Bio-Inspired Spiking Neural Networks for Low-Power Edge Intelligence

SOWMIYA N T; SHYAMALA G A; ENIYAN B; NAVEEN S S; VENKATESAN P; SHYAMALA G A

doi:10.36948/ijfmr.2025.v07i06.60089

Bio-Inspired Spiking Neural Networks for Low-Power Edge Intelligence

Author(s)	Ms. SOWMIYA N T, Ms. SHYAMALA G A, Mr. ENIYAN B, Mr. NAVEEN S S, Mr. VENKATESAN P, Ms. SHYAMALA G A
Country	India
Abstract	Spiking Neural Networks (SNNs) are emerging as a biologically plausible alternative to conventional artificial neural networks for deploying intelligence on low-power edge devices. By mimicking the brain's operational principles, SNNs process information through discrete, asynchronous spikes over time, rather than the continuous, high-precision calculations of traditional deep learning. This event-driven nature means that neurons only compute when they receive input, leading to significant inherent energy savings, especially for sparse data. This review synthesizes recent bio-inspired innovations in SNNs, focusing on architectures and learning algorithms that are suitable for real-world applications. A key development is surrogate gradient training, which overcomes the challenge of training non-differentiable spiking neurons and enables high-performance, deep SNNs directly from data. Furthermore, the synergy between SNNs and novel event-based sensors, such as Dynamic Vision Sensors (DVS), creates a highly efficient pipeline. These sensors naturally output sparse spike streams, eliminating redundant data capture and processing. The ultimate efficiency is realized through hardware-software co-design, where the sparse, event-driven algorithm is deployed on specialized neuromorphic processors. These chips, such as Loihi and TrueNorth, are architected to leverage the sparsity and temporal dynamics of SNNs. Benchmarks consistently demonstrate that such integrated systems (sensors → SNN → neuromorphic hardware) can achieve order-of-magnitude reductions in inference energy compared to standard deep learning on GPUs. This makes SNNs a compelling technology for always-on edge applications, including embedded vision, autonomous navigation, and adaptive real-time control.
Keywords	Spiking Neural Networks, Neuromorphic Computing, Edge Intelligence, Event-Driven Sensing, Energy Efficiency, Surrogate Gradient Learning
Field	Mathematics
Published In	Volume 7, Issue 6, November-December 2025
Published On	2025-11-13
DOI	https://doi.org/10.36948/ijfmr.2025.v07i06.60089

View / Download PDF File

E-ISSN 2582-2160

doi

CrossRef DOI is assigned to each research paper published in our journal.

IJFMR DOI prefix is
10.36948/ijfmr

Downloads

Research Paper Format Copyright Permission Form and Undertaking Form Cover Page Vol 8 Isu 3 Cover Page Vol 8 Isu 2 Cover Page Vol 8 Isu 2

All research papers published on this website are licensed under Creative Commons Attribution-ShareAlike 4.0 International License, and all rights belong to their respective authors/researchers.

CC-BY-SA

About IJFMR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

Bio-Inspired Spiking Neural Networks for Low-Power Edge Intelligence

Share this