Resource allocation with interference coordination for relay-aided cellular orthogonal frequency division multiple access systems

Liang Liang, Gang Feng, Yide ZhangIET Commun 2012 · DOI

TL;DR

This paper proposes a novel Resource Allocation with Interference Coordination (RAIC) scheme for relay-aided cellular OFDMA systems. RAIC dynamically adapts its resource allocation strategy based on traffic load to mitigate interference from relay nodes, thereby improving system throughput and spectral efficiency.

What Problem Does This Paper Address?

The increasing demand for higher data rates in cellular systems, particularly with the advent of 4G and beyond (like LTE-Advanced), necessitates advanced techniques to improve spectral efficiency and coverage. Relay nodes (RNs) are a promising solution, extending coverage and boosting throughput, especially in challenging areas. However, RNs introduce significant intra-cell and inter-cell interference, which can negate the performance gains if not properly managed. Existing interference coordination techniques often don't fully account for the complexities introduced by relay-aided multi-cell environments. This paper addresses the critical challenge of effectively allocating radio resources in such systems to harness the benefits of RNs while minimizing the detrimental effects of interference.

Methods

The core of this paper is the proposed Resource Allocation with Interference Coordination (RAIC) scheme. RAIC is a three-phase approach that dynamically selects one of three resource allocation algorithms based on the current traffic load:

  • Pre-configuration Algorithm (PCA): Used for light traffic loads. It allocates orthogonal spectrum resources to users served by the base station (BS) and adjacent RNs. This phase is an extension of fractional frequency reuse.
  • Semi-static Allocation Algorithm (SSAA): Employed when traffic load increases and orthogonal resources become insufficient. SSAA reuses resource blocks (RBs) for users who haven't been scheduled, based on interference levels.
  • Dynamic Allocation Algorithm (DAA): Applied during heavy traffic loads. DAA clusters users based on their interference levels and then allocates appropriate spectrum resources to these clusters, considering instantaneous channel states. This phase utilizes a graph-based approach for cluster construction, akin to a node coloring problem, and then assigns RBs to these clusters.

The problem of resource allocation is formulated as a maximization of system throughput, considering both BS-RN and RN-MS links in two distinct time slots (slot 1 for BS-RN, slot 2 for BS-MS and RN-MS). The authors also analyze the complexity of their proposed algorithms, noting that the optimization problem is NP-hard, necessitating heuristic approaches.

Datasets

Not discussed in the paper. The evaluation relies on extensive simulation experiments.

Key Insights and Ideas

  • Traffic-Aware Resource Allocation is Crucial: The paper highlights that a one-size-fits-all approach to resource allocation is insufficient in relay-aided OFDMA systems. Dynamically adapting the strategy based on traffic load (light, medium, heavy) is key to balancing interference mitigation and system performance.
  • Interference Classification is Foundational: The authors establish a classification rule for interference levels, which is fundamental to the SSAA and DAA algorithms. This allows for a more nuanced understanding of interference impact.
  • Clustering for Interference Management: The DAA's use of graph-based clustering to group interfering users is an innovative approach to manage interference in dense, relay-aided environments. This allows for more efficient RB reuse within clusters.
  • Trade-off Between Scheduled Users and Throughput: The paper demonstrates that while scheduling more users is desirable, it can increase interference. RAIC aims to find an optimal trade-off, prioritizing throughput and spectral efficiency over simply maximizing the number of scheduled users.

Key Results

The simulation results demonstrate the effectiveness of the proposed RAIC scheme:

  • Improved Throughput: RAIC consistently outperforms non-interference-coordination schemes (SAS, NICS) in terms of throughput across various traffic loads. It achieves throughput comparable to, but lower than, an ideal allocation scheme (IAS) which assumes no interference.
  • Enhanced Spectrum Efficiency: RAIC shows significantly higher spectrum efficiency compared to SAS and NICS, especially under heavier traffic conditions.
  • Better SINR Distribution: The Cumulative Distribution Function (CDF) of Signal-to-Interference-plus-Noise Ratio (SINR) shows that RAIC provides a better SINR distribution than SAS and NICS, indicating effective interference reduction.
  • Scalability with Traffic Load: The RAIC scheme's performance advantage becomes more pronounced as the number of MSs (traffic load) increases, showcasing its ability to handle challenging scenarios.
  • Comparison of RAIC Phases: The paper also compares the performance of individual RAIC phases (PCA, SSAA, DAA) and shows that the full RAIC scheme, by combining these phases, yields the best overall performance. DAA, in particular, shows significant gains under heavy traffic.

Limitations and Future Work

  • NP-Hardness: The authors acknowledge that the resource allocation problem is NP-hard, meaning their proposed algorithms are heuristic and aim for sub-optimal solutions.
  • Simulation-Based: The results are based on extensive simulations, and real-world deployment might introduce additional complexities.
  • Network Topology Assumption: The paper assumes a known network topology and focuses on interference coordination given this topology. Future work could explore dynamic topology adaptation.
  • Specific LTE-A Context: While the paper uses LTE-A as a model, its applicability to other OFDMA systems is mentioned, but further validation in diverse contexts could be beneficial.
  • Future Work: The authors suggest further research into optimizing the trade-off between the number of scheduled users and system performance, especially under heavy traffic conditions. They also imply that exploring more advanced interference coordination techniques within the RAIC framework could yield further improvements.

Summary updated

BibTeX
@article{liang2012resource,
  author = {Liang Liang and Gang Feng and Yide Zhang},
  title = {{Resource allocation with interference coordination for relay-aided cellular orthogonal frequency division multiple access systems}},
  journal = {IET Commun.},
  year = {2012},
  doi = {10.1049/iet-com.2010.0953},
}
Figure