A Stackelberg Game-Theoretic Solution to Win-Win Situation: A Presale Mechanism in Spectrum Market

Wei Bai, Yuli Zhang, Meng Wang, Jin Chen, Han Jiang, Zhan Gao, Donglin Jiao

Abstract

This paper proposes a novel presale mechanism for spectrum allocation to address resource inefficiencies caused by uncertain user demand. By modeling the interaction between a spectrum manager and a user as a Stackelberg game, the authors demonstrate a win-win equilibrium. In this model, the manager increases profit through enhanced resource planning, while the user reduces costs via discounted advance bookings. Theoretical analysis and numerical simulations confirm that this mechanism can improve profits for both parties by approximately 10% compared to traditional models.

Introduction

In contemporary wireless communications, spectrum resources are frequently allocated statically, which often results in significant waste during periods of low demand and severe shortages during demand spikes. While spectrum pooling—where users request resources in real-time—offers a more flexible alternative, it remains plagued by the "uncertainty of demand." Managers struggle to determine optimal reservation levels, and users are often forced to pay premium prices for last-minute resource acquisition.

Existing research typically focuses on optimizing either the supply or demand side in isolation. This paper bridges that gap by introducing a proactive presale mechanism. This mechanism facilitates an "implicit interaction" between the manager and the user: by offering spectrum at a discount for advance booking, the manager obtains a reliable estimate of total demand, thereby reducing risk for both market participants.

The Presale Mechanism and Methodology

The authors employ Stackelberg Game Theory to model the hierarchical relationship within the spectrum market. The interaction is structured as follows:

  1. The Leader (Spectrum Manager): Determines the advance price ($\alpha_0$) and the reservation amount ($m_p$). The manager's goal is to maximize profit by balancing the costs of reserving resources against the higher costs of improvising resources in real-time.
  2. The Follower (User): Decides the booking amount ($s_p$) based on the manager's announced price. The user seeks to maximize their own profit, weighing the benefits of a lower advance price against the risk of over-booking beyond their actual demand.
Schematic of the presale mechanism interaction between the spectrum manager and the user.

Technical Implementation

The model relies on several key technical components:

  • Demand Modeling: User demand ($s_r$) is assumed to follow an exponential distribution, a standard approach for modeling random resource needs in communications.
  • Backward Induction: The Stackelberg Equilibrium (SE) is found by first solving the "Follower Sub-Game" (the user's optimal response) and then using that result to solve the "Leader Sub-Game" (the manager's optimal strategy).
  • Profit Functions: The model incorporates a "capital turnover coefficient" ($k$), acknowledging that advance payments provide immediate liquidity and value to the manager.

To validate the theoretical model, the authors conducted numerical simulations using synthetic parameters (e.g., costs $\beta_0=7, \beta_1=11$; normal price $\alpha_1=18$; unit profit $\gamma=20$).

Key Results and Insights

The primary finding is the existence of a "win-win" region where both the manager and the user achieve higher profits than in a system without presales.

Profit analysis showing the win-win region where both parties benefit from the presale mechanism.

Major Findings:

  • Profit Increase: At the Stackelberg Equilibrium, profits for both parties are approximately 10% higher than in traditional setups.
  • Information through Pricing: The presale acts as a communication tool. The user's booking amount serves as a signal, allowing the manager to allocate resources more precisely and reduce expensive "temporary scheduling."
  • Managerial Over-provisioning: The manager’s optimal reservation ($m_p$) typically exceeds the user's booking amount ($s_p$). This allows the manager to hedge against the high cost of real-time resource acquisition based on the known demand distribution.
  • Equilibrium Uniqueness: The authors mathematically prove that the SE in this model is both unique and optimal, ensuring a stable market interaction.

Conclusion and Future Work

The proposed presale mechanism successfully leverages information sharing through pricing to improve spectrum market efficiency. However, the current model is limited to a single-manager, single-user scenario and assumes that actual demand is independent of price to isolate the information-sharing effect.

Future research will explore multi-user scenarios where competing demands interact within the pool. Additionally, considering more complex demand distributions and integrating the mechanism with dynamic anti-jamming or interference management strategies could further enhance the practical utility of this approach in real-world wireless networks.

Summary updated

BibTeX
@article{bai2019stackelberg,
  author = {Wei Bai and Yuli Zhang and Meng Wang and Jin Chen and Han Jiang and Zhan Gao and Donglin Jiao},
  title = {{A Stackelberg Game-Theoretic Solution to Win-Win Situation: A Presale Mechanism in Spectrum Market}},
  journal = {IEICE Trans. Information and Systems},
  year = {2019},
  doi = {10.1587/transinf.2019edl8095},
}
Figure