PRISM :: Browsing by Author "Yanushkevich, Svetlana N."

Email Notifications Currently Out of Service. Please note we currently have a technical issue with email notifications being sent from PRISM. If you require immediate assistance, please contact us at openservices@ucalgary.libanswers.com. Thank you for your patience.

Browsing by Author "Yanushkevich, Svetlana N."

Now showing 1 - 20 of 36

Open Access
A Machine Learning Predictor and Corrector Framework to Identify and Resolve VLSI Routing Short Violations
(2018-10-24) Fakheri Tabrizi, Aysa; Behjat, Laleh; Rakai, Logan M.; Yanushkevich, Svetlana N.; Dimitrov, Vassil
The growth of Very Large Scale Integration (VLSI) technology provokes new challenges in design automation of Integrated Circuits (ICs). Routability is one of the most challenging aspects in Electronic Design Automation (EDA) that is faced in two consecutive phases of physical design: placement and routing. During placement, the exact locations of circuit components are determined. During routing the paths for all of the wires are specified. Routing is performed in two stages: global routing and detailed routing. Many of the violations that occur during the detailed routing stage stem from ignoring the routing rules during placement. Therefore, detecting and preventing routing violations in the placement stage has become critical in reducing the design time and the possibility of failure. In this thesis, Eh?Predictor, a deep learning framework to predict detailed routing short violations during placement is proposed. In the development of this predictor, relevant features, contributing to routing violations, were identified, extracted, and analyzed. A neural network model that can handle imbalanced data was customized to detect these violations using the defined features. The proposed predictor can be integrated into a placement tool and be used as a guide during the placement process to reduce the number of shorts happening in the detailed routing stage. One of the advantages of this technique is that by using the proposed deep learning-based predictor, global routing is no longer required as frequently. Hence the total runtime for place and route can be significantly reduced. In addition to Eh?Predictor, a detailed routing-aware detailed placement algorithm is developed to improve detailed routability in a relatively short runtime. The proposed technique is referred to as Detailed Routing-aware Detailed Placer (DrDp). DrDp is a heuristic that aims to reduce the local congestion and mitigate routing failure by aligning the connected cells where possible at the final stage of detailed placement process. Experimental results show that Eh?Predictor is able to predict on average 90% of the short violations of previously unseen data with only 5% false alarm rate and considerably reduce computational time, and DrDp can effectively improve the detailed routing quality in a short runtime with no significant change in detailed placement score or total wirelength.
Open Access
A Serial Communication-Based FPGA Co-Emulation Test Bench
(2018-04-23) Cao, Dingcheng; Smith, Michael Richard; Far, Behrouz Homayoun; Yanushkevich, Svetlana N.
FPGA designs are verified first with the simulation testing on a computer to check their behavior, and next verified with the emulation tested in an actual FPGA system to check its functional and timing performance. Co-emulation technology has been introduced to combine the simulation and emulation verification into one step which verifies the design by running the simulation test case directly on an FPGA board. Current co-emulation test benches accelerate the verification speed but need complex hardware to support its parallel communication interface. This research proposes to replace the parallel communication interface with a serial design to reduce the complexity of the hardware design. According to this proposal, the thesis introduces the hierarchy and architecture of the serial communication-based co-emulation test framework. It develops the proposed test bench based on the framework. Finally, it demonstrates the simulation and co-emulation test result to prove the feasibility of the proposed test bench
Open Access
Accelerating Sequence Calculations on Parallel GPU Architecture
(2019-02-28) Hossain, Roksana; Magierowski, Sebastian K.; Messier, Geoffrey G.; Nielsen, Jørgen S.; Yanushkevich, Svetlana N.
In this thesis, I have implemented a GPU (graphics processor unit) based sequencing algorithm that finds a sequence or order among data points to optimize a given objective. I have studied the sequencing algorithm as a path planner for an unmanned aerial vehicle (UAV) and also, a basecaller for a miniature DNA sequencer. Parallel implementation utilizing GPU enables faster processing and decision making that are important when data is quite large and a real-time response is critical e.g., in UAV based transportation. The goal of using UAV in my thesis is to construct a wireless sensor network in a remote location by deploying wireless sensor nodes. The proposed path planner, also known as a sequencer, is designed to find the shortest path that is also a safe path using travelling salesman problem. A path is considered safe when the vehicle would not collide with any obstacle. Two sets of heuristic algorithms, one for generating a sequence of waypoints (sequence generator) and another one for constructing a path between two waypoints (path explorer), are used to find the near optimal solution. The highly-parallel multicore GPU is used for the real-time implementation that offloads compute-intensive portions from the traditional CPU to the GPU to make the decision-making process faster. In this thesis, the parallel execution of the sequence generator and the path explorer achieved a 4.82x and 164x speed-up compared to the CPU-only approach respectively. The second sequencer is a palm-sized miniature DNA sequencer, the so-called MinION device. In the case of the MinION, a vast multitude of DNA strands are introduced into the device and converted into noisy electronic time-series signals; these measurements are essentially physical signatures related to the molecular make-up of the sensed DNA. Among a long "pipeline" of analysis steps to be performed on such measurement sequences, the first, and arguably most intensive, is the so-called basecalling step which analyzes the time-series and converts it into the equivalent monomeric base of the DNA under test using the Viterbi algorithm.
Open Access
Advanced Methods for Efficient Digital Signal Processing and Matrix-Based Computations
(2018-04-12) Gomes Coelho, Diego Felipe; Dimitrov, Vassil S.; Behjat, Laleh; Walus, Konrad; Yanushkevich, Svetlana N.; Jacobson, Michael J.
Modern engineering and scientific problems demand a great amount of data processing power. The type of data that needs to be processed varies from application to application. Image processing, genome matching, physics phenomena simulation, and cryptography are a few examples of processing-power demanding applications. In a wide range of those computationally intensive applications, the arithmetic complexity plays an important role, having direct impact on the implementation performance. In this thesis, we present several methods that are novel contributions of the author to some computationally intensive problems. The introduced methods reduce the overall computing time or other relevant hardware’ and software implementation metrics by decreasing the arithmetic complexity associated with each task. Verified results are shown with peer-reviewed journal papers in reputable journals. In particular, problems on signal processing, eigenvalue computation, and matrix inversion for radar image classification are considered.
Open Access
Automated Gait Trait Analysis and Applications
(2018-09-20) Kozlow, Patrick; Yanushkevich, Svetlana N.; Bartley, Norman R.; Goldsmith, Peter B.
The main focus of this thesis is the development and feasibility testing of a proposed gait biometric screening system based on the Kinect v2 sensor. To achieve contactless gait biometric extraction the system uses a virtual Kinect 3D skeleton to construct models in real time. These models are then used to identify an individual's gait characteristics. The features found from the virtual models are passed through a gait recognition system which provides insight into what type of gait pattern is being observed by the camera. Extensive experiments with different classification methods such as Support Vector Machines, K-Nearest-Neighbors, and Dynamic Bayesian Networks are tested to determine the effectiveness of the system. The proposed gait recognition network is tested using locally collected and publically available databases to validate the results and prove that the system is feasible.
Open Access
A contribution to risk-informed inspection and maintenance planning for unpiggable pipelines subject to internal corrosion
(2020-12-10) Melo Gonzalez, Carlos Alberto; Dann, Markus R.; Hugo, Ronald J.; Groth, Katrina M.; Yanushkevich, Svetlana N.; Ziadé, Paul; Park, Simon S.
Pipelines are the safest transportation mode for hydrocarbons, but internal corrosion is still a major cause of failure for energy pipelines. Inspection and repair strategies are implemented to avoid pipeline failures and their consequences. In-line inspection is the most detailed examination technique for pipelines, but almost half of existing pipelines are unpiggable and cannot be inspected using this technique. Direct assessment, based on models that numerically assess corrosion in a pipeline, was developed to overcome this limitation and to facilitate inspection at certain sites. However, most of these models do not take localized corrosion into consideration, which is the main cause of pipeline failure. Industry standards provide guidance for selecting inspection sites based on the results of direct assessment models. However, this process is based only on the likelihood of pipeline failure and can lead to decisions which imply elevated risk and increased pipeline lifecycle costs. The aim of this dissertation is to expand the state of knowledge in risk-based inspection and maintenance planning for upstream unpiggable pipelines subject to internal corrosion. The research focuses on unpiggable pipelines in production and gathering systems operated in the upstream oil and gas industry. A framework for probabilistic risk and integrity assessment of unpiggable pipelines is developed. The framework combines advanced flow and corrosion models with risk-based inspection and maintenance planning. It also includes uncertainty analysis and lifecycle-cost optimization. An extreme value analysis is developed to model localized corrosion and microbiologically-influenced corrosion, which are the main causes of internal corrosion failure in pipelines. A method for decision optimization of unpiggable pipeline inspections based on the value of information is proposed. For maintenance decision optimization at specified safety levels, the research considers both risk-constrained optimization and lifecycle-cost optimization. The research outcomes provide pipeline operators with a methodology for developing optimal inspection and maintenance plans, while maintaining adequate safety levels.
Open Access
Contributions to Behavioral Authentication Systems
(2021-01-27) Islam, Md Morshedul; Safavi-Naini, Reihaneh; Fong, Philip W. L.; Jacobson, Michael John; Yanushkevich, Svetlana N.; Mannan, Mohammad
Behavioral Authentication (BA) systems authenticate users through their behavioral characteristics. BA systems construct behavioral profiles of users from their well-designed activities, and store profiles in a profile database on the system. For a verification request, a verification algorithm evaluates the request by comparing the provided verification data with the stored profile. In this thesis, we identify a number of shortcomings of these systems that are motivated by the application of these systems in practice. We study these shortcomings and propose solutions to address each. We designed, implemented and evaluated an activity-based BA system for mobile devices that is used to evaluate our proposed systems, experimentally. In more details, we proposed a challenge-response based BA system named DAC (Draw A Circle) and later extended it to eDAC (extended DAC) to improve its accuracy and usability. In both systems, behavioral data are from users’ response to drawing challenge circles. Through extensive analysis and experiments, we chose a set of features that are non-shareable and non-emulatable, and developed a verification algorithm that can successfully authenticate users with overwhelming probability. We studied the effect of database size on verification error, and that verification error increases with the database size. We introduced the notion of scalability of BA systems that requires the error probability of the system to remain (almost) the same as profile database grows; proposed personalization of verification to achieve scalability. To estimate information in BA systems, we used Biometric Information (BI), and Biometric System Entropy (BSE), two different but related approaches used for information measure in biometric-based systems. We studied the applicability of these measures for BA systems. For cryptographic applications, we proposed BAVault, a fuzzy vault based on the profiles in BA systems that can protect a secret key (message) of reasonable length. BAVault ensures profile privacy, even when the key is known. For profile privacy in profile databases and privacy-preserving verification, we proposed a non-cryptographic approach that uses an efficient profile transformation called random projection, projects a profile (verification data) into a lower dimension space and ensures their privacy. The verification is done in the transformed domain using a similar verification algorithm. Finally, we show an attack on BA systems when the verification algorithm uses the outputs of the classifier for verification decision. To impersonate a user of the BA systems, the attacker will utilize the information leakage of the verification algorithm about the output of the classifier. In all the above cases, we implemented our proposed approach and evaluated their performance.
Open Access
Detecting Abnormalities in Thermal Pattern of Faces for Healthcare Applications
(2019-05-14) Ejindu, Oluchukwu Roseline; Yanushkevich, Svetlana N.; Nowicki, Edwin Peter; Alim, Usman Raza
In this work, we propose a novel method of applying deep learning technique in thermal image processing and analysis for healthcare application. It addresses detection of abnormal thermal patterns, thus identifying, in particular, patterns of elevated temperature that indicate fever, hypothermia and related abnormalities. Temperature estimation is performed based on the analysis of regions-of-interest from the thermal images of human faces. Another focus of this work is to investigate thermal effects of alcohol intoxication. We applied the deep learning approach on 16,000 usable images of 40 subjects from a publicly-available Drunk-Sober database. Two Convolutional Neural Network architectures were investigated for the task of processing of two regions of interest - the forehead and the eyes. The accuracy of the neural network classifiers to predict subject’s insobriety using the forehead and eye regions-of-interest reached 95.5% and 96.67%, respectively, compared to the best-known results on the same data using a non-deep neural networks. To boost the accuracy of classification, both the feature-level and the score-level fusion were applied as well, thus improving the accuracy to 96.92%.
Open Access
Determining Speed and Stride Length using an Ultrawide Bandwidth Local Positioning System
(2021-01-13) Singh, Pratham P.; Stefanyshyn, Darren John; Boyd, Jeffrey Edwin; Edwards, William Brent; Ferber, Reed; Yanushkevich, Svetlana N.
There are many modalities that can profile speed and stride length for runners. One such modality includes using wearable technologies. An example of a wearable technology includes a global positioning system-based wearable. However, due to its limitations, an alternative may include a local positioning system-based wearable operating in the ultrawide bandwidth. Considering that a local positioning system is not good at determining gait events such as heel and step count, applying sensor fusion with an inertial measurement unit may be beneficial. Therefore, the purpose of the dissertation was to compare speed and stride length determined from an ultrawide bandwidth local positioning system equipped with an inertial measurement unit to a criterion standard (i.e. the “gold standard”) such as video motion capture and timing gates. The data suggest that the local positioning system used in the project may not be a valid tool without further processing. Using machine learning algorithms, pertinent features from a gait cycle that can better extract speed and stride length were explored. More specifically, using a stepwise linear regression model first and then using a feedforward neural network proved to be quite successful in estimating stride length. Chapter 1 provides an introduction to the project, Chapter 2 provides a review of relevant literature, Chapter 3 provides an insight into the materials and methods used, Chapter 4 shows the results obtained from the methods described earlier, Chapter 5 is a discussion of the results obtained and Chapter 6 concludes with suggestions regarding next steps that should be taken.
Open Access
The development of a platform for hyperthermia induction in small animal cancer models using MRI guided Focused Ultrasound to test drug delivery of thermosensitive liposomes
(2020-12) Siddiqui, Maryam; Pichardo, Samuel; Pichardo, Samuel; Curiel, Laura; Robbins, Stephen M.; Dunn, Jeff F.; Yanushkevich, Svetlana N.; Pike, G. Bruce
Focused ultrasound concentrates acoustic energy on a small volume. In tissues, this can have several bioeffects, including hyperthermia. Hyperthermia occurs when the body temperature is raised above its core value to about 43⁰C and is sustained for several tens of minutes. The increased temperature can cause damage to the tumour cells directly by denaturing proteins and DNA. It also leads to an increase in oxygenation, which leads to cell toxicity and can have synergistic effects for radio-sensitization. Localized hyperthermia can also be used to improve drug delivery to tumours using thermosensitive formulations. In particular, a liposome-based encapsulation of the chemotherapeutic drug, doxorubicin, has been used in studies for targeted drug delivery to tumours. Doxorubicin is commonly used to treat soft tissue sarcomas, which make up 4-8% of childhood cancers. The liposome dissolves when reaching hyperthermic temperatures and releases its load in tissues where localized hyperthermia takes place. A challenge when studying tumour response in preclinical models is performing precise localized hyperthermia delivery as tight temporal-spatial control of the temperature distribution is required. For this research project, Magnetic Resonance Imaging-guided Focused Ultrasound (MRIgFUS) is proposed as a method to deliver localized hyperthermia in small animal models. Focused ultrasound can precisely concentrate mechanical energy that is transformed into heat, and MRI can be used to target treatment location and monitor temperature spatially and temporally. This monitoring can be used to control hyperthermic levels in a tumour model. In this study, we present details on a platform to deliver MRIgFUS in small animal models. For a demonstration of MRIgFUS as an effective platform for localized hyperthermia in small animals, we performed two studies. The first study was conducted in healthy mice (n=30, C57BL/6) for technical development to establish the basic guidelines for MRIgFUS-based hyperthermia in small animals. The second study used a murine model of alveolar soft part sarcoma (ASPS) to demonstrate that MRIgFUS can increase thermosensitive liposomal doxorubicin delivery. ASPS is a rare type of soft tissue sarcoma commonly found in children and adolescents, with tumour location usually in the body’s extremities. A current challenge in the treatment of ASPS is that tumour resection is unfeasible in 45% of patients. We first used a small cohort of mice (n=6, CB17 SCID) to perform an MRI assessment of alveolar soft part sarcoma tumour growth in a lower limb. A comparison study was then conducted to test the delivery of a temperature-sensitive, liposome-encapsulated form of doxorubicin (ThermoDXR) to the tumour site using MRIgFUS-based hyperthermia. Drug concentrations were compared between four treatment groups. This study was done in ASPS-bearing mice (n=5 in each of four groups, CB17 SCID) with tumours growing subcutaneously, just below the skin. Mice in the first and second group received free doxorubicin with no hyperthermia and free doxorubicin with hyperthermia using MRIgFUS, respectively. Mice in the third and fourth groups received ThermoDXR with no hyperthermia, and ThermoDXR with hyperthermia, respectively. This study hypothesized that the treatment group with ThermoDXR and hyperthermia would have the greatest drug concentration at the tumour site, thereby indicating an improvement in targeted drug delivery. As ASPS tumours were implanted subcutaneously, special arrangements were implemented to target them correctly using MRIgFUS-based hyperthermia; the tumour’s position and the focal target for heating must be carefully planned. In this project, we demonstrated the delivery of the drug to ASPS tumours in mice was enhanced in the group using MRIgFUS based hyperthermia in combination with ThermoDXR. However, future experiments must be conducted with larger sample sizes to evaluate significant differences between groups. Additional studies for tumour growth and survival must also be conducted to test this treatment’s therapeutic effects.
Open Access
Development of an Accurate Clock Delay Model with Application in Clock Network Buffer Sizing
(2019-08-09) Farshidi, Ali; Behjat, Laleh; Rakai, Logan M.; Dimitrov, Vassil; Yanushkevich, Svetlana N.
Clock network synthesis is an important stage of the Integrated Circuit (IC) design cycle. The performance of the IC highly depends on the clock network synthesis which makes this stage critical where accuracy is very important. In this thesis, a new delay model is proposed for clock networks that is capable of estimating clock signal delay with significantly improved accuracy in a relatively low runtime. This model is developed using Least square fitting by employing data oriented training. The developed model is formulated in the form of posynomials which makes it a suitable option for application in geometric programming gate and clock network sizing optimization frameworks. The experimental results demonstrate the effectiveness of the proposed delay model in predicting the delay at the timing critical clock sinks in the clock network, i.e. sinks with minimum and maximum delays, and the estimated values are, on average, 20 ps closer than the Elmore values to the reference circuit simulator tool, ngspice. This is while the runtime of the proposed delay model is negligible compared to the ngspice simulations. This helps designers obtain accurate delay estimations in low runtime for quick optimization iterations. In addition, a clock network buffer sizing approach is developed which includes an objective function with geometric programming format considering two competing objectives, power consumption and clock skew. The clock slew and technology constraints are also integrated into this optimization problem. The clock network buffer sizing experiments show significant improvements compared to the initial clock networks in terms of clock skew, up to 183 ps, while the power consumption improves for all test cases, on average by 54%.
Open Access
DTLS with Post Quantum Security for Origin Authentication and Integrity
(2020-09-24) Parveen, Simpy; Safavi-Naini, Reihaneh S.; Ghaderi, Majid; Yanushkevich, Svetlana N.; Safavi-Naini, Reihaneh S.
All public-key cryptography algorithms that are in use today, including RSA (Rivest–Shamir- Adleman) cryptosystem, DSA (Digital Signature Algorithm), and DH (Diffie-Hellman) key agreement, will be broken if quantum computers become a reality. Hence, applications and protocols must be transitioned to quantum-resistant designs. We consider post-quantum security of DTLS (Datagram Transport Layer Security) for source authentication and message integrity. These are essential security requirements for control plane communications in 5G networks. To provide message integrity while avoiding costly post-quantum secure key exchange protocols that rely on unproven computational assumptions, we will use TESLA (Timed Efficient Stream Loss-tolerant Authentication) protocol. TESLA is a data stream authentication protocol that uses symmetric-key cryptographic primitives and a digital signature scheme to achieve security. We first replace the digital signature in TESLA with a hash-based one to achieve post-quantum security, and then carefully revise the DTLS handshake and record layer protocol to include the new TESLA protocol such that it delivers the same properties for DTLS. We argue our design’s security and show our model’s feasibility using an efficient implementation for an open-source DTLS library, called TinyDTLS. Finally, we provide performance measurements for PQ-DTLS compared with original DTLS in authentication and integrity only mode.
Open Access
Event-driven circuit analysis based on logic differential operators
(2003) Xie, Yan; Yanushkevich, Svetlana N.
Open Access
Extraction Of Noise Parameters For Single-Ended Components Inside A Differential Circuit Using Single-Ended Equipment
(2018-07-09) Huang, Yuxiang; Belostotski, Leonid; Okoniewski, Michal M.; Yanushkevich, Svetlana N.
This thesis proposes an approach of investigating electrical and noise parameters of subcomponents inside a fully differential system. To find the single-ended noise parameters, two sets of single-ended noise-parameter measurements and one set of S-parameter measurement are performed. A proof-of-concept PCBs is designed, fabricated, and tested after the algorithms are verified using Matlab. The circuit is designed to have a bandwidth from 500 MHz to 1.5 GHz and be unconditionally stable at all frequencies. The designed circuit has the gain of 9.4 dB at 500 MHz and 9.5 dB at 1.5 GHz in schematic simulations, while in measurements, it has the gain of 8.1 dB at 500 MHz and 4.3 dB at 1.5 GHz. The minimum noise figure is 2.4 dB at 500 MHz and 2.4 dB at 1.5 GHz in simulations, while in measurements, it is 3.1 dB at 500 MHz and 3.2 dB at 1.5 GHz. This thesis presents schematic and measurement results for the electrical and noise parameters. The measurement results are analyzed in Matlab and compared with the relevant single-ended measurement results to verify the operation of the method.
Open Access
Generation of synthetic fingerprints
(2005) Wang, Penglong; Yanushkevich, Svetlana N.
Open Access
Improving Image Classification Through Generative Data Augmentation
(2019-05-15) Nielsen, Christopher Stephen; Okoniewski, Michal M.; Messier, Geoffrey G.; Yanushkevich, Svetlana N.
As the industrial adoption of machine learning systems continues to grow, there is incredible potential to use this technology to revolutionize how medical diagnostic imaging is performed. The ability to accurately classify the information contained within a medical image is of critical importance for clinical implementation. Successful application of machine learning classification algorithms has traditionally relied on the availability of copious amounts of labelled training data. Unfortunately, medical datasets are typically small due to privacy constraints and the large cost associated with annotating the data. To ameliorate this limitation, a training scheme is developed in this thesis which can operate on small-scale datasets by using a generative adversarial network to augment the dataset with synthetic images. Through quantifying the uncertainty in the classification network, training samples are selected to maximize the performance of the classifier while minimizing the amount of required data. Furthermore, privacy constraints are preserved as the images sampled from the generative adversarial network are inherently anonymized. The experimental results demonstrate the efficacy in this approach and viability for application in the medical domain.
Open Access
Integrated Design of Complex Mechanical Products Considering Modeling, Simulation and Optimization Aspects
(2020-04-21) Imaniyan, Davood; Xue, Deyi; Li, Simon; Yanushkevich, Svetlana N.
The presently developed computer-based design systems are not effective for design of complex mechanical products when multiple tools and methods in different schemes have to be employed at different design stages. In this research, a new integrated framework has been introduced for the design of complex mechanical products considering modeling, simulation/evaluation, and optimization aspects. An integrated system for design of complex mechanical products has also been developed. In this system, first a hybrid scheme is introduced for integrated modeling of complex mechanical products considering conceptual design and detailed design stages. In conceptual design, the generic product considering different design solution candidates is modeled in an AND-OR tree. Specific design candidates modeled by AND trees are created from the generic AND-OR tree through tree-based search. The geometric descriptions in a design candidate are then converted into and associated with the geometric model in a CAD system for detailed design. Second, a hybrid simulation method is developed for evaluating different product aspects with different simulation tools that are integrated through the hybrid modeling scheme. Simulations with geometric descriptions are conducted by analysis functions of the CAD system for detailed design. Simulations with non-geometric descriptions are conducted by the knowledge-based systems for conceptual design. Third, a hybrid optimization method is developed to identify the optimal design of the complex mechanical product. For each design candidate, parameter optimization is conducted to obtain the optimal parameter values. The optimal design solution is identified from all design candidates through configuration optimization. The integrated complex mechanical product design system has been implemented using C# and SOLIDWORKS. Various user interfaces were developed for conducting design activities in modeling, simulation/evaluation and optimization aspects. Communication between the symbolic model in conceptual design and the CAD model in detailed design was achieved through TCP/IP client-server structure and SOLIDWORKS API. A case study has been developed to demonstrate the effectiveness of the newly-introduced design approach.
Open Access
Logic Design of NanoICS
(CRC Press, 2004-10-28) Yanushkevich, Svetlana N.; Shmerko, Vlad P.; Lyshevski, Sergey E.
Today's engineers will confront the challenge of a new computing paradigm, relying on micro- and nanoscale devices. Logic Design of NanoICs builds a foundation for logic in nanodimensions and guides you in the design and analysis of nanoICs using CAD. The authors present data structures developed toward applications rather than a purely theoretical treatment. Requiring only basic logic and circuits background, Logic Design of NanoICs draws connections between traditional approaches to design and modern design in nanodimensions. The book begins with an introduction to the directions and basic methodology of logic design at the nanoscale, then proceeds to nanotechnologies and CAD, graphical representation of switching functions and networks, word-level and linear word-level data structures, 3-D topologies based on hypercubes, multilevel circuit design, and fault-tolerant computation in hypercube-like structures. The authors propose design solutions and techniques, going beyond the underlying technology to provide more applied knowledge. This design-oriented reference is written for engineers interested in developing the next generation of integrated circuitry, illustrating the discussion with approximately 250 figures and tables, 100 equations, 250 practical examples, and 100 problems. Each chapter concludes with a summary, references, and a suggested reading section.
Open Access
Modeling of Antenna Arrays and Mutual Coupling between Array Elements using S-Parameters
(2018-03-28) De Silva, Obinamuni Supun Devinda; Belostotski, Leonid; Okoniewski, Michal M.; Fear, Elise C.; Yanushkevich, Svetlana N.
An antenna can be modeled as a two-port network using S-parameters. However, having antenna S-parameters are not sufficient for modeling an array of antennas as that requires mutual coupling. This thesis proposes a network model to represent an antenna array including mutual coupling. Using the proposed model, an N-element antenna array can be modeled as a combination of two-port networks to represent each element and an N-port network to represent mutual coupling. In this thesis, the operation of the proposed model is explained and the concept of complete antenna array scattering matrix representation is introduced. Two techniques to calculate mutual coupling S-parameters are presented: primary method using numerical solutions and general method using general expressions. The proposed model was validated with reference to the radiation efficiency and S-parameters between the physically accessible ports using several arrays: various inter-element lengths, varying the number of elements and configurations, and non-identical array elements.
Open Access
Multimodal Person Recognition using Social Behavioral Biometric
(2018-04-06) Sultana, Madeena; Gavrilova, Marina L.; Yanushkevich, Svetlana N.; Leung, Henry; Li, Zongpeng; Ruhe, Guenther; Yang, Xiaosong
The goal of a biometric recognition system is to make a human-like decision on individuals’ identity by recognizing their physiological and/or behavioral traits. Nevertheless, decision-making process by either a human or a biometric recognition system can be highly complicated due to low quality of data or an uncertain environment. Human brain has an advantage over computer system due to its ability to perform a massive parallel processing of auxiliary information such as visual cues, cognitive and social interactions, contextual and spatio-temporal data. Similarly to a human brain, social behavioral cues can aid the reliable decision-making of an automated biometric system. Being an integral part of human behavior, social interactions are likely to possess unique behavioral patterns. However, the significance of social behavior for automated user recognition has been noted in the scientific community only recently. In this doctoral thesis, a novel person recognition approach is presented that relies on the knowledge of individuals’ social behavior in order to enhance the performance of a traditional biometric system. The social behavioral information of individuals’ has been mined from an Online Social Network (OSN) and fused with traditional face and ear biometrics. This research identified a set of Social Behavioral Biometric (SBB) features from the online social information and proposed a framework to utilize these features for an automated person recognition for the first time. Extensive experiments confirm that human social behavior expressed through OSN can provide a unique insight onto person recognition. Performance of the proposed multimodal approach has been evaluated to determine the effectiveness of fusing social behavioral information. Experimental results on virtual and semi-real databases demonstrate significant performance gain in the proposed method over traditional biometric system. This doctoral research contributes to an emerging research direction in biometric domain as well as opens new frontier of studying social behavior in virtual domain.