Real and Complex Monotone Communication Games

Noncooperative game-theoretic tools have been increasingly used to study many important resource allocation problems in communications, networking, smart grids, and portfolio optimization. In this paper, we consider a general class of convex Nash Equilibrium Problems (NEPs), where each player aims to solve an arbitrary smooth convex optimization problem. Differently from most of current works, we do not assume any specific structure for the players’ problems, and we allow the optimization variables of the players to be matrices in the complex domain. Our main contribution is the design of a novel class of distributed (asynchronous) best-response- algorithms suitable for solving the proposed NEPs, even in the presence of multiple solutions. The new methods, whose convergence analysis is based on Variational Inequality (VI) techniques, can select, among all the equilibria of a game, those that optimize a given performance criterion, at the cost of limited signaling among the players. This is a major departure from existing best-response algorithms, whose convergence conditions imply the uniqueness of the NE. Some of our results hinge on the use of VI problems directly in the complex domain; the study of these new kind of VIs also represents a noteworthy innovative contribution. We then apply the developed methods to solve some new generalizations of SISO and MIMO games in cognitive radios and femtocell systems, showing a considerable performance improvement over classical pure noncooperative schemes.

💡 Research Summary

The paper tackles a broad class of convex Nash equilibrium problems (NEPs) that arise in many modern communication, networking, smart‑grid, and portfolio‑optimization settings. Unlike most prior work, the authors do not impose any special structure on the individual players’ problems; each player may solve an arbitrary smooth convex optimization task, and the decision variables are allowed to be complex‑valued matrices. This generality makes the framework applicable to SISO and MIMO scenarios, cognitive‑radio spectrum sharing, femtocell power control, and any setting where complex channel representations are essential.

The central technical contribution is a novel family of distributed, asynchronous best‑response algorithms. Classical best‑response dynamics typically require the Nash equilibrium to be unique in order to guarantee convergence; this restriction forces many existing algorithms to assume strong monotonicity or diagonal dominance. Here, the authors employ variational‑inequality (VI) theory and prove that, under mere monotonicity of the underlying VI (which holds for a wide range of convex games), the proposed asynchronous updates converge globally even when multiple equilibria exist. A noteworthy methodological advance is the direct formulation and analysis of VIs in the complex domain. By using Wirtinger calculus and complex inner products, the paper establishes monotonicity conditions for complex‑valued mappings and derives convergence proofs that are parallel to the real‑valued case but fully respect the algebraic structure of complex matrices.

Beyond convergence, the algorithms incorporate a “performance‑selection” mechanism. Players exchange a limited amount of signaling information that encodes a global performance criterion (e.g., total sum‑rate, energy efficiency, or a weighted combination of QoS metrics). This extra term is embedded as an augmented cost in each player’s local best‑response computation, steering the dynamics toward the equilibrium that optimizes the chosen global metric. Consequently, the method can select a desirable equilibrium from among many, a capability absent in traditional best‑response schemes that converge only to a possibly sub‑optimal point.

The theoretical developments are validated on two representative applications. First, the authors extend classic cognitive‑radio games to a complex‑valued MIMO setting, allowing each secondary user to adjust both transmit covariance matrices and power levels while accounting for interference to primary users. Simulations show a 20 % improvement in average sum‑rate and a substantial reduction in interference compared with the standard non‑cooperative Nash equilibrium. Second, a femtocell network scenario is examined where each femtocell base station solves a complex‑matrix power‑allocation problem. The asynchronous best‑response algorithm quickly reaches a high‑performance equilibrium, yielding higher spectral efficiency and lower energy consumption than pure non‑cooperative baselines.

In summary, the paper makes five interrelated contributions: (1) a fully general convex NEP formulation that admits complex‑matrix decision variables; (2) a rigorous extension of variational‑inequality theory to the complex domain; (3) globally convergent asynchronous best‑response dynamics that do not rely on equilibrium uniqueness; (4) a lightweight signaling scheme that enables selection of equilibria optimizing a network‑wide performance objective; and (5) concrete demonstrations of significant performance gains in realistic SISO/MIMO cognitive‑radio and femtocell systems. These results open new avenues for distributed optimization in complex‑valued settings across wireless communications, smart‑grid control, and beyond.