Sensor calibration useful for smart beds with multiple sensors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sensor systems for smartbeds face challenges in precise time synchronization across multiple sensors without a single master clock, often relying on network-time-services which offer limited accuracy and require active network connections, making them unsuitable for low-bandwidth situations and prone to network congestion.
Innovation Solution
A system that synchronizes data from multiple sensors by converting sensor signals into modulated signals, combining them into an aggregate signal in the frequency domain, and demodulating to extract synchronized signals, eliminating the need for external timekeeping services and enabling precise time-based analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If network-time-services are used for synchronization, then time synchronization can be achieved, but network connectivity is required and accuracy is limited
Solution Approach 1:
The patent introduces a downstream sampling system as an intermediary that receives signals from multiple sensors and performs centralized sampling. This sampling system acts as a mediator that eliminates the need for each sensor to maintain its own clock synchronization, thereby removing network-time-service dependencies while achieving precise synchronization through the sampling mechanism.
Solution Approach 2:
The patent replaces the network-based time synchronization mechanism (software/service-based) with a hardware-based sampling system that physically samples all sensor signals at predetermined intervals. This substitution eliminates network connectivity requirements and achieves deterministic synchronization through the physical sampling process.
2Adaptability or versatility
If multiple sensors with individual clocks are used, then sensing coverage is improved, but time synchronization becomes difficult without a master clock
Solution Approach 1:
The patent merges the timing function of multiple individual sensor clocks into a single centralized sampling system. Instead of each sensor operating with its own clock, the downstream sampling system consolidates the timing function, sampling all sensor signals simultaneously at predetermined intervals. This merging eliminates synchronization difficulties while preserving the ability to use multiple sensors.
Solution Approach 2:
The downstream sampling system serves as a universal timing mechanism for all sensors in the system. Rather than requiring separate clock synchronization mechanisms for each sensor, the single sampling system provides timing functionality for the entire sensor array, enabling precise synchronization across diverse sensor types without adding complexity.
3Measurement precision
If network-time-service is used for synchronization, then time alignment can be achieved, but network congestion occurs in low-bandwidth situations
Solution Approach 1:
The patent extracts the time synchronization function from the network infrastructure and implements it locally within the sensor system through the downstream sampling system. By removing the dependency on network-time-services, the system eliminates network bandwidth consumption associated with time synchronization while maintaining precise time alignment through local sampling.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach allows for accurate synchronization and calibration of sensor data without network-time-services, improving the accuracy and reliability of sensor data analysis in smartbed systems, especially in low-bandwidth environments.
Implementation Method 1
convert each of the sensor-signals into corresponding modulated-signals
Data Source
AI summary
A smartbed system can have a plurality of sensors, each sensor configured to: sense at least one phenomenon of a sleep session; generate a sensor-signal based on the sensing of the at least one phenomenon; and transmit, to a computer system, the sensor-signal. The computer system can be configured to: receive each of the sensor-signals; convert each of the sensor-signals into corresponding modulated-signals; generate a combined-signal from the modulated-signals; identify an event in the combined-signal comprising determining a common-timestamp for the event in the combined-signal; and create, for each sensor-signal, a synchronized-signal by demodulating the combined-signal.