Image Sensor ADC Resolution Switching With Boundary Error Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Analog-to-digital conversion circuits in CMOS image sensors face non-linear errors due to offset voltages in comparators, leading to precision issues when changing resolution based on input voltage levels.
Innovation Solution
Incorporating a signal correction unit that adjusts the digital signal by using boundary values to correct for non-linear errors, allowing the analog-to-digital conversion circuit to change resolution based on input voltage levels without discontinuity, using equations to generate corrected digital signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If resolution is increased to improve conversion precision, then measurement precision is improved, but power consumption and conversion time increase
Solution Approach 1:
The patent applies dynamic resolution adjustment by changing the resolution of the analog-to-digital converter based on the input voltage level. When the input voltage level is high, a lower resolution is used to reduce conversion time and power consumption. When the input voltage level is low, a higher resolution is used to maintain conversion precision. This dynamic adaptation resolves the contradiction between precision and productivity.
2Productivity
If resolution is decreased to reduce power consumption and conversion time, then productivity is improved, but measurement precision deteriorates
Solution Approach 1:
The patent applies local quality by using different resolution levels for different input voltage ranges. Instead of using a uniform high resolution for all input levels, the system uses high resolution only when necessary (low input voltage) and low resolution when sufficient (high input voltage). This localized adaptation of quality resolves the contradiction between productivity and measurement precision.
3Productivity
If different resolutions are applied based on input voltage level to improve productivity, then productivity is improved, but non-linear errors increase at resolution change boundaries
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing compensation values for the non-linear errors that occur at resolution change boundaries. These compensation values are determined in advance and stored in a lookup table. When a resolution change occurs, the appropriate compensation value is applied to correct the digital signal, preventing the non-linear error from affecting the output. This preliminary preparation resolves the contradiction between productivity improvement and measurement precision maintenance.
4Device complexity
If comparator offset voltage is present to simplify circuit design, then device complexity is reduced, but non-linear errors increase at resolution boundaries
Solution Approach 1:
The patent converts the harmful effect of comparator offset voltage into a beneficial compensation mechanism. Instead of trying to eliminate the offset voltage through complex circuit design, the system accepts its presence and compensates for it using pre-calculated compensation values stored in a lookup table. The offset voltage causes predictable non-linear errors at resolution boundaries, and these errors are corrected by applying the stored compensation values to the digital output signals. This approach resolves the contradiction between device complexity and measurement precision by transforming a circuit design problem into a software compensation problem.
Data Source
AI summary
An analog-to-digital conversion circuit includes an analog-to-digital conversion unit configured to analog-to-digital convert an input voltage and generate a digital signal, a resolution control unit configured to: set a resolution of the analog-to-digital conversion unit to N (N is the natural number) bits, in a case where the input voltage is smaller than a first voltage, and set the resolution of the analog-to-digital conversion unit to N−M (1≦M<N, M is the natural number) bits in a case where the input voltage is larger than the first voltage, and a signal correction unit configured to: generate a corrected digital signal based on a boundary value and the digital signal, in a case where a value of the digital signal is larger than the boundary value and the input voltage is smaller than the first voltage.