Geofencing draws a virtual boundary around a real world location and triggers an action when a device crosses it. It works by comparing a phone's estimated position, usually from GPS, WiFi, or cell towers, against that boundary. Geofences are good at answering whether a device is in an area, but they locate coarsely, often to tens of metres and worse indoors, so they cannot tell which of several nearby providers a person actually means. Location and proximity are related but not the same thing.
What geofencing is
A geofence is a virtual perimeter drawn around a physical place. It might be a circle of a certain radius around a point, or a polygon traced around a building or a neighbourhood. Software watches a device's position and fires an event when the device enters the fence, leaves it, or lingers inside it.
Geofencing is used to send a notification when someone arrives, to start a task when a delivery vehicle reaches a depot, to remind a person of something at a location, or to log when equipment leaves a site. The fence is about place. It answers whether a device is inside a defined area, and it triggers on the crossing.
How geofencing works
A geofence needs two things: a boundary and a position to test against it. The boundary is defined once, as a centre and radius or as a set of coordinates. The position comes from whatever the device can use to locate itself, most often GPS, but also WiFi networks the Wi-Fi Alliance standardizes, nearby cell towers, and sometimes Bluetooth beacons for indoor cases.
On a set schedule or when the device moves, the software estimates the current position and checks it against the boundary. If the device has crossed in or out, the corresponding event fires. The whole design rests on the accuracy of that position estimate, and that accuracy varies a great deal with the source and the environment.
Outdoors with a clear sky, GPS can be reasonably tight. Indoors, in a dense city, or among tall buildings, the estimate drifts, sometimes badly. A fence that assumed street level accuracy can then fire late, fire early, or miss the crossing entirely.
Geofencing versus signal based proximity
The clearest way to understand a geofence is to set it beside signal based proximity. A geofence tests a computed map position against a boundary. Signal based proximity senses a provider's actual broadcast and reasons from the signal itself. The two answer different questions.
| Approach | What it measures | Typical accuracy | The question it answers |
|---|---|---|---|
| Geofence | A computed map position against a boundary | Coarse, often tens of metres, worse indoors | Is the device inside this area |
| Signal based proximity | A provider's short range broadcast, sensed directly | From contact range to tens of metres, by radio | Which specific provider is the device beside |
Where geofences fall short
A geofence is coarse by nature, and coarseness is a problem the moment precision matters. If two providers stand a few metres apart, a geofence that spans tens of metres cannot separate them. Both fall inside the same fence, so the fence cannot say which one a customer is actually next to.
The estimate also lags. Position updates are periodic and can arrive after the device has already moved, so a crossing may register seconds late. Indoors, where GPS is weakest and where many payments happen, the drift is largest. A fence is a good tool for the question it was built for, which is presence in a broad area, and a poor tool for identifying a single nearby provider.
Proximity is not location
The distinction that matters is this. Location says where a device sits on a map. Proximity says which specific thing a device is next to. A geofence is a location tool, and it will always struggle to name the exact provider among several close together, because a map coordinate is not the same as a relationship between two devices.
Parousya's patented proximity method works from the relationship instead of the coordinate. Patent US10657515B2 covers how a provider broadcasts a short range signal, the customer's phone detects it and identifies the nearest provider, and the payment routes through a central server so the two devices never connect directly. There is no map lookup and no boundary to cross. The phone reasons from the signal that is actually present, which is what lets it pick the one right provider a coarse geofence would blur together.
The divisional patent US11392923B2 adds identification without a payment, so presence can be confirmed on its own. Where a geofence tells you a device is somewhere in an area, the proximity method tells you the device is beside a particular provider, and it does so without ever revealing a map position.
Common questions
- What is geofencing in simple terms?
- It is a virtual boundary around a real place. When a phone crosses in or out, an action fires. Software compares the phone's estimated position, usually from GPS, WiFi, or cell towers, against that boundary.
- How accurate is geofencing?
- It depends on the position source. Outdoors with clear GPS it can be reasonably tight, but it is often coarse to tens of metres and degrades badly indoors and among tall buildings, so it cannot reliably separate providers a few metres apart.
- What is the difference between geofencing and proximity?
- A geofence tests a map position against an area boundary, so it answers whether a device is inside a region. Proximity senses a provider's actual signal, so it answers which specific provider a device is next to. Location is not the same as proximity.
- Why is geofencing not precise enough for payments?
- Because it locates coarsely. If several providers sit within the same fence, a geofence cannot say which one the customer means. Picking one exact provider needs the signal based approach the patented proximity method uses.
- How does proximity payment identify the right provider without a map?
- Patent US10657515B2 covers sensing a provider's short range broadcast directly, identifying the nearest provider, and routing the payment through a server. It reasons from the present signal, not from a computed location, so no map position is used or revealed.
Sources
- What is ultra wideband?Precise ranging where geofences are coarse.
- How do robotaxis work?Where picking the right car beats a boundary.
- The proximity methodSignal based identification, end to end.
- Geofence, definedThe short definition and related terms.
- Geofencing vs beaconsA boundary against direct signal presence.
- Rideshare and robotaxiMatching a rider to the right car.