Neural Network Signal Generator for Fast Regulatory Test Calibration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Generating test signals for electrical and optical devices that meet complex, non-linear regulatory requirements is a time-consuming iterative process due to the complex relationship between signal generator inputs and desired characteristics, especially when multiple test signals are needed.
Innovation Solution
A test apparatus and method utilizing an artificial neural network (ANN) to generate test signals by training the ANN with input parameters and calculated characteristics, allowing the signal generator to produce signals that meet specific desired parameters efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If an iterative process with signal analyzer is used to adjust inputs and measure output characteristics, then the test signal meets regulatory requirements, but the process is time consuming
Solution Approach 1:
The patent pre-calculates and stores the complex non-linear relationship between signal generator inputs and output characteristics in a lookup table before actual testing. This preliminary computation eliminates the need for iterative adjustments during signal generation, allowing direct retrieval of input parameters that produce desired output characteristics, thus dramatically reducing signal generation time while maintaining accuracy
Solution Approach 2:
The patent creates a simplified mathematical model or lookup table that copies the essential characteristics of the complex non-linear relationship between inputs and outputs. Instead of performing complex real-time calculations or iterative measurements, the system uses this pre-created representation to directly determine input parameters, replacing the time-consuming iterative process with a simple query operation
2Adaptability or versatility
If multiple component signal generators are combined to generate test signals, then complex signal characteristics can be achieved, but the relationship between inputs and desired characteristics is complex and non-analytic
Solution Approach 1:
The patent introduces an intermediary lookup table that mediates between the multiple component signal generators and the desired output characteristics. This lookup table pre-computes and stores the complex non-linear relationships, serving as a bridge that translates desired output characteristics into appropriate input parameters for each component generator without requiring real-time complex calculations
Solution Approach 2:
The patent performs preliminary computation to establish the complex relationships between multiple component generator inputs and final output characteristics before actual signal generation. By pre-calculating and storing these relationships in a lookup table, the system avoids the complexity of real-time non-analytic calculations while maintaining the ability to generate diverse test signals with precise characteristics
3Reliability
If iterative adjustment and measurement is performed for each test signal, then regulatory requirements are met, but the process is not efficient when large number of test signals are required
Solution Approach 1:
The patent performs preliminary computation to pre-determine the input parameters for multiple different test signals by storing the complex input-output relationships in a lookup table. When multiple test signals are required, the system can directly retrieve pre-calculated parameters from the lookup table without performing iterative adjustments and measurements for each signal, thereby maintaining regulatory compliance while dramatically improving throughput
Solution Approach 2:
The patent creates a comprehensive lookup table that copies and stores the relationships for generating multiple different test signal types. This pre-computed reference enables rapid generation of large numbers of test signals by simple table lookup, replacing the inefficient iterative process while ensuring each signal meets regulatory requirements through pre-validated parameters
Data Source
AI summary
A test apparatus and a method for operating a data processing system to generate a test signal for testing a DUT are disclosed. The apparatus includes a signal generator, artificial neural network, and controller. The signal generator generates a test signal determined by a plurality of signal generator input parameters, X, that are coupled thereto. The test signal is characterized by a plurality of calculated parameters, Y. The artificial neural network has the calculated parameters as inputs and a plurality of outputs connected to the plurality of signal generator inputs. The controller receives desired values for the calculated parameters and couples those desired values to the neural network inputs, thereby causing the test signal generator to generate a test signal having the desired values for the calculated parameters.