Dynamic Memory Allocation for High Dynamic Range Image Sensors

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing high dynamic range imager systems inefficiently utilize memory, as much of the memory remains unused until the first conditional reset, leading to unnecessary storage of zeros and wastage of memory cells, which do not convey relevant information during the majority of the exposure time.

Innovation Solution

The imager system dynamically allocates memory cells based on the exposure time between the first conditional reset and row readout, reducing the number of memory rows needed and allowing for reuse of memory rows, with a memory manager configuring memory allocation and handling defects by skipping unusable rows.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If memory is allocated for all pixel rows from the beginning, then all pixel data can be stored, but much of the memory remains unused until the first conditional reset, leading to wastage of memory cells

Engineering Contradiction:
Improvenumber of memory cellsVSAvoidmemory utilization efficiency
Core Design Contradiction:
Quantity of substanceVSLoss of energy

Solution Approach 1:

The memory is segmented into multiple banks, with each bank dedicated to storing data from specific pixel rows. This segmentation allows the system to allocate memory dynamically based on which pixel rows require storage, rather than allocating the entire memory array from the beginning. As a result, memory cells are only allocated when needed, reducing wastage of memory cells during the exposure period before the first conditional reset.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The memory allocation system is made dynamic, where memory banks are activated or allocated on-demand based on the exposure timing and which pixel rows require conditional reset. The system transitions from a static, pre-allocated memory configuration to a dynamic allocation model where memory resources are assigned based on actual operational needs during the imaging process.

Inventive Principle:
Principle #15Dynamics

2Loss of energy

If memory is allocated dynamically based on exposure time, then memory usage is optimized, but the system complexity increases due to memory management requirements

Engineering Contradiction:
Improvememory utilization efficiencyVSAvoidmemory management complexity
Core Design Contradiction:
Loss of energyVSDevice complexity

Solution Approach 1:

The memory manager is designed as a multi-functional control unit that handles multiple tasks: it tracks exposure timing, determines which pixel rows require conditional reset, allocates appropriate memory banks, and manages data transfer. By consolidating these functions into a single universal controller, the system achieves dynamic memory allocation without proportionally increasing overall system complexity.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The memory management system operates autonomously by self-tracking the exposure timing and automatically determining when memory allocation is needed. The system monitors its own state and triggers memory bank activation based on predefined timing parameters, reducing the need for external control and simplifying the overall architecture.

Inventive Principle:
Principle #25Self-service

3Quantity of substance

If the number of memory rows is reduced, then memory cells are saved and yield is improved, but the system must handle defects more efficiently

Engineering Contradiction:
Improvenumber of memory cellsVSAvoiddefect handling capability
Core Design Contradiction:
Quantity of substanceVSReliability

Solution Approach 1:

The memory system implements local quality control by individually monitoring and managing each memory bank's status. When a defect is detected in a specific memory bank or pixel row, the system can isolate that local area and redirect data storage to alternative healthy memory banks. This localized approach allows the system to maintain overall reliability even with a reduced total number of memory cells, as defects do not compromise the entire memory array.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system incorporates redundancy mechanisms in advance by maintaining a pool of备用 (standby) memory banks that can be activated if defects are detected. This beforehand cushioning ensures that when defects occur, the system has pre-prepared alternative storage locations, maintaining operational reliability without requiring a larger initial memory array.

Inventive Principle:
Principle #11Beforehand cushioning (Prior cushioning)

Data Source

PatentUS9380228B2High dynamic range image sensor system and method thereof
Publication Date: 2016.06.28 HL KLEMOVE CORP
  • US9380228B2 patent drawing
  • US9380228B2 patent drawing
  • US9380228B2 patent drawing

AI summary

A high dynamic range imager system is provided that includes an imager having a pixel array, and memory in communication with the pixel array, the memory comprising a plurality of memory cells, wherein the number of memory cells is approximately the number of whole row-times of exposures between the first conditional reset and row readout.