Dr. Yazan A Alqudah

Car accident rates have increased in recent years, resulting in losses in human lives, properties, and other financial costs. An embedded machine learning-based system is developed to address this critical issue. The system can monitor road conditions, detect driving patterns, and identify aggressive driving behaviors. The system is based on neural networks trained on a comprehensive dataset of driving events, driving styles, and road conditions. The system effectively detects potential risks and helps mitigate the frequency and impact of accidents. The primary goal is to ensure the safety of drivers and vehicles. Collecting data involved gathering information on three key road events: normal street and normal drive, speed bumps, circular yellow speed bumps, and three aggressive driving actions: sudden start, sudden stop, and sudden entry. The gathered data is processed and analyzed using a machine learning system designed for limited power and memory devices. The developed system resulted in 91.9% accuracy, 93.6% precision, and 92% recall. The achieved inference time on an Arduino Nano 33 BLE Sense with a 32-bit CPU running at 64 MHz is 34 ms and requires 2.6 kB peak RAM and 139.9 kB program flash memory, making it suitable for resource-constrained embedded systems.

The advent of cloud computing and the ubiquity of broadband wireless coverage and wide spread usage of smart phones around the world carries the potential for transforming health care services, reducing health care cost and ensuring faster care for urgent cases. To these objectives, we present a cloud-based mobile health monitoring solution that takes advantage of cloud infrastructure and mobile processing capability to address the rising cost of health care monitoring. The solution enables health care providers to remotely analyze, monitor and diagnose patient's data. The solution integrates a powerful data analysis tool, cloud computing and mobile services. This paper presents a proof of concept that has been developed to monitor, record and analyze heart rate. The design enables a physician to develop custom analysis and monitoring to collect key indicator or set alerts without a need for infrastructure implementations to store or transfer the data.

To meet the ever-growing demand for higher data rates, accurate channel models are needed for optimal design and deployment of mobile wireless networks. This work proposes a new method for addressing path loss modeling at 800 MHz of suburban environment based on field measurements. Using generalized linear mixed models, we develop a new statistical model that accounts for the autocorrelation among measurements at the same distance at different times. The proposed method allows linear, quadratic, and cubic relationship forms between the path loss measurements and the natural logarithm of the distance, which is almost unexplored as existing models use a straight line relationship. A comparison study consists of comparing nine propagation models in terms of the mean absolute prediction error. The new model performs over 30% better than the existing models for the considered measurements. We also show that a cubic relationship form between path loss measurements and the logarithm of distance could be more suitable than a straight line form for prediction purposes. The results show that the generalized linear mixed models significantly improve the prediction power regardless of the form of the model (linear, quadratic, or cubic).

With the introduction of autonomous and self-driving cars, innovative research is needed to ensure safety and reliability on the road. This work introduces a solution to understand vehicle behaviour based on sensors data. The behaviour is classified according to driving events. Understanding driving events can play a significant role in road safety and estimating the expense and risks of driving a vehicle. Rather than relying on the distance and time driven, driving events can provide a more accurate measure of vehicle driving consumption. This measure will become valuable as more ride-sharing applications are introduced to roads around the world. Estimating driving events can also help better design the road infrastructure to reduce congestion, energy consumption and pollution. By sharing data from official vehicles and volunteers, crowd sensing can be used to better understand congestion and road safety. This work studies driving events and proposes using machine learning to classify these events into different categories. The acquired data is collected using embedded mobile device motion sensors to train machine learning algorithms to classify the events.

In this paper, a hybrid energy storage system (HESS), combining a battery and a supercapacitor (SC), is studied for dispatching solar power at one hour increments for an entire day for 1 MW grid connected photovoltaic (PV)arrays. HESS relies on PV for charging and not the grid, and hence is immune to fluctuating electricity prices. The battery and SC are intended to supply predetermined constant power, but not to provide ancillary services for the grid operation. To develop a cost-effective energy storage system, a low pass filter (LPF) is used to allocate the power between a battery and a SC. The best cost of the energy storage is calculated based on the time constant of the LPF through extensive simulations. Several rule-based algorithms based on the battery state of charge (SOC) are developed to estimate the grid reference power for each one-hour dispatching period. An economic comparison of using different kinds of the algorithms for estimating the grid reference power are also presented in this study. The objective is to better understand the annual energy storage cost for hourly dis-patching solar power. The actual solar data of four different days as a representative of each season recorded at Oak Ridge National Laboratory are used in the simulations. The relationship between the actual PV cell temperature and the ambient temperature in power output calculations are also considered and their effects on energy storage price calculations are presented in this paper.

Anti-Lock Braking Systems (ABS) and Brake-by-Wire Systems (BBW) are safety-critical applications by nature. Such systems are required to demonstrate high degrees of dependability. Fault-tolerance is the primary means to achieve dependability at runtime and has been an active research area for decades. Fault-tolerance is usually achieved in traditional embedded computing systems through redundancy and voting methods. In such systems, hardware units, actuators, sensors, and communication networks are replicated where special voters vote against faulty units. In addition to traditional hardware and software redundancy, hybrid and reconfiguration-basedapproachestofault-toleranceareevolving.Inthisarticle, we present a reconfiguration-based fault-tolerant approach to achieve high dependability in ABS BBW braking systems. The proposed architecture makes use of other components of less safety-critical systems to maintain high dependability in the more safety-critical systems. This is achieved by migrating safety-critical software tasks from embedded computerhardwarethatrunsintoamalfunctiontootherembeddedcomputinghardwarerunningless-critical software tasks. Or by using a different configuration in terms of the used speed sensors and type of ABS. The proposed architecture is on average 20% more reliable than conventional ABS architectures assuming equal reliabilities of different components.

Exposing engineering students during their education to real-world problems and giving them the chance to apply what they learn in the classroom is a vital element of engineering education. The Embedded Systems course at Princess Sumaya University for Technology (PSUT) is one of the main courses that bridge the gap between theoretical electrical engineering education and the real-world. This paper presents the experience of applying project-based learning to enhance teaching the Embedded Systems course at PSUT. The feedback from students illustrated the effectiveness of this method in enhancing the understanding and the ability of students in applying embedded systems design concepts to solve real-world engineering problems.

The ubiquity of broadband wireless coverage and widespread usage of smart phones around the world carry a potential for transforming health care services, reducing health care cost, and ensuring faster care for urgent cases. To these objectives, we present a mobile-based health monitoring solution that takes advantage of the mobile's increasing processing capability to address the rising cost of health care. The solution enables health care providers to easily analyze and diagnose a patient's data. This is possible due to the low cost in integrating a powerful data analysis tool with the mobile device. This paper presents a proof of concept that has been developed to monitor, record, and analyze the heart rate. The design enables a physician to develop custom analysis and monitoring to collect key indicators or set alerts without a need for infrastructure implementations to store or transfer the data.

Engineering education is constantly challenged to bridge the gap between classroom and real world problems. This paper reports on the experience at PSUT to complement the teaching of digital signal processing through project-based learning. A low cost digital stethoscope is utilized by students to record heartbeat sounds and apply DSP concepts to extract vital information from the signal. The ability to apply concepts learned in the classroom to a real world problem was an important element that motivated students. The feedback from students reinforced the effectiveness of these techniques in teaching concepts of digital signal processing.

WiMAX technology carries the promise of broadband access and wireless coverage. Developing countries throughout the world have been fast at adopting and employing the new technology to bridge the digital divide. The deployment of WiMAX networks enables the validation and testing of the technology. It is imperative that the technology be tested in different environments and the results shared and compared. Jordan provides a unique environment in its architecture, building construction materials, usage model, topology and vegetation. This work considers a mobile WiMAX network operating at 3.5 GHz deployed in Amman, Jordan. The work presents a new model for predicting path loss based on the results of field measurements of signals power and it compares proposed model and measured data to different propagation models.