We will be carrying out essential maintenance activity on this website on Monday 12th October between 0600 and 0900 UK time. You might experience some brief interruption or loss of some functionality. We apologise for any inconvenience this may cause.

New Work Item Proposal - 47-3050 ED1: Semiconductor devices – Neuromorphic devices- Part 7: Evaluation method of synaptic decay time in memristor devices

Source:
IEC
Committee:
EPL/47 - Semiconductors
Categories:
Information management | Standardization. General rules
Comment period start date:
Comment period end date:

Comment by:

Scope

This part of IEC 63550 specifies the test methods for determining synaptic decay time in neuromorphic memristor devices by monitoring the relaxation of a programmed read current state after the removal of programming stimuli. It covers the test configurati on, acquisition of decay data over time, and extraction of synaptic decay time using single -exponential or double-exponential fitting. It is applicable to two-terminal neuromorphic memristor devices without limitation by material system or size, provided that the measured decay can be adequately represented by one of the prescribed fitting models over the observation window. The method does not assume a specific relaxation mechanism; the exponential fitting prescribed in this document is adopted as a phenomenological approximation that enables consistent and comparable decay -time extraction across different device technologies.

Purpose

Memristor-based artificial synapses are emerging as a key building block of neuromorphic hardware for artificial intelligence (AI), as their analogue conductance modulation emulates biological synaptic plasticity. The synaptic decay time, which quantifies how a programmed conductance state relaxes toward its baseline after the programming stimulus is removed, determines whether a device is suitable for long-retention applications such as static pattern recognition or for temporal signal processing based on short-term plasticity. As neuromorphic computing advances from research toward commercialization, published decay data remain difficult to compare because programming conditions, read bias, observation windows, and fitting methods vary widely across labora tories. This document establishes a standardized evaluation method covering the measurement configuration, acquisition of decay data over time, and extraction of synaptic decay time based on single-exponential and double-exponential fitting. It thereby improves the reliability, reproducibility, and comparability of synaptic decay data across different devices and laboratories, and supports technology commercialization and international trade in this emerging market.

Comment on proposal

Required form fields are indicated by an asterisk (*) character.


Please email further comments to: admin.start@bsigroup.com

Follow standard

You are now following this standard. Weekly digest emails will be sent to update you on the following activities:

You can manage your follow preferences from your Account. Please check your mailbox junk folder if you don't receive the weekly email.

Unfollow standard

You have successfully unsubscribed from weekly updates for this standard.

Error