This post is the continuation of the part-1 post. Check it out here.

In this part, we will focus on the mathematics of the algorithm. To get a better idea of how VMD works and also I will try my best to clear all the doubts which you might have encountered in the previous part.

There are few prerequisites before we start the maths. You should be familiar with

- Fourier Transform
- Constrained Optimization

If you think you know the above stuff then let’s test it.

There are three important things that VMD does

- Wiener Filter properties
- Forming analytical signal
- Frequency…

This will be a blog series because it's tough to cover everything in one blog. Please bear with me if things are confusing. Any doubts, the comments section is always waiting for you.

Let's start by asking a simple question.

Can you hear the instruments individually while listening to any music with your eyes close?

Well, it seems we use different filters to focus our attention on a particular instrument. Interestingly, the filters are adaptable and can work while listening to any type of song. This method is known as source separation. There are multiple inputs combined and our ears…

ANN’s are the most fundamental structure of neural networks. The basic ANN structure is known as the perceptron. Perceptron is a simple linear regression with an activation function. Linear Regression is applied for finding the linear relationship between input and output. But most of the real data is non-linear in nature, so to make the regression versatile, we use perceptron with activation. The activation function adds non-linearity to the output making it flexible for non-linear input.

Until the stimuli on our nerves cross a particular threshold, neurons will not activate and won't send any signal to our brain. Similarly, a…

Pursuing Master's in Artificial Intelligence and Data Science. My research area is to combine signal processing ideas with deep learning methods.