Rural Access Disparity Reaches 1.65x in LEO Spectrum Sharing, Research Demonstrates

Dynamic spectrum sharing is crucial for the successful operation of multiple low Earth orbit (LEO) satellite constellations, but current approaches often overlook the issue of equitable access. Seyed Bagher Hashemi Natanzi from Worcester Polytechnic Institute, Hossein Mohammadi and Vuk Marojevic from Mississippi State University, along with Bo Tang, demonstrate that standard spectrum allocation policies can unintentionally worsen the digital divide between urban and rural areas. Their research, based on large-scale simulations of 3GPP-compliant networks, reveals a significant bias favouring users in areas with better satellite connections, a disparity that actually increases with greater bandwidth. To address this, the team developed FairShare, a new framework that enforces geographic fairness without compromising system efficiency, reducing scheduler runtime while actively mitigating the urban-rural access gap. This work offers a vital diagnostic tool for regulators and a practical solution for ensuring equitable spectrum governance in the rapidly expanding field of satellite communications.

DSS Bias and the Rural Digital Divide

Dynamic spectrum sharing (DSS) among multi-operator low Earth orbit (LEO) mega-constellations is essential for coexistence, yet prevailing policies focus almost exclusively on interference mitigation, leaving geographic equity largely unaddressed. This work investigates whether conventional DSS approaches inadvertently exacerbate the rural digital divide, a critical concern for global connectivity. Researchers employed large-scale, 3GPP-compliant non-terrestrial network (NTN) simulations with geographically distributed users to systematically evaluate standard allocation policies. Results uncover a stark and persistent structural bias, identified as SNR-priority, within current DSS mechanisms.

This bias consistently favours users in areas with higher signal-to-noise ratios, effectively prioritising densely populated regions with existing infrastructure. Consequently, rural and remote areas experience significantly reduced access to spectrum resources, hindering their ability to benefit from LEO satellite constellations. The research demonstrates that this disparity is not merely a consequence of propagation conditions, but an inherent feature of the allocation policies themselves.

Ka-band Spectrum Access Disparity in LEO Constellations

The increasing number of low Earth orbit (LEO) satellite constellations, operated by companies such as Starlink, OneWeb and Kuiper, is creating intense competition for shared Ka-band spectrum. Dynamic spectrum sharing (DSS) is becoming essential for these systems to coexist, but current regulations primarily focus on mitigating interference rather than ensuring equitable access across different geographic locations. Researchers have identified a significant disparity in spectrum access between urban and rural users, finding that typical scheduling methods induce an urban-rural access disparity, favouring those with better satellite positioning. This disparity stems from the physical limitations of non-terrestrial networks (NTN), where a user’s signal quality is heavily influenced by satellite elevation geometry.

Rural and remote users experience lower elevation angles and longer signal paths, leading to increased free-space path loss and atmospheric attenuation. Existing fairness metrics are unsuitable for LEO networks as they do not account for these elevation-dependent propagation effects. To address this, a new framework called FairShare has been proposed, a lightweight, quota-based system designed to enforce geographic fairness. Experiments demonstrate that FairShare not only reverses the existing bias, achieving a disparity ratio of 0.72×, but also reduces scheduler runtime by 3.3%. Surprisingly, increasing system bandwidth actually amplifies the disparity, increasing it from 1.0× to 1.65× as resources expand, highlighting the need for algorithmic intervention. The findings suggest that achieving fairness in next-generation satellite networks requires proactive algorithmic solutions, rather than relying solely on increased bandwidth or traditional terrestrial fairness metrics. FairShare provides both a diagnostic metric for regulators and a practical mechanism for equitable spectrum governance, ensuring all users, regardless of location, have fair access to this vital resource.

Urban-Rural Spectrum Access Disparity in LEO DSS

Scientists have demonstrated a structural bias in current dynamic spectrum sharing (DSS) policies for low Earth orbit (LEO) mega-constellations, revealing a disparity in access between urban and rural users. Through large-scale, 3GPP-compliant network simulations, the research team measured a 1.65x urban-rural access disparity induced by standard signal-to-noise ratio (SNR)-priority scheduling. This means rural users experienced considerably less access to spectrum compared to their urban counterparts due to less favourable satellite geometry. Experiments revealed that this disparity isn’t simply a result of limited resources; increasing system bandwidth from 50MHz to 300MHz actually amplified the gap, raising the disparity from 1.0x to 1.65x.

To address this inequity, the researchers developed FairShare, a quota-based framework designed to enforce geographic fairness in spectrum allocation. Tests confirm that FairShare reverses the inherent bias, achieving a disparity ratio of 0.72x, effectively prioritising equitable access for rural users. Importantly, this improvement in fairness was achieved without compromising system efficiency; the team recorded a 3.3% reduction in scheduler runtime. The study meticulously evaluated conventional allocation policies using geographically distributed users and 3GPP TR 38.811-compliant channel models.

Data shows that SNR-priority scheduling and demand-proportional allocation consistently produced disparities of 1.65x and 1.40x respectively, highlighting the policy-inherent nature of the problem. The FairShare framework introduces geographic quotas, partitioning spectrum access to ensure equitable distribution. This intervention not only eliminates urban bias but actively promotes fairness, delivering a positive disparity ratio indicating improved access for rural communities.

FairShare Mitigates Urban-Rural Access Disparity

This research demonstrates that standard spectrum allocation policies within multi-operator low Earth orbit networks create a significant disparity in access between urban and rural users. Simulations reveal that signal-to-noise ratio prioritisation inherently favours those with better satellite geometry, resulting in a 1.65times greater access rate for urban compared to rural areas, a bias that worsens with increased bandwidth. These findings establish a misalignment between conventional approaches and the goal of equitable geographic access to satellite-based connectivity. To address this, researchers developed FairShare, a quota-based framework designed to enforce geographic fairness in spectrum allocation.

Implementation of FairShare not only reversed the observed bias, achieving a 0.72times disparity ratio, but also improved scheduler efficiency by reducing runtime. The study highlights that explicit geographic partitioning is both necessary and sufficient to achieve equitable access, offering a diagnostic metric and enforceable mechanism for regulators to promote fairer spectrum governance. The authors acknowledge that validating these simulation results with real-world low Earth orbit measurements represents a logical next step. Furthermore, they suggest that FairShare establishes a baseline for equitable service access, upon which further enhancements to network performance can be built.