C-DRONE GUIDE · 1 AUGUST 2026
Aerial fibre-optic network audit by drone: poles, overloading, dangling cables, price
Along country roads, through housing estates and out to the last hamlet, France's fibre-optic network very often runs on poles — sometimes Enedis's own, sometimes dedicated supports put up for fibre. That aerial estate, built fast to meet the France Très Haut Débit plan's targets, ages just as fast: cable loops left dangling for a connection that was never made, splice closures left open to the weather, poles loaded well beyond what they were built for — what the press has nicknamed the "tinsel syndrome". Checking the real condition of tens or hundreds of kilometres of line, pole by pole, on foot, takes weeks and misses the point — what shows from above, not from the road. The drone runs the line in a few hours, photographs every pole and compares its load against the previous campaign. Here is what an aerial drone audit documents, who it is for, and prices observed in 2026.
Published on 1 August 2026, reviewed on 13 August 2026 — regulations in force as of August 2026.
The "tinsel syndrome": an aerial network ageing faster than it was built
A significant share of the French FTTH network runs on aerial supports rather than buried ducts — cheaper to deploy, faster to connect, but more exposed to weather and to the build-up of disorder across successive connection campaigns. Every newly connected customer adds a cable to the same pole; when the pull-through is not done properly, the length of cable left dangling for a future connection — instead of being cut to size — piles up pole after pole, into the tangles visible from the street that the press has nicknamed the "tinsel syndrome". Beyond looks, every extra loop adds weight to the pole, exposes extra tension points to wind and ice, and makes the next technician's job harder, since they can no longer tell which cable serves which home.
Arcep, the French telecoms regulator, tracks this decline nationwide: its FTTH network quality observatory, published in February 2026 on data through September 2025, puts the failure rate reported by commercial operators at around 0.12% nationally — an improving figure, but one the regulator explicitly links to higher operating costs in less dense areas, where aerial runs are longer and more exposed to weather hazards (see the Arcep observatory). Those are precisely the areas — scattered housing, rural town centres, hamlets served by a public-initiative network (RIP) — that foot inspection covers worst, for lack of time and of visibility from ground level.
The drone method: running the line, photographing every pole, measuring the sag
The mission follows the proven logic of drone power line inspection, applied this time to telecommunication cables: the drone runs along the line of poles a few metres away, photographs each pole from several angles and assigns it a geo-referenced identifier. The catalogue of defects is specific to a telecom network: unconnected dangling cables and surplus loops, open or poorly remade splice closures, corroded fittings and brackets, a split or leaning wooden pole, a cable fallen or trailing on the ground after a branch fall, a dangerous crossing with a power line on a shared support. Each point is ranked by severity, from a mere cosmetic defect to an immediate hazard for road users.
On extended runs, a second layer sharpens the diagnosis: a photogrammetric model of the corridor, compared from one campaign to the next, tracks how the sag — the cable's droop between two poles — changes over time, and flags a section that has slackened or overloaded since the last pass, before it gives way. A study by Oh and Lee published in 2017 in Sensors demonstrated 3D extraction of aerial cables from multiple drone images, with an altitude accuracy of a few centimetres, sufficient for this kind of monitoring (see the study on Google Scholar). Sorting the defects themselves can draw on the same deep-learning building blocks already validated on power lines: the study by Siddiqui and Park published in 2020 in Energies automatically detects, from an onboard video feed, line components and their defects in a cluttered environment (see the study on Google Scholar) — an automatic sort that pre-selects the poles to examine first along a run of several hundred kilometres, ahead of the human review of the report.
Who orders this audit, and in what framework
Three families of commissioners come up again and again. Infrastructure operators, who own a territory's network layer, use it to map the real condition of their estate ahead of a tinsel-clearing plan or a maintenance review. Local authorities delegating a public-initiative network use it to put hard evidence behind the quality of the network their delegate has delivered, mid-contract or before accepting a new tranche — the same dated-proof logic we detail for buying a drone service under public procurement. Telecom civil-engineering subcontractors, finally, rely on this survey to cost a clean-up operation before launching it, rather than discovering the scale of the disorder pole by pole on site.
The regulatory framework has recently sharpened: Arcep decision n° 2023-2801, adopted on 14 December 2023 as part of the 7th market-analysis cycle for fixed broadband and very-high-speed broadband, places non-discrimination obligations on Orange over access to its civil-engineering infrastructure and its aerial supports; it notably provides that renovating a support that can no longer bear an additional load can be carried out by the infrastructure operator that needs it, at costs advanced and then reimbursed by Orange on the basis of the efficient costs incurred (see the decision on arcep.fr). Precisely documenting which supports are concerned, with dated, geo-referenced photos, is the first item in any renovation file built on that basis. On supports shared with the power network, coordination with Enedis remains essential, in the same logic described for power line inspection; on a run beyond the pilot's direct line of sight, moving to a beyond-visual-line-of-sight flight follows the framework we detail in our BVLOS, STS and SORA guide.
Prices observed in 2026
Ranges observed in 2026 (excl. VAT):
- Audit of an ordinary run (up to 10 km of line, per-pole sheet, ranked report): €1,500 to €3,000.
- Extended network or full RIP sector (beyond 10 km, tapering per-kilometre rate): €150 to €300/km depending on pole density and accessibility.
- Comparative follow-up after a first campaign (same poles, sag-evolution measurement): tapering rate from the second campaign.
- Targeted check after a storm (restricted run, priority on flagged sections): €600 to €1,400, as a rapid callout.
The administrative build-up — the pole owner's agreement, coordination with the towns crossed, the DGAC declaration for beyond-visual-line-of-sight runs — weighs more on the lead time than on the price of the mission itself: a provider already in contact with the relevant infrastructure operator cuts that lead time down to just the weather window. Request a quote stating the approximate run length, the number of poles and the audit's purpose (acceptance, maintenance, dispute).
Frequently asked questions
Can the drone work on a support that also carries a power line?
Yes, but with the same caution as any power line inspection: a safety distance from live conductors, coordination with Enedis or the relevant operator on the shared support. Overflying the telecom side never waives that vigilance once a bare electrical cable shares the same pole.
Is a drone audit enough to launch tinsel-clearing works?
It provides the dated, geo-referenced survey needed to prioritise and cost the operation, but the decision to renovate a support and who funds it then falls under the framework set by the infrastructure operator and, where applicable, the relevant Arcep decision. The audit triggers the file — it does not replace it.