Extracting traces from acquired ultra wideband radar data

Hello,

I have some data acquired from a UWB radar that I would like to try to extract traces from. The data itself is complex valued and the magnitude of it is pictured below.

I have managed to extract the most intense trace itself by finding the maximum value at each column and for my use case it is enough, but I am wondering if I can do it in a smarter way and potentially extract the whole 'helix' shape which you can see in the image?

Answer 1

[Ayush]

Hello Halldór,

For extracting the whole ‘helix’ shape or the data from UWB radar leading up to that shape, there are several signal processing and image processing techniques that you can consider. Although this would depend on the properties/parameters of your data, the noise level as well as the features that led to the ‘helix’ shape. Here are some smart and alternate ways to get to your desired result:

1. Thresholding and Connected Component Analysis: Applying a threshold can help eliminate weaker signals and thus distinguish the 'helix' from any background noise. Once you've applied this threshold, you can then use connected component analysis to detect and isolate the continuous areas that make up the 'helix' pattern in your data.

2. Peak Detection: Instead of just taking the maximum value in each column, you could use more advanced peak detection algorithms that can find multiple peaks if they exist. Please refer to the below documentation to know more about peak detection function in MATLAB:

• https://www.mathworks.com/help/signal/ref/findpeaks.html

3. Hough Transform: If the 'helix' can be approximated by a parametric curve, the Hough transform using the “hough” function can be used to detect such shapes in the data. Please refer to the below documentation to know more about “hough” function:

• https://www.mathworks.com/help/images/ref/hough.html?s_tid=doc_ta

4. Time-Frequency Analysis: Perform a time-frequency analysis (e.g., Short-Time Fourier Transform, Wavelet Transform) to better understand the structure of the 'helix' in both time and frequency domains. Please refer to the below file exchange link to know more about Short-Time Fourier Transform:

• https://www.mathworks.com/matlabcentral/fileexchange/45197-short-time-fourier-transform-stft-with-matlab?s_tid=srchtitle_support_results_1_Short-Time%20Fourier%20Transform%252C%20

Hope it helps!