Once and for All: Scheduling Multiple Users Using Statistical CSI under Fixed Wireless Access

Xin Guan, Zhixin Chen, Yibin Kang, Qingjiang ShiarXiv 2023 · DOI · arXiv

Abstract

Conventional multi-user scheduling schemes are designed based on instantaneous channel state information (CSI), requiring decisions every transmission time interval (TTI) of a few milliseconds. This paper investigates the scheduling problem of a fixed wireless access (FWA) network using only statistical CSI. Leveraging the fixed positions of user terminals, we propose an "once-and-for-all" scheduling approach: making a single high-quality scheduling decision that lasts across multiple TTIs. We address the resulting mixed-integer non-smooth non-convex stochastic problem using sigmoid smoothing, deterministic equivalent approximations, and an accelerated Frank-Wolfe algorithm. Numerical results verify the effectiveness of our proposed scheme over state-of-the-art benchmarks.

The FWA Scheduling Challenge

In modern MU-MIMO cellular networks, base stations must rapidly decide which users to group together and how to allocate power. Traditionally, this relies on Instantaneous Channel State Information (I-CSI). Because I-CSI changes every few milliseconds, scheduling decisions must be made extremely fast, often forcing the use of simple but suboptimal "greedy" algorithms.

Fixed Wireless Access (FWA)—where receivers are stationary devices on homes—offers a unique opportunity. Because these devices do not move, their Statistical CSI (S-CSI) remains stable for hundreds of milliseconds. This paper proposes using this stable data to perform a sophisticated optimization once, rather than making rushed decisions every millisecond.

FWA network architecture with multiple stationary users (CPEs) served by a single base station.

Methodology: Solving the "Mathematical Nightmare"

The authors frame scheduling as a mixed-integer, non-smooth, non-convex stochastic optimization problem. They employ several advanced techniques to find a solution:

  1. Soft Approximation: Hard "on/off" indicator functions are replaced with smooth sigmoid functions, allowing for gradient-based optimization.
  2. Deterministic Equivalent (DE): Using random matrix theory, they derive a closed-form mathematical approximation of the ergodic data rate, avoiding time-consuming simulations.
  3. Implicit Optimization: To handle the fixed-point equations within the DE formulas, the authors use implicit differentiation to derive explicit gradients.
  4. Accelerated Frank-Wolfe (AFW): This projection-free algorithm efficiently handles power and user constraints, ensuring rapid convergence.

Key Results

The proposed algorithm was tested against the state-of-the-art Semi-Orthogonal User Selection (SUS) method using both Rayleigh and high-fidelity Quadriga channel models.

Average relative error of the Deterministic Equivalent approximation as a function of the number of base station antennas.

A critical finding is the accuracy of the Deterministic Equivalent approximation. As shown in the figure above, the relative error drops significantly as the number of base station antennas increases, reaching near-zero at 64 antennas. This confirms the approach is perfectly suited for Massive MIMO systems. Overall, the "once-and-for-all" strategy significantly outperformed traditional greedy benchmarks in total network throughput.

Conclusion and Future Work

By shifting from update frequency to optimization quality, this research demonstrates that S-CSI can drive superior scheduling in FWA networks. Future work will explore the impact of CSI estimation errors and extend the framework to multi-cell environments where inter-tower interference must be considered.

Summary updated

Abstract

Conventional multi-user scheduling schemes are designed based on instantaneous channel state information (CSI), indicating that decisions must be made every transmission time interval (TTI) which lasts at most several milliseconds. Only quite simple approaches can be exploited under this stringent time constraint, resulting in less than satisfactory scheduling performance. In this paper, we investigate the scheduling problem of a fixed wireless access (FWA) network using only statistical CSI. Thanks to their fixed positions, user terminals in FWA can easily provide reliable large-scale CSI lasting tens or even hundreds of TTIs. Inspired by this appealing fact, we propose an \emph{`once-and-for-all'} scheduling approach, i.e. given multiple TTIs sharing identical statistical CSI, only a single high-quality scheduling decision lasting across all TTIs shall be taken rather than repeatedly making low-quality decisions every TTI. The proposed scheduling design is essentially a mixed-integer non-smooth non-convex stochastic problem with the objective of maximizing the weighted sum rate as well as the number of active users. We firstly replace the indicator functions in the considered problem by well-chosen sigmoid functions to tackle the non-smoothness. Via leveraging deterministic equivalent technique, we then convert the original stochastic problem into an approximated deterministic one, followed by linear relaxation of the integer constraints. However, the converted problem is still highly non-convex due to implicit equation constraints introduced by deterministic equivalent. To address this issue, we employ implicit optimization technique so that the gradient can be derived explicitly, with which we propose an algorithm design based on accelerated Frank-Wolfe method. Numerical results verify the effectiveness of our proposed scheduling scheme over state-of-the-art.
BibTeX
@misc{guan2023once,
  author = {Xin Guan and Zhixin Chen and Yibin Kang and Qingjiang Shi},
  title = {{Once and for All: Scheduling Multiple Users Using Statistical CSI under Fixed Wireless Access}},
  year = {2023},
  eprint = {2304.14225},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2304.14225},
}
Figure