MUMBAI, India, Oct. 12 -- Intellectual Property India has published a patent application (202611104557 A) filed by Noida Institute Of Engineering And Technology Niet on August 31, 2026, for Fpga-Based Adaptive Predistortion System With Closed-Loop Calibration For Analog Radio-Over- Fiber.
Inventors include Dr. Niraj Agrawal; and Dr. Vinod Mansiram Kapse.
The application for the patent was published on October 09, 2026, under issue no. 41/2026.
Abstract: ABSTRACT TITLE: FPGA-BASED ADAPTIVE PREDISTORTION SYSTEM WITH CLOSED-LOOP CALIBRATION FOR ANALOG RADIO-OVER-FIBER The present invention provides an FPGA-based adaptive digital predistortion system (100) for linearization of analog radio-over-fiber fronthaul links which comprises a baseband signal conditioning unit (110), an FPGA-based predistortion engine (120) implementing a spline-activated neural computation core (124) as a hardware signal processing pipeline within the programmable logic fabric, a radio-frequency upconversion and electro-optic transmission module (130, 140), an optical receiver and digitization module (160), and a closed-loop adaptive calibration controller (180). The calibration controller (180) continuously monitors error vector magnitude and spectral regrowth metrics, estimates distortion profiles through a distortion profile estimator (182), computes updated predistortion coefficients via a coefficient update engine (184) employing an Adam optimizer in an indirect learning architecture, and transfers said coefficients to the FPGA predistortion engine (120) through a double- buffered weight storage memory (128) with atomic buffer swap at sample boundaries, enabling hitless in-service recalibration for thermally and operationally varying 5G fronthaul environments. [Figure 1]
Disclaimer: Curated by HT Syndication.
