1: \begin{abstract}

2: In this two-parts paper we propose a decentralized strategy, based on a

3: game-theoretic formulation, to find out the optimal precoding/multiplexing

4: matrices for a multipoint-to-multipoint communication system composed of a

5: set of wideband links sharing the same physical resources, i.e., time and

6: bandwidth. We assume, as optimality criterion, the achievement of a Nash

7: equilibrium and consider two alternative optimization problems: 1) the

8: competitive maximization of mutual information on each link, given

9: constraints on the transmit power and on the spectral mask imposed by the

10: radio spectrum regulatory bodies; and 2) the competitive maximization of the

11: transmission rate, using finite order constellations, under the same

12: constraints as above, plus a constraint on the average error probability. In

13: Part I of the paper, we start by showing that the solution set of both

14: noncooperative games is always nonempty and contains only pure strategies.

15: Then, we prove that the optimal precoding/multiplexing scheme for both games

16: leads to a channel diagonalizing structure, so that both matrix-valued

17: problems can be recast in a simpler unified vector power control game, with

18: no performance penalty. Thus, we study this simpler game and derive

19: sufficient conditions ensuring the uniqueness of the Nash equilibrium.

20: Interestingly, although derived under stronger constraints, incorporating

21: for example spectral mask constraints, our uniqueness conditions have

22: broader validity than previously known conditions. Finally, we assess the

23: goodness of the proposed decentralized strategy by comparing its performance

24: with the performance of a Pareto-optimal centralized scheme. To reach the

25: Nash equilibria of the game, in Part II, we propose alternative distributed

26: algorithms, along with their convergence conditions.

27: \end{abstract}

28: