MUMBAI, India, Jan. 2 -- Intellectual Property India has published a patent application (202541123058 A) filed by Sumanth Sakkara; and Dr. Cyril Prasanna Raj P, Belagavi, Karnataka, on Dec. 6, 2025, for 'computation efficient reconfigurable intra and inter pipelined deep neural network hardware architecture for underwater edge processing systems.'

Inventor(s) include Sumanth Sakkara; and Dr. Cyril Prasanna Raj P.

The application for the patent was published on Jan. 2, under issue no. 01/2026.

According to the abstract released by the Intellectual Property India: "A Field Programmable Gate Array (FPGA)-based deep neural network (DNN) accelerator for underwater acoustic communication is disclosed, specifically designed for edge and IoT applications. The system provides a robust, low-power, and scalable solution for deep learning-based underwater IoT communication, addressing the sensitivity of conventional OFDM Fast Fourier Transform (FFT) processors to multipath fading and channel nonlinearities. The core technology features a DNN-based encoder-decoder transceiver architecture that enables joint symbol encoding and data compression while ensuring robust signal reconstruction in bandwidth-limited underwater channels. The hardware is implemented as a pipelined single-neuron architecture that incorporates a novel Cascaded Serial Multiplier Matrix (CSMM) and Zero Padding Algorithm (ZPA). This architecture is designed to eliminate dedicated multipliers, resulting in a significant reduction in hardware resource utilization. The FPGA implementation achieves a 100% reduction in dedicated multipliers, 53.2% fewer Look-Up Tables (LUTs), 49.8% fewer registers, a peak operating frequency of 248 MHz and only 9 mW dynamic power consumption. A nine-stage pipelined DNN is developed to ensure scalable operation across various input sizes and modulation schemes, delivering high throughput, low latency, and energy-efficient real-time performance. The invention provides a highly hardware-efficient and robust deep learning acceleration platform for next- generation underwater communication systems."

Disclaimer: Curated by HT Syndication.