Neurosynaptic Core Circuit for Axonal Signal Processing

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current neuromorphic and synaptronic computation systems lack efficient mechanisms for simulating biological neuronal functions, particularly in handling axonal inputs and generating neuronal outputs, and mapping external inputs and outputs effectively.

Innovation Solution

A neurosynaptic system comprising a delay unit for buffering axonal inputs, a neural computation unit for generating outputs, and a permutation unit for mapping external inputs and outputs, utilizing multiple electronic neurons, axons, and synapse devices to simulate biological neuronal processes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If traditional digital models are used for computation, then manipulation of 0s and 1s is straightforward, but the system cannot function analogously to biological brains

Engineering Contradiction:

Improvebiological brain functionalityVSAvoidcomputation system structure

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system is divided into multiple independent neurosynaptic core circuits, each containing electronic neurons, axons, and synapse devices. These cores can be replicated and interconnected to form larger networks, enabling scalable implementation of biological brain functions while maintaining manageable complexity at each core level.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent creates artificial copies of biological neuronal structures (electronic neurons, axons, and synapse devices) that replicate the functional behavior of their biological counterparts. These electronic copies enable digital systems to simulate biological brain processing without requiring actual biological components.

Inventive Principle:

Principle #26Copying

2Adaptability or versatility

If neuromorphic computation creates connections between processing elements to simulate neurons, then biological brain functionality is achieved, but efficient handling of axonal inputs and generation of neuronal outputs becomes problematic

Engineering Contradiction:

Improveneuronal function simulationVSAvoidcomputation efficiency

Core Design Contradiction:

Adaptability or versatilityVSProductivity

Solution Approach 1:

The delay unit pre-processes and buffers axonal inputs before they reach the neural computation unit. This preliminary action organizes incoming signals in advance, enabling the computation unit to process inputs more efficiently without being overwhelmed by raw input streams.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The permutation unit acts as an intermediary between external inputs/outputs and the internal neural computation unit. It maps and routes signals appropriately, facilitating efficient communication between the external environment and the simulated neuronal system without requiring direct complex connections.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Reliability

If multiple electronic neurons, axons, and synapse devices are used to simulate biological processes, then biological fidelity is improved, but system complexity increases

Engineering Contradiction:

Improvebiological process simulation accuracyVSAvoidnumber of electronic components

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

Each neurosynaptic core circuit is designed as a universal building block that can perform multiple functions: receiving external inputs, processing axonal inputs through delay units, performing neural computations, and generating outputs. This multi-functionality reduces the need for separate specialized components, managing overall system complexity while maintaining biological fidelity.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system employs a hierarchical nested structure where synapse devices are contained within neural computation units, which are contained within neurosynaptic core circuits, which can be interconnected to form larger networks. This nesting allows complex biological processes to be simulated through layered organization, where each level handles specific aspects of neuronal function.

Inventive Principle:

Principle #7Nested doll (Nesting)

Data Source

PatentUS10650301B2Utilizing a distributed and parallel set of neurosynaptic core circuits for neuronal computation and non-neuronal computation

Publication Date: 2020.05.12 INTERNATIONAL BUSINESS MACHINE CORPORATION

AI summary

Embodiments of the invention provide a neurosynaptic system comprising a delay unit for receiving and buffering axonal inputs, and a neural computation unit for generating neuronal outputs by performing a set of computations based on at least one axonal input received by the delay unit. The system further comprises a permutation unit for receiving external inputs to the system, and transmitting external outputs from the system. The permutation unit maps each external input received as either an axonal input to the delay unit or an external output from the system. The permutation unit maps each neuronal output generated by the neural computation unit as either an axonal input to the delay unit or an external output from the system. The neural computation unit comprises multiple electronic neurons, multiple electronic axons, and a plurality of electronic synapse devices interconnecting the neurons with the axons.