Integrate-and-fire Neuron Circuit Using Single-gated Feedback FBFET
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional neuron circuits for spiking neural networks (SNNs) face challenges in achieving small size and low power consumption, which are essential for efficient hardware-based implementations, as they often require numerous transistors and struggle with bidirectional parallel operation and synapse plasticity.
Innovation Solution
A novel integrate-and-fire (IF) neuron circuit utilizing a single-gated feedback field-effect transistor (FBFET) with a capacitor and three transistors, which generates and resets spike voltage through a positive feedback loop, allowing for efficient integration and firing operations with reduced power consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional CMOS-based circuits are used to implement neuron functions (input pulse integration, threshold triggering, spike voltage), then the neuron circuit can achieve required functions, but the circuit size increases and power consumption increases due to using numerous transistors
Solution Approach 1:
The patent combines multiple neuron functions (input pulse integration, threshold triggering, spike voltage generation) into a single FBFET device. The FBFET integrates the channel region, intrinsic region, and gate electrode structure to perform all three functions simultaneously, eliminating the need for separate transistor circuits for each function.
Solution Approach 2:
The FBFET is designed as a universal device that can perform multiple neuron functions: integrating input pulses through capacitance, triggering when threshold is exceeded through the positive feedback loop, and generating spike voltage through drain voltage amplification. This single device replaces what would traditionally require multiple specialized components.
2Reliability
If conventional neuron circuits are used, then sufficient transistor availability is maintained, but the power consumption increases significantly
Solution Approach 1:
The FBFET utilizes its own internal positive feedback loop mechanism to generate and amplify spike voltage automatically when the threshold is exceeded. This self-service mechanism eliminates the need for external high-power amplification circuits, reducing overall power consumption while maintaining reliable spike generation.
3Reliability
If more transistors are used in neuron circuits, then circuit functionality and reliability are improved, but the circuit size becomes larger preventing VLSI integration
Solution Approach 1:
The FBFET structure nests multiple functional regions within a single device: the gate electrode wraps around the intrinsic region, which contains the channel region. This nested configuration allows multiple neuron functions to be packed into a compact area, enabling VLSI integration while maintaining reliable 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
The proposed neuron circuit achieves a high fire frequency of 20 kHz and low power consumption of 7 μW using only four transistors, enabling the possibility of realizing a hardware-based SNN for very-large-scale integration (VLSI) with improved efficiency and reduced power usage.
Implementation Method 1
when the generated potential exceeds a threshold value, the neuron circuit generates and outputs a spike voltage corresponding to the generated potential using a single-gated feedback field-effect transistor (FBFET) connected to the capacitor
Data Source
AI summary
The present disclosure relates to a novel integrate-and-fire (IF) neuron circuit using a single-gated feedback field-effect transistor (FBFET) to realize small size and low power consumption. According to the present disclosure, the neuron circuit according to one embodiment may generate potential by charging current input from synapses through a capacitor. In this case, when the generated potential exceeds a threshold value, the neuron circuit may generate and output a spike voltage corresponding to the generated potential using a single-gated feedback field-effect transistor connected to the capacitor. Then, the neuron circuit may reset the generated spike voltage using transistors connected to the feedback field-effect transistor.