Field programmable gate arrays implementation of Dual Tree Complex Wavelet Transform.

Area of Science:

• Biomedical Signal Processing
• Digital Signal Processing
• Embedded Systems Engineering

Background:

• Biomedical signals (BSs) often exhibit non-stationary characteristics due to physiological system variability.
• Traditional Discrete Wavelet Transform (DWT) for BS analysis has limitations including aliasing, lack of directionality, and shift-variance.
• The Dual Tree Complex Wavelet Transform (DTCWT) offers an improved alternative to DWT, addressing its inherent drawbacks.

Purpose of the Study:

• To propose a novel hardware architecture for implementing the DTCWT algorithm on Field-Programmable Gate Arrays (FPGAs).
• To enable real-time feature extraction and denoising of biomedical signals using an efficient DTCWT implementation.
• To design a scalable architecture capable of processing multi-channel biomedical data in parallel.

Main Methods:

• Development of a novel hardware architecture for DTCWT implementation on FPGAs.
• Optimization of the DTCWT algorithm to utilize minimal hardware resources, specifically one adder and one multiplier.
• Design for parallel processing of N channels to accommodate multi-channel biomedical data acquisition systems.

Main Results:

• The proposed FPGA architecture successfully implements the DTCWT with high efficiency.
• The architecture achieves real-time processing capabilities for biomedical signal analysis.
• The design demonstrates scalability for handling multiple input channels concurrently.

Conclusions:

• The novel FPGA architecture provides an efficient and effective hardware solution for DTCWT-based biomedical signal processing.
• This implementation facilitates the development of advanced portable real-time medical devices for rapid patient diagnosis.
• The optimized resource utilization and parallel processing capabilities make the architecture suitable for various medical applications.