to more complex heterogeneous network scenarios could be a valuable future direction.
Summary updated