MUMBAI, India, Sept. 7 -- Intellectual Property India has published a patent application (202641033719 A) filed by Srm Institute Of Science And Technology Tiruchirappalli on March 20, 2026, for Spiking Neural Network Using Leaky Integrate-And-Fire (lif) Neuron Model Simulation.
Inventors include G. Akshaj; Suvas T. V; and Dr. K. S. Vishvaksenan.
The application for the patent was published on September 04, 2026, under issue no. 36/2026.
Abstract: This is a software-defined model of neuromorphic computation implementation based on the Leaky Integrate-and-Fire (LIF) model as a type of neuron in Spiking Neural Networks (SNNs). It simulates the critical biological neuron firing mechanisms by means of integration of membrane potentials, leakage, threshold-dependent spikes and a reset mechanism. Discrete-time numerical integration is used to update the neuron membrane potential with applied input current raising the potential and a leak term lowering the potential to the resting point. It also accepts various input stimuli (e.g. constant current injection and spike-train-controlled currents e.g. Poisson spike sequences) allowing the neuron to be studied in different conditions. When the membrane potential exceeds a specified threshold the neuron fires a spike output and the potential becomes reset, with a periodic firing mechanism being produced. Both membrane potentials and spike train responses have been simulated, along with toxin simulation and optional raster plot visualization of multi-neuron networks and provide a clear understanding of the dynamics of spiking. The simulator demonstrates a computationally and scalable framework of the research of spike-based processing in the context of SNN behavior validation when all the emulated components exist in the code. It can be used as the backbone of neuromorphic research and has many more extensions to synaptic learning rules, multilayer SNNs, and future neuromorphic hardware acceleration.
Disclaimer: Curated by HT Syndication.
