Last modified:   01.11.2021

Ionuț Fîciu is currently PhD researcher at Doctoral School of Electronics, Telecommunications & Information Technology, University Politehnica of Bucharest since October 2020 and his main research interest consists on efficient algorithms for acoustic applications. He received in 2018 his B. Sc. degree from the Faculty of Electronics, Telecommunications, and Information Technology, Politehnica University of Bucharest, Romania as a valedictorian, and he also has a master’s degree in Advanced Digital Imaging Techniques obtained with the dissertation thesis called "Deep neural networks for environmental sounds classification", obtained in 2020. Since 2018 he also works as a consultant software developer in the IT industry.

Currently, Ionuț is involved in the D.A.S.T.I. research project with the main interest focused on a new approach that exploits the impulse response decomposition based on the nearest Kronecker product and low-rank approximations, which fits very well for sparse system identification problems, having the following goals: developing the convergence analysis of the decomposition-based algorithms, developing multidimensional decomposition-based algorithms, developing computationally efficient versions of the RLS-NKP algorithms, and developing multichannel decomposition-based algorithms.

During his B. Sc. Ionuț took part in the “IoTurk” research project, focused on Image Outlining and Tagging by Unsupervised Refinement Kernel as a Machine Learning & Computer Vision Technician. This project was developed in partnership with the Faculty of Electronics, Telecommunications, and Information Technology, Politehnica University of Bucharest. The project scope included the implementation of a new solution for pixel-level semi-automation annotation of large datasets, without any constraints based on the application of the number of identified classes. During this project he published his first research article, below-mentioned.

For 5 years he was also an involved volunteer in Electronics Students League (LSE) - Education Department. Participation in different volunteering actions, including:

· Tutoring (2015-2019): guiding younger students by providing useful materials and information related to the learning process;

· Volunteering to the entrance exam simulation, 2015 edition - responsible check-in and payments, 2016 and 2017 editions - member of the general committee;

· Presentation of the faculty of Electronics, Telecommunications and Information Technology in high schools, 2015 and 2016 editions.

Grades:

· B.Sc. diploma exam average: 10.00

· B.Sc. graduation average: 9.53 (valedictorian distinction)

· M.Sc. dissertation exam average: 10.00

· M.Sc. graduation average: 9.70

Contact:

Email: ionut.ficiu22@gmail.com

Research interests:

Adaptive filtering algorithms, adaptive systems; sparse system identification; acoustic signal processing; adaptive noise control; network and acoustic echo cancellation; recurrent neural networks.

Programming/configuration and testing technologies:

C/C++, C#, Python 3.x, Matlab, SQL (and associated tools/ IDEs/ development frameworks).

Journal papers:

1. I.-D. Fîciu, C.L. Stanciu, C. Anghel, and C. Elisei-Iliescu, “Low-Complexity Recursive Least-Squares Adaptive Algorithm Based on Tensorial Forms,” Applied Sciences, vol. 11(18), p. 8656, September 2021, DOI: 10.3390/app11188656, WOS: 000699216700001– ISI Q2.

Conference paper:

1. I.-D. Ficiu, C. Stanciu, C. Anghel, C. Paleologu, and L. Stanciu, “Combinations of Adaptive Filters within the Multilinear Forms”, in Proc. International Symposium on Signals, Circuits and Systems (ISSCS), 15-16 July, 2021.

2. I.-D. Ficiu, L.-M. Dogariu, C.-L. Stanciu, and C. Paleologu, “Identification of multilinear forms using combinations of adaptive algorithms,” in Proc. the Sixth International Conference on Advances in Signal, Image and Video Processing (SIGNAL), Valencia, Spain, June 2021.

3. I.-D. Ficiu, C. Elisei-Iliescu C.-L. Stanciu, and C. Paleologu, “ Variable-Regularized Low-Complexity RLS Adaptive Algorithms for Bilinear Forms,” in Proc. the The Thirteenth International Conference on Adaptive and Self-Adaptive Systems and Applications (ADAPTIVE), Porto, Portugal, 18-22 April, 2021.

4. I. Ficiu, R. Stilpeanu, C. Toca, A. Petre, C. Patrascu and M. Ciuc, “Automation Annotation of Object Instances by Region-Based Recurrent Neural Networks”, 2018 IEEE 14th International Conference on Intelligent Computer Communication and Processing (ICCP 2018), Sep. 2018, Cluj-Napoca, Romania.

Research grants and fellowships

1. Now: Research project (as PhD researcher): D.A.S.T.I. (New Decomposition-Based Algorithms for Sparse System Identification) – Funding Institution: UEFISCDI, Project code: PN-III-P1-1.1-TE-2019-0529 (http://www.comm.pub.ro/dasti/).

2. 2017 – 2018: Research project (as machine learning & computer vision technician): IoTurk (Image Outlining and Tagging by Unsupervised Refinement Kernel) – Funding Institution: UEFISCDI, Project code: PN-III-P2-2.1-PED-2016-0292 (http://imag.pub.ro/ioturk/).

Collaborators:

· Cristian Lucian STANCIU: http://www.comm.pub.ro/cstanciu/

· Constantin PALEOLOGU: www.comm.pub.ro/paleologu

· Cristian ANGHEL: www.comm.pub.ro/canghel

· Camelia ELISEI-ILIESCU: www.comm.pub.ro/camelia_elisei