Browsing by Author "Electrical and Computer Engineering, School of Engineering and Technology"
Now showing 1 - 10 of 178
Results Per Page
Sort Options
Item 3-Level Residual Capsule Network for Complex Datasets(IEEE, 2020-02) Bhamidi, Sree Bala Shruthi; El-Sharkawy, Mohamed; Electrical and Computer Engineering, School of Engineering and TechnologyThe Convolutional Neural Network (CNN) have shown a substantial improvement in the field of Machine Learning. But they do come with their own set of drawbacks. Capsule Networks have addressed the limitations of CNNs and have shown a great improvement by calculating the pose and transformation of the image. Deeper networks are more powerful than shallow networks but at the same time, more difficult to train. Residual Networks ease the training and have shown evidence that they can give good accuracy with considerable depth. Residual Capsule Network [15] has put the Residual Network and Capsule Network together. Though it did well on simple dataset such as MNIST, the architecture can be improved to do better on complex datasets like CIFAR-10. This brings us to the idea of 3-Level Residual Capsule which not only decreases the number of parameters when compared to the seven-ensemble model, but also performs better on complex datasets when compared to Residual Capsule Network.Item 3D Centroidnet: Nuclei Centroid Detection with Vector Flow Voting(IEEE, 2022-10) Wu, Liming; Chen, Alain; Salama, Paul; Dunn, Kenneth W.; Delp, Edward J.; Electrical and Computer Engineering, School of Engineering and TechnologyAutomated microscope systems are increasingly used to collect large-scale 3D image volumes of biological tissues. Since cell boundaries are seldom delineated in these images, detection of nuclei is a critical step for identifying and analyzing individual cells. Due to the large intra-class variability in nuclei morphology and the difficulty of generating ground truth annotations, accurate nuclei detection remains a challenging task. We propose a 3D nuclei centroid detection method by estimating the "vector flow" volume where each voxel represents a 3D vector pointing to its nearest nuclei centroid in the corresponding microscopy volume. We then use a voting mechanism to estimate the 3D nuclei centroids from the "vector flow" volume. Our system is trained on synthetic microscopy volumes and tested on real microscopy volumes. The evaluation results indicate our method outperforms other methods both visually and quantitatively.Item A Highly Miniaturized, Chronically Implanted ASIC for Electrical Nerve Stimulation(IEEE, 2022) Shah, Jay V.; Quinkert, Christopher J.; Collar, Brett J.; Williams, Michael T.; Biggs, Ethan N.; Irazoqui, Pedro P.; Electrical and Computer Engineering, School of Engineering and TechnologyWe present a wireless, fully implantable device for electrical stimulation of peripheral nerves consisting of a powering coil, a tuning network, a Zener diode, selectable stimulation parameters, and a stimulator IC, all encapsulated in biocompatible silicone. A wireless RF signal at 13.56 MHz powers the implant through the on-chip rectifier. The ASIC, designed in TSMC’s 180 nm MS RF G process, occupies an area of less than 1.2 mm2. The IC enables externally selectable current-controlled stimulation through an on-chip read-only memory with a wide range of 32 stimulation parameters (90 – 750 μA amplitude, 100 μs or 1 ms pulse width, 15 or 50 Hz frequency). The IC generates the constant current waveform using an 8-bit binary weighted DAC and an H-Bridge. At the most power-hungry stimulation parameter, the average power consumption during a stimulus pulse is 2.6 mW with a power transfer efficiency of ~5.2%. In addition to benchtop and acute testing, we chronically implanted two versions of the device (a design with leads and a leadless design) on two rats’ sciatic nerves to verify the long-term efficacy of the IC and the full system. The leadless device had the following dimensions: height of 0.45 cm, major axis of 1.85 cm, and minor axis of 1.34 cm, with similar dimensions for the device with leads. Both devices were implanted and worked for experiments lasting from 21–90 days. To the best of our knowledge, the fabricated IC is the smallest constant-current stimulator that has been tested chronically.Item A Theoretical Study on Porous-Silicon Based Synapse Design for Neural Hardware(IEEE, 2021-12) Sikder, Orthi; Schubert, Peter; Electrical and Computer Engineering, School of Engineering and TechnologyPorous silicon (po-Si) is a form of silicon (Si) with nanopores of tunable sizes and shapes distributed over the bulk structure. Although crystalline Si (c-Si) is already established as one of the most advantageous and promising elements for its technological significance, the additional key aspect of po-Si is its large surface area with respect to its small volume which is beneficial for surface chemistry. In this work, we explore the design of a po-Si based synaptic device and investigate its potential for neuromorphic hardware. First, we analyze several electrical properties of po-Si through density functional theory (Ab Initio/ first principle) calculation. We show that the presence of intra-pore dangling states appears within the bandgap region of po-Si. While the bandgap of the po-Si is well known to be higher than c-Si yielding low carrier density and higher resistance, the appearance of these dangling states can significantly participate in electronic transport through hopping mechanism. Then, we analyze the electric-field driven modulation in the dangling bond through controlled intra-pore Si-H bond dissociation. Such modulation of the dangling state density further allows the tenability of the po-Si conductance. Finally, we theoretically evaluate the current-voltage characteristics of our proposed po-Si based synaptic devices and determine the possible range of obtainable conductivity for different porosity. Our analysis signifies that the integration of such devices in the synaptic fabric can enable significantly denser and energy-efficient neuromorphic hardware.Item A-MnasNet and Image Classification on NXP Bluebox 2.0(ASTES, 2021-01) Shah, Prasham; El-Sharkawy, Mohamed; Electrical and Computer Engineering, School of Engineering and TechnologyComputer Vision is a domain which deals with the challenge of enabling technology with vision capabilities. This goal is accomplished with the use of Convolutional Neural Networks. They are the backbone of implementing vision applications on embedded systems. They are complex but highly efficient in extracting features, thus, enabling embedded systems to perform computer vision applications. After AlexNet won the ImageNet Large Scale Visual Recognition Challenge in 2012, there was a drastic increase in research on Convolutional Neural Networks. The convolutional neural networks were made deeper and wider, in order to make them more efficient. They were able to extract features efficiently, but the computational complexity and the computational cost of those networks also increased. It became very challenging to deploy such networks on embedded hardware. Since embedded systems have limited resources like power, speed and computational capabilities, researchers got more inclined towards the goal of making convolutional neural networks more compact, with efficiency of extracting features similar to that of the novel architectures. This research has a similar goal of proposing a convolutional neural network with enhanced efficiency and further using it for a vision application like Image Classification on NXP Bluebox 2.0, an autonomous driving platform by NXP Semiconductors. This paper gives an insight on the Design Space Exploration technique used to propose A-MnasNet (Augmented MnasNet) architecture, with enhanced capabilities, from MnasNet architecture. Furthermore, it explains the implementation of A-MnasNet on Bluebox 2.0 for Image Classification.Item A-MnasNet: Augmented MnasNet for Computer Vision(IEEE, 2020-08) Shah, Prasham; El-Sharkawy, Mohamed; Electrical and Computer Engineering, School of Engineering and TechnologyConvolutional Neural Networks (CNNs) play an essential role in Deep Learning. They are extensively used in Computer Vision. They are complicated but very effective in extracting features from an image or a video stream. After AlexNet [5] won the ILSVRC [8] in 2012, there was a drastic increase in research related with CNNs. Many state-of-the-art architectures like VGG Net [12], GoogleNet [13], ResNet [18], Inception-v4 [14], Inception-Resnet-v2 [14], ShuffleNet [23], Xception [24], MobileNet [6], MobileNetV2 [7], SqueezeNet [16], SqueezeNext [17] and many more were introduced. The trend behind the research depicts an increase in the number of layers of CNN to make them more efficient but with that the size of the model increased as well. This problem was fixed with the advent of new algorithms which resulted in a decrease in model size. As a result, today we have CNN models which are implemented on mobile devices. These mobile models are small and fast which in turn reduce the computational cost of the embedded system. This paper resembles similar idea, it proposes a new model Augmented MnasNet (A-MnasNet) which has been derived from MnasNet [1]. The model is trained with CIFAR-10 [4] dataset and has a validation accuracy of 96.89% and a model size of 11.6 MB. It outperforms its baseline architecture MnasNet which has a validation accuracy of 80.8% and a model size of 12.7 MB when trained with CIFAR-10.Item Abandoned Mine Voids for Pumped Storage Hydro(Juniper Publishers, 2019) Schubert, Peter J.; Izadian, Afshin; Wheeler, J. W.; Electrical and Computer Engineering, School of Engineering and TechnologyPumped Storage Hydro (PSH) is geographically limited but can expand greatly if abandoned subsurface coal mines are leveraged for the lower reservoir. Such lands are already permitted, generally less desirable, and found in regions eager for job creation. Vertical stacking of the upper and lower reservoirs is an efficient use of the land. Water can be raised by electric pumps as part of energy arbitrage; however, water can also be raised with Hydraulic Wind Turbines. HWTs are far less costly than traditional electric turbines, and start-up at lower wind speeds - thereby extending their geographic range. The HWT masts can serve double duty as tent poles to support translucent architectural fabric over the surface lake. This prevents evaporation and ingress of wildlife, and provides an interior space useful for non-electric revenue, such as aquaculture and greenhouses. Water cycled through the system can, in some cases, supplement local sources. Seepage through water tables replenishes clean water. Subsurface water is cool and can be circulated through server farms in data centers which represents a potential revenue source that can be started up well in advance of the primary energy storage operation. Combined, these factors bring an innovative solution to site selection, design, and engineering for PSH which promises accelerated commissioning and permitting, and low-cost operation. The bottom line for communities in Coal Country is more jobs and cheaper power.Item Adaptive Rule-Based Energy Management Strategy for a Parallel HEV(MDPI, 2019) Bagwe, Rishikesh Mahesh; Byerly, Andy; dos Santos, Euzeli Cipriano, Jr.; Electrical and Computer Engineering, School of Engineering and TechnologyThis paper proposes an Adaptive Rule-Based Energy Management Strategy (ARBS EMS) for a parallel hybrid electric vehicle (HEV). The aim of the strategy is to facilitate the aftermarket hybridization of medium- and heavy-duty vehicles. ARBS can be deployed online to optimize fuel consumption without any detailed knowledge of the engine efficiency map of the vehicle or the entire duty cycle. The proposed strategy improves upon the established Preliminary Rule-Based Strategy (PRBS), which has been adopted in commercial vehicles, by dynamically adjusting the regions of operations of the engine and the motor. It prevents the engine from operating in highly inefficient regions while reducing the total equivalent fuel consumption of the vehicle. Using an HEV model developed in Simulink®, both the proposed ARBS and the established PRBS strategies are compared over an extended duty cycle consisting of both urban and highway segments. The results show that ARBS can achieve high MPGe with different thresholds for the boundary between the motor region and the engine region. In contrast, PRBS can achieve high MPGe only if this boundary is carefully established from the engine efficiency map. This difference between the two strategies makes the ARBS particularly suitable for aftermarket hybridization where full knowledge of the engine efficiency map may not be available.Item Administrative Policy for Stochastic Democracy(AIAA, 2018-09) Schubert, Peter J.; Sommer, Joe; Electrical and Computer Engineering, School of Engineering and TechnologyPrior studies of stochastic democracy have compared it to other forms of governance, demonstrated how to scale up or scale down as population changes, and developed an algorithm for start-up on Day 1. Left unanswered is the administrative policy for regulating the statutes developed by the legislative bodies. As the aim of stochastic democracy is design of a corruption-resistance form of managing human affairs the implementation of the activities of the government must also be robust against undue influence, bribery, and abuse of power. Decision-makers in a stochastic democracy by design cannot be “career” politicians, however, the bureaucrats of the government agencies or departments or ministries are advantageously retained across the changes in the legislative bodies. This quality invites corruption, the answer to which cannot be simply to apply oversight or policing. In this paper is developed an integrated structure which supplants the Byzantine-derived corporate-style hierarchy. Seven principles are applied to the bureaucracy and their integration and practice described herein as administrative policy, the principles being: transparency of regulatory process; not-less-than time limits; disclosure of change proposers; inclusion of economic externalities; open debate and notices of intent; chairmanship and participant selection; and periodic but stochastic changes in the number of agencies at each level of governance. This latter enforces either consolidation or expansion, within high and low limits, the re-organization of which will shuffle the reporting structure of the regulatory bureaucracy and disrupt entrenched habits and possible corrupting schemes. When complementing the legislative functions this work rounds-out the formation of a corruption-resistant, scalable form of truly representative governance for space habitats and societies of arbitrary size.Item Advanced Image Processing for Defense and Security Applications(SpringerOpen, 2011-03-09) Du, ElizaYingzi; Ives, Robert; Nevel, Alan van; She, Jin-Hua; Electrical and Computer Engineering, School of Engineering and Technology