Abstract
Cooperative transmission can greatly improve communication system performance by taking advantage of the broadcast nature of wireless channels. Most previous work on resource allocation for cooperation transmission is based on centralized control. In this paper, we propose two share auction mechanisms, the SNR auction and the power auction, to distributively coordinate the resource allocation among users. We prove the existence, uniqueness and effectiveness of the auction results. In particular, the SNR auction leads to a fair resource allocation among users, and the power auction achieves a solution that is close to the efficient allocation.
Introduction
In cooperative wireless communications, "relay" nodes act as virtual antennas to help source-destination pairs transmit data more reliably. While this significantly enhances performance, relay power is a finite, shared resource that must be managed carefully. Traditional solutions rely on centralized control, where a single controller requires global knowledge of every channel state. This approach creates massive signaling overhead and fails to scale in dense networks.
This paper addresses the fundamental challenges of distributed resource allocation: determining when a relay is helpful for a specific user and how the relay should divide its power among multiple competitors. By applying auction theory, the authors create a decentralized market where the relay acts as an auctioneer and users act as bidders, eliminating the need for a central controller with global knowledge.
System Model and Methodology
The proposed framework utilizes a two-phase cooperative transmission protocol. In Phase 1, the source broadcasts its signal to both the destination and the relay. In Phase 2, the relay forwards the received signal to the destination using the Amplify-and-Forward (AF) protocol. The destination then combines the signals from both phases to maximize the data rate.
To manage this process distributively, the authors introduce two "share auction" mechanisms:
- SNR Auction: Users pay based on the Signal-to-Noise Ratio (SNR) improvement they receive. This mechanism is designed to promote fairness, ensuring that the marginal utility is equalized among users.
- Power Auction: Users pay based on the actual amount of relay power they consume. This mechanism is designed to promote efficiency, maximizing the total network throughput.
Unlike "winner-take-all" auctions, these share auctions allocate power proportionally based on the size of a user's bid relative to the total pool. Users update their bids iteratively based on limited feedback from the relay, a process proven to converge to a unique Nash Equilibrium.
Performance Evaluation
The mechanisms were evaluated using numerical simulations of various network topologies, including a controlled two-user scenario and a larger 20-user random distribution. A key finding is the inherent trade-off between fairness and efficiency.
As shown in the comparative analysis, the Power Auction (Delta-RSum-Power) closely tracks the performance of the optimal, centralized VCG auction (Delta-RSum-VCG), achieving approximately 95% of the maximum possible rate increase. In contrast, the SNR Auction (Delta-RSum-SNR) provides a lower total rate increase but ensures that the benefits of cooperation are distributed more equitably among the participating users.
Key Insights and Conclusion
The research demonstrates that distributed auctions can effectively manage complex, non-convex optimization problems in wireless networks. By using a simple threshold policy, users can locally decide "when to relay" based on the current price of resources.
While the current model focuses on single-relay scenarios and the AF protocol, it provides a robust foundation for future work in multi-relay environments and dynamic channel conditions. The proven geometric convergence of the bidding process makes these auction mechanisms highly suitable for real-time implementation in next-generation wireless systems.
Summary updated