IoT Interfaces, Location and Privacy · Study deck

Indoor Positioning: Wi-Fi, UWB, and Sensor Fusion

A Bluetooth Low Energy (BLE) deployment can estimate position.

UX Uma is your guide for this deck.

locationawarenessindoor

After studying this chapter

Learning objectives

You will be able to:

design and maintain a Wi-Fi fingerprint map
explain UWB time-of-flight accuracy and deployment limits
combine radio and motion evidence for indoor-outdoor transitions
Explain: Walls and other features shape those patterns, so AP3's weak value may be as informative as AP1's strong value.

Major section

Wi-Fi Fingerprinting

Walls and other features shape those patterns, so AP3's weak value may be as informative as AP1's strong value.

During the online phase, the device measures a new vector and compares it with the stored fingerprints.
The closest match supplies an estimate, not a direct range.
Device scans visible APs and measures RSSI.

Major section

Wi-Fi Fingerprinting (continued)

Moved access points, changed furniture, crowds, device calibration, or stale surveys can shift the pattern, so confidence and a survey-maintenance rule belong with the result.

Temporal variations: Different occupancy (empty vs. Crowded) affects propagation.
Device variations: Different phones have different RSSI calibration.
Machine learning (automatically detect and adapt to environmental changes).

Major section

UWB Time-of-Flight Precision

Traditional one-way ToF problem: If transmitter and receiver clocks are not synchronized, measuring $t_{\text{receive}} - t_{\text{transmit}}$ gives nonsense (clocks may differ by milliseconds).

Tag → Anchor: Tag sends pulse at $t_1$ (tag clock).
Example with numbers: Tag sends pulse at $t_1 = 0$ .
Anchor waits $T_{\text{reply}} = 500$ µs, then replies.

Diagram showing UWB positioning with anchors at known positions and tag at unknown position.

Major section

UWB Time-of-Flight Precision (continued)

Signal travels to anchor (distance $d = 5$ m): arrives at $\frac{5}{3 \times 10^8} = 16.67$ ns.

Tag receives at $t_4 = 500 \text{ µs} + 33.34 \text{ ns} = 500.03334$ µs.
The timing calculation becomes a position only when several measured distances meet in a known geometry.
Three distances can solve the two-dimensional position; the fourth adds redundancy for checking noise and obstructed paths.

Major section

Indoor-Outdoor Sensor Fusion

Scenario: A delivery driver app must track location seamlessly as the driver moves from outdoor parking (GPS) to indoor warehouse (Wi-Fi/BLE) for package pickup.

Building floor plan constraint: Driver must be in accessible walkway.
HDOP increases from 1.2 to 6.5 (poor geometry).
GPS provides absolute position outdoors but fails indoors.

Indoor positioning combines GPS handoff, Wi-Fi fingerprints, BLE beacons, and IMU motion in a fusion filter, then applies map constraints before exposing a usable place estimate.

Major section

Indoor-Outdoor Sensor Fusion (continued)

Result: The fused sensors and map constraint can support a Loading Dock 3 assignment; validate position error against calibrated site measurements before reporting an accuracy value.

IMU provides relative movement but drifts over time; Wi-Fi fingerprinting gives zone-level position but requires a pre-surveyed database; BLE beacons provide proximity constraints.
A Wi-Fi site map gives 2–5 m, while BLE gives 1–3 m.
The worked examples have named the evidence sources.

Major section

Indoor-Outdoor Sensor Fusion (continued)

No source in the delivery-driver example remains trustworthy across the whole journey. Figure: Indoor positioning combines GPS handoff shows how complementary measurements enter one estimate while their accuracy and drift determine how much influence each receives.

On the data sources side of Figure: Indoor positioning combines GPS handoff, GPS gives a point with an error band of 3–10 m; that band is 10 m: 3 m = 7 m wide.
The IMU fills motion between fixes, but its drift grows with time.
The next step is turning those measurements into a repeatable implementation path.

Major section

Hospital Asset Tracking Design

Room-level accuracy (identify which room equipment is in).

Decision: The listed BLE design fits the $50,000 infrastructure cap, but its stated 5–7-year battery life misses the 10-year requirement.
Key Insight: UWB provides finer positioning than BLE but has a higher infrastructure cost and shorter listed battery life.
For hospital asset tracking, room-level accuracy may be sufficient; validate room assignment performance and the tag power plan against the deployment requirements.

Major section

RSSI Calibration Neglect

Illustrative deployment failure: A hypothetical retail example shows an 8 m estimate for a beacon actually 3 m away when the path-loss model is uncalibrated; use measured RSSI and reference distances for any deployment claim.

The Cost: A wrong path-loss exponent can produce large site-dependent range and position errors.
This calibrated value should replace the default n=2.5 in your deployment.
Datasheet values are laboratory measurements in free space—real buildings have walls, metal, people, and furniture that change RF propagation.

Major section

Key Takeaway

The choice between BLE beacons (cheap, 1-3m), Wi-Fi (free infrastructure, 3-5m), and UWB (expensive, 10-30cm) depends on accuracy requirements, budget, and existing infrastructure.

The most robust real-world systems use sensor fusion, combining multiple technologies through Kalman filters to smooth transitions and handle individual sensor failures.
The key insight: no single positioning technology works everywhere -- seamless location awareness requires intelligent fusion of multiple sources.
Indoor positioning requires fundamentally different technologies from outdoor GPS because satellite signals cannot penetrate buildings.

Major section

For Kids: Meet the Sensor Squad!

Indoor positioning is like finding your way in a building without any windows to see the sky!

The Sensor Squad was running a treasure hunt inside a huge shopping mall.
But there was a problem -- GPS didn't work indoors!
"Don't worry!" said Temperature Terry. "We have INDOOR tools!".

Major section

For Kids: Meet the Sensor Squad! (continued)

"My GPS says I'm OUTSIDE!" cried the LED, standing right in the middle of the food court. "The satellite signals can't get through the roof!".

Each one continuously shouted: "I'm at the food court!" or "I'm at the shoe store!" When Lila's phone heard a beacon, it knew she was nearby.
If I measure the signal strength from THREE beacons, I can figure out exactly where I am -- just like GPS uses three satellites!" said Sammy.
This PATTERN is unique to the pet store!".

Deck summary

Key takeaways

Walls and other features shape those patterns, so AP3's weak value may be as informative as AP1's strong value.

Moved access points, changed furniture, crowds, device calibration, or stale surveys can shift the pattern, so confidence and a survey-maintenance rule belong with the result.
Traditional one-way ToF problem: If transmitter and receiver clocks are not synchronized, measuring $t_{\text{receive}} - t_{\text{transmit}}$ gives nonsense (clocks may differ by milliseconds).
Signal travels to anchor (distance $d = 5$ m): arrives at $\frac{5}{3 \times 10^8} = 16.67$ ns.
Room-level accuracy (identify which room equipment is in).

Retrieval practice

Recall check 1 of 5

UX Uma says: answer from memory, then check your reasoning.

Q1In an illustrative warehouse planning exercise, 500 pallet tags need aisle-level detection. A site trial confirms BLE meets that need; assumed installed totals are BLE $65,000 and UWB $280,000, and shelving obstructs camera coverage. With a $70,000 budget, which option fits these stated inputs?

AUWB with 10-30cm accuracy for precise pallet location

BGPS receivers on each pallet for global tracking

CBLE beacons for low-cost 1-3 m zone detection.

DCamera-based tracking with computer vision

Show answer

Answer: C Under the stated trial and illustrative quotes, BLE meets the aisle-level task within budget; UWB exceeds the budget.

Retrieval practice

Recall check 2 of 5

UX Uma says: answer from memory, then check your reasoning.

Q2A driver walks indoors as GPS fades. Motion sensing continues briefly, while a Wi-Fi fingerprint gives a rough zone. What should the app do?

AKeep last outdoor GPS point as current during indoor walking.

BTrust motion for the whole indoor route because inertial tracking remains accurate across the walk.

CSnap to Wi-Fi zone and conceal its rough uncertainty.

DBridge with motion, correct by Wi-Fi zone and map, and show uncertainty.

Show answer

Answer: D Bridge the gap with motion, correct it with the Wi-Fi zone and map, and show uncertainty.

Retrieval practice

Recall check 3 of 5

UX Uma says: answer from memory, then check your reasoning.

Q3You're deploying BLE beacons in a museum for indoor wayfinding. Identifiers are unique and phone scanning works, but beacons are 30 meters apart and visitors' locations jump between distant exhibits as reflected signals vary. What is the most likely cause?

ABeacon identifiers duplicated across several exhibits, confusing the map lookup

BMultipath and sparse beacon geometry making RSSI distances unstable

CPhone Bluetooth permission denied, so no beacon packets are scanned

DBLE transmit power is too high, preventing nearby phones from hearing it

Show answer

Answer: B Indoor RF propagation suffers from multipath effects—signals bounce off walls, floors, ceilings, metal objects, and people.

Retrieval practice

Recall check 4 of 5

UX Uma says: answer from memory, then check your reasoning.

Q4You're implementing Wi-Fi fingerprinting for indoor positioning in an office building. Accuracy degrades significantly after access points are relocated and the stored fingerprint map is not resurveyed. What is the most likely cause?

AThe office moved from 2.4 GHz to 5 GHz, which always blocks fingerprinting

BAccess points slowly lost transmit power through normal hardware aging

CA floor-plan or access-point change invalidated the RSSI fingerprint map

DPhone Wi-Fi receivers wore out after repeated scans by the app

Show answer

Answer: C Wi-Fi fingerprinting maps RSSI patterns to physical locations during a calibration phase.

Retrieval practice

Recall check 5 of 5

UX Uma says: answer from memory, then check your reasoning.

Q5What raw RSSI, range and motion observations feed indoor position fusion before a fix exists?

APosition Fusion and Confidence

BRadio and Sensor Observations

CWayfinding and Fallback

Show answer

Answer: B Raw indoor observations belong together because RSSI, ranges, and motion samples are evidence inputs rather than a finished position.

Q6Complete the IoT dashboard widget creator with DOM manipulation:

Aconst widget = document.createElement('div');

Bconst widget = document.newElement('div');

Cconst widget = document.querySelector('div');

Dconst widget = new Element('div');

Show answer

Answer: A document.createElement() creates DOM elements. innerHTML with template literals allows dynamic HTML content.

Print reference

Answers 1 of 2

Answer key.

C · Under the stated trial and illustrative quotes, BLE meets the aisle-level task within budget; UWB exceeds the budget.
D · Bridge the gap with motion, correct it with the Wi-Fi zone and map, and show uncertainty.
B · Indoor RF propagation suffers from multipath effects—signals bounce off walls, floors, ceilings, metal objects, and people.
C · Wi-Fi fingerprinting maps RSSI patterns to physical locations during a calibration phase.

Print reference

Answers 2 of 2

Answer key.

B · Raw indoor observations belong together because RSSI, ranges, and motion samples are evidence inputs rather than a finished position.
A · document.createElement() creates DOM elements. innerHTML with template literals allows dynamic HTML content.