Imperial College London

ProfessorDaniloMandic

Faculty of EngineeringDepartment of Electrical and Electronic Engineering

Professor of Machine Intelligence
 
 
 
//

Contact

 

+44 (0)20 7594 6271d.mandic Website

 
 
//

Assistant

 

Miss Vanessa Rodriguez-Gonzalez +44 (0)20 7594 6267

 
//

Location

 

813Electrical EngineeringSouth Kensington Campus

//

Summary

 

Overview

Danilo's current research interests are in the area of blind source separation and extraction especially for brain science. He is also working on extending his previous results from real valued adaptive filters to multivariate adaptive filters, especially for complex signals.

Danilo has recently had a keen interest in the exciting area of detecting the nature of a signal, such as their linearity, nonlinearity, or their deterministic or stochastic behaviour, together with signal modality.

Danilo's research  interests have mostly been in  adaptive linear and nonlinear signal processing. His  work has focused on nonlinear time series analysis an prediction, especially using neural networks, and nonlinear adaptive filters. Among neural networks, he is mostly interested in recurrent and modular neural networks, their learning algorithms, convergence, and choice of optimal parameters.


His previous work includes adaptive real-time processing of surface seismic waves, DSP hardware architectures, and development of computer aided devices for Braille reading. He is also interested in optimization and biomedical signal processing.

 

Current research projects:

 

Nonlinear Multidimensional Adaptive Modelling, System Identification and Prediction.

Especially complex-valued and hyper-complex neural networks for natural phenomena, such as in modelling of environmental signals, complex industrial plants and trajectory tracking.

 

Wind Modelling and Prediction

Using advanced nonlinear multivariate predictors. Simultaneous modelling of all the components of multidimensional and multivariate signals.

 

One- and multi-dimensional adaptive denoising

Using adaptive signal processing algorithms and wavelets, with applications to biomedical signals, such as fMRI and medical images

 

Blind Source Separation and Extraction

Especially on-line adaptive algorithms for Electroencephalogram, MEG, and fMRI, especially based on some fundamental properties of signals, such as theor nonlinear, stochastic or nonstationary nature.

 

Brain Signal Processing

Signal conditioning and classification in various EEG based mental tasks, such as decision making processes.

 

Sensor Fusion

Combining information from sensors measuring different physical quantities, especially in modelling and detection of sleep stages, where a combination of EEG, EOG, ECG and respiratory signal is processed. Also sensor fusion in navigation, and robotics.

 

Fit for Duty

Data fusion based techniques for fatigue detection.

 

Signal Modality Characterisation

Based on the recently developed DVV method, this new exciting area of research can be used to asses a qualitative performance of adaptive filters as well as for initial characterisation of signal's nature, as for instance, linear, nonlinear, deterministic or stochastic.

Collaborators

Prof David Sharp, submitted a joint grant proposal, Department of Medicing, Ear EEG for trauma, 2013

Prof Mary Morrell, joint BRU grant on Ear-EEG, The Royal Brompton Hospital, Sleep Clinic, 2012

Prof A. Cichocki, RIKEN Brain Science Institute, sabbatical, 2011 - 2011

Prof Nikola Kasabov. This collaboration resulted in jointly co-editing the Springer Handbook of Bio- and Neuro-Informatics, Springer, 2014.http://www.springer.com/engineering/computational+intelligence+and+complexity/book/978-3-642-30573-3, Aukland University of Technology, Bio- and neuro-informatics, 2010

Prof Akira Hirose. This collaboration resulted in a joint special issue in IEEE Transactions on Neural Networks, and several joint special sessions and tutorials in international conferences., University of Tokyo, complex neural networks, 2009

Dr Dragan Obradovic, Dr Hans-Georg Zimmermann, Dr Andras Szabo.This collaboration has resulted in a jointly edited book with D. Obradovic "Signal Processing Techniques for Knowledge Extraction and Information Fusion", Springer 2008, Siemens Corporate Research, Munich, Germany, Signal Processing for Smart Grid., 2008

Prof Jianting Cao. This collaboration resulted in a best student paper award in ISNN'10 and several joint articles, Saitama Institute of Technology, Brain computer interface, 2008

Prof Anthony Kuh. This collaboration resulted in an edited book (Springer 2008), several jointly organized special sessions in conferences, and several articles, University of Hawaii, Signal Processing for Smart Grid, 2008

Prof Aaron Williamon. This collaboration resulted in our appearance in the Channel 4 documentary (3 episodes), "Music and Emotion", and two proposal submissions, Royal College of Music, 2008

Dr Marco Leo. This collaboration resulted in two joint special sessions in international conferences, and several joint articles, ISSIA, Bari, Italy, Data and sensor fusion, 2008

Prof Kazuyuki Aihara. This collaboration resulted in a successful joint grant proposal and several articles in international journals, University of Tokyo, wind forecasting, complexity science, 2008

Dr Preben Kidmose. Dr Kidmose is a Visiting Academic at CSP, and this collaboration has resulted in a patent application, and a shortlisted entry to the Annual BCI Competition in 2012, Aarhus University, Denmark, Brain computer interfacce, 2008

Dr Tomasz Rutkowski. This collaboration was supported by an international grant (Royal Society), has resulted in several jointly organized special sessions in conferences, and a best paper award (ICIC'09), Tsukuba University, Japan, Brain computer interface, 2008

Prof Ali Sayed. This has resulted in jointly organized special sessions in international conferences, joint articles, and Prof Sayed's plenary at our annual conference within the UDRC, University of California, Los Angeles, CA, USA, adaptive filters, 2008

Prof Marc Van Hulle. This is a long-lasting collaboration, supported by several exchange grants and participation in a joint FP7 project (Neuroprobes), KU Leuven, Belgium, Machine learning for brain computer interface, 2008

Prof Scott Douglas, we have two 3-hour tutorials in foremost signal processing conferences (ICASSP, ICWCS), and several articles. Prof Douglas has also been appointed a Visiting Academic at CSP, Southern Methodist University, Dallas, TX, USA, adaptive filtering, 2008

Prof Andrzej Cichocki, Dr Tomasz Rutkowski.This has resulted in a Visiting Academic position for D. Mandic, a joint Best Paper Award in ICIC'09, and an RIKEN Award for D. Mandic, RIKEN, Brain Science Institute, Wakoshi, Saimata, Japan, 2008

Dr Michael Ungstrup, Dr Mike Lind Rank.This work has resulted in a working Ear-EEG prototype, two pending patents (Danish and USA) (shortlisted for the best Brain Computer Interface in 2012), and several articles, Widex A/S, the third biggest European hearing aids manufacturer, Tension measures in hearing aids, Ear-EEG, 2007 - 2012

Guest Lectures

Tutorial: "Complex valued signal processing: Algorithms and application" (a 3-hour tutorial), International Conference on Acoustics, Speech and Signal Processing, Vancouver, 2013

Plenary talk: "Multivariate time-frequency analysis", WIRN Symposium, Vietri, Italy, 2012

Tutorial: "Quaternion valued signal processing", EUSIPCO (European Signal Processing Conference)a 2-hour tutorial, Barcelona, Spain, 2012

Keynote talk. 22nd Italian Workshop on Neural Networks, WIRN, Italian Neural Network Society, Salerno, Italy, 2012

Guest Lecture: "Frequency estimation in smart grid", Siemens Corporate Research, Munich, Germany, 2012

Plenary talk: "Multivariate Empirical Mode Decomposition", Institute for Mathematics and Applications (IMA)A Hot Topics Workshop sponsored by the NSF, Minneapolis, Minnessota, USA, 2011

Participation in Channel 4 programme "Music and Emotion". Three episodes were recorded in San Diego, Princeton, Montreal, Imperial, and Amsterdam. About 15 minutes were shown from our Lab, Channel 4, London, UK, 2011

Tutorial: "Adaptive Signal Processing of Noncircular Complex Signals" (a three-hour tutorial), The 7th IEEE International Symposium on Wireless Communication Systems (ISWCS’10),, York, UK, 2010

Featured talk: "Statistical Signal Processing for Noncircular Complex Signals", IEEE Chapter Denmark, DTU Lyngby, Denmark, 2010

Keynote: "Signal Processing for Vector Sensors: Wind Energy Applications", The Marie Curie SYSWIND winter school, Dublin, 2010

Plenary talk: "Signal Processing for Vector Sensors: Crossroads of Tools and Ideas", International Conference on Signal Processing Systems (ICSPS’10), Dalian, China, 2010

Plenary talk: "Data Driven Time-Frequency Decompositions for Information Fusion”,, European Workshop on Advanced Predicitive Sensory-Motor Control (DRIVESCO),, Judkrante, Lithuania, 2009

Guest Lecture: "Data Driven Time Frequency Analysis", IET & NPL Seminar”Statistical EM Methods for Analysing Complex Systemsand Structures”,, National Physical Laboratories (NPL), London, UK, 2008

Guest Lecture: "Challenges in Complex Valued Signal Processing: Noncircularity and WidelyLinear Models”,, IEEE C&S Society, Chapter Japan, Tokyo Institute of Technology, Tokyo, Japan, 2008

Research Staff

Cheong Took,C

Leong,WY

Looney,D

Xia,Y