Abstract
In a secondary spectrum market, primaries set prices for their unused channels to secondaries. The payoff of a primary depends on the availability of unused channels of its competitors. We consider a model where a primary can acquire its competitor's channel state information (C-CSI) at a cost. We formulate a game between two primaries where each primary decides whether to acquire C-CSI or not and then selects its price based on that. We first characterize the Nash Equilibrium (NE) of this game for a symmetric model where the C-CSI is perfect. We show that the payoff of a primary is independent of the C-CSI acquisition cost. We then generalize our analysis to allow for imperfect estimation and cases where the two primaries have different C-CSI costs or different channel availabilities. Our results show interestingly that the payoff of a primary increases when there is estimation error. We also show that surprisingly, the expected payoff of a primary may decrease when the C-CSI acquisition cost decreases when primaries have different availabilities.
Introduction
In dynamic spectrum access, primary license holders—such as TV broadcasters or telecom operators—can lease temporarily unused spectrum channels to secondary users. To maximize profit, a primary must set a competitive price. However, a primary typically operates under uncertainty regarding whether its competitors also have channels available for lease at any given moment.
A primary could potentially deploy sensors or pay a third party to acquire Competitor Channel State Information (C-CSI). This paper investigates whether such information is worth the cost. If a primary knows a competitor is active, it might lower its price to remain competitive. If the competitor also possesses this knowledge, it can trigger a "race to the bottom" price war. This research analyzes how the option to acquire C-CSI—considering acquisition costs, estimation errors, and differing channel availabilities—affects pricing strategies and overall payoffs in a duopoly market.
Model and Methodology
The scenario is modeled as a non-cooperative Bayesian game between two primary license holders and one secondary buyer. In each time slot, a primary with an available channel makes two simultaneous decisions:
- Information Acquisition: Whether to pay a cost ($s$) to acquire C-CSI (decision $Y$) or not (decision $N$).
- Pricing: Selecting a price for their channel, typically drawn from a probability distribution to prevent predictable undercutting.
The notations used throughout the analysis are summarized in the table below:
The authors solve for the Bayesian Nash Equilibrium (NE) and prove the existence of a $[T, p]$ class of strategies. In this framework, a primary acquires C-CSI with probability $p$ if the cost is below a certain threshold $T$, and does not acquire it otherwise. The study begins with a symmetric model (perfect information, equal costs, equal availability) and generalizes to handle imperfect C-CSI estimation, unequal acquisition costs, and heterogeneous channel availabilities.
Key Insights and Results
The analysis yields several counter-intuitive findings regarding the value of information in spectrum markets:
- Information Neutrality in Symmetric Markets: In the basic symmetric model, the expected payoff of a primary is independent of the C-CSI acquisition cost. The ability to acquire information about a competitor does not increase expected profits; it merely shifts the equilibrium pricing strategy.
- The Profitability of Uncertainty: Introducing errors into C-CSI estimation (accuracy $q_s < 1$) actually increases the primaries' payoffs. Uncertainty dampens aggressive price wars; if a primary is not entirely certain of a competitor's presence, it is less likely to drop prices drastically.
- Adverse Effects of Cheaper Information: When primaries have different baseline channel availabilities, reducing the cost of C-CSI can actually decrease the expected payoff for the primary with lower availability. This contradicts the general economic intuition that cheaper information is always beneficial.
- Equilibrium Randomization: There is no Nash Equilibrium where both primaries choose to acquire C-CSI 100% of the time. To remain competitive, they must randomize both their decision to acquire information and their pricing distributions.
The authors proved that the unique Nash Equilibrium strategy is a $[T, p]$ strategy, where the threshold $T$ peaks when uncertainty about a competitor's availability is highest (at $q = 0.5$). Furthermore, when estimation error is introduced, the threshold $T$ for acquiring information decreases, yet the overall expected payoff strictly increases compared to the perfect information scenario.
Conclusion and Future Work
While this study provides a rigorous mathematical foundation for understanding information acquisition in spectrum markets, it primarily focuses on a duopoly. While the authors provide some asymptotic results for $n > 2$ primaries, a comprehensive multi-player analysis remains a subject for future research. Additionally, the current model assumes a static, one-shot game per time slot. Real-world spectrum markets are continuous and dynamic. Future work could explore models where primaries and secondaries take a long-term view, incorporating elements like reputation, learning, and strategic behavior across multiple time slots.
Summary updated