patent.fyi
Applications  /  US 2026/0354752
United States Patent Application Publication

US 2026/0354752 A1

Published
Nov 12, 2026
Filed
Dec 29, 2025
Earliest date on the face
Jul 6, 2023
Classification
A61B 5/4809
First published here
Nov 19, 2026

Generated automatically. Not reviewed by a person. Not legal advice or an opinion.

Section 1 · machine-generated

What if?

Questions about how individual parts of claim 1 might be changed. A question may describe something different from what the claim language recites, or a version of it; no view is offered on which.

No entries.

Section 2 · machine-generated

Variations

Further arrangements of the same subject matter, described here as of the publication date at the foot of this page.

Other field of use

Adaptive sleep scheduling in crew cabins

A wall-mounted bedside unit in a vessel crew cabin contains a radar module aimed at the bunk, a touchscreen, a dimmable lamp, and a speaker. Each crew member selects a profile and enters the start of the next duty period. The controller estimates sleep onset and wake time from breathing and movement, then compares those times with the profile's stored targets and recent history. When the difference exceeds a stored threshold, the controller shifts the target bedtime or wake time within the available off-duty period and schedules the lamp and speaker accordingly.

Component substitution

Under-mattress piezoelectric sensing for schedule adjustment

A piezoelectric film strip beneath the mattress at torso level replaces the radar module, and a shielded lead connects the strip to a high-impedance amplifier in the bedside housing. The controller extracts respiratory motion from the low-frequency signal and flags larger transients as body movement. It estimates sleep and wake intervals from changes in respiratory regularity and the frequency of movement, then compares the resulting sleep duration with the stored target and recent nights. A threshold crossing updates the schedule and shifts the timing of the bedside lamp and speaker.

Parameter or scale variation

Bounded schedule shifts across seven nights

A bedside radar module supplies breathing and movement measurements in 30-second epochs, and the controller stores seven completed nights in a rolling buffer. Each morning, the controller compares the median estimated sleep-onset time with the current target bedtime on a time axis that begins at local noon. When the difference exceeds 20 minutes, the controller moves the target bedtime 10 minutes toward the median, subject to a minimum scheduled sleep window of seven hours and a fixed wake time. The controller starts a 30-minute lamp dimming ramp before the revised bedtime and a 15-minute brightening ramp before wake time.

Independent sleep schedules for two sleepers

Two radar modules mount on the headboard, one above each mattress half, and point down toward their respective sleeping positions. The controller alternates 50-millisecond measurement slots between the modules and applies a separate distance gate to each return stream before extracting breathing and motion features. It stores two sleep histories and compares each sleeper's estimated sleep onset with that sleeper's target and recent baseline. A threshold crossing updates the corresponding schedule and shifts the timing of that side's baffled reading lamp and pillow speaker.

Combination with existing systems

Bedside scheduling with motorized blackout shades

The bedside controller connects to a room automation hub over a local network, and the hub controls a motorized blackout shade with position feedback. The controller derives sleep onset and wake time from radar measurements and updates the stored schedule when the difference from the target exceeds a history-based threshold. It sends revised shade commands to the hub to close the shade at bedtime and begin opening it at the target wake time. The bedside lamp starts its wake ramp at the scheduled time if the hub reports that the shade remains closed.

Automated or data-driven variant

Model-based sleep and wake interval estimation

The controller converts radar measurements into 30-second epochs and computes respiratory rate, respiratory interval variability, and gross movement energy for each epoch. A gradient-boosted tree model trained on labeled sleep and wake epochs assigns a probability of sleep, and the controller applies a five-epoch median filter to the probability sequence. The controller counts epochs above a stored probability threshold to estimate total sleep time and compares that estimate with the nominal duration and the median of the previous seven nights. When the difference from the nominal duration exceeds a threshold derived from the recent history, the controller updates the target bedtime and shifts the timing of the lighting or audio sequence.

Cite aspatent.fyi, US 2026/0354752 A1, claim 1, first published 2026-11-19T03:31:31-05:00, https://patent.fyi/US20260354752/, SHA-256 674ecb3c5412b6d5ef5f4c14252b015b05c0b6f8a6044a4ba2c2289254fafb21
First published
Permanent addresshttps://patent.fyi/US20260354752/
Content fingerprint (SHA-256)674ecb3c5412b6d5ef5f4c14252b015b05c0b6f8a6044a4ba2c2289254fafb21
Methodv2026-11-18.2 · automated, not reviewed by a person
Correctionscorrections@patent.fyi