Off-campus UNL users: To download campus access dissertations, please use the following link to log into our proxy server with your NU ID and password. When you are done browsing please remember to return to this page and log out.
Non-UNL users: Please talk to your librarian about requesting this dissertation through interlibrary loan.
Innovative Framework of Trustworthy Computing and Transmission for Electrocardiogram Physiological Signals
Innovative integration of Diagnosis computing, Steganography, and Transmission for Electrocardiograms is proposed. Most of today’s internet-based healthcare applications rely on the cloud, i.e. they execute computation within datacenters/cloud and data storage. This cloud-based approach brings strong operational benefits, but also comes with important limitations in terms of loss of privacy, cost, availability, and latency. To address the challenges posed by pure cloud-based deployments, alternative distributed architectures are emerging that emphasize decentralized and loosely coupled interactions. The increasing popularity of Edge Computing and Internet of Things-based wearables coupled with body sensors offer us a unique idea of integrating the computational resource distribution of Diagnosis, Steganography, and Transmission tasks in the healthcare domain. In Diagnosis-Steganography-Transmission (DST) framework for health monitoring and real-time diagnosis especially designed for Coronary Artery Disease diagnostic purposes work by extracting the patient’s health state by a real-time execution of a preliminary diagnostic algorithm in the local embedded computing platform before the ECG data is transmitted over wireless networks for further analysis. Local pre-diagnosis assists the operation of the communication module in deciding when and how much data should be transmitted and the given quality. The novelty of DST is that the unequal importance of physiological signals (e.g., ECG) feature offers an inherent connection between computing distribution of diagnosis, Unequal Embedding Strength (UES)-based steganography, and Unequal Error Protection-based communication. Our UES-based Unequal Steganography Embedding algorithm provides security to the patient’s confidential data with low Wavelet-based Weighed Percentage Root-mean-square Difference (WWPRD) (< 0.5%). A lower WWPRD signifies a reduction in error of the received ECG waveform, thus a higher diagnosis quality. By DST, the steganography embedder, source encoder, and channel encoder in the communication module effectively reconfigure the intricacy of the control factors to match the energy constraints while maintaining ECG reconstruction quality. Moreover, the diagnosis module also reconfigures the complexity of the process of diagnosis to match communication bandwidth constraints. Furthermore, in DST, energy is a function of diagnosis depth that further saves energy because the severity of the patient’s health decides diagnosis depth, which decides the number of bits needs to be transmitted. Additionally, instead of always-on communication, intermittent communication based on local pre-diagnosis significantly saves energy - the communication energy consumption depends on the transmission time and by intermittent communication, a significant amount of energy (up to 99.3% in typical application set-up) has been saved.
Sahu, Neerja, "Innovative Framework of Trustworthy Computing and Transmission for Electrocardiogram Physiological Signals" (2020). ETD collection for University of Nebraska - Lincoln. AAI27955810.