Transistor Error Adaptation via Least Mean Square Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for compensating for variations in transistor characteristics in displays are inefficient, leading to suboptimal performance due to poor determination of compensation parameters.
Innovation Solution
A method and system that measure error values between target and actual currents driven by transistors, adjusting compensation parameters using a modified least mean square error adaptation algorithm, applying voltage adjustments to the transistor gate based on these parameters to improve transistor performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional compensation methods are used to determine compensation parameters, then the compensation process can be implemented, but the determination of compensation parameters performs poorly leading to suboptimal transistor performance
Solution Approach 1:
The patent implements a feedback mechanism where the error value (difference between target and actual currents) is measured and used to iteratively adjust compensation parameters. The measured error feeds back into the parameter adjustment process, allowing the system to converge on optimal compensation values that improve transistor performance while achieving precise parameter determination.
Solution Approach 2:
The patent transforms the static compensation parameter determination into a dynamic iterative process. Compensation parameters are not fixed but are continuously adjusted based on measured error values through multiple iterations, allowing the system to adapt and converge on optimal values rather than relying on fixed traditional methods.
2Measurement precision
If compensation parameters are adjusted iteratively to improve accuracy, then better transistor performance is achieved, but the compensation process requires more time and computational resources
Solution Approach 1:
The patent applies partial action by performing a limited number of iterations (e.g., 3-5 iterations) rather than continuing until perfect convergence. This provides sufficiently accurate compensation parameters without requiring excessive time and computational resources, achieving an optimal balance between precision and efficiency.
Solution Approach 2:
The patent performs preliminary measurements of transistor characteristics (such as threshold voltage and mobility) before the iterative compensation process. These preliminary actions provide initial estimates that accelerate convergence, reducing the number of iterations needed and thereby reducing the overall time required while maintaining high accuracy.
Data Source
AI summary
A method for compensating for characteristics of a transistor. In some embodiments, the method includes: measuring an error value, the error value being a difference between: a target current and a current driven by the transistor when the transistor is controlled by a compensated control signal based on an input control signal; adding to a first compensation parameter a first adjustment; adding to a second compensation parameter a second adjustment; and applying to a gate of the transistor a voltage equal to the sum of: the second compensation parameter, and the product of: the first compensation parameter, and an uncompensated drive voltage.