Library Storage and networks2018Design paperCorpus record

Helium: A Decentralized Wireless Network

Helium. Amir Haleem, Andrew Allen, Andrew Thompson.

The Helium paper for a wireless network built from independently owned hotspots. Coverage is the commodity. Proof-of-coverage is the paper's way of checking that a radio is where it says it is and that it can be heard. Later changes of purpose and token are not this document.

Helium's paper proposes a wireless network built by people who deploy hotspots, with proof-of-coverage as a way to show that a hotspot is where it claims and that it can see its neighbours.

The five-minute read

The resource is radio coverage, not hash power

Miners in this design are hotspots. They are paid, in the paper's story, for providing network coverage that devices can actually use.

Proof of coverage is a radio challenge

A hotspot is challenged to send a beacon. Neighbours who hear it submit receipts. The geometry of who heard whom is the evidence that the radio is real and roughly where it says.

Location is asserted and then tested

The operator says where the hotspot is. The challenges are supposed to make a false location expensive. The quality of that test is the quality of the whole incentive.

A network token coordinates two sides

Deployers earn for coverage. Devices pay to send data. The paper uses one token economy to face both. That coupling is a design choice with the usual price risk.

Useful coverage and gamed coverage diverge

A proof that radios can hear each other is not a proof that a paying device wanted that location. The paper's mechanism and the later argument about gaming should be held apart.

One action, walked through

An operator asserts a hotspot's location and identity and brings it online.
The protocol selects a challenger. The hotspot emits a beacon at the time and frequency required.
Nearby hotspots that witness the beacon submit receipts, signed and time-bounded.
The protocol scores the proof based on who heard it and whether that fits the asserted geography.
Rewards follow the score. A device that wants to send data uses the network under a separate payment path the paper specifies.

The argument, unpacked

Radio proofs are physical and still gameable

Antennas, attenuation and clusters of hotspots placed to witness each other can satisfy a geometric test without serving a single sensor in a useful place. The paper's proof raises the cost of lying. It does not define 'useful' as 'a customer paid'. Later design changes were an admission of this gap.

Witnesses are the security, and also the collusion set

A hotspot alone cannot easily prove its own coverage. It needs neighbours. Neighbours can be friends. The protocol has to make collusion more expensive than honest witnessing, which is an economic claim on top of the radio claim.

Coverage is local

A global token and a local radio footprint are different maps. Paying the same reward schedule everywhere will overbuild some cities and skip others. The paper's marketplace has to face geography. A uniform emission curve does not.

What has to be true

Beacon receipts are hard to forge from a distant antenna farm, under the parameters of the proof.
Enough honest hotspots exist in a region for witnesses to mean something.
Device traffic exists. A perfect proof of coverage with no devices is a radio hobby with a subsidy.
Location assertions are checked often enough that a moved hotspot does not earn on an old claim forever.

What happened after the paper

Helium deployed a large hotspot network, then had to revise rewards because proof-of-coverage was gamed and because coverage did not match demand. The network later looked to new backhaul and a different token structure. The paper is the original coverage-proof thesis. The revisions are evidence about where the thesis was thin, not footnotes to ignore.

What to check before you use the idea

Does the proof show that a paying device can use this hotspot, or only that other hotspots heard it?
How expensive is it to place witnesses that satisfy the geometry without providing useful coverage?
Are rewards tied to transferred data, to coverage proofs, or to both, in the version being cited?
What changed in the reward function after the initial deployment?

Terms

Hotspot: A radio the operator deploys, which both provides coverage and participates in proofs.
Proof of coverage: A challenge-and-witness protocol meant to show a hotspot's radio is present where asserted.
Beacon: The radio transmission a challenged hotspot sends for neighbours to hear.
Witness: A nearby hotspot that submits a receipt saying it heard the beacon.

The problem the paper names

A telecom network is usually owned by the operator who also owns the spectrum and the towers. Helium's paper proposes that many owners run hotspots, that users pay for data, and that the protocol checks coverage instead of trusting a self-report.

What the design proposes

Hotspots assert a location and participate in challenges that a radio in that place should be able to witness.
Proof-of-coverage is the substitute for a central site survey.
A two-sided market, in the paper, pays for coverage and charges for network use. Both sides have to exist.

How the mechanism is specified

Challenges, witnesses and a consensus among hotspots are specified so that a hotspot cannot cheaply claim coverage it does not provide.
The physical layer matters. A proof about a radio is not a proof about a financial transaction, and the paper should not be read as one.
Later migrations of the network's data plane are subsequent decisions. The original paper is about the wireless network as specified.

What this page does not treat as proven

Proof-of-coverage can be gamed if the challenge design is weak. The paper's mechanism is only as good as its radio assumptions.
A dense map of hotspots is not the same as useful coverage for a paying device.
We do not describe token emissions or hotspot income.

Why a venture studio still reads it

This is the DePIN paper on our desk when a venture says 'the device proves itself'. Helium's standard is a challenge another radio can witness. A self-reported GPS pin is not that standard.

This is Blockchain Lab's reading of a public design paper. It is not the paper, not a copy of it, and not an offer of tokens, equity, custody or a partnership. Later network behaviour can diverge from the text. Nothing here is investment, legal or technical advice.

Research status: Design paper. Last reviewed: 1 October 2026. This is a reading of a public paper, not investment, legal or security advice.