RFID, NFC & UWB · Study deck

UWB Indoor Positioning Systems

Ultra-wideband (UWB) is a radio method that uses very short signals to measure distance.

Radio Remi is your guide for this deck.

anchor-geometrygdopindoor-positioning

Radio Remi, the module guide, in a scene from this chapter.

After studying this chapter

Learning objectives

You will be able to:

Explain why anchor geometry and line-of-sight quality can matter more than nominal radio precision.
Compare TWR, TDoA, and angle-assisted UWB architectures without treating any one mode as universally best.
Use GDOP as an error-amplification signal rather than as a magic accuracy guarantee.
Design a site survey that records anchor coordinates, visibility, obstruction classes, calibration evidence, and retest triggers.

Major section

Start With the Story

A dot on a map may look exact, yet a poor anchor layout or blocked path can place that dot on the wrong side of an aisle.

When confidence falls, the safe response may be to warn, slow down, or ask for another source instead of guessing.
One site test cannot promise the same result in every room or season.
That claim depends on anchor placement, synchronization, geometry, line-of-sight quality, filtering, confidence, and how the application reacts when uncertainty grows.

Major section

Eddie's Math Bridge: Link Budget Before Geometry

The mathematical gist.: The chapter's 30 m by 20 m floor has a 36.06 m diagonal.

Geometry cannot amplify a range measurement that the anchor never detected.

Major section

Position Estimates as Evidence

A UWB location estimate should be treated as an evidence record, not as the exact physical truth.

The record should say what was measured, which assumptions were used, and how confident the engine is in the result.
Applications should respond differently to high-confidence, low-latency estimates than to stale or low-confidence estimates.
For example, an asset map can show a broad uncertainty area, while a robot safety workflow may require a recent estimate with enough visible anchors and a low residual before it acts.

Major section

Anchors Turn Ranges Into Coordinates

A UWB positioning system is a set of fixed anchors at known coordinates and mobile tags to be located.

Ranging gives distances or arrival-time differences; the positioning system combines several of those measurements into an (x, y) or (x, y, z) estimate.
The two big system questions are geometry and capacity.
Geometry asks whether anchor placement gives the solver enough independent viewpoints.
For an asset map, a broad low-confidence region may be acceptable.

Major section

Anchor Geometry and GDOP

Anchor placement sets the shape of the positioning problem.

Collinear or near-collinear anchors can make a small distance error appear as a large position error in the weak direction.
The: GPS receivers calculate GDOP to estimate position quality inference is bounded by the: Geometry Affects Positioning Accuracy observation.

Anchor geometry sets accuracy: clustered anchors produce a large error region (high GDOP), while well-spread anchors intersect sharply for low GDOP.

Major section

Anchor Geometry and GDOP (continued)

The geometry factor is not the whole error budget, but it is a useful warning sign.

If the geometry factor is high, improving firmware or filter settings may not fix the installation.
A 30 m x 20 m work area might use anchors at known map coordinates and a tag mounted on a cart.
GDOP does not predict every error; it warns when the installation makes small timing errors expensive.

Major section

Measurement Quality Pipeline

This pipeline prevents a common failure: a location server publishes a coordinate even when the underlying evidence is weak, and the application treats the coordinate as certain.

A coordinate without quality fields is not enough for automation or safety decisions.

UWB positioning measurement pipeline from raw time observations through quality checks, anchor selection, solver, residual review, confidence class, and application policy.

Major section

Architecture Choices

To place architecture choices on firm evidence, separate tag exchanges from clear distance using it. Anchor clock identifies the later check.

TWR positioning uses message exchanges between a tag and anchors.
The trade-off is that tags and anchors spend airtime on each exchange, so update rate and tag count must be planned.

UWB positioning architecture choices comparing TWR, TDoA, and angle-assisted approaches by tag behavior, anchor needs, synchronization, solver evidence, and likely fit.

Major section

Capacity Often Picks the Positioning Method

The reason large deployments often favor TDoA over TWR is channel airtime.

In TWR, every position needs a back-and-forth exchange between the tag and each anchor, so airtime scales with tags x anchors x update rate.
In TDoA, a tag transmits one blink and synchronized anchors listen, so one short transmission can support the fix.
The tradeoff is that anchor synchronization, timestamp quality, and calibration become release-critical evidence.

Major section

Capacity Often Picks the Positioning Method (continued)

If 200 tags need a 1 Hz update, the system must budget about 200 tag reports per second before retries, management frames, and rejected observations.

If the workflow can accept 0.2 Hz for slow-moving assets, the load drops to 200 x 0.2 = 40 reports per second.
That change may decide whether TDoA blinks are sufficient, whether TWR exchanges are practical only for a smaller tool fleet, or whether the project needs smaller zones with different update policies.
A simple review estimate is 0.15 x 2.5 + 0.20 + 0.10 = 0.675 m.

Major section

Site Survey and Deployment Process

A site survey should produce an installable coordinate system and an evidence trail.

Both output and anchors need evidence.
Reopen site survey and deployment process whenever anchors changes.
A later change to shelving, timing, firmware, or tag placement therefore reopens the affected evidence instead of silently inheriting the original acceptance.

UWB positioning deployment review flow from requirements through coordinate frame, obstruction survey, anchor plan, calibration, acceptance tests, release evidence, and retest triggers.

Major section

Common Review Failures

Treating accuracy as a fixed product property: Position quality changes with geometry, line-of-sight, mounting, timing, update rate, and filtering.

Optimizing the center and ignoring edges: Edges, corners, doorways, lifts, racks, and vehicle paths often have weaker geometry than the open middle of a room.
Hiding NLOS with smoothing: Filters can make bad data look stable.
Forgetting the coordinate frame: A precise tag estimate is useless if the anchor map, floor level, axis direction, or application map is misaligned.

Deck summary

Key takeaways

A dot on a map may look exact, yet a poor anchor layout or blocked path can place that dot on the wrong side of an aisle.

The mathematical gist.: The chapter's 30 m by 20 m floor has a 36.06 m diagonal.
A UWB location estimate should be treated as an evidence record, not as the exact physical truth.
A UWB positioning system is a set of fixed anchors at known coordinates and mobile tags to be located.
Anchor placement sets the shape of the positioning problem.

Retrieval practice

Recall check 1 of 3

Radio Remi says: answer from memory, then check your reasoning.

Q1Why is UWB positioning not simply 'indoor GPS'?

AIt estimates position from timed observations, anchor geometry, and clock quality

BBecause it uses exactly the same satellites that outdoor GPS receivers use

CBecause its short pulses make radio timing precision the main site-acceptance measure.

DBecause peer-to-peer ranging lets a tag determine its room coordinates without surveyed reference points.

Show answer

Answer: A UWB estimates position from timed radio observations, anchor geometry, and clock quality with explicit confidence, unlike satellite GPS.

Retrieval practice

Recall check 2 of 3

Radio Remi says: answer from memory, then check your reasoning.

Q2A UWB system has six anchors, but all of them are mounted along one long wall of a warehouse. Tags near the opposite side produce unstable depth estimates even when several anchors are visible. What is the best diagnosis?

AThe system has too many anchors; UWB positioning requires exactly three anchors for 2D

BThe anchor geometry has high GDOP in the depth direction, so small measurement errors are amplified

CSwitching from TWR to TDoA automatically fixes the position error

DThe application should smooth the coordinate more aggressively

Show answer

Answer: B Anchors on one wall do not surround the tag, so the solver has poor leverage for depth.

Retrieval practice

Recall check 3 of 3

Radio Remi says: answer from memory, then check your reasoning.

Q3A warehouse must locate thousands of asset tags with modest update rates and long tag battery life. Which positioning approach fits, and what is the deciding factor?

ASS-TWR with several anchors per tag, because avoiding anchor synchronization would simplify the warehouse installation.

BAoA from a single anchor, because reducing installed anchor count lowers cabling and survey costs.

CRaise transmit power so distant tags need fewer retries, then keep the existing two-way exchange schedule.

DTDoA with synchronized anchors: each tag blinks once and anchors timestamp it, so airtime scales better than TWR.

Show answer

Answer: D TDoA's single-blink, listen-at-all-anchors model minimizes per-tag airtime and energy, so it scales to many tags.

Print reference

Answers

Answer key.

A · UWB estimates position from timed radio observations, anchor geometry, and clock quality with explicit confidence, unlike satellite GPS.
B · Anchors on one wall do not surround the tag, so the solver has poor leverage for depth.
D · TDoA's single-blink, listen-at-all-anchors model minimizes per-tag airtime and energy, so it scales to many tags.