High-Altitude Platform Stations (HAPS)

• High-Altitude Platform Stations (HAPS) are quasi-stationary aerial platforms in the stratosphere offering ultra-wide wireless coverage and energy-efficient operations.
• HAPS integrate large mMIMO arrays, photovoltaic power systems, and advanced backhaul links to dynamically supplement terrestrial networks in urban environments.
• Performance analyses indicate HAPS achieve high capacity utilization and reduce power consumption by approximately 55% compared to dense small-cell deployments.

High-Altitude Platform Stations (HAPS) are quasi-stationary aerial systems operating in the stratosphere, typically at altitudes around 20 km above ground, designed to provide ultra-wide-area wireless communications, data processing, and network services. Leveraging large payloads, photovoltaic energy autonomy, and advanced multi-antenna payloads, HAPS function as super-macro base stations (SMBSs) or overlays to terrestrial radio access networks (RANs). Their intrinsic characteristics—wide footprint, elevated line-of-sight probability, and energy efficiency—make HAPS a compelling solution for managing dynamic and unpredictable mobile traffic in dense urban environments, where conventional RAN densification leads to over-provisioning and energy inefficiency (Kement et al., 2022).

1. HAPS Platform Architecture and Network Integration

HAPS-SMBS platforms are deployed at approximately 20 km altitude with a coverage radius up to 35 km (minimum elevation angle 30°), enabled by large cylindrical mMIMO antenna arrays, substantial onboard computing and storage, and solar/battery-based power. The energy model is governed by the differential equation:

$$\frac{\mathrm{d}E_{\rm bat}}{\mathrm{d}t} = P_{\rm solar}(t) - P_{\rm comm}(t) - P_{\rm prop}(t) - P_{\rm avionics}(t)$$

where each term denotes, respectively, energy harvested by photovoltaic panels, energy consumed by communications, propulsion, and avionics subsystems.

The integration with terrestrial RANs is realized via an overlay design: the HAPS super-macro base station serves as a fallback for users whose resource requests are blocked or deferred by conventional ground base stations. The HAPS maintains wireless backhaul/fronthaul connections to a ground gateway node, using RF and/or free-space optical (FSO) links. Handover procedures encompass both horizontal (intra-terrestrial) and vertical (HAPS–BS) mobility, requiring new thresholding logic for vertical handover to mitigate ping-pong effects in strong LoS channels. Control is coordinated by a joint RAN controller for user-association and resource allocation (Kement et al., 2022).

2. Analytical Performance Models: Link, Capacity, and Energy

Path-Loss and Link-Budget

For HAPS-user LoS connections, the free-space path loss (FSPL) is:

$$L_{\rm FSPL}(d, f)_{\rm dB} = 20\log_{10}(d) + 20\log_{10}(f) + 20\log_{10}\!\left(\frac{4\pi}{c}\right)$$

Ground links to terrestrial BSs typically exhibit:

$$\mathrm{PL}(d) = K + 10\,n\,\log_{10}(d) + \chi_\sigma$$

with path-loss exponent $n=2$ for LoS (HAPS) and $n=4$ for NLoS (BS), $K$ as frequency-constant, and $\chi_\sigma$ log-normal shadowing.

Capacity and Spectral Efficiency

User rate and system sum-rate are derived from the instantaneous SNR:

$$R_i = B\,\log_2\left(1 + \gamma_i\right),\qquad C_{\rm sys} = \sum_{i\in\mathcal{U}} R_i$$

where spectral efficiency $\eta_{\rm spec}$ and capacity utilization $\eta_{\rm util}$ are

$$\eta_{\rm spec} = \frac{C_{\rm sys}}{B} \quad (\mathrm{bit/s/Hz}),\qquad \eta_{\rm util} = \frac{\sum_{i\in\mathcal{U}} R_i}{C_{\rm total}}$$

with $C_{\rm total}$ the aggregate RAN capacity.

Energy Consumption Model

Total HAPS-SMBS power consumption is given by

$$P_{\rm HAPS} = P_{\rm tx} + P_{\rm elec} + P_{\rm prop}$$

for transmission, electronics, and propulsion/avionics, respectively. The energy efficiency is implicitly measured as throughput per joule (bits/J). No explicit convex resource-allocation problem is formulated, but such a problem could minimize $P_{\rm HAPS}$ under rate and link constraints (Kement et al., 2022).

3. Urban Simulation Framework and Case-Study Parameters

The reference case paper simulates an $8\times8$ km$^2$ urban grid with 14,000 uniformly-distributed users ($\approx 219$/km$^2$), served by 36 macro-BSs (700 m radius, 1 Gbps each, IMT-2020-compliant). User traffic demand is modeled per-slot (1 min, 1,440 slots/day) as $D_i\sim|\mathcal{N}(0,20)|$ Mbps (mean $\approx 16$ Mbps), with random waypoint mobility. HAPS parameters include:

• Altitude: 20 km
• Coverage: 35 km radius
• Communication payload: 2–20 Gbps capacity
• Aggregate HAPS SMBS power: e.g., 140.6 kW for 2 Gbps capacity (Kement et al., 2022)

4. Comparative Performance of HAPS-SMBS vs. RAN Densification

Performance is evaluated using user-served ratio, capacity utilization, and total network power. For “Original + 2 Gbps HAPS” and “Original + 49 small-cells”, the observed metrics are:

Scenario	Capacity	Users Served	Cap. Util.	Power
Original + 2 Gbps HAPS	38 Gbps	100%	71.2%	140.6 kW
Original + 49 SC	85 Gbps	100%	31.3%	314.5 kW

Key operational regimes show HAPS solutions maintain 100% user service under higher average demand ($\bar D$), with higher capacity utilization, until a critical threshold marking saturation (Kement et al., 2022). Notably, HAPS-SMBS achieves:

• Double the capacity utilization (phase 1) relative to densification.
• Total network power consumption reduced by approximately 55%.
• Hardware deployment footprint orders-of-magnitude lower.

HAPS is therefore most advantageous for highly bursty, spatially unpredictable traffic.

5. Resource Management, Scalability, and Limitations

Resource management for HAPS-integrated RANs requires:

• Joint user association (dynamic selection of terrestrial BS vs HAPS).
• Adaptive ON/OFF scheduling of HAPS sectors.
• mMIMO-based beam-steering for interference mitigation.
• Refined vertical handover thresholds to prevent unnecessary HAPS-terrestrial ping-pong events (Kement et al., 2022).

A single HAPS can replace dozens to hundreds of terrestrial BSs in urban hotspots, providing scalable, persistent coverage over up to 500 km$^2$. The green energy model (solar + Li-ion battery) permits near-zero carbon operation and long endurance.

Persistent open challenges include:

• Optimizing handover and mobility control to match HAPS’s unique channel temporal statistics.
• Coordinated radio resource management across HAPS and terrestrial infrastructure for interference and spectrum efficiency.
• Onboard power constraints and the need for advanced PV, battery, or wireless power transfer solutions.
• Regulatory and certification hurdles for stratospheric operation.

6. Future Directions and Practical Implications

Recommended research priorities include:

• Design of integrated HAPS–terrestrial RAN controllers for seamless traffic offload and load balancing.
• Formal power-minimization resource allocation subject to dynamic rate and QoS constraints.
• Prototyping of full-scale HAPS-SMBS with authentic solar-battery subsystems and measurement of real-world power/traffic profiles.
• Regulatory engagement to establish airspace, frequency, and operational standards for HAPS deployment.

The empirical and analytical findings demonstrate that solar-powered HAPS at stratospheric altitudes can absorb transient and unpredictable urban mobile traffic far more sustainably than small-cell densification approaches, while ensuring full coverage and user service continuity (Kement et al., 2022).