Memristive Crossbar for Wavelet Transform Data Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Resource-constrained devices, particularly in the Internet of Things (IoT), face challenges with data compression due to high computational intensity, energy consumption, and inefficiencies in conventional computing architectures, leading to bottlenecks and increased complexity.
Innovation Solution
Implementing a memristive crossbar that sets conductances to act as coefficients for wavelet transformation matrices, enabling efficient two-dimensional discrete wavelet transform (DWT) for data compression, thereby reducing energy usage and physical space requirements while alleviating the von Neumann bottleneck.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional computing architectures are used for data compression, then data compression can be performed, but energy consumption increases and processing speed decreases due to the von Neumann bottleneck
Solution Approach 1:
The patent merges memory and computation into a single integrated structure using a memristive crossbar array. The crossbar simultaneously stores data (in memristor conductance states) and performs computation (via parallel matrix operations), eliminating the separation between memory and processor that causes the von Neumann bottleneck. This integration allows data compression operations to be performed directly on stored data without repeated data movement between separate memory and computation units.
Solution Approach 2:
The patent replaces conventional digital computing mechanics with analog computing mechanics. Instead of using discrete digital circuits to perform sequential arithmetic operations, the system uses analog voltage and current signals flowing through the memristive crossbar to perform parallel matrix multiplications. This analog approach naturally computes wavelet transforms and other compression operations in a single pass, dramatically reducing computation time and energy consumption compared to sequential digital processing.
2Productivity
If conventional computing architectures are used for data compression, then compression can be achieved, but device complexity increases due to memory hierarchy and multi-core requirements
Solution Approach 1:
The patent merges memory and computation into a single integrated structure using a memristive crossbar array. The crossbar simultaneously stores data (in memristor conductance states) and performs computation (via parallel matrix operations), eliminating the separation between memory and processor that causes the von Neumann bottleneck. This integration allows data compression operations to be performed directly on stored data without repeated data movement between separate memory and computation units.
Solution Approach 2:
The memristive crossbar array serves multiple functions: it acts as both memory storage and computation engine, and can perform various linear algebra operations (matrix multiplication, transposition, inversion) by reconfiguring the same hardware structure. This universal capability eliminates the need for separate specialized units for different compression algorithms or processing stages, simplifying the overall system architecture while maintaining high compression performance.
3Speed
If memory size and hierarchy are increased to improve system performance, then processing capability improves, but energy consumption and area increase
Solution Approach 1:
The patent merges memory and computation into a single integrated structure using a memristive crossbar array. The crossbar simultaneously stores data (in memristor conductance states) and performs computation (via parallel matrix operations), eliminating the separation between memory and processor that causes the von Neumann bottleneck. This integration allows data compression operations to be performed directly on stored data without repeated data movement between separate memory and computation units.
Solution Approach 2:
The patent changes the fundamental operating parameters of the computing system by transitioning from digital voltage levels to analog conductance states. Memristors store data as continuous conductance values rather than discrete binary states, enabling direct analog computation. This parameter change allows the system to perform complex compression operations with fewer physical components, reducing area while maintaining or improving performance.
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
This approach results in improved data processing speed, energy efficiency, and reduced physical space utilization compared to conventional hardware and software compression methods, making it suitable for resource-constrained IoT devices.
Implementation Method 1
Conductances of memristors may be set such that the memristors of the crossbar act as coefficients of a wavelet transformation coefficient matrix with respect to voltage signals applied to input rows of the crossbar
Data Source
AI summary
Data compression using a memristive crossbar is enabled. Conductances of memristors may be set such that the memristors of the crossbar act as coefficients of a wavelet transformation coefficient matrix with respect to voltage signals applied to input rows of the crossbar. The memristors may act as coefficients of the transpose of the wavelet transformation coefficient matrix when voltage signals are applied to input columns of the crossbar. Hence, the memristive crossbar may be used to implement a two dimensional (2D) discrete wavelet transform (DWT) on two dimensional data (e.g., image data) encoded in the voltage signals. The resulting currents in the columns of the memristive crossbar may be integrated and converted to voltage signals that are fed back into columns of the memristive crossbar such that the rows of the memristive crossbar output electronic signals that correspond to the image data compressed in accordance with Haar 2D-DWT image compression.