High-voltage transmission line corridor mapping for electromagnetic interference planning
Overhead high-voltage corridors generate corona-discharge noise and Fresnel-zone hazards that degrade nearby base-station and microwave-link performance. SAR polarimetry and optical line-detection algorithms map pylon locations and conductor geometry from orbit, giving network planners a documented spatial record before site selection or frequency coordination.
Sensors
- TanDEM-X (DLR/Airbus, X-band SAR): Bistatic interferometric pair at 9.65 GHz; single-pass coherence preserves fine structure on metallic conductors and lattice pylons. Spotlight mode reaches 1 m ground resolution. InSAR-derived elevation models resolve conductor sag geometry at span level, though thermal sag variation is not captured in a single acquisition.
- ALOS-2 PALSAR-2 (JAXA, L-band SAR): 1.27 GHz L-band penetrates vegetation canopy, making it the preferred band for detecting conductors obscured by tree cover along corridor edges. Stripmap mode delivers 3 m resolution; ScanSAR wide-area mode drops to 60 m. Full polarimetric (quad-pol) data support Pauli and Freeman-Durden decomposition to separate wire-like double-bounce returns from distributed clutter.
- Sentinel-1 (ESA, C-band SAR): 5.405 GHz C-band; IW mode at 5 x 20 m resolution with 6-day revisit (two-satellite constellation). Free and open archive from 2014. Dual-pol (VV+VH) limits decomposition depth compared with quad-pol systems, but the dense time series supports change detection along corridors and is cost-effective for initial corridor delineation over large regions.
- Pleiades Neo (Airbus, optical): 0.3 m panchromatic resolution in tasked acquisitions. At this scale, pylon shadow geometry and conductor catenary curves are directly measurable in stereo pairs. Cloud cover is the primary constraint; optical data complement SAR rather than replace it in temperate or tropical climates.
Why a power line is a radio-frequency problem, not just a civil one
Corona discharge occurs when the electric field gradient at a conductor surface exceeds the breakdown strength of surrounding air, typically above 230 kV on transmission lines. The resulting partial discharge emits broadband radio-frequency noise from below 1 MHz up through the VHF band, with significant energy in the frequency ranges used by LTE 700 and 800 MHz rural base stations. A receiver sited within a few hundred metres of a high-voltage corridor without accounting for this source will show elevated noise floor figures that frequency-coordination models built on terrain and clutter alone cannot explain.
The second mechanism is geometric. Microwave point-to-point links sharing a corridor with transmission infrastructure must maintain Fresnel-zone clearance. The first Fresnel zone radius at mid-span on a 10 km, 7 GHz link is roughly 23 m. A conductor sagging into that zone introduces diffraction loss and, under certain atmospheric conditions, multipath. Neither effect appears in a propagation model unless the conductor position is known with reasonable accuracy.
What SAR polarimetry actually detects on a transmission line
Steel lattice pylons produce a strong double-bounce return in the HH channel at L-band: the signal bounces from the ground to the vertical leg of the structure and back to the sensor. In Pauli decomposition of quad-pol ALOS-2 data, this signature appears in the blue channel (odd-bounce) and red channel (double-bounce) depending on pylon orientation relative to the flight path. Conductors themselves, being thin cylindrical targets, appear as coherent line features in high-resolution SAR imagery rather than as point scatterers; their detection depends on the ratio of conductor diameter to resolution cell size.
At X-band (TanDEM-X Spotlight, 1 m), a 30 mm diameter conductor subtends a fraction of a resolution cell but still produces a detectable coherent return along its length when the look geometry is favourable. The bistatic coherence of TanDEM-X is particularly useful: metallic structures maintain high coherence between the two antennas, while vegetation decorrelates, giving a discrimination layer that complements amplitude-based detection. L-band ALOS-2 adds the ability to see conductors beneath the canopy edge of corridor vegetation, which X-band cannot penetrate.
Conductor sag: what the geometry tells you and what it cannot
Catenary sag in a transmission-line span is a function of conductor temperature, mechanical tension, ice loading and span length. At rated current and high ambient temperature, sag on a 400 m span of typical ACSR conductor can reach 8 to 12 m relative to the attachment points. At low load and low temperature, the same span may sag only 4 to 6 m. TanDEM-X single-pass InSAR can resolve the conductor elevation profile at span level when the acquisition geometry is suitable, but the result represents conditions at the moment of the overpass, not the worst-case thermal sag that determines regulatory ground clearance or Fresnel-zone encroachment.
This is an honest limit of the satellite method. The practical use is not to replace dynamic sag monitoring (which requires continuous sensors or physics-based load-flow modelling) but to establish the corridor geometry, confirm pylon spacing and approximate attachment-point heights, and flag spans where even the static sag profile places conductors within a defined buffer of a co-located microwave path. That is a meaningful filter before a field survey team is dispatched.
Optical line detection: where it helps and where it fails
At Pleiades Neo's 0.3 m resolution, conductors are not resolved as physical objects but their shadows on the ground are measurable in low-sun acquisitions, and pylon structures are directly visible and measurable in stereo. Automated line-detection algorithms, typically variants of the Hough transform or deep-learning edge detectors trained on infrastructure imagery, extract corridor centrelines and pylon positions with planimetric accuracy of 1 to 3 m in clear conditions. Stereo processing adds height: pylon attachment-point elevations derived from Pleiades Neo stereo pairs are accurate to roughly 1 to 2 m in open terrain.
Cloud cover is the binding constraint in tropical and maritime climates. A single Pleiades Neo acquisition over a 200 km corridor in equatorial West Africa may be partially or entirely obscured. The practical workflow combines SAR-derived corridor delineation, which is cloud-independent, with optical stereo for height attribution in cloud-free windows. Neither sensor alone is sufficient for a production-quality corridor map in difficult climates.
From corridor map to interference-planning input
The output of the mapping process is a GIS layer of pylon centroids with height attributes, corridor polygons with buffer zones sized to the corona-discharge noise radius at the operating voltage, and, where conductor geometry has been resolved, span-level sag profiles. This layer is ingested into radio-planning tools such as ICS Telecom or Atoll as an additional clutter and noise-source layer, supplementing the terrain and building models that those tools already consume.
The corona-discharge buffer radius is not a fixed number. Published ITU-R guidance and peer-reviewed studies suggest that conducted and radiated interference from 400 kV lines can be significant at distances up to 100 m from the outer conductor in the HF and low-VHF bands, falling off more steeply at higher frequencies. The satellite-derived corridor polygon allows a network planner to identify candidate base-station sites that fall within this buffer and flag them for additional field measurement before frequency coordination is finalised.
Satellize can run this corridor-extraction and buffer-generation workflow on client-licensed commercial imagery combined with open Sentinel-1 time series, applying the same analytics infrastructure used in the Tonga crop-estimation programme to a very different physical target. The deliverable is a GeoPackage or Shapefile ready for direct import into standard radio-planning environments, accompanied by a sensor-and-method provenance report.
Archive depth and update cadence
Sentinel-1's open archive runs from April 2014 for European coverage and from late 2016 for most of the rest of the world. This is long enough to detect new transmission-line construction, corridor expansions and pylon replacements through coherence-change analysis. ALOS-2 archive coverage is sparser but extends to 2014 globally. TanDEM-X global DEM data were acquired primarily between 2010 and 2015 and are available commercially through DLR/Airbus.
For most network-planning applications, a one-time corridor map with annual refresh is adequate: transmission infrastructure changes slowly. Where a client is planning a new base-station cluster near an existing corridor, a single high-resolution acquisition pair (TanDEM-X Spotlight plus Pleiades Neo stereo) combined with a Sentinel-1 time series for change context provides the necessary inputs. The cost of a targeted acquisition over a 50 km corridor segment is a fraction of a single field survey day, which is the relevant comparison.
Typical figures
| Best spatial resolution (SAR) | TanDEM-X Spotlight: ~1 m; ALOS-2 Spotlight: 1–3 m; Sentinel-1 IW: 5 x 20 m |
| Best spatial resolution (optical) | Pleiades Neo panchromatic: 0.3 m; stereo height accuracy ~1–2 m in open terrain |
| Revisit cadence | Sentinel-1: 6 days (two-satellite); ALOS-2: 14 days; TanDEM-X/Pleiades Neo: tasked on demand, typically 1–5 days depending on latitude and cloud |
| SAR frequency bands used | X-band 9.65 GHz (TanDEM-X); L-band 1.27 GHz (ALOS-2 PALSAR-2); C-band 5.405 GHz (Sentinel-1) |
| Minimum detectable pylon spacing | Pylons detectable at 5 m resolution or better; conductor catenary detectable as line feature in TanDEM-X Spotlight and Pleiades Neo |
| Conductor sag resolution | Span-level profile from TanDEM-X InSAR; represents static snapshot at acquisition time only, not dynamic thermal sag |
| Cloud penetration | SAR sensors (all three) are cloud-independent; Pleiades Neo optical is cloud-limited |
| Archive depth | Sentinel-1: from 2014; ALOS-2: from 2014; TanDEM-X global DEM: 2010–2015 baseline acquisition |
| Delivery formats | GeoPackage, Shapefile, GeoTIFF (height raster), CSV pylon centroid table, PDF provenance report |
Analytics Satellize can run
| Pylon centroid map with height attributes | SAR double-bounce detection (Pauli decomposition on ALOS-2 quad-pol) combined with optical structure extraction from Pleiades Neo stereo | GeoPackage point layer with pylon height, span length and voltage-class attribution where publicly available |
| Corridor polygon with corona-discharge buffer zones | Automated line-detection (Hough-transform or learned edge detector) on SAR amplitude and optical imagery; buffer radius parameterised by voltage class | Polygon GIS layer at 1:5,000 scale, importable into ICS Telecom, Atoll or equivalent radio-planning tools |
| Conductor sag profile per span | TanDEM-X single-pass InSAR elevation extraction along detected conductor centreline; comparison against attachment-point heights from stereo optical | Per-span sag table (GeoPackage linestring with elevation attribute) and flagged list of spans within user-defined Fresnel-zone buffer of co-located microwave paths |
| Corridor change detection report | Sentinel-1 coherence-change analysis across multi-year archive; new construction or pylon replacement appears as coherence loss followed by stable high-coherence point scatterers | Annual change report as PDF with GIS difference layer showing added or removed infrastructure |
| Base-station site interference-risk score | Spatial intersection of candidate site coordinates with corona-discharge buffer polygons and Fresnel-zone encroachment flags; scored by proximity band | Ranked site list in CSV with risk band (within 50 m, 50–100 m, 100–200 m of outer conductor) and recommended field-measurement action |
| Sensor and method provenance report | Documented acquisition dates, sensor modes, processing chain and known limits (cloud cover, sag snapshot caveat, resolution floor) | PDF report suitable for inclusion in frequency-coordination submissions or regulatory filings |
Who does the work
We can get this done for you. Satellize runs its own analyst desk and a strong science team. You do not buy a data feed and work out what it means; our people source the imagery, run the analysis described on this page, and hand you the answer with its confidence limits stated. Discuss this requirement.