TL;DR: An RTK network is a set of interconnected, precisely surveyed base stations that stream real-time GNSS corrections to rovers, turning meter-level GPS into centimeter-level position data. Single-baseline RTK draws from the nearest station; Network RTK (VRS) interpolates from several stations for more consistent accuracy over a wider area. Station density, not marketing claims, is what actually determines real-world accuracy.
Whether guiding drones across farmland or survey robots through dense urban sites, RTK is what turns a GNSS receiver’s meter-level position into a centimeter-level one. This guide covers what an RTK network is, how it works, and what to look for when choosing one.
Key takeaways:
- An RTK network is a set of interconnected base stations delivering real-time GNSS correction data to rovers over a wide geographic area.
- Single-baseline RTK draws corrections from the nearest physical base station, with a practical ceiling around 20-35 km. Network RTK (VRS) interpolates corrections from several surrounding stations to extend consistent accuracy further.
- Station density, roughly 30-40 km spacing for a dense network, is what actually governs real-world accuracy, more than the correction method itself.
- Private, self-owned networks generally offer more consistent uptime, redundancy, and support than crowd-sourced or public networks.
What Is an RTK Network?
RTK (real-time kinematic) positioning compares the carrier-phase signal a rover receives from GNSS satellites against the same signal received by a fixed, precisely surveyed base station. That comparison cancels out most of the error in the rover’s raw GNSS measurement, bringing accuracy from several meters down to one to two centimeters.
An RTK network extends this by connecting multiple base stations into shared infrastructure. Instead of one rover depending on a single nearby station, the network can hand off corrections between stations or interpolate them from several, maintaining accuracy over a much larger area than any one base station could cover alone.
Single-Baseline RTK vs. Network RTK (VRS)
| Approach | How it works | Best for |
|---|---|---|
| Single-baseline RTK | The rover receives corrections directly from one nearby physical base station. Accuracy degrades roughly 1-1.5 cm per 10 km of added distance, with a practical ceiling around 20-35 km. | Simple, traceable deployments close to a station. |
| Network RTK (VRS) | Software interpolates a virtual reference station near the rover from several surrounding physical stations. | Wider-area coverage, redundancy if one station goes down. |
Both approaches are what the industry calls OSR (Observation Space Representation): both work with real measurements from physical stations, just combined differently. SSR (State Space Representation), also called PPP-RTK, takes a different approach entirely, modeling individual GNSS error sources (orbit, clock, ionosphere, troposphere, signal bias) rather than interpolating station observations, which lets it cover continental-scale areas with fewer stations at a lower accuracy tier than RTK.
Neither single-baseline nor Network RTK is categorically “better.” The deciding factor for either one is station density, not the method. For a full technical breakdown of why, including how atmospheric conditions actually limit accuracy and how to evaluate a provider’s real-world (not just median) performance, see Demystifying Network RTK: What Drives Correction Accuracy and Why Density Matters.
How RTK Networks Work
Base Stations
Base stations are surveyed precisely enough that their position is known to within one or two centimeters. Each station continuously logs GNSS observations and atmospheric data, then streams that data to the network in real time.
RTK-Enabled Rovers
A rover receiver compares its own raw carrier-phase measurements against the base station’s corrections to resolve the ambiguity in the carrier signal, canceling out most atmospheric, orbital, and signal-propagation error. The result is roughly 100x more precise than a standalone GNSS fix.
Network Access
Most organizations aren’t set up to run their own base station and NTRIP infrastructure, which is why RTK service providers exist. The Point One RTK Network, for example, operates 3,000+ self-owned base stations across the US, Europe, Australia, and Canada at roughly 30-40 km spacing, streaming corrections over a single NTRIP connection, including L-Band satellite delivery in areas without cellular coverage. See the live coverage map for current station locations.
Benefits of RTK Networks
- Dense coverage. Multiple interconnected stations reduce signal obstructions and maintain accuracy across a wider service area than a single base station can.
- Real-time, accurate corrections. Continuous correction broadcasts give rovers immediate centimeter-level accuracy rather than requiring post-processing.
- No base station management. Running your own station means surveying, calibration, monitoring, and ongoing maintenance. A managed network shifts that overhead to the provider; see our full build-your-own vs. managed RTK cost breakdown for a real three-year TCO comparison.
- Scalability. Networks can add coverage without you deploying new hardware. TRIC Robotics, for example, uses Point One’s network to run farm-scale autonomous robots without a single on-site base station.
- Redundancy. If one station goes down, Network RTK can draw corrections from surrounding stations instead of interrupting the whole operation.
- Cost efficiency. A subscription model avoids the upfront capital cost of purchasing and maintaining base station hardware.
Public vs. Private RTK Networks
Public RTK networks, including state DOT and community-run networks, offer free or low-cost access and can be a reasonable starting point for casual or non-critical use. They typically rely on crowd-sourced or government-operated stations with variable maintenance schedules and no dedicated support line, so uptime and station density can vary significantly by region.
Private networks own the infrastructure end to end, which generally means more consistent uptime, tighter calibration and security control, and direct technical support. Point One’s network, for instance, runs on professionally installed stations with dual cellular modems and four SIM slots per station, five days of battery backup, and automated position-integrity monitoring that pulls a station offline if it detects physical displacement, rather than relying on a human to notice. The tradeoff versus a public network is cost: private networks run on a subscription.
How to Choose an RTK Network
Accuracy track record, base station density near your operating area, uptime history (not just a target number), constellation support (GPS, GLONASS, Galileo, BeiDou), and standards-based integration (NTRIP, RTCM) are the factors that matter most. For a full evaluation framework, including the exact questions to ask a provider about coverage, uptime, accuracy, integration, and cost, see our complete NTRIP service provider guide.
Frequently Asked Questions
What is the difference between GPS and RTK?
GPS provides standalone position accuracy within several meters. RTK adds real-time correction data from a base station to bring that same GNSS signal down to centimeter-level accuracy.
Does RTK need an internet connection?
Not necessarily. RTK corrections can be delivered over cellular data, radio frequency, or L-Band satellite, so a network with satellite delivery can maintain accuracy even without cellular coverage.
What is a Virtual Reference Station (VRS)?
A VRS is a synthetic reference point that Network RTK software generates from several nearby physical base stations. It’s a delivery format, not a specific correction method. VRS data can come from either Network RTK interpolation or an SSR model, and the two have different accuracy characteristics even though they look identical to a rover.
Do I need to own a base station to use RTK?
No. Subscribing to an RTK network gives you access to a provider’s base station infrastructure over NTRIP, without the cost or maintenance of surveying and running your own station. See our DIY vs. managed RTK breakdown for when self-hosting still makes sense.
Does station density matter more than the correction method?
Yes. Both single-baseline and Network RTK can deliver 1-3 cm accuracy when backed by a dense network (roughly 30-40 km station spacing). Sparse networks produce similar average accuracy but with much longer error tails, which is where real-world systems actually fail.
Access an RTK Network Built for Reliability
Choosing an RTK network comes down to accuracy, coverage, uptime, and how well it integrates with the hardware you already run. The Point One RTK Network is a fully self-owned, receiver-agnostic network offering both single-baseline and Network RTK/VRS on one platform, with 99.9% uptime and delivery over cellular or L-Band satellite.