The radar chip at the center of this story is smaller than a grain of rice. It operates at 60 gigahertz, sweeps a slice of spectrum five times wider than your home Wi-Fi, and draws about as much power as a hearing aid. It was designed to do one job: notice a thing in front of a moving car before the car hits it. Adaptive cruise control. Blind-spot alerts. The unglamorous plumbing of automotive safety.

That same chip is now being pointed at a sleeping infant’s chest to count its breaths.

Run the US grant record forward and the shift is hard to miss. Patents that pair radar with the words respiration, heartbeat, or vital sign trickled out for decades: six grants between 2014 and 2016, eleven from 2017 to 2019, thirteen from 2020 to 2022. Then, from January 2023 through May 2026, forty-six. The last three years produced more contactless-vital-sign radar patents than the entire decade before them combined. Twenty-six distinct organizations hold a piece of the recent crop, and they do not come from one industry. They come from three that have never had much reason to talk to each other: carmakers, consumer-electronics firms, and hospital-bed manufacturers.

When three fields start filing the same patent, something in the adjacent space just became buildable.

What the radar actually sees

Strip away the marketing and every one of these inventions is doing the same physically improbable thing: measuring the millimeter-scale rise and fall of a human chest from across a room, using nothing but reflected radio waves.

The mechanics are unforgiving. A breath moves the chest wall several millimeters. A heartbeat moves it a fraction of a millimeter, buried underneath the breathing motion and its echoes. Pulling the heart out of that signal is the entire game. Infineon’s 2023 grant (US 11,585,891) describes transmitting frequency chirps, catching the reflections, and fitting an ellipse to the in-phase and quadrature components of the return to recover a clean displacement signal, then throwing away any measurement whose amplitude and phase imbalance say it has been corrupted by stray motion. A companion Infineon patent wraps the whole thing in a Kalman filter to track the vital sign as the target shifts. This is signal processing borrowed wholesale from radar’s day job, repurposed to watch a rib cage instead of a bumper.

The hardest sub-problem has a tidy answer in a 2024 University of Florida grant (US 11,944,462). Its title is “Heart rate measurement using adaptive harmonics filtering,” and the trick is a comb of notch filters tuned to the respiration rate and its harmonics. Breathing, and the ringing overtones of breathing, get carved out of the spectrum; what survives is the heartbeat. The filter’s notch depths adjust on the fly as the breathing gets deeper or shallower. It is a small, specific, clever idea, and it is the kind of idea that separates a radar that detects “something alive over there” from one that reports a number a clinician would trust.

The professor who has been at this for twenty years

That Florida patent is not a fresh arrival. Its priority date runs back to April 2016, and its named inventors are Jenshan Lin, a University of Florida electrical engineer, and Linda Hayward, a physiologist in the same university’s veterinary medicine college. The pairing tells you something: this was never a pure circuits problem. You need someone who understands the radio and someone who understands what a body does.

Lin has been chasing contactless cardiopulmonary sensing since the mid-2000s, long before any of it was fashionable. His lab built a Ka-band Doppler radar that detected a heartbeat at two meters on sixteen microwatts of transmitted power, and verified an early system against an infant simulator. In review papers with titles like “Sensing of Life Activities at the Human-Microwave Frontier,” he mapped the path from bench instruments to the micro-radars now showing up in shipping products. The first US patent to put radar and vital signs in the same document was granted in 1985. The science is forty years old. What changed is not the physics. What changed is that a 60-gigahertz radar that used to cost a fortune and fill a board now costs a few dollars and fits under a fingernail, because the automotive industry ordered tens of millions of them.

Why the carmakers showed up

The forcing function has a bureaucratic name: child presence detection. Starting in 2023, Euro NCAP, the body whose star ratings move European car sales, began awarding points to vehicles that can tell a child has been left in a hot cabin. The rules are specific in a way that rules out the easy answers: the system has to keep watching for at least fifteen minutes after the engine is off, and it has to catch a sleeping infant hidden under a blanket in a footwell, where a camera sees nothing. Detecting an actual sign of life, a heartbeat or a breath, through fabric and seat foam is a job radio waves can do and optics cannot.

So the automotive supply chain piled in. Luxembourg’s IEE, a seat-sensor maker, holds grants (US 11,230,293 and US 11,644,561) on decomposing a cabin into range, Doppler, and angle, then deciding whether a flicker of motion is a passenger’s pulse or a phone buzzing on the seat. Hyundai Mobis, GM, and the chipmakers Infineon, Qualcomm, and Intel are all in the recent record. The Israeli firm Vayyar, which raised a $108 million round at a valuation north of a billion dollars from Koch’s venture arm, sells a single radar-on-chip that does child-presence detection and seat-belt reminders at once. Vayyar’s own origin is the tidy irony of this whole story: the company started building radio-wave sensors to scan for breast cancer, pivoted the technology into cars for safety regulation, and is now watching it circle back toward health.

The same chip, three buildings over

Here is the part an R&D director should sit with. The automotive mandate is paying to industrialize a sensor that two adjacent industries want for entirely different reasons, and they get to ride the same cost curve down.

In the consumer aisle, Google’s second-generation Nest Hub put a 60-gigahertz Soli radar on the nightstand and used it for sleep sensing. Google’s 2024 grant (US 12,127,825) describes filtering out static clutter, running a spectral analysis of the room, and reporting heart and breathing rates only when the scene says monitoring is allowed. In published work, the system tracked sleeping users’ heart rate to within about 1.7 beats per minute, with no watch, no chest strap, nothing touching the body.

In the hospital, Hill-Rom, now part of Baxter, holds a 2024 grant (US 11,877,844) that mounts radar sensors in the frame of a patient bed, above the mattress, to monitor breathing rate, tidal volume, and chest expansion, and then feeds that reading back to the bed’s air bladders that rotate the patient and loosen lung secretions. The monitor and the therapy close a loop, with no electrodes to fall off and no wires to tangle.

A camera-and-electrode patient monitor competes against other patient-monitor companies. A radar one competes against nobody and everybody, because the silicon inside it is being amortized by Toyota-scale order volumes, not by hospital procurement. The contactless monitor that needs no adhesive, sees through a blanket, and protects patient privacy because it captures motion rather than images has, for the first time, a bill of materials set by the car industry. That is the convergence worth watching. Forty years after the first patent, the chest-reading radar finally got cheap enough to be everywhere, and it got cheap because of a regulation about hot cars.

Method note: Counts come from roughly 9.3 million US utility patents granted by the USPTO, current through late May 2026, filtered to grants whose text pairs “radar” with “respiration,” “heartbeat,” or “vital sign.” The three-year and decade windows are by grant publication date; the 26-organization figure counts distinct assignees, including variant spellings, on grants issued since January 2023. Patent contents are drawn from the granted claims and abstracts cited inline. The earliest matching US grant dates to 1985. External figures on Euro NCAP child-presence rules, Vayyar’s funding, the Nest Hub Soli radar, and Jenshan Lin’s research come from Euro NCAP, Times of Israel, Nature Scientific Reports, and University of Florida sources, respectively. A keyword match is a starting point, not proof of a shared mechanism; the patents featured here were read individually to confirm they describe the same physical measurement.