SkyRadar Blog

In this tutorial, we will have a detailed look at one of the most powerful classes of machine learning and Artificial Intelligence algorithms that exists: the Bayesian Classifiers.

SkyRadar's NextGen Pulse Radar operates in the x-band, just like en-route radars in ATC, as well as military and marine radars. It includes all the important features needed by ATCOs and ATSEPs and reaches up to Artificial Intelligence. Its modular structure allows to customize it on university requirements as well.

In the first part of this series, we introduced the general concepts needed for understanding the Hidden Markov Models Classifiers. Namely: Bayesian Logic, The concept of Bayesian Classifiers and Bayesian networks. All these notions are now grouped to form a new type of classifier which can accurately model and classify time-series data such as the RADAR data.

In this article we will explore a very special class of classifiers, the Hidden Markov Model Classifiers (HMMs). They are mainly a statistical and probabilistic model but they have found their entry in the world of Machine Learning since they can be trained and classify data.

The recent crash of Ukrainian Airlines Flight 752, downed by a surface-to-air (SAM) Iranian missile after a tragic human error from a military RADAR operator underlines dramatically the need for ADS-B, combined with automatic computer-based recognition of RADAR targets.

In this article we shall perform the anatomy of a simple but efficient convolutional network (CNN), the LeNet-1 neural network.

RADAR data classification mainly relies on convolutional neural networks. In this article, we shall detail and explain the main operations performed by Convolution networks in order to classify RADAR data.

A relevant technique to classify RADAR object is to use a colored 2D map representing range, speed or frequency against time and color the map with power intensity. Here we represent examples of such typical RADAR data.

Fast reaction and decision-making of the RADAR operator is a key factor in ATC. There is a need to develop techniques for the automatic extraction and fine-granulated classification of RADAR objects, allowing for faster and better decision making and more effective processes.