All work

Research

IEEE Xplore, 2026 · 2026

ECG denoising

A Verilog implementation of a cascaded filter architecture that cleans a real-time ECG signal in hardware, with a reported 19.92 dB improvement in signal-to-noise ratio.

ECG denoising
IEEE Xplore, 2026 — Published2026

Specification

As built, not as pitched

Reported gain
+19.92 dB SNR
Implementation
Verilog RTL
Design and analysis
MATLAB
Verification
ModelSim, captured ECG
Preserved feature
QRS complex
Published
IEEE Xplore, Mar 2026
With
Navya, Sphatika, Megha
Organisation
IEEE Xplore, 2026
Year
2026
Stage
Published
Role
RTL design, Filter architecture, Signal validation
Delivered
Cascaded Verilog filter chain, MATLAB analysis, ModelSim verification

An ECG is a small signal in a noisy room: baseline wander from breathing, mains hum from the wiring, and muscle noise from the patient. Diagnoses are made on the shape of a wave that is easy to lose.

The work specifies a cascade of filters — baseline removal, mains rejection and smoothing — designed in MATLAB and then written in Verilog so it can run on acquisition hardware instead of a desktop. Simulated against captured ECG data in ModelSim, the chain reports a 19.92 dB improvement in signal-to-noise ratio while preserving the QRS complex.

Published through IEEE Xplore in March 2026 with Navya, Sphatika and Megha. The DOI and author record will be linked here once verified.

Next project

Education software

AcademicOS