1: \begin{abstract}

2: Weighted sum-rate (WSR) maximization plays a critical role in communication

3: system design. This paper examines three optimization methods for

4: WSR maximization, which ensure convergence to stationary points: two

5: block coordinate ascent (BCA) algorithms, namely, weighted sum-minimum

6: mean-square error (WMMSE) and WSR maximization via fractional programming

7: (WSR-FP), along with a minorization-maximization (MM) algorithm, WSR

8: maximization via MM (WSR-MM). Our contributions are threefold. Firstly,

9: we delineate the exact relationships among WMMSE, WSR-FP, and WSR-MM,

10: which, despite their extensive use in the literature, lack a comprehensive

11: comparative study. By probing the theoretical underpinnings linking

12: the BCA and MM algorithmic frameworks, we reveal the direct correlations

13: between the equivalent transformation techniques, essential to the

14: development of WMMSE and WSR-FP, and the surrogate functions pivotal

15: to WSR-MM. Secondly, we propose a novel algorithm, WSR-MM+, harnessing

16: the flexibility of selecting surrogate functions in MM framework.

17: By circumventing the repeated matrix inversions in the search for

18: optimal Lagrange multipliers in existing algorithms, WSR-MM+ significantly

19: reduces the computational load per iteration and accelerates convergence.

20: Thirdly, we reconceptualize WSR-MM+ within the BCA framework, introducing

21: a new equivalent transform, which gives rise to an enhanced version

22: of WSR-FP, named as WSR-FP+. We further demonstrate that WSR-MM+ can

23: be construed as the basic gradient projection method. This perspective

24: yields a deeper understanding into its computational intricacies.

25: Numerical simulations corroborate the connections between WMMSE, WSR-FP,

26: and WSR-MM and confirm the efficacy of the proposed WSR-MM+ and WSR-FP+

27: algorithms.

28: \end{abstract}