Knowledge Base

Geodata Requirements: Macro Cells vs Small Cells in 5G | LuxCarta

Written by LuxCarta | Apr 15, 2026 8:13:02 AM

Yes. Macro cells and small cells require different geodata specifications because they operate at different heights relative to the clutter layer, cover vastly different geographic footprints, and use different propagation models. Macro cells tolerate 5 to 10 m resolution terrain and clutter data; small cells at street level, especially at mmWave frequencies, require sub-2 m 3D building data, per-tree vegetation height, and in some cases wall-level obstacle data.

What Geodata Does a Macro Cell Deployment Need?

Macro cells mount antennas at 20 to 50 m height, typically above most or all of the surrounding clutter layer. The antenna looks down over buildings and vegetation, making rooftop-level accuracy the critical requirement.

Minimum macro cell geodata stack:

  • DTM: 5 to 10 m resolution, ≤2 m vertical RMSE
  • LULC/Clutter: 10 to 25 m resolution, standard morphology classes (8 to 12)
  • Clutter heights / DSM: 5 to 10 m resolution for height-above-terrain of each clutter pixel
  • 3D Buildings (urban areas): LOD1 (box model) per-building heights at the macro cell coverage scale

For sub-6 GHz macro cell planning, LiDAR-precision data is not required. Satellite-derived DSMs at 5 to 10 m resolution, combined with LULC at the same resolution, provide sufficient accuracy for empirical and semi-deterministic models like COST-Hata, 3GPP UMa, and Atoll/Planet's standard urban propagation models.

What Geodata Does a Small Cell or mmWave Deployment Need?

Small cells mount below or within the clutter layer, on street poles, building facades, or low rooftops at 5 to 15 m. The antenna is surrounded by obstructions on all sides, making the exact geometry of nearby buildings and vegetation the dominant factor in coverage prediction.

Minimum small cell geodata stack:

  • DTM: 1 to 2 m resolution, ≤1 m vertical RMSE
  • LULC/Clutter: 50 cm to 1 m resolution, 15 to 19 classes including street-level features
  • 3D Buildings: LOD2 per-building models with individual roof geometry, not just box heights
  • 3D Vegetation: Per-tree height with trunk/canopy separation, critical for street-level propagation
  • DSM: 1 to 2 m resolution consistent with building data
  • Road Network: Centerline and width for street canyon propagation modeling

For mmWave small cells at 26 to 28 GHz, additional data that is optional for macro cells becomes important:

  • Wall/Fence data: obstacles at street level can impose significant diffraction loss at 26 GHz
  • Rooftop structures: parapets, elevator shafts, and mechanical equipment become relevant obstructions when Fresnel zones are narrow

Key Differences at a Glance

Specification Macro Cell Small Cell (sub-6 GHz) Small Cell (mmWave)
DTM resolution 5 to 10 m 2 to 5 m 1 to 2 m
Building data LOD1 heights LOD1 to LOD2 LOD2 with roof geometry
LULC resolution 10 to 25 m 5 to 10 m 50 cm to 1 m
Vegetation Clutter class (generic) 3D tree height recommended 3D tree height and canopy required
Wall/fence data Not needed Not needed Beneficial
Propagation model Empirical / semi-empirical Semi-deterministic Deterministic / ray-tracing

How LuxCarta Addresses This

LuxCarta provides geodata at the full specification range required by both macro and small cell deployments. Building footprints with per-building heights at 93%+ capture rate meet the precision threshold for 5G small cell planning at sub-6 GHz. For mmWave deployments, LuxCarta's 3D vegetation data (individual tree polygons with trunk/canopy separation) addresses the street-level attenuation modeling gap that causes systematic errors in small cell predictions at 26 GHz. LuxCarta's wall and fence extraction capability, presented at IGARSS 2024 with 80.31% precision and 86.32% recall, provides the next level of granularity for dense urban mmWave small cell environments where street furniture and boundary structures contribute meaningfully to the obstacle geometry.

Frequently Asked Questions

Can I use the same geodata dataset for both macro and small cell planning?

Yes, if the dataset meets the small cell specification, which is more demanding. A high-resolution dataset (1 to 2 m DTM, LOD2 buildings, 3D vegetation) satisfies both macro and small cell requirements. The reverse is not true: macro-cell-grade data (10 m DTM, LOD1 buildings) is insufficient for small cell planning and will produce unreliable predictions at street level.

Does iBwave require different geodata from outdoor planning tools like Atoll?

iBwave is used primarily for indoor and venue-scale planning, where the relevant geodata is building interior geometry (floor plans, wall materials, room dimensions) rather than outdoor 3D city models. Outdoor-to-indoor penetration loss modeling in iBwave does use external building facade data, but the primary dataset for iBwave is architectural, not geospatial in the traditional RF planning sense.

What resolution is good enough for sub-6 GHz small cell planning?

For sub-6 GHz small cell planning (3.5 GHz typical), 2 to 5 m building height data with LOD1 to LOD2 geometry, and LULC at 5 to 10 m resolution, provides sufficient accuracy for most deployment scenarios. Going to 1 m resolution adds incremental accuracy for dense urban areas where building geometries are complex and small cell spacing is tight (under 200 m inter-site distance).

LuxCarta provides AI-powered 3D geospatial data solutions for telecom, simulation, and smart city applications worldwide. Learn more at luxcarta.com or explore on-demand extraction at BrightEarth.