Neuron Circuit Biristor Voltage Divider Energy Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current neuromorphic computing systems, such as spiking neural networks, face challenges in reducing energy consumption due to the complexity and energy-intensive nature of their hardware components, particularly in mimicking the human brain's efficient neuronal circuits.
Innovation Solution
A neuron circuit incorporating a bistable resistor (biristor) and two transistors functions as a voltage divider, reducing the magnitude and pulse width of output voltage, thereby decreasing energy consumption by modulating the operating voltage using resistance values, and is integrated into neuromorphic circuits and data processing devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a complex circuit with capacitor, integrator, comparator, and reset circuit is used for LIF neuron operation, then the neuron circuit can perform leaky integrate-and-fire function, but the device complexity increases and energy consumption increases
Solution Approach 1:
The patent combines multiple functional components (capacitor, integrator, comparator, reset circuit) into a single biristor-based circuit. The biristor itself performs the integration function through its voltage-current characteristics, while the voltage divider circuit integrates the comparator and reset functions, thereby reducing the number of discrete components while maintaining LIF neuron functionality
Solution Approach 2:
The biristor is designed to perform multiple functions simultaneously: it acts as the core computing element for integration, provides voltage scaling through the voltage divider mechanism, and enables the firing threshold comparison. This multi-functionality eliminates the need for separate dedicated components for each function, reducing overall circuit complexity
2Reliability
If a complex circuit with capacitor, integrator, comparator, and reset circuit is used for LIF neuron operation, then the neuron circuit can perform leaky integrate-and-fire function, but the energy consumption increases
Solution Approach 1:
By merging multiple functional blocks into a unified biristor-based architecture, the patent reduces the total number of active components that consume energy. The voltage divider circuit shares the power supply voltage efficiently, and the biristor's inherent voltage-current characteristics eliminate the need for continuous powering of separate integrator and comparator circuits
Solution Approach 2:
The patent changes the operating parameters of the circuit by using the voltage divider ratio to control the effective threshold and integration characteristics. This allows the circuit to operate with lower voltage and current levels compared to traditional LIF implementations, thereby reducing energy consumption while maintaining functional reliability
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 configuration effectively decreases energy consumption in neuromorphic hardware by adjusting output voltage magnitude and pulse width, enhancing the efficiency of neuromorphic computing systems.
Implementation Method 1
a voltage divider that is enabled by the collector signal, performs voltage division on an operating voltage by using values of resistances included therein
Data Source
AI summary
According to an embodiment of the present disclosure, a neuron circuit may be provided. The neuron circuit includes a biristor that includes a collector electrode receiving a constant input current from a first synapse circuit and an emitter electrode connected with a ground and outputs a collector signal through the collector electrode, and a voltage divider that is enabled by the collector signal, performs voltage division on an operating voltage by using values of resistances included therein, and outputs an output voltage corresponding to a result of the voltage division to a second synapse circuit.