Skip to content. Skip to main navigation.

Profiles

avatar

Ioannis D. Schizas received the diploma in Computer Engineering and Informatics (with honors) from the University of Patras, Greece, in 2004, the M.Sc. in Electrical and Computer Engineering from the University of Minnesota, Minneapolis, in 2007 and the Ph. D. in Electrical and Computer Engineering from the University of Minnesota, Minneapolis, in June 2011. Since August 2011 he has been an Assistant Professor at the Electrical Engineering department at the University of Texas at Arlington. His general research interests lie in the areas of statistical signal processing, wireless sensor networks and data dimensionality reduction. Specifically, he is working on distributed signal processing using wireless ad hoc sensor networks, as well as sparsityaware information processing with applications to (distributed) data dimensionality reduction and data clustering

University of Minnesota
PhD
University of Minnesota
MSc
University of Patras, Greece
5-year Diploma
Computer Engineering and Informatics
September 2011
January 2013
Assist Professor
The University of Texas at Arlington
July 2013
Ongoing
Member
IEEE Geoscience and Remote Sensing
September 2011
Ongoing
Member
IEEE SPS
September 2011
Ongoing
Member
IEEE
Data clustering and cleansing

Design and analysis of a generalized algorithmic framework that can extract the different informative portions within a data set that may be large and spatially scattered, while cleansing the data from all corrupted/irrelevant components. Traditional statistical inference techniques rely on a data model which is assumed to characterize the behavior of all the available data, while there may be a few outliers. The proposed research agenda focuses on developing universal algorithms that have learning capabilities and can identify informative portions within a data vector sequence without relying on pre-specified data models. Further, the project involves the design of a novel framework that allows the extraction and elimination of irrelevant data affecting an arbitrary in size data portion. The proposed research will introduce benefits in a wide span of applications including analysis of ecological and climatic data, medical imaging, data mining and machine learning.

Distributed statistical signal processing

Development of distributed signal processing algorithms with applications to statistical inference, denoising, dimensionality reduction and compression.

Wireless sensor networks

Development and analysis of distributed power-aware algorithms for multi-target tracking and denoising using networks of sensors.

2014
G.  Ren, V. Maroulas and I. D. Schizas, ``Joint Distributed Sensor Selection and Multi-Target Tracking,'' IEEE Transactions on Aerospace and Electronic Systems,  submitted Jan. 2014. \newline Available: http://www-ee.uta.edu/Online/schizas/mtt.pdf
Journal Article
Submitted
2013
A. Aduroja, I. D. Schizas and V. Maroulas, ``Distributed Principal Component Analysis in Sensor Networks,'' Proc. of IEEE Intl. Conf. on Acoustics, Speech and Signal Processing (ICASSP) , May 26-31, 2013.
Conference Paper
Published
2013
I. D. Schizas, `` Adaptive Distributed Sparsity-Aware Matrix Decomposition,'' Proc. of IEEE Intl. Conf. on Acoustics, Speech and Signal Processing (ICASSP) , May 26-31, 2013.
Conference Paper
Published
2013
J. Chen and I. D. Schizas, ``Distributed Sparse Canonical Correlation Analysis in Clustering Sensor Data,'' Proc. of the Asilomar Conf. on Signals, Systems and Computers, Pacific Grove, CA, Nov. 2013
Conference Paper
Accepted
2013
I. D. Schizas, ``Distributed Data Cleansing via a Low-Rank Decomposition,'' Proc. of IEEE Global Conf. on Signal and Information Processing (GlobalSIP), Dec.  3-5, 2013.
Conference Paper
Submitted
2013
G. Ren and I. D. Schizas, ``Distributed Sensor-Informative Tracking of Targets,'' Proc. of the IEEE Intl. Workshop on Comp. Advances in Multi-Sensor Adaptive Processing, Saint Martin, Dec. 2013.
Conference Paper
Submitted
2013
I. D. Schizas, `` Distributed Informative-Sensor Identification using Sparsity-Aware Matrix Factorization,'' IEEE Trans. on Sig. Proc., vol. 61, no. 18, pp. 4610--4624, Sep. 2013. 
Journal Article
Published
2012
I. D. Schizas and G. B. Giannakis, ``Covariance-Domain Sparsity for Data Compression and Denoising,'' IEEE Transactions on Signal Processing, May 2012.
Journal Article
Published
2012
I. D. Schizas, ``Distributed Informative-Sensor Determination via  Sparsity-Cognizant Matrix Decomposition,'' Proc. of IEEE Workshop on Statistical Signal Processing, August 5-8, 2012.
Conference Paper
Published
2011
I. D. Schizas and G. B. Giannakis, ``Covariance-Domain Sparsity for Data Compression and Denoising,'' IEEE Transactions on Signal Processing.
Journal Article
Submitted
2011
 I. D. Schizas and G. B. Giannakis, `` Eigenspace Sparsity for Compression and Denoising,'' Proc. of Intl. Conf. on Acoustics, Speech and Signal Processing, Prague, Czech Republic, May 22-27, 2011. 
Conference Paper
Published
2010
A. Ribeiro, I. D. Schizas, S. I. Roumeliotis and G. B. Giannakis, ``Kalman Filtering in Wireless Sensor Networks: Incorporating communication cost in state estimation problems,'' IEEE Control Systems Magazine, April 2010.
Journal Article
Published
2009
H. Zhu, I. D. Schizas and G. B. Giannakis, ``Power-Efficient Dimensionality Reduction for Distributed Channel-Aware Kalman Tracking Using Wireless Sensor Networks,'' IEEE Transactions on Signal Processing, August 2009. 
Journal Article
Published
2009
I. D. Schizas, G. Mateos and G. B. Giannakis, `` Distributed LMS for Consensus-Based In-Network Adaptive Processing,'' IEEE Transactions on Signal Processing, June 2009.
Journal Article
Published
2009
G. Mateos, I. D. Schizas and G. B. Giannakis, ``Distributed Recursive Least-Squares for Consensus-Based In-Network Adaptive Estimation,'' IEEE Transactions on Signal Processing, Nov. 2009.
Journal Article
Published
2009
G. Mateos, I.D. Schizas and G. B. Giannakis, ``Performance Analysis of the Consensus-Based Distributed LMS Algorithm,'' EURASIP Journal on Advances in Signal Processing, Oct. 2009.
Journal Article
Published
2009
G. Mateos, I. D. Schizas and G. B. Giannakis, `` Closed-form MSE perfomance of the distributed LMS algorithm,''Proc. of 13th DSP Workshop, Marco Island, FL, January 4-7, 2009.
Conference Paper
Published
2009
I. D. Schizas, G. B. Giannakis and N. D. Sidiropoulos, ``Exploiting Covariance-domain Sparsity for Dimensionality Reduction,'' Proc. of 3rd Intl. Workshop on Comp. Advances in Multi-Sensor Adapt. Proc., ArubaIsland, Dec. 13-16, 2009.
Conference Paper
Published
2008
I. D. Schizas, G. B. Giannakis and N. Jindal, ``Distortion-Rate Bounds for Distributed Estimation with Wireless Sensor Networks, EURASIP Journal on Advances in Signal Processing, 2008.
Journal Article
Published
2008
I. D. Schizas, A. Ribeiro and G. B. Giannakis, `` Consensus in Ad Hoc WSNs with Noisy Links - Part I: Distributed Estimation of Deterministic Signals,'' IEEE Transactions on Signal Processing, January 2008.
Journal Article
Published
2008
I. D. Schizas, G. B. Giannakis, S. I. Roumeliotis and A. Ribeiro, ``Consensus in Ad Hoc WSNs with Noisy Links - Part II: Distributed Estimation and Smoothing of Random Signals,'' IEEE Transactions on Signal Processing, April 2008.
Journal Article
Published
2008
I. D. Schizas,G. Mateos and G. B. Giannakis, `` Stability analysis of the consensus-based distributed LMS algorithm,''Proc. of Intl. Conf. on Acoustics, Speech and Signal Processing, Las Vegas, NV, March 30-April 4, 2008.
Conference Paper
Published
2007
A. Ribeiro,I.D. Schizas, J.-J. Xiao, G. B. Giannakis, and Z.-Q. Luo '' Distributed Estimation Under Bandwidth and Energy Constraints,'' in Wireless Sensor Networks: Signal Processing and Communications Perspectives (A. Swami, Q. Zhao, Y. Hong, and L. Tong, eds.), Wiley, February 2007.
Book Chapter
Published
2007
I. D. Schizas, G. B. Giannakis and Z.-Q. Luo, ``Distributed Estimation Using Reduced-Dimensionality Sensor Observations,'' IEEE Transactions on Signal Processing, August 2007.
Journal Article
Published
2007
I. D. Schizas,A. Ribeiro and G. B. Giannakis, `` Consensus-Based Distributed Parameter Estimation in Ad Hoc Wireless Sensor Networks with Noisy Links,''Proc. of Intl. Conf. on Acoustics, Speech and Signal Processing, Honolulu, HI, April 15-20, 2007.
Conference Paper
Published
2007
I. D. Schizas,G. B. Giannakis and A. Ribeiro, ``Distributed MAP and LMMSE Estimation of Random Signals Using Ad Hoc Wireless Sensor Networks with Noisy Links,'' Proc. of SPAWC, Helsinki, Finland, June 17- 20, 2007.
Conference Paper
Published
2007
 I. D. Schizas,G. B. Giannakis, Stergios I. Roumeliotis and A. Ribeiro, ``Any-time Optimal Distributed Kalman Filtering and Smoothing,'' Proc. of Wrkshp. on Statistical Signal Processing, Madison, WI, August 26-29, 2007.
Conference Paper
Published
2007
 H. Zhu, I. D. Schizas andG. B. Giannakis, `` Power-Efficient Dimensionality Reduction for Distributed Channel-Aware Kalman Tracking Using Wireless Sensor Networks,'' Proc. of Wrkshp. on Statistical Signal Processing, Madison, WI, August 26-29, 2007.
Conference Paper
Published
2007
 G. Mateos, I. D. Schizas and G. B. Giannakis, ``Distributed Least-Mean Square Algorithm Uisng Wireless Ad Hoc Networks,'' Proc. of 45th Allerton Conf., Univ. of Illinois at U-C, Monticello, IL, Sept. 26-28, 2007.
Conference Paper
Published
2007
I. D. Schizas, G. Mateos and G. B. Giannakis, ``Distributed Recursive Least-Squares Using Wireless Ad Hoc Sensor Networks,'' Proc.of 41st Asilomar Conf. on Signals, Systems, and Computers, Pacific Grove, CA, Nov. 4-7, 2007.
Conference Paper
Published
2006
I. D. Schizas, A. Ribeiro, and G. B. Giannakis '' Dimensionality Reduction, Compression and Quantization for Distributed Estimation with Wireless Sensor Networks,'' in Wireless Communications (P. Agrawal, D. M. Andrews, P. J. Fleming, G. Yin, and L. Zhang, eds.), vol. 143 of IMA Volumes in Mathematics and its Applications, pp. 259--296, Springer, New York, 2006. 
Book Chapter
Published
2006
I. D. Schizas, G. B. Giannakis and Z.-Q. Luo, ``Optimal Dimensionality Reduction for Multi-Sensor Fusion in the Presence of Fading and Noise,'' Proc. of Intl. Conf. on Acoustics, Speech and Signal Processing, Toulouse, France, May 15-19, 2006. 
Conference Paper
Published
2006
I. D. Schizas, A. Ribeiro and G. B. Giannakis, ``Distributed Estimation with Ad-Hoc Wireless Sensor Networks,'' Proc. of XIV EuropeanConf. Signal Processing Conference, Florence, Italy, Sep. 4-8, 2006.
Conference Paper
Published
2006
I. D. Schizas and G. B. Giannakis, `` Consensus-Based Distributed Estimation of Random Signals with Wireless Sensor Networks,'' Proc. of 40th Asilomar Conf. on Signals, Systems, and Computers, Pacific Grove, CA, Oct. 29-Nov. 1, 2006.
Conference Paper
Published
2005
I. D. Schizas, G. B. Giannakis and N. Jindal, ``Distortion-Rate Analysis for Distributed Estimation with Wireless Sensor Networks,'' Proc. Of 43rd Allerton Conf., Univ. of Illinois at U-C, Monticello, IL, Sept. 28-30, 2005.
Conference Paper
Published
2005
I. D. Schizas, G. B. Giannakis and Z.-Q. Luo, ``Distributed Estimation Using Reduced Dimensionality Sensor Observations,'' Proc. of 39th Asilomar Conf. on Signals, Systems, and Computers, Pacific Grove, CA, Oct. 30-Nov. 2, 2005.
Conference Paper
Published
November 2013

Distributed Informative-Sensor Identification and Tracking via Sparsity-Aware Covariance Factorization

IEEE
Antennas and Propagation Society Technical Seminar, Fort Worth Chapter, Nov. 2013.

Invited
April 2013

Distributed Determination of Informative Network Nodes via Sparsity-Cognizant Covariance Decomposition

Invited
March 2013

Distributed Informative-Sensor Identification via Sparsity-Aware Covariance Decomposition

Invited
September 2012
Ongoing
Distributed Informative Sensor Selection via Sparse Covariance Factorization
Other
Not Specified
Wireless sensor networks (WSNs) are widely used in applications such as environmental monitoring, surveillance and health monitoring of large structures. A major challenge when using WSNs is their limited life expectancy. However, in practice the phenomena of interest in a sensed field are quite localized and affect a small number of sensors. This property of locality induces sparsity in sensor data statistical descriptors such as the covariance matrix. This project involves the development and analysis of efficient algorithms that uncover sparse structures in the sensor data covariance, and determine in a distributed fashion which sensors acquire informative measurements and have to remain active. Identifying the `informative' sensors can lead to significant energy savings. The research focuses on the novel utilization of covariance sparsity in developing distributed informative sensor selection algorithms that have the ability to adaptively learn the statistical behavior of the sensed field, without relying on a priori known data models. The investigators consider the design and performance analysis of a generalized sparsity-aware framework that is capable to analyze the sensor data covariance matrix into sparse factors. Norm-one regularization mechanisms are employed to estimate and identify the support (nonzero entries) of the unknown sparse covariance factors. The factors' support is used to identify the informative sensors. A novel blending of the sparsity-based factorization techniques with distributed optimization tools is also studied to obtain distributed algorithms that enable sensors to collaboratively learn the underlying sparse covariance structure, even with noisy inter-sensor communications and overlapping sources. The research also involves adaptive implementations that track online the set of informative sensors in non-stationary settings formed by mobile sources, using stochastic approximation toolboxes properly tuned to exploit sparsity.
September 2012 -
August 2015
Distributed Informative Sensor Selection via Sparse Covariance Factorization
$205,662
June 2012 -
May 2013
Energy-Efficient Sensor Networks via Distributed Active Sensor Selection
$10,000
Ongoing
Guohua is working on distributed multi-target tracking problems.
Doctoral
Ongoing
Jia is working on sparsity-aware signal processing with applications to big data processing.
Doctoral
Ongoing
Abiodun is working on distributed data denoising problems.
Master's
Spring 2015
EE 5369 - Distributed Estimation Theory
Office Hours (also by appointment)
DayStartEnd
Tuesday3:30PM4:30PM
Thursday3:30PM4:30PM
The course presents major theoretical toolboxes for designing estimators and analyzing their performance. Specifically, the course will touch upon Cramer-Rao bound theory and present important estimators such as the maximum likelihood estimator, least-squares estimator and minimum mean-square error estimator to name a few. Tracking of time-varying processes and online estimation techniques will also be considered. After traditional theory is covered, the focus will move in distributed estimation with applications in sensor networks. Decentralized optimization tools such as the alternating direction method of multipliers will be studied and applied in deriving distributed estimators and tracking algorithms. Different modern distributed techniques will be considered and compared along with applications in sensing, data compression, data denoising and multi-target tracking.  The goal of this course is to help graduate students acquire the necessary theoretical background to tackle estimation problems that appear in many engineering applications, especially in networks of sensors.
Last Updated on December 31, 2014, 2:04 pm
Fall 2014
EE 5362 - Digital Communications
Office Hours (also by appointment)
DayStartEnd
Tuesday3:00PM4:00PM
Thursday3:00PM4:00PM
The course presents fundamental principles underlying the transmission and reception of digital information, and studies the different parts of a modern digital communication system. Specifically, the course will touch upon different digital modulation schemes, as well as design and performance analysis of optimum receivers for additive white Gaussian noise (AWGN) channels. Some concepts of information theory and channel coding will also be studied. Further, techniques for carrier and symbol synchronization will be presented. Communication over bandlimited channels will also be explored, and the effects of intersymbol interference (ISI) and channel equalization techniques will be studied. The goal of this course is to help graduate students acquire the necessary theoretical background to i) understand the components of a digital communication system, ii) be able to design a digital communication system, and iii) analyze its performance both analytically and numerically.
Last Updated on August 8, 2014, 11:54 am
Spring 2014
EE 5369 - Ee 5369-001
The course presents major theoretical toolboxes for designing estimators and analyzing their performance. Specifically, the course will touch upon Cramer-Rao bound theory and present important estimators such as the maximum likelihood estimator, least-squares estimator and minimum mean-square error estimator to name a few. Tracking of time-varying processes and online estimation techniques will also be considered. After traditional theory is covered, the focus will move in distributed estimation with applications in sensor networks. Decentralized optimization tools such as the alternating direction method of multipliers will be studied and applied in deriving distributed estimators and tracking algorithms. Different modern distributed techniques will be considered and compared along with applications in sensing, data compression, data denoising and multi-target tracking.  The goal of this course is to help graduate students acquire the necessary theoretical background to tackle estimation problems that appear in many engineering applications, especially in networks of sensors.
Last Updated on April 24, 2014, 5:08 pm
Fall 2013
EE 5362 - Digital Communications
Office Hours
DayStartEnd
Tuesday3:00PM4:00PM
Thursday3:00PM4:00PM
The course presents fundamental principles underlying the transmission and reception of digital information, and studies the different parts of a modern digital communication system. Specifically, the course will touch upon different digital modulation schemes, as well as design and performance analysis of optimum receivers for additive white Gaussian noise (AWGN) channels. Some concepts of information theory and channel coding will also be studied. Further, techniques for carrier and symbol synchronization will be presented. Communication over bandlimited channels will also be explored, and the effects of intersymbol interference (ISI) and channel equalization techniques will be studied. The goal of this course is to help graduate students acquire the necessary theoretical background to i) understand the components of a digital communication system, ii) be able to design a digital communication system, and iii) analyze its performance both analytically and numerically.
Last Updated on August 7, 2013, 4:36 pm
Fall 2013
EE 5362 - Digital Communications-Web Session
Office Hours
DayStartEnd
Tuesday3:00PM4:00PM
Thursday3:00PM4:00PM
The course presents fundamental principles underlying the transmission and reception of digital information, and studies the different parts of a modern digital communication system. Specifically, the course will touch upon different digital modulation schemes, as well as design and performance analysis of optimum receivers for additive white Gaussian noise (AWGN) channels. Some concepts of information theory and channel coding will also be studied. Further, techniques for carrier and symbol synchronization will be presented. Communication over bandlimited channels will also be explored, and the effects of intersymbol interference (ISI) and channel equalization techniques will be studied. The goal of this course is to help graduate students acquire the necessary theoretical background to i) understand the components of a digital communication system, ii) be able to design a digital communication system, and iii) analyze its performance both analytically and numerically.
Last Updated on August 7, 2013, 4:38 pm
Spring 2013
EE 5362 - DIGITAL COMMUNICATIONS
The course presents fundamental principles underlying the transmission and reception of digital information, and studies the different parts of a modern digital communication system. Specifically, the course will touch upon different digital modulation schemes, as well as design and performance analysis of optimum receivers for additive white Gaussian noise (AWGN) channels. Some concepts of information theory and channel coding will also be studied. Further, techniques for carrier and symbol synchronization will be presented. Communication over bandlimited channels will also be explored, and the effects of intersymbol interference (ISI) and channel equalization techniques will be studied. The goal of this course is to help graduate students acquire the necessary theoretical background to i) understand the components of a digital communication system, ii) be able to design a digital communication system, and iii) analyze its performance both analytically and numerically.
Last Updated on August 7, 2013, 4:03 pm
Fall 2012
EE 5368 - WIRELESS COMMUNICATION SYSTEMS
The course presents fundamental principles underlying the wireless transmission and reception of information, and studies the different parts of a modern wireless communication system. Specifically, the course will touch upon different digital modulation schemes, as well as the design and performance analysis of a transmission and reception end. The concept of diversity and its impact on reception performance (probability of symbol detection error) will be discussed. Channel capacity and channel coding will also be studied. Further, techniques for adaptive modulation and channel equalization used in state-of-the-art wireless systems will be presented. Communication using orthogonal frequency division multiplexing (OFDM), as well as spread spectrum techniques will also explored. Topics in multi-user systems, random access, cellular systems and ad hoc networks will also be covered. The goal of this course is to help graduate students to i) learn about different wireless communication technologies; ii) understand the basic components of a wireless communication system; iii) be able to design basic components in a wireless communication system; and iv) analyze its performance both analytically and numerically.
Last Updated on August 7, 2013, 4:03 pm
Spring 2012
EE 5362 - DIGITAL COMMUNICATIONS
The course presents fundamental principles underlying the transmission and reception of digital information, and studies the different parts of a modern digital communication system. Specifically, the course will touch upon different digital modulation schemes, as well as design and performance analysis of optimum receivers for additive white Gaussian noise (AWGN) channels. Some concepts of information theory and channel coding will also be studied. Further, techniques for carrier and symbol synchronization will be presented. Communication over bandlimited channels will also be explored, and the effects of intersymbol interference (ISI) and channel equalization techniques will be studied. The goal of this course is to help graduate students acquire the necessary theoretical background to i) understand the components of a digital communication system, ii) be able to design a digital communication system, and iii) analyze its performance both analytically and numerically.
Last Updated on August 7, 2013, 4:03 pm
Fall 2011
EE 5369 - Estimation Theory
The course presents major theoretical toolboxes for designing estimators and analyzing their performance. Specifically, the course will touch upon Cramer-Rao bound theory and present important estimators such as the maximum likelihood estimator, least-squares estimator and minimum mean-square error estimator to name a few. Tracking of time-varying processes and online estimation techniques will also be considered. The goal of this course is to help graduate students acquire the necessary theoretical background to tackle estimation problems that appear in many engineering applications.
Last Updated on August 7, 2013, 4:03 pm
August 2004
Ongoing
Reviewer for Journals
    IEEE Transactions on Signal Processing, IEEE Journal Of Selected Topics of Signal Processing IEEE Transactions on Image Processing, IEEE Transactions on Information Theory, IEEE Transactions on Communications, IEEE Journal on Selected Areas in Communications, IEEE Communications Letters, IEEE Signal Processing Letters, IEEE Signal Processing Magazine, IEEE Transactions on Robotics, Elsevier Signal Processing, EURASIP Journal on Advances in Signal Processing, EURASIP Journal on Wireless Communications and Networking
Volunteered
December -
Reviewer for Journals
IEEE Transactions on Signal Processing, IEEE Journal Of Selected Topics of Signal Processing IEEE Transactions on Image Processing, IEEE Transactions on Information Theory, IEEE Transactions on Communications, IEEE Journal on Selected Areas in Communications, IEEE Communications Letters, IEEE Signal Processing Letters, IEEE Signal Processing Magazine, IEEE Transactions on Robotics, Elsevier Signal Processing, EURASIP Journal on Advances in Signal Processing, EURASIP Journal on Wireless Communications and Networking
Uncategorized