- Ottawa, Ontario, Canada
- Michel K. Kulhandjian is a Research Scientist in the School of Electrical Engineering and Computer Science at the Uni... moreMichel K. Kulhandjian is a Research Scientist in the School of Electrical Engineering and Computer Science at the University of Ottawa. He is currently working on investigating new encoding and decoding schemes for Sparse Code Multiple Access (SCMA) in collaboration with Prof. Claude D’amours. He is also employed as an embedded software designer at Harris Corporation in Ottawa, where he is working on designing and developing software platform for Air-to-Ground, Ground-to-Ground IP-based ATM voice communication systems. He received his M.S. and Ph.D. degrees in Electrical Engineer from the State University of New York at Buffalo in 2007 and 2012, respectively. He had previously received his B.S. degree in Electronics Engineering and Computer Science (Minor), with “Summa Cum Laude” from the American University in Cairo (AUC) in 2005. He was employed at Alcatel-Lucent, in Ottawa, Ontario, in 2012, to conduct Redit
Research Interests:
In this work, we develop a theoretical framework for reliable digital recording system identification from digital audio files alone, for forensic purposes. A digital recording system consists of a microphone and a digital sound... more
In this work, we develop a theoretical framework for reliable digital recording system identification from digital audio files alone, for forensic purposes. A digital recording system consists of a microphone and a digital sound processing card. We view the cascade as a system of unknown transfer function. We expect the same manufacturer and model microphone-sound card combinations to have very similar/near identical transfer functions, bar any unique manufacturing defect. Input voice (or other) signals are modeled as non-stationary processes. The technical problem under consideration becomes blind deconvolution with non-stationary inputs, as it manifests itself in the specific application of digital audio recording equipment classification. Experimental results demonstrate over 99.2% accuracy in identification of the recording devices.
Research Interests:
Research Interests:
In this paper, we propose a fast decoder algorithm for uniquely decodable (errorless) code sets for overloaded synchronous optical code-division multiple-access (O-CDMA) systems. The proposed decoder is designed in a such a way that the... more
In this paper, we propose a fast decoder algorithm for uniquely decodable (errorless) code sets for overloaded synchronous optical code-division multiple-access (O-CDMA) systems. The proposed decoder is designed in a such a way that the users can uniquely recover the information bits with a very simple decoder, which uses only a few comparisons. Compared to maximum-likelihood (ML) decoder, which has a high computational complexity for even moderate code lengths, the proposed decoder has much lower computational complexity. Simulation results in terms of bit error rate (BER) demonstrate that the performance of the proposed decoder for a given BER requires only 1-2 dB higher signal-to-noise ratio (SNR) than the ML decoder.
Research Interests:
In this work, we develop a theoretical framework for reliable digital recording system identification from digital audio files alone, for forensic purposes. A digital recording system consists of a microphone and a digital sound... more
In this work, we develop a theoretical framework for reliable digital recording system identification from digital audio files alone, for forensic purposes. A digital recording system consists of a microphone and a digital sound processing card. We view the cascade as a system of unknown transfer function. We expect the same manufacturer and model microphone-sound card combinations to have very similar/near identical transfer functions, bar any unique manufacturing defect. Input voice (or other) signals are modeled as non-stationary processes. The technical problem under consideration becomes blind deconvolution with non-stationary inputs, as it manifests itself in the specific application of digital audio recording equipment classification. Experimental results demonstrate over 99.2% accuracy in identification of the recording devices.
Research Interests:
Research Interests:
Research Interests:
Research Interests: Image Processing, Data Mining, Digital Media, Iterative Methods, Statistical Analysis, and 15 moreAuthentication, Cross Correlation, Correlation, Data Hiding, IEEE, Information Hiding, Reliability, Spread Spectrum, Bit Error Rate, Autocorrelation, Spread Spectrum Communication, Digital Signatures, Steganalysis, Image Steganography and Steganalysis, and Least squares method
ABSTRACT
Research Interests:
Research Interests:
... mail: {mingli,mkk6,pados,batalama}@buffalo.edu Michael J. Medley Air Force Research Laboratory/RIGF 525 Brooks Rd Rome, NY 13441 E-mail: michael.medley@rl.af.mil ABSTRACT We consider the problem of passive spread ...
Research Interests:
In the last lecture, we discussed unique decoding of RS codes and briefly went through the Berlekamp-Welch algorithm. In today’s lecture we will study Berlekamp-Welch algorithm in more detail. Recall that the [n, k, n−k+1]q Reed-Solomon... more
In the last lecture, we discussed unique decoding of RS codes and briefly went through the Berlekamp-Welch algorithm. In today’s lecture we will study Berlekamp-Welch algorithm in more detail. Recall that the [n, k, n−k+1]q Reed-Solomon code regards a message as a polynomial P (X) of a degree at most k−1, and the encoding of a message m = (m0,..., mk−1) is (P (α1),..., P (αn)). Here, mi ∈ Fq, k ≤ n ≤ q, and P (X) = ∑ k−1 i=0 miX i. Now let us look at the decoding problem of Reed-Solomon codes. Suppose we are given distinct values α1,..., αn where αi ∈ Fq with received word y = (y1,..., yn) ∈ F n q and parameters k and e < n−k+1 2, where e is an upper bound on the number of errors which occurred during transmission. Our goal is to find a polynomial P (X) ∈ Fq[X] of degree at most k − 1, such that P (αi) ̸ = yi for at most e values of i ∈ [n] (assuming such a P (X) exists). Although this problem is quite non-trivial one, a polynomial time solution can be found for this problem. Thi...
Last lecture we saw the proof of the Singleton bound which claims that for any (n, k, d)q code, k ≤ n − d + 1. In today’s lecture we will study Reed-Solomon codes. These codes meet the Singleton bound, i.e. satisfy k = n − d + 1 (but have... more
Last lecture we saw the proof of the Singleton bound which claims that for any (n, k, d)q code, k ≤ n − d + 1. In today’s lecture we will study Reed-Solomon codes. These codes meet the Singleton bound, i.e. satisfy k = n − d + 1 (but have the unfortunate property that q ≥ n). Note that this implies that the Singleton bound is tight, at least for q ≥ n. 1 Reed-Solomon Codes We begin with the definition of Reed-Solomon codes. Definition 1.1 (Reed-Solomon code). Let Fq be a finite field and Fq[x] denote the Fq-space of univariate polynomials where all the coefficients of x are from Fq. Pick {α1, α2,...αn} distinct elements (also called evaluation points) of Fq and choose n and k such that k ≤ n ≤ q. We define an encoding function for Reed-Solomon code as RS: Fk q → Fnq as follows. A message m = (m0, m1,..., mk−1) with mi ∈ Fq is mapped to a degree k − 1 polynomial. where m ↦ → fm(x), ∑k−1 fm(x) = mix i. (1)
In this paper, we present some of the interesting questions raised in the “5G and Beyond” workshop discussions. We propose to explore alternative solutions in other discipline such as algebraic combinatorics and discrete optimization.... more
In this paper, we present some of the interesting questions raised in the “5G and Beyond” workshop discussions. We propose to explore alternative solutions in other discipline such as algebraic combinatorics and discrete optimization. Construction of matrices that possess certain desirable properties in discrete tomography can be exploited for the 5th generation (5G) multiuser waveform design. Moreover, we propose to investigate thoroughly the random graph theory developed by Erdős and A. Rényi for 5G multiuser waveform design. Furthermore, we explore using the “Best Match” algorithm, which has a polynomial time complexity, in 5G communications.
Research Interests:
In this paper, we propose a waveform design that is robust to noncoherent detection for quadrature phase-shift keying (QPSK) and quadrature amplitude modulation (QAM) schemes. Our proposed complex constellation waveforms are based on the... more
In this paper, we propose a waveform design that is robust to noncoherent detection for quadrature phase-shift keying (QPSK) and quadrature amplitude modulation (QAM) schemes. Our proposed complex constellation waveforms are based on the quaternary complex Hadamard matrices. The advantage of the proposed waveforms is that they have the largest noncoherent minimum distance in the complex domain and meet the maximum Hamming distance in their equivalent binary domain.
Research Interests:
We present our ongoing work on developing a multi-source energy harvesting buoy that could be used to power underwater acoustic sensor nodes deployed on the surface of a sea or an ocean. The buoy harvests energy from three different... more
We present our ongoing work on developing a multi-source energy harvesting buoy that could be used to power underwater acoustic sensor nodes deployed on the surface of a sea or an ocean. The buoy harvests energy from three different sources; wind, solar and wave energy. The collected energies are combined optimally through a power management circuit, which can deliver the harvested energy either to a gateway sensor node or replenish its rechargeable battery for future use. A prototype of the multi-source energy harvester is presented with some preliminary results.
Research Interests:
: This paper considers the problem of blind active spreadspectrum (SS) steganalysis defined as the extraction of hidden data with no prior information. We first develop a multisignature iterative generalized least-squares (M-IGLS) core... more
: This paper considers the problem of blind active spreadspectrum (SS) steganalysis defined as the extraction of hidden data with no prior information. We first develop a multisignature iterative generalized least-squares (M-IGLS) core procedure to seek unknown messages hidden in image hosts via multi-signature direct-sequence spread-spectrum embedding. Neither the original host nor the embedding signatures are assumed available. Then, cross-correlation enhanced MIGLS (CC-M-IGLS), a procedure described herein in detail that is based on statistical analysis of repeated independent M-IGSL processing of the host, is seen to offer most effective hidden message recovery. In fact, experimental studies show that the proposed CC-M-IGLS active SS steganalysis algorithm can achieve probability of error close to what may be attained with known embedding signatures and host autocorrelation matrix.
Research Interests:
Improved low-density spreading (LDS) code designs based on the Gaussian separability criterion are conceived. We show that the bit error rate (BER) hinges not only on the minimum distance criterion, but also on the average Gaussian... more
Improved low-density spreading (LDS) code designs based on the Gaussian separability criterion are conceived. We show that the bit error rate (BER) hinges not only on the minimum distance criterion, but also on the average Gaussian separability margin. If two code sets have the same minimum distance, the code set having the highest Gaussian separability margin will lead to a lower BER. Based on the latter criterion, we develop an iterative algorithm that converges to the best known solution having the lowest BER. Our improved LDS code set outperforms the existing LDS designs in terms of its BER performance both for binary phase-shift keying (BPSK) and for quadrature amplitude modulation (QAM) transmissions. Furthermore, we use an appallingly low-complexity minimum mean-square estimation (MMSE) and parallel interference cancellation (PIC) (MMSE-PIC) technique, which is shown to have comparable BER performance to the potentially excessive-complexity maximum-likelihood (ML) detector. T...
Research Interests:
In this paper, we examine the bit error rate (BER) performance of low-density spreading (LDS) systems employing the minimum mean square error (MMSE) and parallel interference cancellation (MMSE-PIC) detector in frequency-selective fading... more
In this paper, we examine the bit error rate (BER) performance of low-density spreading (LDS) systems employing the minimum mean square error (MMSE) and parallel interference cancellation (MMSE-PIC) detector in frequency-selective fading channels while additionally proposing a precoding scheme for multipath channels which effectively transforms the multipath channel into a single path channel. Numerical studies illustrate that the precoding scheme is effective and that reliable communication is possible, even in harsh multipath conditions.
Research Interests:
Research Interests:
We consider the problem of designing a fast decoder for antipodal uniquely decodable (errorless) code sets for overloaded synchronous code-division multiple access (CDMA) systems where the number of signals K_{max}^a is the largest known... more
We consider the problem of designing a fast decoder for antipodal uniquely decodable (errorless) code sets for overloaded synchronous code-division multiple access (CDMA) systems where the number of signals K_{max}^a is the largest known for the given code length L. The proposed decoder is designed in a such a way that the users can uniquely recover the information bits with a very simple decoder, which uses only a few comparisons. Compared to maximum-likelihood (ML) decoder, which has a high computational complexity for even moderate code length, the proposed decoder has a much lower computational complexity. Simulation results in terms of bit error rate (BER) demonstrate that the performance of the proposed decoder only has a 1-2 dB degradation at BER of 10^{-3} when compared to ML.
Research Interests:
We develop a two-dimensional (2-D) direction-of- arrival (DOA) estimation scheme for coherent wideband source signals using coherent signal subspace method based auxiliary-vector (CSSM-AV) basis. Computation of the basis is carried out by... more
We develop a two-dimensional (2-D) direction-of- arrival (DOA) estimation scheme for coherent wideband source signals using coherent signal subspace method based auxiliary-vector (CSSM-AV) basis. Computation of the basis is carried out by a modified version of the orthogonal CSSM-AV filtering algorithm. The proposed method reconstructs the signal subspace using a cross-correlation matrix after which the modified CSSM-AV algorithm is employed to estimate the azimuth and elevation angles. Then, successive orthogonal maximum cross- correlation auxiliary vectors are calculated to form a basis for the scanner-extended signal subspace. This technique is very efficient in reducing the algorithm complexity. Since it does not require that the eigenvectors be determined in order to find the signal subspace and yields a superior resolution performance for closely spaced sources even when the number of samples is low. Specifically, the complexity of the proposed 2-D DOA estimation algorithm com...
Research Interests:
We begin with the definition of Reed-Solomon codes. Definition 1.1 (Reed-Solomon code) . Let Fq be a finite field andFq[x] denote theFq-space of univariate polynomials where all the coefficients of x are fromFq. Pick {α1, α2, ...αn}... more
We begin with the definition of Reed-Solomon codes. Definition 1.1 (Reed-Solomon code) . Let Fq be a finite field andFq[x] denote theFq-space of univariate polynomials where all the coefficients of x are fromFq. Pick {α1, α2, ...αn} distinct elements (also calledevaluation points ) of Fq and choosen and k such thatk ≤ n ≤ q. We define an encoding function for Reed-Solomon code as RS : Fq → F n q as follows. A message m = (m0, m1, ..., mk−1) with mi ∈ Fq is mapped to a degree k − 1 polynomial.
In this paper, we study American sign language (ASL) hand gesture recognition using Doppler radar. A set of ASL hand gesture motions are captured as micro- Doppler signals using a microwave X-band Doppler radar transceiver. We apply joint... more
In this paper, we study American sign language (ASL) hand gesture recognition using Doppler radar. A set of ASL hand gesture motions are captured as micro- Doppler signals using a microwave X-band Doppler radar transceiver. We apply joint time-frequency analysis and observe the presence of the micro- Doppler signatures in the spectrogram. The micro- Doppler signatures of different hand gestures are analyzed using Matlab. Each hand gesture is observed to contain unique spectral characteristics. Based on unique spectral characteristics, we investigate the classification of ASL essential short phrases including emergency signals. For recognizing and characterizing the presence of micro-Doppler signatures in spectrogram we explore deep convolution neural network (DCNN) algorithm. We show that the DCNN algorithm can classify different sign language gestures based on the presence of micro- Doppler signatures in the spectrogram with fairly high accuracy. Experimental results reveal that ut...
Step 1: Let y′ i = Din(yi), 1 ≤ i ≤ N . Step 2: Run Dout on y′ = (y′ 1, . . . , y′ N). By the properties of Din, for any fixed i, there is an error at y′ i with probability ≤ γ 2 . Each such error is independent, since errors in BSCp... more
Step 1: Let y′ i = Din(yi), 1 ≤ i ≤ N . Step 2: Run Dout on y′ = (y′ 1, . . . , y′ N). By the properties of Din, for any fixed i, there is an error at y′ i with probability ≤ γ 2 . Each such error is independent, since errors in BSCp itself are independent by definition. Because of this, and by linearity of expectation, the expected number of errors in y′ is ≤ γN 2 . Taken together, those two facts allow us to conclude that, by the Chernoff bound, the probability that the total number of errors will be more than γN is at most e− γN 6 . Since the decoder Dout fails only when there are more than γN errors, this is also the decoding error probability. Expressed in asymptotic terms, the error probability is 2−Ω( γN n .
In the last decade, wireless communication service has experienced explosive growth while communication technologies have progressed generation by generation. Code-division multiplexing (CDM) or code-division multiple-access (CDMA) is... more
In the last decade, wireless communication service has experienced explosive growth while communication technologies have progressed generation by generation. Code-division multiplexing (CDM) or code-division multiple-access (CDMA) is seen as a promising basic technology for 3G/4G cellular communications networks. In such rapidly growing communication systems in which higher number of users share the same channel becomes very challenging problem since it introduces multiple-access interference (MAI). In this dissertation, we investigate the overloaded code-division multiplexing where the number of multiplexed signals exceeds the code (signature) length L. We propose overloaded code design framework where code set satisfy "errorless" (uniquely decodability) property in noiseless multiplexed transmission. In this proposed framework we aim to identify the maximum number of codes/signatures that can be potentially appended to a Sylvester-Hadamard matrix of order L, while maint...
In this paper, we study the classification of human activity on the surface of a body of water using sonar. In particular, we investigate the classification of three different swimming styles; freestyle, butterfly, and backstroke.... more
In this paper, we study the classification of human activity on the surface of a body of water using sonar. In particular, we investigate the classification of three different swimming styles; freestyle, butterfly, and backstroke. Experiments are conducted in a swimming pool to capture acoustic micro-Doppler signatures produced by the different swimming styles. Two acoustic hydrophones are used underwater; one to transmit a single tone signal in the direction of a swimmer and the other to receive the reflected waveform from the swimmer's body. We apply joint time-frequency analysis on the received acoustic signal to extract the micro-Doppler signatures present in the spectrogram. Each of these swimming style activities presents their own unique micro-Doppler signatures. To classify the acoustic micro-Doppler signatures, we explore a deep convolution neural network (DCNN) algorithm. Spectrogram can be considered as an image in which case applying DCNN can serve well for feature r...
In this work, we develop a theoretical framework for reliable digital recording system identification from digital audio files alone, for forensic purposes. A digital recording system consists of a microphone and a digital sound... more
In this work, we develop a theoretical framework for reliable digital recording system identification from digital audio files alone, for forensic purposes. A digital recording system consists of a microphone and a digital sound processing card. We view the cascade as a system of unknown impulse response. We expect the same manufacturer and model microphone-sound card combinations to have very similar/near identical impulse responses, bar any unique manufacturing defect. Input voice (or other) signals are modeled as non-stationary processes. The technical problem under consideration becomes blind deconvolution with non-stationary inputs, as it manifests itself in the specific application of digital audio recording equipment classification. We propose a conditionally maximum-likelihood (CML) algorithm to estimate underlying systems impulse response together with a novel nearest neighborhood algorithm for recording system identification. Experimental results demonstrate over 99.2% acc...
Research Interests:
Mobile healthcare, or mHealth, is one of the key pillars of information and communication technologies for healthcare that consists of telemedicine, telehealth, eHealth, and mHealth. In the past two decades, mobile health has become a... more
Mobile healthcare, or mHealth, is one of the key pillars of information and communication technologies for healthcare that consists of telemedicine, telehealth, eHealth, and mHealth. In the past two decades, mobile health has become a transformative concept for healthcare delivery innovations on a global scale. The success was based on the market-driven strategies that utilised the advances in mobile communications, computing, and sensor technologies, especially in recent years. Those market-driven mobile health systems were also closely associated with the global proliferation of smartphones, and based on the correlated usage principle of the smartphone applications for healthcare and wellbeing. However, the global commercial success of the smartphone-based mHealth model was not widely translated into successful scaled-up and tangible healthcare benefits, especially in low- and-middle income countries, compared to the consumer mobile health markets. The numerous healthcare challeng...
We focus on a multiway relay channel (MWRC) network where two or more users simultaneously exchange information with each other through the help of a relay node. We propose for the first time to apply ternary uniquely decodable (UD) code... more
We focus on a multiway relay channel (MWRC) network where two or more users simultaneously exchange information with each other through the help of a relay node. We propose for the first time to apply ternary uniquely decodable (UD) code sets that we have developed to allow each user to uniquely recover the information bits from the noisy channel environment. One of the key features of the proposed scheme is that it utilizes a very simple decoding algorithm, which requires only a few logical comparisons. Simulation results in terms of bit error rate (BER) demonstrate that the performance of the proposed decoder is almost as good as the maximum-likelihood (ML) decoder. In addition to that through simulations, we show that the proposed scheme can significantly improve the sum-rate capacity, which in turn can potentially improve overall throughput, as it needs only two time slots (TSs) to exchange information compared to the conventional methods.
Research Interests:
We consider the problem of extracting blindly data embedded over a wide band in a spectrum (transform) domain of a digital medium (image, audio, video). We develop a novel multicarrier/signature iterative generalized least-squares... more
We consider the problem of extracting blindly data embedded over a wide band in a spectrum (transform) domain of a digital medium (image, audio, video). We develop a novel multicarrier/signature iterative generalized least-squares (M-IGLS) core procedure to seek unknown data hidden in hosts via multicarrier spread-spectrum embedding. Neither the original host nor the embedding carriers are assumed available. Experimental studies on images show that the developed algorithm can achieve recovery probability of error close to what may be attained with known embedding carriers and host autocorrelation matrix.
Research Interests:
We consider the problem of designing binary antipodal uniquely decodable (errorless) code sets for overloaded code-division multiplexing applications where the number of signals K is larger than the code length L. Our proposed errorless... more
We consider the problem of designing binary antipodal uniquely decodable (errorless) code sets for overloaded code-division multiplexing applications where the number of signals K is larger than the code length L. Our proposed errorless code set design aims at identifying the maximum number of columns that can be potentially appended to a Sylvester-Hadamard matrix of order L, while maintaining the errorless code property. In particular, we derive formally the maximum number of columns that may be appended to the Sylvester-Hadamard matrix of order L = 8 and use this result as a seed to produce an infinite sequence of designs in increasing L. In the noiseless transmission case, a simple algorithm is developed to uniquely decode all signals. In additive white Gaussian noise (AWGN), a slab-sphere decoding scheme can be utilized for efficient and effective decoding.
Research Interests:
We consider the problem of passive spread-spectrum steganalysis where the objective is to decide the presence or absence of spread-spectrum hidden data in a given image (a binary hypothesis testing problem). Unlike conventional... more
We consider the problem of passive spread-spectrum steganalysis where the objective is to decide the presence or absence of spread-spectrum hidden data in a given image (a binary hypothesis testing problem). Unlike conventional feature-based approaches, we describe an unsupervised (blind) low-complexity approach based on generalized least-squares principles that may enable rapid high-volume image processing. Extensive experiments on image sets and comparisons with existing steganalysis techniques demonstrate most satisfactory classification performance measured in probability of correct detection versus induced false alarm rate.
Research Interests:
This paper considers the problem of blindly extracting data embedded over a wide band in a spectrum (transform) domain of a digital medium (image, audio, video). We first develop a multi-signature iterative generalized least-squares... more
This paper considers the problem of blindly extracting data embedded over a wide band in a spectrum (transform) domain of a digital medium (image, audio, video). We first develop a multi-signature iterative generalized least-squares (M-IGLS) core procedure to seek unknown data hidden in hosts via multi-signature direct-sequence spread-spectrum embedding. Neither the original host nor the embedding signatures are assumed available. Then, cross-correlation enhanced M-IGLS (CC-M-IGLS), a procedure described herein in detail that is based on statistical analysis of repeated independent M-IGLS processing of the host, is seen to offer most effective hidden message recovery. Experimental studies on images show that the proposed CC-M-IGLS algorithm can achieve recovery probability of error close to what may be attained with known embedding signatures and host autocorrelation matrix.