MUMBAI, India, July 13 -- Intellectual Property India has published a patent application (202621068921 A) filed by Symbiosis International Deemed University on June 01, 2026, for Gan-Discriminator-Refined Semantic Embedding System With State-Space Sequence Modelling For Hierarchical Ddos Detection In Industrial Iot Networks.
Inventor includes Dr. Sagar Dhanraj Pande.
The application for the patent was published on July 10, 2026, under issue no. 28/2026.
Abstract: ABSTRACT GAN-DISCRIMINATOR-REFINED SEMANTIC EMBEDDING SYSTEM WITH STATE-SPACE SEQUENCE MODELLING FOR HIERARCHICAL DDOS DETECTION IN INDUSTRIAL IOT NETWORKS A system (100) and method for hierarchical detection and classification of Distributed Denial of Service (DDoS) attacks in Industrial Internet of Things (IIoT) environments is disclosed. The system comprises a Flow Serialisation Module (110) that converts raw network flow records into structured natural-language strings, a Semantic Embedding Layer (120) that encodes the strings using a frozen large language model with Matryoshka Representation Learning, a GAN Discriminator Refinement Module (130) whose adversarially pre-trained frozen discriminator (134) provides penultimate hidden-layer activations as attack-aware feature refinements concatenated with the semantic embeddings, a Mamba Selective State-Space Model Block (140) that processes temporal sequences of refined embeddings with linear- time complexity through input-dependent selective gating, a Cross-Feature Attention Module (150) producing a context vector through multi-head attention and temporal mean-pooling, and a Hierarchical Dual-Head Classifier (160) with a Binary Head (162) gating a Multiclass Head (164) via a confidence threshold. The system achieves simultaneous detection of network-layer and hardware-layer DDoS attacks in resource-constrained IIoT environments. [
Disclaimer: Curated by HT Syndication.
