학술논문

A Low-Power ABR Characteristic Waveform Automatic Detection Algorithm Design and FPGA Implementation
Document Type
Conference
Source
2024 International Conference on Microelectronics (ICM) Microelectronics (ICM), 2024 International Conference on. :1-5 Dec, 2024
Subject
Bioengineering
Components, Circuits, Devices and Systems
Computing and Processing
Engineered Materials, Dielectrics and Plasmas
General Topics for Engineers
Photonics and Electrooptics
Power, Energy and Industry Applications
Robotics and Control Systems
Energy consumption
Accuracy
Power demand
Filtering
Filtering algorithms
Hardware
Microelectronics
Object recognition
Character recognition
Biological neural networks
Auditory Brainstem Response
Recognition of Characteristic Waveform
U-Net
Filter
FPGA
Language
ISSN
2159-1679
Abstract
Portable, low-power ABR characteristic waveform automatic identification devices have broad application prospects in clinical and scientific research. This paper proposes an FPGA-based ABR characteristic waveform automatic identification device. The device integrates an ABR automatic identification algorithm, which outputs the latencies of waves I, III, and V with the ABR waveform as input into the system. The algorithm consists of two parts: a filtering algorithm and a U-Net neural network algorithm, achieving an average accuracy of 91.96% on a self-built dataset. Compared to using only the U-Net neural network algorithm, this method reduces the energy consumption by 19.64% on the self-built dataset, which is of great practical value in scenarios that require long-term or repeated ABR characteristic waveform automatic detection tasks.