Neuromorphic System Using Global Self-Controller for Reset and Lateral Inhibition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing CMOS-based neuron circuits face challenges with large area requirements and high power consumption due to the size of integrate capacitors and the number of devices, leading to complexity and accuracy limitations in neuromorphic systems.
Innovation Solution
A neuromorphic system incorporating a neuromorphic device with a thin semiconductor structure, doped regions, and a gate electrode connected to a synapse array, utilizing a global self-controller to generate control signals and implement homeostasis, reset, and lateral inhibition functions, while minimizing area and power consumption through charge accumulation and switching mechanisms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If CMOS-based neuron circuits use traditional integrate capacitors and comparators, then neuron functions can be implemented, but the area and power consumption increase significantly
Solution Approach 1:
The patent extracts the integrating function from traditional large capacitors and implements it using the intrinsic capacitance of neuromorphic device gate insulating films, thereby eliminating the need for separate large-area integrate capacitors while maintaining the neuron's integrating function
Solution Approach 2:
The patent merges the neuron circuit components by using the neuromorphic device structure itself to perform multiple functions - the gate insulating film serves as both the device insulation layer and the integration capacitor, reducing the total number of separate components and overall circuit area
2Reliability
If CMOS-based neuron circuits use traditional integrate capacitors and comparators, then neuron functions can be implemented, but power consumption increases
Solution Approach 1:
The neuromorphic device utilizes its own gate insulating film capacitance for integration purposes, making the device self-sufficient and eliminating the need for additional power-hungry comparator circuits and large capacitors, thereby reducing overall power consumption while maintaining functionality
3Adaptability or versatility
If the number of devices in neuron circuits is increased, then more functions can be implemented, but the area and complexity increase
Solution Approach 1:
The neuromorphic device is designed to perform multiple functions simultaneously - serving as the neuron body, integration element, and firing threshold detector through its gate insulating film capacitance and switching characteristics, thereby reducing the need for separate dedicated components for each function
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
The system achieves a high degree of integration and low power consumption, enabling efficient implementation of neuromorphic functions with improved accuracy and systematic operation, allowing for the organic performance of reset and lateral inhibition operations.
Implementation Method 1
a neuromorphic device configured to accumulate charges according to the signal input from the synapse array
Implementation Method 2
to allow a current to suddenly flow when a potential due to the accumulated charges reaches a threshold value
Implementation Method 3
a first switching device configured to be located between one end of the neuromorphic device and the output circuit and to control an electrical connection between the neuromorphic device and the output circuit
Data Source
AI summary
Provided is a neuromorphic system using a neuron circuit. The neuromorphic system includes: one or two or more neuron circuits configured to output a firing signal according to signals input from a synapse array; a homeostatic circuit for each neuron circuit; and a global self-controller configured to generate and provide control signals for the neuron circuits by using the firing signal output from the neuron circuits. The neuron circuit includes a neuromorphic device and an output circuit that outputs the firing signal of the neuromorphic device. The global self-controller generates and supplies a reset signal to the neuromorphic device of the fired neuron circuit, and the global self-controller generates and supplies a lateral inhibition signal to the neuromorphic device of the non-fired neuron circuit. The homeostatic circuit alleviates inhibition of other neurons by the neurons with a predominant firing function.