Product Marketing Manager, Novelda

Kjetil Meisal's profile picture

Tackling sleep monitoring with non-contact sensor technology

Respiration rate is an important vital sign that can provide insight into a person’s general state of health and quality of sleep. Breathing rate and breathing patterns are also considered good indicators of underlying medical conditions. While current technologies in today’s market promise accurate and dependable monitoring, the majority are less than ideal - either they are not accurate and require continual adjustment or are uncomfortable for subjects to use.

This article explores the various established methods for measuring and analyzing respiration to understand their limitations when applied to the requirements for daily health care and fitness tracking. Overcoming these challenges is not easy but ultra-wideband (UWB) impulse radar can serve as the basis of sensors such as Novelda’s XeThru -- enabling sensors to achieve an effective range up to 2.8m and capable of ‘seeing through’ obstacles, such as clothing, while also being low cost and easy to use.

Sleep monitoring for healthcare

Breathing rate is the number of breaths a person takes per minute, and is best measured when a person is at rest. The rate may increase with fever, illness, and with other medical conditions. The most common method for measuring breathing rate is by physical assessment, observing a person’s chest and counting the number of breaths during one minute. Depth of breathing can be determined with a spirometer, a device that measures lung function based on the volume of air breathed in and out. In its simplest form doctors use a spirometer to detect conditions like asthma. Breathing rate on its own provides limited information, but breathing patterns - measuring rate, amplitude and other characteristics - provide far more valuable information, which can be used for medical diagnostics as well as the evaluation of sleep quality.

Most of the sleep monitoring technologies that exist today are invasive and require the subject to be connected to the measuring equipment. This is certainly true for the spirometer described above but even the simple electromechanical measurement of breathing rate will typically require an elastic strap to be tightened around the subject’s chest. Alternative acoustic techniques require a device to be connected to a subject’s neck, while capacitive techniques require a special mattress or sensing unit to be installed in the bed or on a subject’s body. These methods are mainly used in professional monitoring situations and provide accurate breathing pattern data. However, a fundamental challenge still remains, which is the physical connection of a sensor to the body, causing stress to the monitored subject. Any associated discomfort will in turn affect the subject’s breathing and potentially invalidate the data.

Consumer sleep monitoring

Until recently, even simple respiration tests have been confined to the doctor’s office; more extensive sleep monitoring has been the preserve of clinics and hospitals and then only for patients referred after preliminary diagnosis of a medical condition. The advent of smartphones and similar high-tech gadgetry, including devices such as heart rate and blood-sugar monitors, coupled with individuals who have become much more conscious of and concerned about their health and well-being has raised expectations of what should be possible away from a medical environment. Consequently the market has been flooded with healthcare and fitness monitoring devices introduced to provide consumers with the means to track their physical activity and manage their personal health. These products use technology similar to their medically qualified counterparts, while further refinements of medical designs address a trend towards providing complete health assessments by monitoring sleep quality and breathing patterns while at rest.

With an appreciation of the limitations of current solutions, and understanding that consumers expect health and fitness products to be comfortable, safe and easy to use, it is clear that a truly unobtrusive sleep monitor needs to meet some pretty demanding design criteria. It should be able to accurately measure and record breathing from a distance and its placement shouldn’t be critical as long as the person is within a reasonable detection zone. The person’s orientation within that zone shouldn’t matter and the monitor should operate reliably despite reasonable obstructions e.g. through a duvet.

The technology should be suitable for use with people of all ages and sizes but should also be non-intrusive and preserve the privacy of the person being monitored – this typically precludes monitoring using any form of video surveillance. It should also be possible to monitor more than one person concurrently. Further considerations in terms of the technical performance of a sensor that delivers on these goals are that it should be: safe to use, low cost, low power and ideally, for ease of integration into an end-equipment design, should also be small in size and provide a digital interface.

Of the many products in today’s market capable of measuring breathing rates, the majority take the form of fitness trackers involving body-worn contact sensors built into wristbands or chest straps. However, sleeping with a monitoring device connected to the body is uncomfortable and battery-operated units also require regular charging. More advanced sleep monitors, which provide a means of non-contact sensing, avoid these issues and provide quality sleep monitoring with greater comfort. These use technologies such as capacitive sensing, where sensor units are placed under bed sheets, or continuous wave (CW) radar solutions that monitor breathing movement from the bedside. While they don’t require direct contact, capacitive solutions generally need to operate in close proximity to a subject and their reliability can be affected by temperature and humidity variations. CW radar can operate more remotely but cannot distinguish between movement due to respiration and other body movements (as discussed further below).

Radar technology would seem to be the ideal choice for systems that are non-intrusive and can monitor breathing from a distance, especially as its signals can pass through materials such as clothing or bedding. Even then there are limitations. While CW Doppler radar is highly sensitive to movement and can easily detect the frequency of repetitive movements like breathing, it only provides phase information and cannot measure absolute distance. Not being able to measure absolute distance means a CW-based radar system cannot distinguish between other body movements, such as hands or feet, and chest movement. Hence a CW system is less capable of resolving actual chest movement, which is necessary to provide reliable data throughout the night.

What is needed is an approach that can accurately measure distance, sufficient to differentiate between shallow and normal breathing, and precisely track a subject’s breathing patterns from their chest movement (Figure 1). Achieving this with a low power, low cost solution is the real challenge.


Figure 1. Reliable detection of respiration from chest movement requires a radar system that is capable of absolute distance measurement.

A Solution using UWB impulse radar

In pursuing a solution to address these criteria, Novelda concluded that an electromagnetic sensor using the principles of radar should be able to meet all the technical requirements provided other concerns could be overcome. The previously noted limitation of CW radar, not being able to measure absolute distance, does not apply when using ultra-wideband (UWB) impulse radar techniques, which, by emitting and sampling signal pulses, can achieve highly accurate distance measurements determined by the time differences between transmitted and received pulses.

Furthermore, by using what is essentially a spread spectrum approach and employing digital signal processing (DSP) to recover the return signal, UWB radar can operate at much lower power levels than conventional radar. This overcomes the potential consumer concern of not wanting a high-power radar sitting on their bedside table - this technique allows operation at power levels less than 1/1000th the power of a Bluetooth headset. The spread spectrum nature of UWB also means it can coexist with other RF systems without causing, or being affected by, interference. For example, CW Doppler radar operates at much higher power levels, which can interfere with WiFi and radio signals. Conversely while UWB operates at very low power levels, DSP techniques can reliably extract the signal from the noise in much the same way as ADSL delivers broadband Internet connectivity from ordinary phone lines.

Radar is considered to be a complex and expensive technology, typically deployed in high-end market segments. Certainly this has been true in the past for traditional systems constructed from discrete components and using costly ceramic substrates. In most markets, especially consumer, the widespread adoption of integrated circuit technology has enabled products to be produced in volume and offered at affordable prices. The high level of integration possible today enables complete system-on-chip solutions with consequent benefits in size and power consumption (Figure 2).


Figure 2. Novelda’s UWB impulse radar IC integrates a complete transmitter and receiver circuit with all the necessary timing and signal processing elements.

So finally …

However you label it, health, fitness and wellness are of increasing concern to most people as evidenced by the plethora of electronic gadgetry available today. Many fitness tracker and similar devices measure activity, such as distance, speed or steps taken, but their ability to measure the body’s response is often limited to heart rate. Breathing rate and particularly breathing patterns are further useful performance indicators but, as we have seen, their measurement is more complex and potentially more intrusive.

In investigating alternatives to respiration tracking techniques that have traditionally been confined to medical practices, Novelda has concluded that UWB impulse radar overcomes both the technical and useability challenges to provide a solution for measuring and analyzing breathing rates and patterns, all without contact and without being blocked by obstructions such as clothing or bedding. Indeed, this sensing technology enables compact respiration tracking designs such as Novelda's own XeThru sensor module.