Current Mirror Filter Circuit for Image Sensor Noise Reduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional filter circuits for image sensor modules face challenges in setting a wide range of filter constants without relying on input signal driving capacity and struggle with integration due to device scale restrictions, while also being prone to errors and malfunctioning with continuous pulse noise.

Innovation Solution

A filter circuit utilizing first and second current mirror circuits with transistors, a capacitor, and input transistors to control charging and discharging, allowing for adjustable filter constants and effective noise elimination without increasing device scale.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Volume of moving object

If RC filters are integrated into an image sensor IC, then the device scale is reduced, but the filter constant cannot be set in a wide range due to chip area restrictions

Engineering Contradiction:
Improvedevice scaleVSAvoidfilter constant range
Core Design Contradiction:
Volume of moving objectVSAdaptability or versatility

Solution Approach 1:

The patent changes the fundamental parameters of the filter circuit by replacing the conventional RC configuration with a current mirror-based active filter. This allows the filter constant to be controlled by current values rather than resistance and capacitance products, enabling a wide range of filter constants to be achieved within the same chip area. The filter constant is determined by the ratio of mirror currents and capacitor values, providing flexibility without increasing device scale.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent substitutes the passive mechanical RC filter system with an active electronic system using current mirror circuits. This replacement allows for dynamic control of the filter constant through current regulation, eliminating the need for large physical RC components and enabling precise filter constant setting in a compact integrated circuit format.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If conventional RC filters are used, then extraneous noise is eliminated, but the device scale increases due to the need for discrete components

Engineering Contradiction:
Improvenoise eliminationVSAvoiddevice scale
Core Design Contradiction:
ReliabilityVSVolume of moving object

Solution Approach 1:

The patent merges the filter functionality with the existing image sensor IC circuitry by using current mirror circuits that are already present in the IC. This integration eliminates the need for separate discrete RC components, reducing device scale while maintaining effective noise elimination through the active filtering mechanism.

Inventive Principle:
Principle #5Merging (Combining)

3Adaptability or versatility

If charge and discharge circuits are switched based on input signal level, then filter constant can be controlled, but the input signal driving capacity must be sufficiently high

Engineering Contradiction:
Improvefilter constant controlVSAvoidinput signal driving capacity
Core Design Contradiction:
Adaptability or versatilityVSPower

Solution Approach 1:

The patent introduces current mirror circuits as intermediary elements that decouple the input signal from the charge/discharge control function. The current mirrors act as buffers, allowing the input signal to control the filter operation without requiring high driving capacity. The mirror currents provide the necessary drive strength while the input signal only needs to control the switching of transistors.

Inventive Principle:
Principle #24Intermediary (Mediator)

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

Enables accurate setting of filter constants across a wide range, reduces device scale, and effectively eliminates extraneous noise, enhancing image sensor accuracy and integration capabilities.

Implementation Method 1

a first current mirror circuit which is built with a plurality of transistors and which passes a first mirror current according to a constant current

Methodology Applied
Scientific EffectCurrent mirror effect:

Implementation Method 2

a second current mirror circuit which is built with a plurality of transistors and which, operating in inverted synchronism with the first current mirror circuit, passes a second mirror current according to a constant current

Methodology Applied
Scientific EffectCurrent mirror effect:

Implementation Method 3

a capacitor which is charged by the first mirror current and which is discharged by the second mirror current

Methodology Applied
Scientific EffectCapacitance: Capacitance

Data Source

PatentUS7773271B2Filter circuit, and image sensor, image sensor module, and image reading apparatus provided therewith
Publication Date: 2010.08.10 ROHM CO LTD
  • US7773271B2 patent drawing
  • US7773271B2 patent drawing
  • US7773271B2 patent drawing

AI summary

A filter circuit has: a first and a second current mirror circuit that are each built with a plurality of transistors and that respectively pass a first and a second mirror current according to a constant current; an input terminal via which an input pulse signal is fed in; a first transistor that, in response to the signal fed in via the input terminal and by using a supply voltage, makes the first current mirror circuit operate; a second transistor that, operating in inverted synchronism with the first transistor, in response to the signal fed in via the input terminal and by using a ground voltage, makes the second current mirror circuit operate; a capacitor that is charged by the first mirror current and that is discharged by the second mirror current; and an output terminal via which a voltage at one end of the capacitor is fed out. Thus, the filter circuit permits its filter constant to be set accurately in a wide range without relying on the driving capacity of the input signal, and can be integrated without an undue increase in device scale.