Propagation and Rate-Aware Cell Switching Optimization in HAPS-Assisted Wireless Networks

Cell switching is a promising approach for improving energy efficiency in wireless networks; however, existing studies largely rely on simplified models and energy-centric formulations that overlook key performance-limiting factors. This paper revisits the cell switching concept by redefining its modeling assumptions and mathematical formulation, explicitly incorporating realistic propagation effects such as building entry loss (BEL) and atmospheric losses relevant to non-terrestrial networks (NTN), particularly high-altitude platform station (HAPS). Beyond proposing a new cell switching strategy, the conventional energy-focused problem is reformulated as a multi-objective optimization framework that jointly minimizes power consumption, unconnected users, and data rate degradation. Through this reformulation, the proposed methods ensure that energy-efficient operation is achieved without compromising user connectivity and data rate performance, thereby inherently supporting sustainability objectives for sixth-generation (6G) networks. To solve this reformulated problem, two complementary approaches are employed: the weighted sum method (WSM), which enables flexible and adaptive weighting mechanism, and the {ε-constraint-inspired method (εCM), which converts connectivity and rate-related objectives into constraints within the conventional energy-focused problem. Moreover, unlike prior work relying only on simulations, this study combines system-level simulations with Sionna-OpenAirInterface (OAI) based emulation on a smaller network to validate the proposed cell switching concept under realistic conditions. The results show that, compared to the conventional approach, WSM reduces rate degradation for up to 70% for high-loss indoor users and eliminates the 44% drop for low-loss indoor users.

💡 Research Summary

The paper revisits the concept of cell switching in the context of high‑altitude platform stations (HAPS) assisting terrestrial wireless networks. While prior work has largely treated cell switching as a pure energy‑saving problem—switching off lightly loaded base stations (BSs) to reduce power consumption—it has ignored realistic propagation effects that can dramatically affect user connectivity and data rates after a BS is turned off. The authors therefore introduce a more comprehensive model that explicitly incorporates building entry loss (BEL) and atmospheric attenuation (gas and rain) into the path‑loss calculations for both terrestrial (TN) and non‑terrestrial (NTN) links. The TN link follows the 3GPP Urban Macro (UMa) model with separate expressions for line‑of‑sight (LoS) and non‑LoS conditions, while the NTN link to the HAPS‑super‑macro BS (HAPS‑SMBS) adds free‑space loss, atmospheric loss, BEL, and a user‑type specific loss term.

Beyond the propagation model, the authors adopt the EARTH power model to capture the dynamic power consumption of each BS, including transmit power, circuit power, efficiency, and sleep power. Users are allocated to the BS that offers the highest SINR, provided the BS has enough resource blocks (RBs) and the received signal exceeds a reference sensitivity. The conventional energy‑focused cell switching formulation (EFM) minimizes total network power subject only to load constraints and binary on/off decisions for the small cells (SBSs). This formulation, however, can lead to a surge in unconnected users and dissatisfied users whose post‑switching data rates fall below their pre‑switching rates.

To address these shortcomings, the paper reformulates the problem as a multi‑objective optimization that simultaneously minimizes (i) total power consumption, (ii) the number of unconnected users, and (iii) the number of dissatisfied users. Two solution strategies are proposed:

1. Weighted Sum Method (WSM) – The three objectives are combined into a single scalar objective using normalized weights α, β, and ν. By adjusting these weights, network operators can prioritize energy savings, connectivity, or quality‑of‑service according to operational policies. The resulting problem remains a mixed‑integer non‑linear program (MINLP).

2. ε‑Constraint Inspired Method (εCM) – The original power‑minimization problem is retained as the primary objective, while the connectivity and rate objectives are enforced as explicit constraints (e.g., post‑switching unconnected users must not exceed a threshold ε₁, and post‑switching data rates must be at least as high as pre‑switching rates ε₂). This approach guarantees service quality while still seeking power reduction.

Both methods are NP‑hard; therefore, the authors evaluate three algorithmic approaches within the εCM framework: an exhaustive search (for small instances), a greedy heuristic, and a genetic algorithm (GA) as a meta‑heuristic. The GA encodes the SBS on/off vector as chromosomes and evolves solutions through selection, crossover, and mutation.

The performance evaluation uses a 1 km × 1 km area populated with one HAPS‑SMBS, one macro BS (MBS), and multiple SBSs. Users are divided equally into three categories: high‑loss indoor (high BEL), low‑loss indoor (low BEL), and outdoor (no BEL). BEL is varied from 0 to 30 dB to assess sensitivity, while atmospheric loss is held constant. Each user is assigned a single RB, ensuring a uniform traffic demand per simulation interval. User mobility follows a random‑walk model.

Key findings include:

• WSM reduces data‑rate degradation for high‑loss indoor users by up to 70 % compared with the conventional EFM, and completely eliminates the 44 % rate drop observed for low‑loss indoor users.
• εCM achieves comparable power savings (≈15–20 % reduction relative to EFM) while guaranteeing that the number of unconnected users does not increase and that post‑switching rates meet or exceed pre‑switching rates.
• The inclusion of realistic BEL and atmospheric losses leads to more prudent SBS switching decisions; many SBSs that would be switched off under a pure power model remain active to preserve coverage for indoor users.
• The hybrid validation framework—large‑scale system‑level simulations combined with a Sionna‑OpenAirInterface (OAI) emulation of a smaller network—demonstrates that the proposed algorithms perform consistently under realistic PHY‑MAC conditions, not just abstract link‑budget calculations.

The paper’s contributions are fourfold: (1) derivation of propagation loss models that incorporate BEL and atmospheric attenuation for HAPS‑assisted networks; (2) reformulation of cell switching as a multi‑objective problem that jointly addresses energy, connectivity, and rate performance; (3) development of two complementary solution methods (WSM and εCM) and their algorithmic implementations; (4) validation through both system‑level simulations and a state‑of‑the‑art Sionna‑OAI emulator, a combination rarely seen in the literature.

In summary, by integrating realistic propagation effects and multi‑objective optimization, the study provides a practical pathway toward energy‑efficient yet service‑aware cell switching in future 6G networks that leverage HAPS for wide‑area offloading. The results suggest that sustainability goals can be met without sacrificing user experience, offering valuable guidance for network operators planning heterogeneous terrestrial‑non‑terrestrial deployments.