Indoor positioning for care homes

Indoor positioning systems for care homes

Indoor positioning locates people and objects inside buildings where GPS drops out, so every alert reaches staff with a room, bed, or zone attached.

Most guides are written for warehouses and factories. This one is written for care, with room-level accuracy and no cameras in resident rooms.
Room-level accuracyNo cameras in resident roomsWorks with existing devices
Indoor positioning systems for care homes

The basics

What an indoor positioning system does

An indoor positioning system is hardware around the building plus software that reads the position data. GPS weakens through walls, floors, and roofs, so it cannot place someone reliably indoors.

Indoor positioning fills that gap with local signals. The software puts each reading on a digital floor plan, so instead of raw coordinates, staff see people and objects on a live map.

Tags

Small devices attached to people or objects that emit a detectable signal.

Anchors and sensors

Fixed reference points placed around the building that detect the tag signals, the way satellites work outdoors.

Location engine

Software that turns the detected signals into a position and tracks movement as readings change.
What an indoor positioning system does

Technologies compared

Five technologies, different tradeoffs

There is no universally best option. Each technology trades accuracy against infrastructure, range, and cost, and higher precision usually means more hardware to install and power. Match the technology to the job, not to the spec sheet.

UWB

10 to 30 cm in line-of-sight conditions. Dedicated anchors and tags and higher upfront cost. Best for exact-position work like robotics and small-tool tracking.

BLE

Room-level readings around 1 to 3 metres. Low-cost beacons, easy to scale, coin-cell tags that last years. Good for room-to-zone tracking of people and assets.

WiFi

Broader at around 3 to 5 metres, and can reuse existing access points. Suits general indoor navigation where sub-metre precision is not critical.

RFID

Reports zone or reader-point presence. Cheap passive tags with no battery, but you need a reader network. Good for choke-point presence and inventory.

Ultrasound

High precision in controlled rooms, set up room by room with dedicated beacons and receivers. Used where an exact position matters.

Where it is used

From generic tracking to care operations

Indoor positioning shows up wherever GPS is weak and location has to support movement, assets, or site operations. In care-home operations, the same data does more: it maps to rooms and beds, prioritises alerts, and writes the records managers used to reconstruct by hand.
Generic indoor positioning
Locate tools, equipment, or people somewhere indoors.
Care-mapped operations
Search a named resident or asset and see it on the facility floor plan.
Generic indoor positioning
Real-time location visibility as a raw data feed.
Care-mapped operations
Location plus room-level context and active alerts on one live view.
Generic indoor positioning
Alarms fire on every raw location event.
Care-mapped operations
Rule-based alerts, like out of bed for over 15 minutes at night, so a notification means something.
Generic indoor positioning
Analytics explain movement and facility usage.
Care-mapped operations
Automatic visit logs, response times, and incident reports build in the background.

Choosing a system

How to choose the right system

There is no single best technology, only the best fit for a defined task. Match accuracy, update speed, infrastructure, software, and operating constraints to the job in front of you.
01

Start with requirements

Name the problem first: resident safety, staff safety, assets, navigation, or workflow visibility. Then define who gets alerted, what they do, and how fast the handoff must happen.
02

Match accuracy to the job

Enough accuracy is whatever the task needs, and no more. Care alerts usually only need to answer which room and which bed, so room-to-zone precision is often plenty.
03

Check infrastructure and software

Confirm network dependency, digital maps, power and mounting limits, integrations like nurse call, and policy limits on data, cloud, and privacy before accuracy even matters.

Planning a rollout

How to plan an implementation

Implementation runs in four moves: define the use case, survey the site, map infrastructure to the building, then validate in a pilot before wider rollout. You decide to scale from what the pilot's impact report shows, not from a number set on a whiteboard.
01

Define the use case

Pick one problem with a clear operational payoff. Record today's baseline: how incidents are found now, rough response times, and hours spent on manual reporting.
02

Survey the environment

Walk the actual rooms, beds, corridors, and zones alerts will point to. The walk-through captures both the physical layout and how the setting works today.
03

Plan infrastructure and maps

Digitise the floor plan first, define each tracked room or bed, then place hardware to cover those areas without dead zones.
04

Run a pilot and validate

Confirm coverage with a walk test, check alerts reach staff on existing devices, and measure response times against the baseline. The impact report decides rollout.

The software layer

What the software has to do

Hardware locates a tag. The software turns that location into alerts worth acting on and a view people can run the floor from. Because each sensor is linked to a room or bed, every insight is anchored to a real place.

Detect and filter

Virtual zones detect entry and exit. Rules like out of bed for over 15 minutes at night keep notifications to events that actually need someone to step in.

Map context to places

Each reading lands on a digitised floor plan tied to a specific room or bed, so an alert says room 4, bed 2, not zone 3.

One operations view

A single live map shows residents, caregivers, vehicles, assets, and active alerts together, instead of five tools that each show a slice. The single-view senior monitoring comparison weighs how systems combine resident activity, live alerts, and one operational view.
What the software has to do

Accuracy and reliability

What actually affects accuracy

Headline accuracy is a best case. Four things decide what you get in a working building, and calibration matters as much as the technology.

Positioning technology

Sets the accuracy ceiling. UWB reaches 10 to 30 cm in line of sight; BLE and WiFi suit broader room- or zone-level visibility.

Environmental interference

Radio, light, and sound interference degrade the signal. How much depends on the positioning method's tolerance for it.

Infrastructure layout

Tags report to anchors around the building, so position quality depends on how well that network covers where people actually move.

Calibration

Thresholds that ignore the resident's baseline produce false alerts, and frequent false alarms desensitise staff until real alerts get ignored.

Before you commit

What to check before you commit

Before you sign, confirm the system delivers the accuracy you need, keeps alerts actionable, fits daily workflows, and can prove value in a real pilot with clear terms.

Accuracy fit

Judge precision against the real task, not the marketing claim. Confirm coverage with a walk test in the actual building.

Alert reliability

Test alert quality under real conditions, including ordinary movement and shift traffic. Actionable alerts versus total alerts beats raw event volume.

Workflow fit

Alerts, location context, and reporting all have to slot into existing routines. A system can be accurate and still not fit how staff work.

Pilot and contract

Agree pilot scope, support ownership, and exit terms up front. Let the final impact report, not a sales-meeting number, decide rollout.

Guardian pilot

See Guardian work in one ward

We install Guardian in one ward and map the floor plan, priority rooms, and alert rules around how the team works.

Over 6 to 8 weeks, Guardian records response times and incidents in the background. You get an impact report and use your own data to decide whether to roll out.

Room-level alerts

Staff see the resident's name and room on the devices they already use.

Automatic records

Guardian records visits, incidents, and response times without adding shift-end paperwork.

Proof before rollout

Compare the pilot results with your baseline, then decide what to scale.
Guardian Insight showing live care alerts and room-level activity

Guardian

Test Guardian in one ward or team

Run a 6 to 8 week pilot alongside daily care. Use your own response times, incident records, and impact report to decide whether to roll out.
Request a pilot

FAQ

Common questions about indoor positioning

Can indoor positioning work without GPS? +
Yes. Indoor positioning is a local, signal-based way to locate people and assets indoors without GPS.

GPS relies on satellite signals, and those weaken or drop out indoors. Metal, concrete, and brick block or scatter them, so GPS cannot place someone reliably inside.

Local signals replace satellites. Indoor positioning uses RFID, Bluetooth, WiFi, or ultra-wideband, with UWB reaching about 10 to 30 cm in line-of-sight conditions. A UWB setup places tags on people or objects and anchors around the building, then calculates position from the signals between them.
Do you need dedicated hardware? +
Dedicated hardware is usually required. The exact hardware depends on the positioning technology and the accuracy you need.

Typical hardware has two parts:
  • Mobile tags worn by people or attached to assets being tracked.
  • Fixed points (anchors, beacons, or sensors) placed around the building to receive tag signals.

The central system uses those signals to calculate position. A live internet connection is separate: in local IoT setups, gateways and smart devices communicate on site, so positioning can continue without an active internet link.
How accurate can indoor positioning be? +
Indoor positioning accuracy ranges from room-level to centimetre-level. The useful planning number depends on the technology, device support, and the building.

Planning ranges:
  • WiFi: 5 to 15 metres typically; room-level with RSSI fingerprinting, and under 1 metre with RTT/FTM on supported devices.
  • Standard Bluetooth: about 3 to 5 metres; BLE AoA or channel sounding reaches roughly 0.5 to 1 metre in good conditions.
  • Bluetooth 6.0 channel sounding: up to 10 to 30 cm in theory.
  • UWB: centimetre-level, the tightest of the group.

Lab accuracy is a best case. Deployed accuracy often degrades by 20 to 50% or more in indoor-positioning research because buildings change signal behaviour. Common causes are multipath reflections from walls and metal, people blocking the signal path, moved furniture, and non-line-of-sight between tags and anchors.

High-accuracy modes need compatible hardware and line-of-sight planning: Wi-Fi RTT depends on access-point support and density; BLE AoA and channel sounding need compatible Bluetooth hardware; UWB performs best with line of sight.
Can indoor positioning work across multiple floors? +
Yes, when the system can assign each location to the correct level.

Multi-floor positioning is used in buildings where every level can look alike, such as multi-storey car parks.

The system has to identify the floor level as well as the horizontal position. Floor-by-floor maps and location logic keep second-floor room 12 separate from third-floor room 12.
Do you need indoor maps first? +
Indoor maps are optional for basic tracking, but required for navigation, wayfinding, and map-based location display.

Positioning locates a tag. Navigation adds route guidance on top, so the map requirement depends on the feature you plan to show.
  • Basic tracking: no map is needed if a tag location can be shown as a zone or device reading.
  • Navigation and wayfinding: a map is required for routes and directions.
  • Map-based display: a map is required for a moving marker or room-level view.

For care alerts, a digitised floor plan gives room context without full navigation. During setup, each sensor is assigned to the relevant room or bed, which turns a raw alert into a clear destination like room 12, bed 2.
Aleks Timm

Author

Aleks Timm

Aleks Timm leads Guardian and builds privacy-first operations technology for care homes and home care providers. Teams get location-aware alerts they can act on, clearer situational awareness, and measured insight into how care work actually runs.

Read More