TL;DR
Get workout gear delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
Nottingham Trent University researchers and Integrated System Technologies Ltd developed a prototype that combines radar, low-resolution thermal sensors and smart plugs to monitor activity at home. Tests in a mock domestic setting found that combining sensor data improved activity-recognition performance; the study does not establish how well the system works in occupied homes or whether it can detect medical emergencies reliably.
Nottingham Trent University researchers, working with Integrated System Technologies Ltd, have developed a home-monitoring prototype that combines radar, thermal sensors and smart plugs to track older residents’ daily activity without cameras or wearable devices. A study published in Sensors reports improved activity-recognition performance when the sensor data were combined in a mock home, while whether the system can reliably identify emergencies in real households remains unestablished.
The prototype uses millimeter-wave radar to detect movement, low-resolution thermal sensors to identify broad postures such as sitting, standing, walking and lying down, and smart plugs to record appliance use associated with routines such as making tea or preparing food. The researchers say the sensors are placed discreetly at selected locations in a home. Thermal output is kept pixelated, and the system is designed to avoid capturing facial features or identifying a resident.
Its AI system learns an individual’s activity patterns over time and looks for departures from those patterns. The researchers describe possible signals including a fall, prolonged inactivity, changes in mobility or sleep, and missing household routines. If the system flags a possible emergency, the intended response is to alert care professionals or family members, who could check on the resident remotely or in person. The report does not describe this as a diagnostic tool.
For the study, the team used a mock domestic environment with six room layouts, including bedrooms and living areas with different furniture arrangements. They recorded a series of activities to test recognition of movement and posture. The researchers report that combining the sensor streams improved performance compared with relying on sensors individually. The supplied report does not give specific accuracy figures, participant numbers or results from long-term use in occupied homes.
Monitoring Without Cameras or Wearables
For older adults who want to remain in their homes, a system that notices unusual activity could offer another way for family members or care professionals to check on well-being. The proposed approach is designed to work in the background, without requiring a resident to wear or remember to charge a device, and without using conventional video cameras. That could make it more acceptable to some people, although individual preferences and consent would matter.
The potential benefit is still a research aim, not a demonstrated outcome. The reported test examined activity recognition in a mock setting; it did not show that the prototype prevents injury, reduces care needs or allows people to live independently for longer. Before such a system could guide care, its performance in ordinary homes, including how often it misses events or raises false alerts, would need to be assessed.
Products Worth Considering
BowFlex Results Series SelectTech 552 Adjustable Dumbbell
Levoit Vital 100S-P Air Purifier
BowFlex Results Series SelectTech Adjustable Dumbbells (Pair)
As an affiliate, we earn on qualifying purchases.
How the Three Sensors Work Together
The study, titled “Privacy-Preserving Ambient Sensing for Activities of Daily Living: Multimodal Radar-Thermal Human Activity Recognition and Smart Plug Appliance Recognition,” was published in the journal Sensors in 2026. The research involved NTU’s School of Architecture, Design and the Built Environment and School of Science and Technology, alongside Integrated System Technologies Ltd. The publication’s DOI is 10.3390/s26165066.
Each sensor type supplies a different kind of information: radar detects movement, thermal sensing estimates posture, and smart plugs register appliance use. The team’s approach is to combine those signals rather than treat any one device as a complete account of someone’s day. Lead researcher Dr. Yangang Xing said the system could also flag gradual changes in routine or mobility, not just sudden events. Those possibilities require validation beyond the mock-environment tests described in the report.
“By combining these three technologies, we can establish a complete picture of how someone is coping at home.”
— Dr. Yangang Xing, lead researcher from Nottingham Trent University
Real-Home Performance Still Unknown
The published account describes testing in a mock domestic environment, not a long-term trial with older residents living in their own homes. It does not report how accurately the system detected falls or other medical emergencies, how frequently it might generate false alarms, or how well it performs across different household conditions and individual routines. The number and characteristics of people involved in the activity testing are also not specified in the supplied report.
It is also unclear when or whether the prototype could become commercially available, what it would cost to install or maintain, and how residents would control data access and sharing. The system is presented as a means of flagging possible concerns for a human follow-up, not as a replacement for emergency services, clinical assessment or personal care.
Further Testing Before Home Use
The next evidence needed is testing in occupied homes over time, with clear reporting on recognition accuracy, missed events, false alerts and the effect of different layouts and routines. Researchers and any future developers would also need to explain how alerts are reviewed, who can access sensor data, and how residents give or withdraw consent. The supplied report does not announce a trial schedule, product launch or regulatory milestone.
For now, the confirmed development is a research prototype and a study reporting better activity-recognition performance from combined sensor inputs in a mock setting. Whether that result translates into dependable support for older people living independently remains an open question.
Key Questions
What does the home sensor prototype monitor?
It combines radar to detect movement, low-resolution thermal sensors to recognize broad body postures, and smart plugs that register appliance use. Together, those signals are intended to help identify changes in a resident’s activity patterns.
Does the system use cameras or wearable devices?
The reported prototype is designed to work without cameras or wearable devices. Its thermal sensors produce pixelated, low-resolution output intended to show posture without displaying facial features or personal identity.
Has it been shown to detect heart attacks, strokes or falls reliably?
No reliability result for detecting those medical events is provided in the report. The researchers identify them as possible events the system aims to flag, but the reported testing took place in a mock home and focused on recognizing movement and posture.
Could the system help older people stay at home longer?
That is a potential benefit described by the researchers, not an outcome established by the study. The reported results do not show that the prototype extends independent living or replaces care services.
When will the system be available?
The report gives no release date or commercial availability plan. Further real-home testing and information about costs, data handling and alert performance would be needed to assess practical use.
Source: rss
