Open Access. Powered by Scholars. Published by Universities.®
- Keyword
-
- Wireless communication systems (5)
- Artificial intelligence (4)
- Machine learning (4)
- Broadband communication systems (2)
- Communication (2)
-
- Computer algorithms (2)
- Computer software (2)
- Deep Learning (2)
- Deep learning (Machine learning) (2)
- Internet of things (2)
- Optical communications (2)
- Speech processing systems (2)
- Wireless Communication (2)
- AI-Synthesized (1)
- Anomaly detection (Computer security) (1)
- Artificial Intelligence (1)
- Autoencoder (1)
- Autonomous agent (1)
- BCI (1)
- Bispectral Analysis (1)
- Blockchain (1)
- Blockchains (Databases) (1)
- Brain-computer interfaces (1)
- CCA (1)
- Chest (1)
- Chest X-ray (1)
- Coincidence (1)
- Communities (1)
- Complementary (1)
- Computer Security (1)
Articles 1 - 17 of 17
Full-Text Articles in Engineering
Domain Aware Deep Learning For Wireless Physical Layer, Shuvam Chakraborty
Domain Aware Deep Learning For Wireless Physical Layer, Shuvam Chakraborty
Legacy Theses & Dissertations (2009 - 2024)
Wireless receiver design for OFDM systems is well investigated with classical signal processing tools, which lack the capacity to extract intrinsic channel effects in received signal and lead to high decoding error in receiver. Current deep learning techniques have shown improvement in such cases. But these models are mostly being developed as black box without any anchor to the theory of wireless signal propagation, which leads to surface level information gain and lacks generalizability. We propose deep learning models where the hyperparameters and learning objectives are derived from domain knowledge of wireless signal propagation. These models not only increase the …
Ai-Synthesized Speech : Generation And Detection, Ehab Alsayed Albadawy Abdrabuh
Ai-Synthesized Speech : Generation And Detection, Ehab Alsayed Albadawy Abdrabuh
Legacy Theses & Dissertations (2009 - 2024)
From speech to images, and videos, advances in machine learning have led to dramatic improvements in the quality and realism of so-called AI-synthesized content. While there are many exciting and interesting applications, this type of content can also be used to create convincing and dangerous fakes. We seek to develop forensic techniques that can distinguish a real human voice from a synthesized voice. We observe that deep neural networks used to synthesize speech introduce specific and unusual artifacts not typically found in human speech. Although not necessarily audible, we develop various detection algorithms to measure these artifacts and be able …
Dynamic Instance-Wise Decision-Making For Machine Learning, Yasitha Warahena Liyanage
Dynamic Instance-Wise Decision-Making For Machine Learning, Yasitha Warahena Liyanage
Legacy Theses & Dissertations (2009 - 2024)
In a typical supervised machine learning setting, the predictions on all test instances are based on a common subset of features discovered during model training. However, using a different subset of features that are most informative for each test instance individually may improve not only the quality of prediction but also the overall interpretability of the model. To this end, in this dissertation, we study the problem of optimizing the trade-off between instance-level sparsity and the quality of prediction using a dynamic instance-wise decision-making approach. Specifically, this approach sequentially reviews features one at a time for each data instance given …
Augmented Communications : A Solution For Overcoming High Spatial Correlation Of The Massive-Miso Vlc Channel, Monette Khadr
Augmented Communications : A Solution For Overcoming High Spatial Correlation Of The Massive-Miso Vlc Channel, Monette Khadr
Legacy Theses & Dissertations (2009 - 2024)
A key challenge for future wireless networks is to come upon a riveting compromise between spectral efficiency, complexity, and energy efficiency. The challenge is also intensified due to the pace at which the Internet-of-Things (IoT) technology is arriving, causing an upheaval to pre-existing network infrastructures in terms of elevating spectrum scarcity. To keep pace with the exploding data demand forecasts, a circumvention is required. One realization is by utilizing the high-band spectrum and the rich body of knowledge on multiple-input multiple-output (MIMO) technologies. One of the prominent high frequency technologies is visible light communications (VLC). VLC provide a large unregulated …
Mixed Carrier Communication For Sixth-Generation Networks :, Ahmed Fahmy Mahmoud Hussein
Mixed Carrier Communication For Sixth-Generation Networks :, Ahmed Fahmy Mahmoud Hussein
Legacy Theses & Dissertations (2009 - 2024)
Recently, research on sixth-generation (6G) wireless networks has gained significant interest. By 2030, it is expected that 6G will introduce revolutionary applications and services. Thus, 6G is likely to expand across all available spectrum, including terahertz (THz) and optical frequency bands. Although 5G will offer a massive upgrade to the spectrum, the technology does not provide solutions to support a vast multitude of services and devices simultaneously.Motivated by the heterogeneity of wireless technologies, devices, and services, the Mixed Carrier Communication (MCC) concept is introduced for the first time. MCC is a novel concept that supports the 6G vision by enabling …
Increasing Performance Of Classifiers For Ssvep-Based Brain-Computer Interfaces Using Extension Methods, Ethan Douglas Webster
Increasing Performance Of Classifiers For Ssvep-Based Brain-Computer Interfaces Using Extension Methods, Ethan Douglas Webster
Legacy Theses & Dissertations (2009 - 2024)
Brain-computer interfaces (BCI) provide an alternative communication method that does not require standard physical mediums (speech, typing, etc.). These systems have been implemented to provide additional communication and control options for people with certain motor disabilities. Classification is an important part of BCI systems and consists of inferring user commands from brain activity. Supervised classification methods often achieve higher accuracy, but unsupervised classification methods are useful when training is not practical for the user. This thesis focuses on unsupervised classification algorithms used for a BCI speller application and presents extensions for two existing classifiers that improve classification accuracy and thus …
Sequentially-Closed And Forward-Closed String Rewriting Systems, Yu Zhang
Sequentially-Closed And Forward-Closed String Rewriting Systems, Yu Zhang
Legacy Theses & Dissertations (2009 - 2024)
In this dissertation we introduce the new concept of sequentially-closed string rewriting systems which generalizes forward-closed string rewriting systems and monadic string rewriting systems. We also investigate subclasses and properties of finite and regular sequentially-closed systems and forward-closed systems.
Invariant-Based Online Software Anomaly Detection And Selective Regression Testing, Yizhen Chen
Invariant-Based Online Software Anomaly Detection And Selective Regression Testing, Yizhen Chen
Legacy Theses & Dissertations (2009 - 2024)
Software has been extensively used in various domains to provide online services. With the growing popularity of these types of applications, the quality of the software has a great impact on many of our daily activities [1]. Reliable software executions that deliver expected outcomes are essential for quality services. Software is considered abnormal when its behavior deviates from what is expected at any point during its execution. When anomalous behavior propagates to an exit point of the software and produces an incorrect output or an unexpected termination of the execution, it is considered a software failure. An anomaly may or …
Design And Simulation Of A Voltage Controlled Current Source For Electrical Impedance Tomography Applications, Farial Nur Maysha
Design And Simulation Of A Voltage Controlled Current Source For Electrical Impedance Tomography Applications, Farial Nur Maysha
Legacy Theses & Dissertations (2009 - 2024)
Electrical impedance tomography (EIT) is a simple, non-invasive and ionizing radiation-free imaging technology with potential application to medical diagnostics such as lung function, cardiac output, breast cancer, and cysts. Of the above imaging techniques, lung imaging has developed into the prime application for EIT. Because it presents an ill-posed inverse problem, EIT requires high-precision instrumentation and this thesis studies a new method for obtaining a high-precision current source and voltmeter for EIT. This thesis describes various simulation studies performed on the voltage-controlled current source (VCCS). The output impedance (Zo) of various types of Howland current source (HCS) including the basic …
Efficient Detection Of Diseases By Feature Engineering Approach From Chest Radiograph, Avishek Mukherjee
Efficient Detection Of Diseases By Feature Engineering Approach From Chest Radiograph, Avishek Mukherjee
Legacy Theses & Dissertations (2009 - 2024)
Deep Learning is the new state-of-the-art technology in Image Processing. We applied Deep Learning techniques for identification of diseases from Radiographs made publicly available by NIH. We applied some Feature Engineering approach to augment the data from Anterior-Posterior position to Posterior-Anterior position and vice-versa for all the diseases, at the same point we suppressed ‘No Finding’ radiographs which contributed to more than 50% (approximately 60,000) of the dataset to top 1000 images. We also prepared a model by adding a huge amount of noise to the augmented data, which if need be can be deployed at rural locations which lack …
Communications Using Deep Learning Techniques, Priti Gopal Pachpande
Communications Using Deep Learning Techniques, Priti Gopal Pachpande
Legacy Theses & Dissertations (2009 - 2024)
Deep learning (DL) techniques have the potential of making communication systems
Emotion Forecasting In Dyadic Conversation : Characterizing And Predicting Future Emotion With Audio-Visual Information Using Deep Learning, Sadat Shahriar
Emotion Forecasting In Dyadic Conversation : Characterizing And Predicting Future Emotion With Audio-Visual Information Using Deep Learning, Sadat Shahriar
Legacy Theses & Dissertations (2009 - 2024)
Emotion forecasting is the task of predicting the future emotion of a speaker, i.e., the emotion label of the future speaking turn–based on the speaker’s past and current audio-visual cues. Emotion forecasting systems require new problem formulations that differ from traditional emotion recognition systems. In this thesis, we first explore two types of forecasting windows(i.e., analysis windows for which the speaker’s emotion is being forecasted): utterance forecasting and time forecasting. Utterance forecasting is based on speaking turns and forecasts what the speaker’s emotion will be after one, two, or three speaking turns. Time forecasting forecasts what the speaker’s emotion will …
Autonomous Spectrum Enforcement : A Blockchain Approach, Maqsood Ahamed Abdul Careem
Autonomous Spectrum Enforcement : A Blockchain Approach, Maqsood Ahamed Abdul Careem
Legacy Theses & Dissertations (2009 - 2024)
A core limitation in existing wireless technologies is the scarcity of spectrum, to support the exponential increase in Internet-connected and multimedia-capable mobile devices and the increasing demand for bandwidth-intensive services. As a solution, Dynamic Spectrum Access policies are being ratified to promote spectrum sharing for various spectrum bands and to improve the spectrum utilization. This poses an equally challenging problem of enforcing these spectrum policies. The distributed and dynamic nature of policy violations necessitates the use of autonomous agents to implement efficient and agile enforcement systems. The design of such a fully autonomous enforcement system is complicated due to the …
Synchronous Subgraphs In Networks, Navita Jain
Synchronous Subgraphs In Networks, Navita Jain
Legacy Theses & Dissertations (2009 - 2024)
Community detection is a central problem in network analysis. The majority of existing work defines communities as subsets of nodes that are structurally well-connected and isolated from the rest of the network. Apart from their underlying connectivity, nodes in real-world networks exhibit temporal activity: user posts in social networks, browsing activity on web pages and neuron activations in brain networks to name a few. While edges encode potential for community interactions, participation in the community can be quantified by synchronized member activity. Given both the network structure and individual node activity, how can we detect communities that are both well-connected …
Effective Entity Resolution Methodology For Improving Data Quality And Reliability Of Service-Oriented Applications, Ewa Musial
Legacy Theses & Dissertations (2009 - 2024)
This dissertation proposes new paradigms for improving the testing, reliability of service-oriented applications as well as the quality of data. Since it is difficult to track information flowing through the multiple tiers of an application, testing service-oriented systems can be very challenging. We present a methodology for testing service-oriented applications that takes into account all the components, including services, external services, and data components. The results of our experiments demonstrate that this approach greatly improves the effectiveness of testing service-oriented applications.
Ensuring The Effectiveness Of Information Security Policy : The Development And Validation Of An Information Security Policy Model, Vivid Vicki Chen
Ensuring The Effectiveness Of Information Security Policy : The Development And Validation Of An Information Security Policy Model, Vivid Vicki Chen
Legacy Theses & Dissertations (2009 - 2024)
The purpose of this dissertation is to develop and test a conceptual model of an Information Security Policy (ISP) and to measure the benefits that accrue to organizations that implement and deploy such policies. As a result of rapid changes in technology, the importance of computer Information Security Policy (ISP) has increased dramatically. In recent decades, governments and private enterprises have increasingly come to store ever greater amounts of information on computers and on networks. Unfortunately, storing information in this manner not only makes firms engaged in cutting-edge technology vulnerable to hackers, but may also jeopardize customer / employee relations, …
Graphene-Based Post-Cmos Architecture, Sansiri Tanachutiwat
Graphene-Based Post-Cmos Architecture, Sansiri Tanachutiwat
Legacy Theses & Dissertations (2009 - 2024)
The semiconductor industry relies on CMOS technology which is nearing its scaling limitations. In order to continue the historical growth rate of the device density of digital logic chips, novel nanomaterials and nanodevices will need to be developed.