Causal modeling improves radio network scheduling and resource use
From Confounding to Counterfactuals: A Cross-Layer Causal Model for RAN Scheduling
Networking and Internet ArchitectureMachine Learning
Summary
Allocating resources in wireless networks can be tricky because the network environment affects both the scheduling decisions and performance, causing errors in understanding what helps user devices. The authors propose a method that uses a detailed causal model of different network layers to better predict how resource changes impact each user. Their approach uses counterfactual thinking—asking what would happen to a user if their resources were changed—to improve scheduling choices. Tests show this method predicts outcomes more accurately and boosts overall network performance compared to standard techniques.
What this means in practice
- •For network schedulers: Improve scheduling algorithms by incorporating causal models that separate confounding effects and identify user-specific responses to resource allocations.
- •For wireless network operators: Optimize resource distribution across users in radio networks to enhance overall performance while reducing misallocation caused by biased data associations.
Authors
Artin Elhamirad, Christo Kurisummoottil Thomas, Ali Tajer
Abstract
Data-driven control in radio access networks (RANs) selects resource allocations using associations between control actions and key performance indicators (KPIs) learned from network telemetry. Because the shared radio environment influences both the scheduler's decisions and the resulting performance, a bias known as confounding, such associations overstate the effect of an allocation and misdirect resources toward user equipment (UEs) that cannot benefit. Moreover, individual UEs respond heterogeneously to the same decision, so even a corrected average effect does not identify which UEs gain from additional resources. While causal inference has recently been applied to RANs, existing approaches do not model the cross-layer mechanisms that connect an allocation to its outcome. In this paper, a hierarchical structural causal model (H-SCM) spanning the PHY, MAC, and network layers is proposed that separates cell-load confounding from UE-specific response heterogeneity. In particular, a counterfactual scheduling problem is formulated in which physical resource block (PRB) allocations are selected using per-UE counterfactual goodput rather than observational conditional averages. This problem is solved via a pipeline combining constrained causal discovery, structural-equation fitting, and abduction-action-prediction. Simulations demonstrate that naive observational estimation overstates the full-PRB response by 73.3%, whereas adjustment for cell-load confounding reduces the relative error to 9.8%. Held-out interventions are predicted with a coefficient of determination of 0.988. Counterfactual scheduling achieves a mean relative loss in total cell goodput below 0.5% against the ground-truth optimum, compared with 2.0-2.9% for a neural-network baseline.