Blood flow analysis is performed by passing a high frequency ultrasonic wave into the blood vessels through a transducer (transmitter). The reflected signal; The transducer of the receiver has a different frequency due to the Doppler principle. This signal is passed to a DSP processor to find the frequency spectrum. Due to the high frequency of the ultrasonic wave, the resolution of the frequency spectrum output will not be good. Therefore, advance in the FFT Zoom technique, in which a very small frequency change can be obtained due to the formation of clots with a good resolution. It can be used to locate the initial presence of a blood clot. All these tasks must be achieved with a single DSP chip to make the system profitable and energy efficient and therefore widely accepted.
The Fast Fourier Transform (FFT) is one of the most commonly used algorithms in digital signal processing and is widely used in applications such as image processing and spectral analysis.
The purpose of this application note is to investigate efficient partitioning / parallelization schemes for 1-D FFTs in the parallel processing DSP TMS320C40. The partitioning of the FFT algorithm is important in two special cases:
For large FFT calculations in which the input data does not fit into the RAM of the chip available in the processor. In this case, execution must be performed with off-chip data, resulting in performance degradation. As a consequence, the execution time grows exponentially with the FFT size.
ADVANTAGES
1. Higher resolution frequency domain
2.Reduced the cost and complexity of the hardware
ZOOM FFT
Zoom-FFT is a process in which an input signal is mixed to the base band and then decimated, before being passed to a standard FFT. The advantage is, for example, that if you have a sampling frequency of 10 MHz and require at least 10Hz resolution over a small frequency band (say 1 KHz), then you do not need a FFT of 1 Mega Point, just decimate By a factor of 4096 and use a 256-point FFT that is obviously faster.