Five error factors to check to avoid failure in centimeter-level GPS positioning

Centimeter-level positioning is in high demand at sites where location information is used for work: civil engineering, construction, infrastructure inspection, asset management, agriculture, cultural property recording, as-built verification, and so on. For conventional map display or general position logging, errors on the order of a few meters may not cause major problems in some cases, but tasks such as verifying construction positions, working close to boundaries, recording pipes or buried objects, or aligning with point clouds and photos—where differences of a few centimeters can lead to rework or misjudgment—require a completely different approach to positioning accuracy.

At the same time, the phrase “centimeter-level positioning (cm level accuracy; half-inch accuracy)” can be misunderstood as meaning that simply acquiring the equipment and applying corrections will always yield stable, high accuracy. In practice, sites often experience unstable results, slight coordinate shifts when re-measuring the same spot, different results between morning and afternoon, or worse accuracy precisely where measurements are needed. In many cases the cause is not a device malfunction but insufficient understanding of the assumptions behind positioning and of the site environment.

Achieving centimeter-level positioning requires more than merely owning a receiver that supports high accuracy. The result depends on multiple factors—satellite visibility, how open the sky is above, reflection environment, the quality of correction information, how the device is set up, and operational procedures. In other words, accuracy is not determined solely by a device’s catalog specifications but by the product of site conditions and operational quality.

This article organizes and explains five error factors that field practitioners searching for “centimeter-level positioning” should always check to avoid failures in high-precision positioning centered on GPS. Rather than merely listing error types, it explains why errors occur, in which site conditions they are likely to occur, how to detect them, and how to mitigate them—presented in a form that can be applied in practice. It is useful both for those considering introduction and for those already operating systems but experiencing result variability.

Table of contents

• What centimeter-level GPS positioning means (centimeter-level positioning (cm level accuracy; half-inch accuracy))

• Error factor 1: bias in satellite geometry and reception environment

• Error factor 2: multipath from reflected radio waves

• Error factor 3: correction data quality and unstable communications

• Error factor 4: variability in equipment setup and observation procedures

• Error factor 5: insufficient understanding of site conditions and lax operational judgment

• Practical mindset to stabilize centimeter-level positioning

• Implementation perspectives to make high-precision positioning useful on site

What centimeter-level GPS positioning means (centimeter-level positioning (cm level accuracy; half-inch accuracy))

First, it is important to grasp that centimeter-level GPS positioning (cm level accuracy; half-inch accuracy), in practical use, is not achieved by single-receiver standalone observations alone. General satellite positioning computes location from signal travel times from satellites, but without correction this includes errors from the atmosphere, satellite orbit information, receiver clock bias, and environmental reflections, typically resulting in practical errors on the order of meters. That may be acceptable for recording positions on a map, but it is insufficient for high-precision tasks like construction verification or surveying assistance.

This is where using correction information to cancel error components as much as possible, and utilizing carrier-phase information for further refinement, becomes important. Only then can one aim for horizontal position accuracy on the order of centimeters. However, this is only true when conditions are met. The high-precision mechanism is powerful but sensitive to prerequisites. In other words, centimeter-level positioning (cm level accuracy; half-inch accuracy) is not “stable because it’s high performance,” but rather “high performance works when prerequisites are observed.”

Also, the required accuracy on site is not uniform. For some tasks, only planar position matters and height can be approximate. For others, vertical error is a major concern. Whether you are recording a single point or continuously collecting while moving, or combining with photos or point clouds, the acceptable error characteristics change. Don’t judge by the phrase “centimeter-level positioning (cm level accuracy; half-inch accuracy)” alone—unless you clarify what needs centimeter accuracy relative to what, device selection and operational design will be misaligned.

Importantly, errors do not always stem from a single cause. If satellite geometry is poor at a given time, reflections from nearby walls are strong, communications are unstable, and setup height is not managed, it becomes hard to identify the primary cause. When poor accuracy occurs on site, instead of immediately suspecting the device, structurally isolating error factors is the first step to stable operation.

Error factor 1: bias in satellite geometry and reception environment

The first thing to check is sky visibility and satellite geometry. GPS positioning uses multiple satellites to compute position, but more satellites do not automatically mean better results. What matters is whether the satellites are captured with good spatial balance. For example, if visible satellites are biased toward one direction, the computed position tends to be unstable; this is the problem of satellite geometry.

This problem commonly occurs near mountain edges, close to engineered slopes on development sites, in urban canyon between buildings, under or near elevated structures, and in locations surrounded by trees. Even if the sky is not completely blocked, environments that are significantly open only in one direction produce directional bias in received satellites. As a result, the position may appear to be determined but drift slowly, take a long time to converge, or fail to yield consistent values on re-observation.

Note that a field worker’s perception of “the sky is visible” does not equal sufficient reception for satellite positioning. Even if it looks open to the eye, low-elevation satellites might be blocked or surrounding structures may carve out significant portions of the sky, degrading actual positioning conditions. When aiming for centimeter-level positioning (cm level accuracy; half-inch accuracy), a much stricter assessment than for general navigation is necessary.

Countermeasures include checking not only directly above the measurement point but also around it for sky openness. Be aware of tall obstacles in the surroundings and whether satellite tracks may be blocked at certain times. If measurement is unavoidable in a harsh environment, consider changing observation timing, securing a nearby location with better reception and combining auxiliary methods, or planning for re-observation and judgment rules to absorb those constraints operationally.

For mobile positioning, it is often the speed of environmental change rather than walking or vehicle speed itself that causes problems. Entering from an open area into proximity with buildings, passing under trees, or moving along slopes—rapid changes in reception conditions can destabilize continuous positioning. Therefore, static point logging and mobile measurements require different checks even with the same equipment.

Satellite geometry and reception environment form the foundation of positioning. If this foundation is poor, no amount of other optimization will stabilize results. To prevent failures in high-precision positioning, first develop the ability to judge whether a site is suitable for measurement.

Error factor 2: multipath from reflected radio waves

A representative factor that disrupts centimeter-level positioning (cm level accuracy; half-inch accuracy) is multipath caused by radio-wave reflections. This occurs when signals from satellites are reflected by the ground, walls, metal surfaces, water, vehicles, fences, etc., and arrive at the receiver delayed relative to the direct signal. Receivers calculate distances assuming the direct path from the satellite; when reflected waves mix in, the receiver accepts signals that traveled a longer path, distorting the distance estimate.

Multipath is troublesome because it is hard to detect visually. The receiver may appear to continue positioning, so in the field the problem is not obvious, while the result subtly deviates. Moreover, that deviation is not constant and varies with satellite positions and the environment. Thus you may measure the same place at different times and get slightly different results, or multiple quick measurements may not converge.

Pay particular attention around buildings with metal cladding, near guardrails or fences, in sites with parked trucks or heavy machinery, near puddles or water surfaces, and around structures with many glass faces. Even under open sky, a surrounding environment with many reflective surfaces degrades positioning quality. In other words, having an open sky and having an environment with few reflections are different conditions; high-precision positioning requires checking materials and placement of surrounding surfaces as well as the sky.

The basic countermeasure is to increase distance from reflection sources. Even differences of tens of centimeters to several meters (tens of cm to several m) can change the impact, so if possible shift position slightly and re-measure to compare stability. Also set the receiver at an appropriate installation height so that it is less likely to pick up unnecessary reflections from low positions. Measurements close to the ground are more susceptible to surrounding objects, so avoid careless handheld use or unstable placement in many situations.

Multipath is not something to be accepted as “a small shift that can be tolerated.” If data obtained with an expectation of centimeter-level accuracy is actually unstable due to reflections, inconsistency may appear when overlaying photos, drawings, or point clouds in later processes. The worst part is that such inconsistencies are often discovered only during post-processing, making root-cause tracing difficult—so detecting anomalies on site is critical.

In practice, focus on whether a measured value should be adopted, not merely whether positioning succeeded. Don’t be reassured by a fixed positioning state alone—check whether there are reflection surfaces nearby, whether re-observation yields small differences, and whether small position shifts cause unnatural jumps in values to greatly reduce multipath risks.

Error factor 3: correction data quality and unstable communications

In centimeter-level positioning (cm level accuracy; half-inch accuracy), the quality of correction information greatly affects results. High-precision is achieved by using reference observations and error models, but if those corrections are unstable, even good reception conditions will not yield expected results. In the field, positioning failures are often attributed to satellites or equipment, but in many cases the reception and applicability of correction data are the real cause.

For example, in operations where corrections are received via communication links, delays, interruptions, reconnects, and low signal strength directly affect positioning stability. While urban areas are often assumed to have no problem, communication quality can vary significantly by location near underground structures, in mountainous regions, ports, deep development sites, and sites with many temporary installations. Even temporary communication instability can make maintaining a fixed solution difficult, preventing continuous high-precision recording.

Also, merely receiving correction data does not guarantee success. If the distance from the reference station is large or environmental differences between the site and reference are significant, the error correlation weakens and corrections become less effective. Particularly in regions with large altitude differences or differing meteorological conditions, or in wide-area operations, theoretical usability may not translate into practical stability. Therefore, centimeter-level positioning (cm level accuracy; half-inch accuracy) requires consideration of whether the correction actually functions adequately for the site, not just that corrections are used.

A commonly overlooked point on site is equating “communication is connected” with “positioning quality is stable.” Even when a communication indicator is active, delays, missing packets, or switching of correction streams can affect results. Conversely, if temporary instability is not noticed on site, poor data may be recorded as normal. This type of failure is hard to reproduce later and difficult to trace.

Countermeasures start by treating the communication environment as part of the positioning system. Don’t stop at checking only the positioning device—pre-identify which areas on site have stable communication, where connections drop, and where interruptions are likely during movement. For critical points, don’t immediately adopt the first value—confirm short-term stability and re-observe if necessary to detect variation caused by corrections.

Correction data are essential to high-precision positioning but can also be a weakness. Ignoring site communication situations and correction applicability can lead to apparently successful positioning that fails to meet required accuracy. If you plan to use centimeter-level positioning (cm level accuracy; half-inch accuracy) in operations, do not treat correction data as a black box—actively assess its quality.

Error factor 4: variability in equipment setup and observation procedures

In high-precision positioning, how equipment is set up and how observations are made often affects results more than the equipment itself. The same place, same time, and same receiver can yield different coordinates if setup differs. Commonly overlooked operational issues include recording antenna position without a clear reference, inconsistent handling of installation height, and differing tilt or holding methods. When seeking centimeter-level accuracy (cm level accuracy; half-inch accuracy), such operational differences cannot be ignored.

For example, when holding the device by hand, even if it appears stationary, slight body sway or changes in the grip cause subtle movement of the antenna. Observers themselves often do not notice this movement. Sites that prioritize speed to record many points quickly tend toward such operation, but if centimeter-level accuracy is required, steps to improve repeatability are necessary. Use stable supports, clearly define observation reference points, and standardize handling of installation height and measurement points; otherwise, human-induced errors before positioning will contaminate results.

Defining what position is being recorded is also important. Whether you want a point on the ground, a corner of a structure, the point directly beneath the device center, or the photographic position affects the meaning of coordinates. In practice, the definition of which point is measured is sometimes vague, causing later mismatches with drawings or point clouds. This is not a precision issue but an observation design issue. No matter how accurately you measure, ambiguous reference definitions render the data unusable.

Vertical handling is another frequent source of confusion. Even if the planar position is good, inconsistent vertical input or reference plane treatment breaks overall product consistency. When multiple people take measurements, one may include the support structure height, another may use the device body height, and yet another may interpret the contact point with the ground differently—such discrepancies are common. These are not algorithmic issues but errors stemming from insufficient on-site rules.

As a countermeasure, document observation procedures—even if simplified—and standardize them so that anyone produces similar results. Define when to start observations, how long to confirm stability, what constitutes the measurement point, how to manage installation height, and when re-observation is required, all in practical site language. In many cases, reproducible operations matter more than having more advanced equipment.

Sites that fail in centimeter-level positioning (cm level accuracy; half-inch accuracy) do not necessarily lack technical understanding. Often the system is understood but daily observation procedures vary among operators. Therefore, improving accuracy often comes not from buying new equipment but from standardizing observation procedures.

Error factor 5: insufficient understanding of site conditions and lax operational judgment

Finally, beyond specific technical factors, insufficient understanding of site conditions and lax operational judgment are major causes of failure. This may seem abstract but is among the most common practical failure modes. Even if there are individual problems like poor satellite geometry, reflections, corrections, or setup issues, recognizing these as anomalies on site and deciding to re-measure can prevent major failures. Conversely, if anomalies go unnoticed and measurements are adopted, errors propagate into later processes.

For example, recording measurements hurriedly for time reasons even when the solution has not stabilized, not re-observing despite many nearby reflectors, operating under the same assumptions in the morning and afternoon even though reception conditions change, or applying the same level of quality checks to both critical and reference points—such judgments are not uncommon. Time pressure on site is real, but in high-precision positioning, “just take it” is often the most dangerous decision.

Misunderstanding centimeter-level positioning (cm level accuracy; half-inch accuracy) as omnipotent is also problematic. Satellite positioning is very effective but struggles in indoor or semi-underground locations, in areas with strong sky blockage, or in strongly reflective environments. If you lack the idea of combining other observation methods, checking against known points, or cross-checking with photos and point clouds, you may misjudge the technology’s strengths and weaknesses. High-precision positioning should be positioned as part of the overall site recording method, not a stand-alone solution.

Also, failing to reverse-engineer the needed accuracy from the product side causes failures. For example, even if you think you measured at centimeter-level, later mapping, photo organization, point-cloud processing, coordinate transformations, or sharing may break consistency, leaving final deliverables below expectations. High-precision positioning alone is meaningless unless the entire operation is designed for high accuracy. Conversely, if required accuracy is defined across the workflow, you can see where strictness is necessary and where simplification is possible.

The important countermeasure is to treat positioning as quality management, not just device operation. To enable on-site decisions about whether to observe, whether to adopt a value, or whether to re-observe, staff must understand “what states are suspicious.” Do not leave that judgment to individual intuition—share it as site rules and check items. Sites with stable high-precision positioning operations invariably have clear decision criteria.

Practical mindset to stabilize centimeter-level positioning

We have covered five error factors, but in practice what matters is not memorizing them individually but arranging site procedures so they can be checked in sequence. Differences in real-world performance come more from operational design than theory. Keep in mind a three-stage approach to quality: before, during, and after positioning.

Before positioning, assess the site reception environment. Check sky openness, surrounding reflectors, communication status, movement routes, and locations of critical measurement points to identify difficult areas. If you know trouble spots beforehand, you can respond calmly when values become unstable on site. Entering a site without any assessment makes isolating causes of anomalies difficult.

During positioning, monitor stability. Don’t assume it is finished once a solution is locked—check for small short-term variations, whether re-observation yields similar values, and whether the result is consistent with the surroundings. For critical points, don’t blindly accept single measurements; checking consistency across several observations significantly improves judgment accuracy.

After positioning, verify consistency with deliverables. Compare with known points to ensure there is nothing unnatural, check that photo and drawing relationships do not feel off, and confirm that relative relationships with other points are intact. In centimeter-level projects, a single value can look plausible yet reveal anomalies in the overall consistency. Therefore, design workflows that include post-processing quality checks rather than treating the job as finished on site.

Also, gaining accuracy and achieving stable reproducibility are slightly different. Practical needs prioritize reproducible quality that anyone can achieve on any site rather than a one-time high accuracy under ideal conditions. To do this, standardize observation methods, share what environments to avoid, and verbalize decision criteria. In adopting high-precision positioning, operational stability matters more than the peak accuracy value.

Implementation perspectives to make high-precision positioning useful on site

Making centimeter-level GPS positioning (cm level accuracy; half-inch accuracy) useful on site requires not only knowing error factors but integrating the process into daily work without undue burden. No matter how accurate a system is, if preparation is cumbersome, operation requires expert knowledge, and results vary among operators, it will not be widely adopted. For practitioners, what matters is reproducing the required accuracy when needed, as simply and reliably as possible.

From this viewpoint, the future of high-precision positioning is less about receiver performance competition and more about ease of field implementation. Factors that strongly affect adoption are: ease of checking positioning results, ease of integration with photos, point clouds, and drawings, ability to standardize operations across different observers, and portability on site. In short, it’s no longer enough to ask whether centimeter-level accuracy can be achieved—the question is whether that accuracy can be fully utilized within business workflows.

If you want to make high-precision positioning more practical on site, handle positioning results together with photography and record-keeping, and operate more flexibly without relying solely on specialist equipment, options like LRTK—a smartphone-mounted GNSS high-precision positioning device—are promising. Such solutions make it easier to incorporate high-precision positioning in an easy-to-handle form, enabling location recording as an extension of everyday tasks and making it easier to equalize recording quality across operators.

Centimeter-level positioning (cm level accuracy; half-inch accuracy) does not succeed simply by introducing equipment. However, by correctly understanding error factors, developing operations suited to your sites, and choosing easy-to-use systems, the reliability of location information can dramatically improve. To reduce rework, enhance the value of records, and improve on-site decision-making accuracy, review how high-precision positioning fits your operations after addressing error factors. Adopting a system such as LRTK, which considers field implementation, is a shortcut to turning centimeter-level positioning into a truly usable technology.