US 9729693

Patent Summary: US 9,729,693

Title: Determining measurement confidence for data collected from sensors of a wearable device

Assignee: Zepp Inc.

Inventors: Joseph Munaretto

Filing Date: November 8, 2016

Issue Date: August 8, 2017

Abstract:
Systems and methods for determining measurement confidence for data collected from sensors of a wearable device are disclosed. In one implementation, a confidence measurement that a wearable device is worn by a user can be determined by checking if the device is in motion, comparing sample voltages from a light emitter to thresholds to determine the surface it is reflecting on, calculating a signal quality metric from a pulse oximeter, and comparing this metric to thresholds of typical biometric data. Other methods include frequency transforming sensor data, performing probabilistic modeling on the transformed data, and determining a confidence measurement using a signal quality estimation based on the modeled data.

Plain-Language Overview of Independent Claims:

This patent has three independent claims (1, 11, and 16) that are substantially similar, each describing a method, an apparatus (the wearable device itself), and a system (the wearable device plus an analysis component) for determining with confidence that a wearable device is actually being worn by a user.

Claim 1 (Method): This claim outlines a step-by-step method to confirm a wearable device is on a user. First, it checks for a low voltage reading from a light sensor when it's off, to ensure it's not pointing at a bright light like the sun. Second, it checks for a higher voltage reading when the light sensor is on, to confirm it's against a dark surface like skin and not in a dark room. Third, it calculates a "signal quality metric" from the sensor data over a short period. Finally, if this metric is above a certain threshold, the device concludes it is being worn by a user.

Claim 11 (Apparatus): This claim describes the physical wearable device itself, which includes a body to be worn, a sensor, a memory, and a processor. The processor is programmed to execute the same multi-step verification method outlined in Claim 1: checking voltages with the light sensor off and on, calculating a signal quality metric, and confirming the device is worn if the metric is high enough.

Claim 16 (System): This claim describes a broader system that includes the wearable device (with its body and sensor) and a separate "analysis component" that contains a memory and processor. This analysis component performs the same series of checks as in the other claims: it analyzes the two voltage readings and the signal quality metric from the wearable device's sensor to determine if the device is being worn by a user.

Litigation Status:

As of April 2026, US Patent 9,729,693 is the subject of litigation. It is listed as an exhibit in a patent infringement lawsuit filed on April 21, 2026, in the U.S. District Court for the Eastern District of Texas: Zepp Inc. et al v. Oura Health Oy (Case 2:2026cv00316). No corresponding dockets were found for the Court of Appeals for the Federal Circuit (CAFC) for 2026.