Real-Time Variance Algorithm for Low-Power Gyroscope Monitoring

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for calculating Allan variance and Allan deviation are processor and memory intensive, making real-time variance analysis challenging in low-power, miniaturized devices like gyroscopes and atomic clocks, which limits their implementation due to high overhead and the need for expensive test instruments or post-processing.

Innovation Solution

A real-time algorithm that performs variance calculations using a data processing unit with a processor and memory device, where only necessary data points are recalculated, and previous values are reused, reducing processing and memory overhead by using bitwise logic and hierarchical calculations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If full Allan variance calculations are performed for all data points, then measurement precision is improved, but processing time and computational overhead increase exponentially

Engineering Contradiction:

Improvevariance calculation precisionVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent segments the full variance calculation into discrete exponentially increasing averaging times (tau values) spaced on a logarithmic x-axis. Instead of calculating all N/2 possible averaging values, the method selects specific tau points (1, 2, 4, 8, etc. - octaves or decades) to calculate variance, reducing the number of calculations while maintaining measurement precision at critical time scales.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent performs partial calculations by computing variance only at selected discrete tau values rather than all possible averaging times. This partial action approach calculates sufficient statistics at key points (octaves/decades) to characterize device stability without the excessive computational burden of complete variance analysis across all data point combinations.

Inventive Principle:

Principle #16Partial or excessive action

2Productivity

If variance calculations are performed regularly in real-time, then monitoring capability is improved, but processor and memory overhead increase

Engineering Contradiction:

Improvereal-time monitoring capabilityVSAvoidprocessor and memory overhead

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent performs preliminary calculations by pre-selecting the discrete tau values (octaves or decades) that will be calculated. The data array is initialized with these predetermined tau points, and as new data arrives, only the necessary rows corresponding to selected tau values are recalculated. This preliminary structuring enables real-time updates without recalculating all possible variance values.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent segments the variance calculation into discrete tau values stored in a data array where each row corresponds to a specific averaging time. This segmentation allows the system to update only specific rows (tau values) in real-time rather than recalculating the entire variance spectrum, reducing processor and memory overhead while maintaining monitoring capability.

Inventive Principle:

Principle #1Segmentation

3Loss of time

If discrete exponentially increasing averaging times are used, then processing time is reduced, but measurement coverage is limited

Engineering Contradiction:

Improvecalculation timeVSAvoidvariance measurement coverage

Core Design Contradiction:

Loss of timeVSAdaptability or versatility

Solution Approach 1:

The patent changes the parameter of averaging time to follow an exponential progression (octaves or decades) rather than linear steps. This parameter transformation allows the system to cover a wide range of time scales with fewer calculation points, reducing processing time while maintaining adaptability across different device stability characteristics through the logarithmic spacing of tau values.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS10445405B2Method of performing variance calculations in real-time for in-situ instant variance measurements for low power applications

Publication Date: 2019.10.15 HONEYWELL INTERNATIONAL INC

AI summary

A system for performing real-time variance analysis of data measurements comprises a data source, and a data processing unit in operative communication with the data source. The data processing unit is configured to receive data measurement values output from the data source and comprises a processer, and a memory device that includes instructions executable by the processor for a real-time algorithm that performs variance calculations for the data measurement values. The real-time algorithm performs a process comprising: initializing a data array for tracking a variance of individual data measurement values received from the data source; receiving a data measurement value from the data source; determining the number of averaging times to be recalculated; and repeating the process when the next data measurement value is received from the data source. The data processing unit is configured to output real-time variance measurements based on the variance calculations for the data measurement values.