Listen to audio version of this blog post:
LiDAR data is technically excellent for telecom RF planning, delivering point-cloud accuracy of 5 to 15 cm vertical and sub-meter horizontal that far exceeds what most propagation models require. However, its high collection cost, limited geographic coverage, and rapid obsolescence in changing urban environments make it cost-effective only for very specific, limited-area use cases, not as a general-purpose national planning dataset.
What Is LiDAR and How Is It Used in Telecom Planning?
LiDAR (Light Detection and Ranging) uses pulsed laser light from an aircraft or drone to measure distance to the ground and objects, generating a dense 3D point cloud. For telecom planning, this point cloud is processed into:
- Digital Terrain Models (DTM): bare-earth elevation after filtering out buildings and vegetation returns
- Digital Surface Models (DSM): first-return surface including all above-ground features
- Building footprints and heights: derived from the elevation difference between DTM and DSM
- Vegetation height maps: canopy height models showing tree heights per pixel
The resulting datasets are highly accurate and spatially detailed, exactly what deterministic 3D propagation models require for ray-tracing calculations.
What Are LiDAR's Advantages for RF Planning?
When LiDAR data is available for the planning area, the advantages are real:
- Vertical accuracy: Airborne LiDAR typically delivers 5 to 15 cm RMSE vertical accuracy, significantly better than satellite-derived DSM (typically 1 to 3 m RMSE)
- Penetration under canopy: Full-waveform LiDAR can capture ground returns through tree canopy, enabling accurate bare-earth DTM even in forested areas
- Point density: High-density surveys (8+ points per m²) resolve individual trees, parapets, rooftop structures, and other small features that affect mmWave propagation
- No imagery dependency: Unlike photogrammetric methods, LiDAR does not require cloud-free optical imagery
For small cell planning at 26 GHz, where centimeter-level obstacles affect the link budget, LiDAR accuracy is genuinely valuable.
What Are LiDAR's Limitations for Telecom Planning?
Despite its technical quality, LiDAR has significant practical limitations that constrain its suitability as the primary geodata source for most telecom planning projects:
Coverage Gaps
- LiDAR is primarily collected by government programs or specialized contractors for specific regions. National LiDAR coverage exists for only a minority of countries, primarily Western Europe, the US, and parts of Australia.
- Countries across Africa, Asia, the Middle East, and Latin America, which are major 5G investment areas, have little or no LiDAR coverage.
- Even where national programs exist, data is rarely current. Many European LiDAR datasets are 5 to 10 years old.
Cost
- Commissioning a new LiDAR survey runs approximately €500 to €1,500 per km² for standard-density airborne surveys, with drone-based surveys costing more for small areas with mobilization overhead.
- A 100 km² urban planning area could easily represent a €50,000 to €150,000 data acquisition cost before processing.
- This cost is difficult to justify for initial network planning, particularly when the data will be used once and then become outdated as the city develops.
Currency and Refresh Rate
- Buildings are constructed and demolished continuously. A LiDAR dataset collected 3 years ago may be missing significant new construction.
- Satellite-derived building extraction can be re-run on fresh imagery whenever new builds become relevant to network planning. LiDAR surveys cannot be refreshed economically at the same frequency.
Format and Processing Complexity
- Raw LiDAR point clouds require significant post-processing to become usable planning layers (DTM, DSM, building footprints with heights).
- This processing typically requires specialized software (such as LAStools, ArcGIS 3D Analyst, or ENVI) and skilled operators.
When Is LiDAR Worth the Cost for Telecom?
LiDAR is worth the investment in a narrow set of scenarios:
| Scenario | LiDAR Value | Reason |
|---|---|---|
| Indoor / sub-building planning | High | iBwave-type planning requires measured building geometry |
| Single-campus private network (hospital, factory, stadium) | High | Small area, high accuracy requirement |
| Dense mmWave small cell network (downtown core, under 5 km²) | High | Link budget tight; rooftop structures and parapets matter |
| Urban macro cell network (50 to 500 km²) | Low to Medium | Satellite 3D data is accurate enough; LiDAR cost is prohibitive |
| National / regional coverage (more than 1,000 km²) | Low | LiDAR is economically impractical; satellite data is the only viable option |
| Emerging market deployments | Very Low | LiDAR data typically unavailable; satellite is the only source |
The common thread: LiDAR is cost-effective when the planning area is very small, the frequency is mmWave, and the propagation environment is geometrically complex enough to justify centimeter-level accuracy.
How Does Satellite-Derived 3D Data Compare?
Modern AI-based extraction from satellite imagery has narrowed the gap with LiDAR substantially:
| Metric | Airborne LiDAR | Satellite-Derived (AI) |
|---|---|---|
| Vertical accuracy (building heights) | 5 to 15 cm RMSE | 1 to 2 m RMSE typical |
| Building capture rate | 95 to 99% | 93%+ (LuxCarta) |
| Geographic availability | Limited (government programs) | Global |
| Currency | As of survey date | Re-extractable from recent imagery |
| Cost | €500 to €1,500/km² | Fraction of LiDAR cost |
| Delivery time | Weeks to months | Days (on-demand) |
| Integration complexity | High (raw point cloud processing) | Low (ready-to-use SHP/GeoTIFF) |
For the vast majority of macro cell, small cell, and FWA planning projects, satellite-derived 3D data provides sufficient accuracy at a fraction of the cost and with global geographic coverage that LiDAR cannot match.
Is There a Hybrid Approach?
Yes. A hybrid approach uses satellite-derived data for the broad planning area and augments with LiDAR for the most critical micro-environments:
- Initial planning: Use satellite-derived LOD2 buildings, DTM, and vegetation data for city-wide coverage and macro cell optimization
- Small cell hotspot siting: Acquire LiDAR or drone survey data for the 2 to 5 km² high-priority commercial district where mmWave small cell density is highest
- Verification: Use drive test data to validate predictions and identify the rare cases where satellite data is insufficient
This approach captures 90%+ of the cost savings from satellite data while retaining LiDAR precision where the link budget demands it.
How LuxCarta Addresses This
LuxCarta builds its 3D building extraction on satellite imagery using an AI pipeline (U-Net CNN architecture) that achieves 93%+ building capture rate and 87.67% precision / 88.44% recall, metrics benchmarked by customers including Telefónica Deutschland and Samsung Networks Europe.
This approach delivers consistent, globally available 3D building data that can be refreshed from new satellite imagery as cities grow, without the cost or coverage limitations of LiDAR. For countries where LiDAR has never been collected, which represents the majority of global 5G deployment targets, satellite-derived data is not a compromise; it is the only viable path.
LuxCarta also uses satellite stereo photogrammetry to generate Digital Surface Models (DSM) and Digital Terrain Models (DTM) at horizontal resolutions matched to the source imagery. Its 3D mesh building extraction technique (published at SPIE 2023) achieves 90%+ accuracy with 4x productivity versus manual digitization, enabling faster delivery of high-quality data even in complex urban environments.
The BrightEarth platform makes this data available on-demand, allowing engineers to acquire exactly the layers they need (buildings, DTM, vegetation, LULC) for a specific area without commissioning a LiDAR survey or waiting for a long procurement cycle.
Frequently Asked Questions
What vertical accuracy does satellite-derived DSM achieve compared to LiDAR?
Satellite stereo photogrammetry typically delivers DSM with 1 to 2 m RMSE vertical accuracy depending on the source imagery resolution and stereo overlap quality. This compares to 5 to 15 cm for airborne LiDAR. For most 5G macro cell and sub-6 GHz small cell planning, 1 to 2 m vertical accuracy is entirely sufficient. For mmWave small cells in dense urban environments, LiDAR accuracy is more valuable but must be weighed against the cost.
Can drone-based LiDAR be a cost-effective middle ground?
Drone-based LiDAR surveys can be cost-effective for very small areas (a few city blocks to a few km²) where aircraft mobilization costs are prohibitive. However, drone surveys are still time-intensive, require permits in many jurisdictions, and produce data that covers only the surveyed area. They are useful for high-priority mmWave small cell corridors but do not scale to city-wide or regional planning.
How often should LiDAR data be refreshed for telecom planning?
In fast-growing cities, building turnover can be significant within 3 to 5 years. LiDAR surveys are rarely refreshed more frequently than every 5 years due to cost, meaning that operators using older government LiDAR datasets may be planning on geodata that no longer reflects the current environment. Satellite-derived extraction can be re-run from recent imagery much more economically, making it better suited to keeping pace with urban change.
Does LiDAR data have licensing restrictions for commercial telecom use?
Many government-collected LiDAR datasets are provided under open licenses (such as the UK National LiDAR Programme or USGS 3DEP in the US) that permit commercial use, but this varies by jurisdiction. Some national datasets require licensing fees for commercial derivative use. LuxCarta's satellite-derived products are delivered with fully transferable commercial licenses, including for military and commercial derivative use.
Is LiDAR data available for most 5G deployment markets?
No. LiDAR national coverage exists reliably for the UK, Netherlands, Denmark, parts of Germany, Australia, and the US. Major 5G investment markets including India, Southeast Asia, the Middle East, Africa, and Latin America have negligible LiDAR coverage. For these markets, satellite-derived geodata is the only scalable option for national or regional network planning.
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.