Abstract
With energy-efficient resource allocation, mobile users and base stations have different objectives. While the base station strives for an energy-efficient operation of the complete cell, each user aims to maximize its own data rate. To obtain this individual benefit, users may selfishly adjust their reported channel state information (CSI). This paper models this interaction as a non-cooperative game, identifying a unique Nash Equilibrium and demonstrating that small-cell architectures provide inherent resilience against such selfish behavior.
Introduction
In modern cellular networks, such as LTE, the base station (BS) manages transmit power allocation based on Channel State Information (CSI) reported by mobile devices. A fundamental "principal-agent" conflict exists: the BS aims for maximum Energy Efficiency (EE)—bits delivered per Joule—while individual users seek to maximize their own Signal-to-Noise Ratio (SNR) for higher data rates. Since the BS cannot independently verify local channel conditions, users have an incentive to "selfishly" report degraded channel states to trick the BS into allocating more power.
Methodology
The study approaches this conflict in two stages. First, an energy-efficiency utility function is defined, and a low-complexity "inverse water-filling" algorithm is proposed to prioritize users with the best channel conditions. Second, the interaction is modeled as a non-cooperative game where users strategically report channel gains. Using i.i.d. Rayleigh fading models and extensive numerical simulations, the authors identify the Nash Equilibrium—a stable state where no user can unilaterally improve their SNR.
Key Insights: The Resilience of Small Cells
A critical finding is the existence of a unique Nash Equilibrium where all users report a specific "fake" CSI value, forcing the BS to distribute power equally ($P/K$). However, the impact of this selfishness varies with the cell's power budget.
As illustrated in Figure 4, while selfish reporting (equilibrium) reduces cell utility compared to honest reporting (actual), the degradation is less severe in "small cells" with lower power constraints ($P=0.1W$). In fact, the actual energy efficiency in the selfish scenario is higher for $P=0.1W$ than for $P=1W$, suggesting that small-cell architectures are inherently more robust against strategic manipulation.
Results: The User's Perspective
From the user's perspective, reporting false CSI is a rational strategy to achieve higher individual SNR.
Figure 6 visualizes this tradeoff. The honest reporting point (white square) corresponds to the region of maximum network utility (red). However, users naturally gravitate toward the Nash Equilibrium (blue square), which yields higher SNR for both individuals at the expense of overall network efficiency.
Conclusion
The research demonstrates that while selfish CSI reporting poses a challenge to energy-efficient resource allocation, its impact is mitigated in high-density user environments and small-cell deployments. Future work may consider the impact of discrete CSI reporting levels and the inclusion of user-side energy consumption, which could further align user incentives with network-wide energy goals.
Summary updated