Signal Analytics is one of the key research domain of Cyber Physical Systems (CPS) where smart networked systems with embedded sensors, processors and actuators that are designed to sense and interact with the physical world and support real-time, guaranteed performance in safety-critical applications.
The main focus of the Signal Analytics Group is to carry out R&D in the following domains:
Machine Learning and Signal Processing
An increasing number of applications require the joint use of signal processing and machine learning techniques on temporal and sensor data. Mining of processed signal can discover interesting patterns.
The specific patterns in the signals can indicate some of the actionable outcomes such as failure of a device, forecasting, identification etc. The IoT enabled sensors produce variety of data streams at high velocity and may send multiple signals simultaneously over time. It is require to process multi-dimensional time-order streams. Therefore, novel ways to find specific patterns in the signals generated by many sources are also required.
Chemometrics and Machine Learning
Chemometrics is an analytical chemical field that uses mathematical tools to design or select experimental procedures. Such procedures are used to provide the maximum relevant chemical information by analysing chemical data and to obtain knowledge about chemical systems. Electrochemical, acoustic and Spectral analysis with pattern recognition and machine learning techniques can enhance the power of sensing schemes. Development of effective analysis tools is a challenge for effective intelligent sensor development for some of the basic problems like water quality determination, soil analysis, milk quality and food quality assessment.