Energy-Aware Competitive Power Allocation for Heterogeneous Networks Under QoS Constraints

Giacomo Bacci, Elena Veronica Belmega, Panayotis Mertikopoulos, Luca Sanguinetti

Abstract

This work proposes a distributed power allocation scheme for maximizing energy efficiency in the uplink of orthogonal frequency-division multiple access (OFDMA)-based heterogeneous networks (HetNets). The user equipment (UEs) in the network are modeled as rational agents that engage in a non-cooperative game where each UE allocates its available transmit power over the set of assigned subcarriers so as to maximize its individual utility (defined as the user's throughput per Watt of transmit power) subject to minimum-rate constraints. In this framework, the relevant solution concept is that of Debreu equilibrium, a generalization of Nash equilibrium which accounts for the case where an agent's set of possible actions depends on the actions of its opponents. Since the problem at hand might not be feasible, Debreu equilibria do not always exist. However, using techniques from fractional programming, we provide a characterization of equilibrial power allocation profiles when they do exist. In particular, Debreu equilibria are found to be the fixed points of a water-filling best response operator whose water level is a function of minimum rate constraints and circuit power. Moreover, we also describe a set of sufficient conditions for the existence and uniqueness of Debreu equilibria exploiting the contraction properties of the best response operator. This analysis provides the necessary tools to derive a power allocation scheme that steers the network to equilibrium in an iterative and distributed manner without the need for any centralized processing. Numerical simulations are then used to validate the analysis and assess the performance of the proposed algorithm as a function of the system parameters.

Introduction

As the telecommunications industry moves toward 5G and beyond, network capacity requirements are increasing thousandfold, yet this must be achieved under extremely tight energy budgets. Small-cell heterogeneous networks (HetNets) address this by bringing access points closer to users, but they introduce significant interference management challenges. Traditional spectral efficiency approaches often lead to a "cascade of power increases," causing battery depletion and network instability. Furthermore, most decentralized models fail to account for strict Quality of Service (QoS) constraints, such as the minimum data rate required for applications to function. This paper addresses the challenge of maximizing energy efficiency (bits per Joule) while guaranteeing these minimum rate requirements in a multi-user, multi-cell environment without a central controller.

Methodology

The proposed framework utilizes non-cooperative game theory, specifically the concept of a Debreu equilibrium (or Generalized Nash Equilibrium). Unlike standard Nash games, the Debreu framework handles coupled constraints where one user's power choice affects the feasibility of another user's QoS requirements.

Key technical components include:

  • Utility Function: Defined as throughput per Watt of transmit power, incorporating both radiated power and "circuit power" (the fixed energy cost of device electronics).
  • Fractional Programming: Employed to transform the non-convex energy efficiency maximization problem into a tractable form.
  • Water-Filling Operator: The optimal power profile is shown to follow a water-filling pattern, where the water level depends on the Lambert W function, circuit power, and QoS rate thresholds.
  • Contraction Mapping: Mathematical proofs establish conditions for the existence and uniqueness of a unique equilibrium, ensuring the iterative algorithm's convergence.

Key Insights: The Impact of Circuit Power

A critical finding of this research is the role of "circuit power." Most existing models ignore the fixed energy cost of keeping a radio active. This work demonstrates that as circuit power increases, users are naturally incentivized to transmit at higher radiative powers to "justify" the fixed energy expenditure. This significantly shifts the equilibrium point of the network.

Utility and rate functions for different normalized circuit power levels (a) $\mu_k \cdot p_{c,k} = 1$ and (b) $\mu_k \cdot p_{c,k} = 10$.

Furthermore, the Debreu framework proves more realistic for modern networks than standard Nash equilibria because it forces rational agents to operate within the feasible set of the entire network's QoS limits. The resulting algorithm is computationally light; a device only requires its own signal-to-interference-plus-noise ratio (SINR) feedback to determine its optimal power allocation.

Simulation Results and Performance

The algorithm was evaluated using numerical simulations based on standardized 3GPP models, including a 24-tap channel model and a topology featuring a macrocell area with five small cells and up to 40 users.

Key results include:

  • Convergence: The distributed algorithm typically reaches equilibrium within 20 iterations, making it suitable for real-world LTE/5G slot durations.
  • Performance Gains: The HetNet configuration achieved an Area Spectral Efficiency (ASE) exceeding 600 b/s/Hz/km², significantly outperforming macro-only setups.
  • Antenna Scaling: Increasing receive antennas from 2 to 8 at small cells resulted in a five-fold increase in utility, highlighting the benefits of Massive MIMO for energy efficiency.
  • Robustness: The algorithm demonstrated the ability to quickly find new equilibria in response to user mobility or changes in base station association.

Conclusion

While the proposed scheme offers a robust, distributed solution for energy-efficient power allocation, certain limitations remain. In highly congested scenarios, the game may become infeasible if all users cannot simultaneously meet their minimum rate constraints. Future work will explore recovery mechanisms for such cases, as well as the impact of high-speed mobility on channel stability and the potential benefits of advanced interference-cancellation receivers.

Summary updated

Abstract

This work proposes a distributed power allocation scheme for maximizing energy efficiency in the uplink of orthogonal frequency-division multiple access (OFDMA)-based heterogeneous networks (HetNets). The user equipment (UEs) in the network are modeled as rational agents that engage in a non-cooperative game where each UE allocates its available transmit power over the set of assigned subcarriers so as to maximize its individual utility (defined as the user's throughput per Watt of transmit power) subject to minimum-rate constraints. In this framework, the relevant solution concept is that of Debreu equilibrium, a generalization of Nash equilibrium which accounts for the case where an agent's set of possible actions depends on the actions of its opponents. Since the problem at hand might not be feasible, Debreu equilibria do not always exist. However, using techniques from fractional programming, we provide a characterization of equilibrial power allocation profiles when they do exist. In particular, Debreu equilibria are found to be the fixed points of a water-filling best response operator whose water level is a function of minimum rate constraints and circuit power. Moreover, we also describe a set of sufficient conditions for the existence and uniqueness of Debreu equilibria exploiting the contraction properties of the best response operator. This analysis provides the necessary tools to derive a power allocation scheme that steers the network to equilibrium in an iterative and distributed manner without the need for any centralized processing. Numerical simulations are then used to validate the analysis and assess the performance of the proposed algorithm as a function of the system parameters.
BibTeX
@article{bacci2015energy,
  author = {Giacomo Bacci and Elena Veronica Belmega and Panayotis Mertikopoulos and Luca Sanguinetti},
  title = {{Energy-Aware Competitive Power Allocation for Heterogeneous Networks Under QoS Constraints}},
  journal = {IEEE Trans. Wireless Communications},
  year = {2015},
  url = {https://openreview.net/forum?id=kS5QkMpIuj_},
}
Figure