Embedded System Research Articles

Embedded System for the Simultaneous Study of SAHS and Cardiac Arrhythmia

This paper presents a proposed embedded system for the simultaneous study of Sleep Apnea/Hypopnea Syndrome (SAHS) and related cardiac arrhythmias, which can affect about 1 billion people worldwide. Its objective is to become a tool to research the relational causes of these disruptions. It can record biomedical signals and vital parameters related to cardiac and respiratory activity commonly used in clinical practice. The system is designed to be used in sleep studies, providing greater comfort to users by integrating two devices into one. The proposed system is battery-powered and has Bluetooth wireless communication for remote studies. The design is based on specific electronic components such as Analog Front-End and System on Chips, and best-practice guidelines for PCB design were followed. A hybrid-scheduler was selected as software architecture to ensure the task execution at specific intervals. The systems signal quality recording sub-system was tested using a multiparametric simulator, commercial holter device and acquired real records. The results are similar to other works.

34‐1: Invited Paper: Future AR‐HUD System

The automotive head‐up display (HUD) was introduced by General Motors in 1988 to prevent driver distraction. Augmented reality (AR) technology has made the HUD more attractive for ADAS visualization, with the ability to overlay driving and sensed information on the road. Early versions of AR‐HUD products have limited functionality due to conventional optic technology and a limited perception of the environment. Future AR‐HUD will require a large field of view, multiplane capability, real‐time sensor fusion and correction, and an embedded system implementation. This paper reviews the future enabling technologies and reference implementations for AR‐HUD.

Energy-Aware Scenario-Based Mapping of Deep Learning Applications Onto Heterogeneous Processors Under Real-Time Constraints

To cope with the increasing demand for deep learning applications in embedded systems, emerging embedded devices tend to equip multiple heterogeneous processors, including GPU and deep learning hardware accelerator, called neural processing unit (NPU). It becomes popular to run multiple deep learning (DL) applications simultaneously to provide several functionalities. In this work, we assume that applications have real-time constraints that may vary at run time. While extensive studies have been conducted recently to find an efficient mapping of multiple DL applications on various hardware platforms, they do not consider the constraints imposed by the NPU and the associated software development kit (SDK) in a real embedded platform. In this paper, we propose a novel energy-aware mapping methodology of multiple DL applications onto a real embedded system that has multiple heterogeneous processors. The objective is to minimize energy consumption while satisfying the real-time constraints of all applications. In the proposed scheme, we first select Pareto-optimal mapping solutions for each application. Then mapping combination is explored, considering the scenario that indicates the dynamism of applications while satisfying the constraints. Also, we reduce energy consumption by tuning the frequency of processors. We could satisfy up to <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$40\%$</tex-math></inline-formula> higher deadline constraints and reduce the energy consumption by <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$22\%$</tex-math></inline-formula> <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$\sim$</tex-math></inline-formula> 31% compared to the static mapping methods with real-life applications and different scenarios on a real platform.

A Novel Exploitative and Explorative GWO-SVM Algorithm for Smart Emotion Recognition

Emotion recognition or detection is broadly utilized in patient-doctor interactions for diseases such as schizophrenia and autism and the most typical techniques are speech detection and facial recognition. However, features extracted from these behavior-based emotion recognitions are not reliable since humans can disguise their emotions. Recording voices or tracking facial expressions for a long term is also not efficient. Therefore, our aim is to find a reliable and efficient emotion recognition scheme, which can be used for non-behavior-based emotion recognition in real-time. This can be solved by implementing a single-channel electrocardiogram (ECG) based emotion recognition scheme in a lightweight embedded system. However, existing schemes have relatively low accuracy. For instance, the accuracy is about 82.78% by using a least squares support vector machine (SVM). Therefore, we propose a reliable and efficient emotion recognition scheme—exploitative and explorative grey wolf optimizer-based SVM (X-GWO-SVM) for ECG-based emotion recognition. Two datasets, one raw self-collected iRealcare dataset, and the widely-used benchmark WESAD dataset are used in the X-GWO-SVM algorithm for emotion recognition. Leave-single-subject-out cross-validation yields a mean accuracy of 93.37% for the iRealcare dataset and a mean accuracy of 95.93% for the WESAD dataset. This work demonstrates that the X-GWO-SVM algorithm can be used for emotion recognition and the algorithm exhibits superior performance in reliability compared to the use of other supervised machine learning methods in earlier works. It can be implemented in a lightweight embedded system, which is much more efficient than existing solutions based on deep neural networks.

An Enhanced Anti-surge Circuit for New Energy Vehicle Electronics System

This paper presents an input surge protection circuit for automotive electronics systems with an adjustable overvoltage protection threshold suitable for high input voltage applications. Compared with the conventional overvoltage protection circuit, it boasts a new high-voltage bandgap reference circuit design without pre-regulation, which takes up less space and can be used on a limited PCB board, expanding the practical application range of the chip. The gate protection circuit can monitor the gate-source voltage of the power MOS in real-time and control the charge pump’s operating mode to improve the system’s safety. The design and simulation of the circuit adopt the 0.18 μm 5V/25V BCD process. The simulation results show that the bandgap reference voltage varies by only 0.38 mV in the input voltage range of 3.5 V-25 V. The quiescent current is 146 uA at 5 V supply voltage. The power MOS can be turned on quickly within 50 us. The overvoltage response time is 599 ns, permitting the circuit to be shut down when a surge occurs.

Analytical Calculation of Air Gap Magnetic Field of SPMSM with Eccentrically Cut Poles Based on Magnetic Pole Division

In the design process of surface-mounted permanent magnet motor (SPMSM) for industrial robots and computer numerical control (CNC) machine tools, to pursue the sinusoidal nature of the back electromotive force, the magnetic poles in the form of eccentric pole cutting structure are often used. To analyze the no-load air gap magnetic field of the SPMSM with eccentrically cut poles simply and accurately, a subdomain model magnetic field analytical calculation method based on equal-area integral block processing of permanent magnets is proposed. The problem that the traditional subdomain analysis model cannot be directly applied to the SPMSM with eccentrically cut poles of unequal thickness is solved. The proposed method considers the influence of stator slotting and the actual permeability of permanent magnets, and directly obtains the fundamental wave and harmonic components of the no-load air gap flux density by solving the subdomain model. The finite element method (FEM) is used to directly calculate the air gap magnetic field for verification. The results of the analytical method and the no-load air gap magnetic density calculated by the FEM are consistent, which verifies the accuracy of the proposed analytical method and can quickly guide the design of the SPMSM with eccentrically cut poles.

Federated Deep Reinforcement Learning for Energy-Efficient Edge Computing Offloading and Resource Allocation in Industrial Internet

Industrial Internet mobile edge computing (MEC) deploys edge servers near base stations to bring computing resources to the edge of industrial networks to meet the energy-saving requirements of Industrial Internet terminal devices. This paper considers a wireless MEC system in an intelligent factory that has multiple edge servers and mobile smart industrial terminal devices. In this paper, the terminal device has the choice of either offloading the task in whole or in part to the edge server, or performing it locally. Through combined optimization of the task offload ratio, number of subcarriers, transmission power, and computing frequency, the system can achieve minimum total energy consumption. A computing offloading and resource allocation approach that combines federated learning (FL) and deep reinforcement learning (DRL) is suggested to address the optimization problem. According to the simulation results, the proposed algorithm displays fast convergence. Compared with baseline algorithms, this algorithm has significant advantages in optimizing the performance of energy consumption.

Review Paper on Smart Guidance Belt For Blinds

Abstract: This project seeks to create a simplified version of a smart wearable belt that can assist blind people with navigation, using a sensible and compact design for the electronic equipment. The design includes an embedded system, and could potentially incorporate machine learning algorithms to detect potholes and other irregularities on the road. The goal is to develop a smart belt that can help blind people travel independently without the use of a cane. The Electronic Blind Mobility Aid is a technology that uses sensors put on a belt worn around the waist to allow blind persons to travel independently. This advanced technology can help many people with visual impairments to travel independently. Blind people often face obstacles that can make it difficult for them to travel alone, but a blind belt can extend their range of perception by serving as an obstacle detector. Visually impaired people often use their hearing to compensate for their reduced eyesight, and can recognize sound sources. This technology enables blind persons to travel independently and receive obstacle alerts via headphones and speakers using sensors installed on a belt that is worn around the waist. Blind people can utilise this approach every day to find potential hazards, roadblocks, and pathways. In the end, this technology will serve to increase the independence of blind individuals and give them more confidence when they travel. To notify users of potential dangers, the device has a buzzer that produces vibration signals and emits a warning sound. Both the sound frequency and the vibration frequency rise as the space between the elastic gloves and the barriers gets smaller. The design incorporates a buzzer and motor in addition to the capability to notify the proper person in case of an emergency.. The goal of this study is to provide a low-cost, effective means of helping blind individuals travel more easily, quickly, and confidently. This design provides a quick-response, cost-effective, and portable solution. This device can let blind persons traverse their environment more safely and confidently by employing sound and vibration cues to notify the user of impediments.

Vision Controlled Motorised Wheelchair

Abstract: Even in this modern day, it is very difficult for physically challenged people to move from one place to other without others help. To give them an independent feel, a smart eye-controlled system is provided. A novel design of an embedded cardiorespiratory monitoring system for wheelchair users is proposed. This system gets the input from eye. An eye-ball controlled wheelchair is a type of assistive technology designed for individuals with paralysis or limited mobility. The wheelchair is controlled through the movement of the user's eyes, allowing them to navigate the chair without the use of their hands or arms. The system consists of a set of sensors that detect eye movements, a microcontroller that interprets the signals, and a motor controller that translates the signals into movement. The user wears a headset with a camera that tracks the movement of their eyes, which is then relayed to the sensors. The wheelchair can be programmed to respond to different eye movements, such as blinking or sustained gaze in a particular direction. The system is designed to be intuitive and easy to use, with minimal training required. One of the primary benefits of the eye-ball controlled wheelchair is increased independence for individuals with limited mobility. It allows them to move around their environment and interact with others without needing assistance. Additionally, the system is non-invasive and does not require surgery or implants. The movement of the wheel chair will be controlled by these inputs. Depending on this input the wheel chair will move forward, left, right or backward. With the help of this system the user can move independently without others help. It also included with obstacle detection system to ensure the safety for the user. If any obstacle is detected then the indication can be given with the help of LCD.The entire system is composed of a sensor node, and a cloud server. The former is used to obtain continuous vitalsign signals, and the latter processes the sampled signals to estimate the temperature level, implemented in an embedded system to achieve a fully integrated radar system. MATLAB image processing is used to sense the movement of the eye ball’s direction using eye ball sensor (Straight Command, Left/Right Command, Stop Command). The system is cost effective and thus can be used by patients spread over a large economy range

Design of Women Safety Device with Fingerprint Sensor

bstract: The safety of women has always been a global concern, and despite various measures adopted, the number of cases of misconduct against women is steadily rising. To address this issue, a new approach in the form of a device has been proposed. The purpose of this gadget is to give women a tool to safeguard their security and safety in the incident of an accident. To ensure that no woman feels helpless in the face of such crime issues, defense mechanisms must be proposed in large-scale productions. The suggested gadget is an embedded system designed on an Arduino board and a NodeMCU module with GPS that delivers a warning sound will be heard in the event of danger and an emergency message is sent with the user's location. To safeguard women's safety, this technology follows the location and transmits a message to the police control room, family, and friends. Moreover, a fingerprint module built into the system makes sure that only authorized users are utilizing the equipment. The smartphone will deem it dangerous and automatically send an SMS to the appropriate authority if the user does not touch their fingertips on the module that stores fingerprints within a predetermined amount of time. In conclusion, the proposed device could be a valuable tool in ensuring the safety and security of women. With the integration of technology, this device could provide a sense of security to women and help prevent crimes against them.

Fuel Monitoring System

Abstract: In recent times, there has been a significant increase in the number of vehicles, particularly those running on fuel. Unfortunately, the rising fuel prices have led to instances where some petrol pump owners attempt to deceive their customers. They manipulate the fuel data or display incorrect information when customers request fuel filling. To address this problem, we have developed a solution called the Smart Fuel Meter, which accurately measures the amount of fuel dispensed into the vehicle's tank and calculates the corresponding cost based on the current fuel price. The Smart Fuel Meter incorporates Internet of Things (IoT) technology, utilizing components such as the ESP12E and an embedded system to enable efficient fuel quantity measurement and fraud detection. To enhance user accessibility and transparency, we have also implemented a web-based interface that allows customers to view real-time data regarding the actual volume of fuel dispensed and the associated cost. This web application updates the fuel price regularly, ensuring accurate cost calculations based on the latest fuel prices.

Simulated annealing‐based high‐level synthesis methodology for reliable and energy‐aware application specific integrated circuit designs with multiple supply voltages

SummaryIntegrated circuits have become more vulnerable to soft errors due to smaller transistor sizes and lower threshold voltage levels. Energy reduction methods make circuits more error‐prone since even the smallest amounts of environmental radiation can cause a bit flip. Furthermore, redundancy‐based error detection and correction methods induce higher costs and area overhead. Thus, several conflicting parameters may need to be considered during embedded system design (e.g., area, performance, energy, and reliability). High‐level synthesis (HLS) is the most practical design step to consider all these parameters as complexity increases at lower levels. HLS can be viewed as a multi‐objective optimization problem of finding a set of Pareto‐optimal designs, allowing designers to choose the ones that best fit the requirements. Moreover, the number of synthesis options superlinearly affects the search space growth, necessitating efficient optimization methods. In this study, we propose simulated annealing (SA)‐based HLS methods for multi‐Vcc application‐specific integrated circuit design, aiming to optimize energy consumption and reliability under the area and latency constraints. Furthermore, we use duplication to improve design reliability as much as the constraints allow. We compared our methods against genetic algorithm (GA)‐based and integer linear programming (ILP)‐based methods and showed their effectiveness in finding optimum or near‐optimum results in a short running time. SA‐based methods achieved up to 21.20% reliability improvement on average and up to 38% energy reduction on average, while preserving the reliability value against the GA‐based metaheuristic counterpart under joint reliability and energy optimization on four HLS benchmarks.

Photonic elementary cellular automata for simulation of complex phenomena

Cellular automata are a class of computational models based on simple rules and algorithms that can simulate a wide range of complex phenomena. However, when using conventional computers, these ‘simple’ rules are only encapsulated at the level of software. This can be taken one step further by simplifying the underlying physical hardware. Here, we propose and implement a simple photonic hardware platform for simulating complex phenomena based on cellular automata. Using this special-purpose computer, we experimentally demonstrate complex phenomena, including fractals, chaos, and solitons, which are typically associated with much more complex physical systems. The flexibility and programmability of our photonic computer present new opportunities to simulate and harness complexity for efficient, robust, and decentralized information processing using light.

Evaluation of the energy viability of smart IoT sensors using TinyML for computer vision applications: A case study

TinyML technology emerges from the intersection of Machine Learning, Embedded Systems, and Internet of Things (IoT), and presents itself as a solution for various IoT fields. For this technology to be successfully applied to embedded devices, it is essential that these devices have adequate energy efficiency. To demonstrate the viability of TinyML technology on embedded devices, field re- search and real experiments were conducted. An embedded system was installed in a turnstile of a Federal Institute, in which a TinyML computer vision model for people detection was implemented. The device counts the number of people, analyzes the battery level, and sends data in real-time to the cloud. The prototype showed promising results, and studies were conducted with a lithium battery and three in series. In these experiments, voltage consumption was analyzed every hour, and the results were presented through graphs. The camera sensor prototype had a consumption of 1.25 volts/hour, while the prototype without the camera sensor showed a longer-lasting consumption of 0.93 volts/hour. This field research will contribute to the advancement of applications and studies related to TinyML in conjunction with IoT and computer vision.

Noise Metrics and Subjective Assessment of Automotive HVAC Systems

HVAC systems are of critical importance in ensuring passengers' thermal comfort inside the car cabin as well as safety requirements for defogging functions. These systems involve various components and subcomponents such as blowers, thermal exchangers or actuators, with a wide range of well-known technologies and also new ones on recently introduced innovative products. Currently, within established electrification trends worldwide, the HVAC system is becoming the most important embedded system that can induce major contribution of noise and vibration. These NVH issues can emerge through different transfer paths inside the car cabin possibly causing significant discomfort to passengers. During developments, the NVH issues are mastered and contained by both suppliers according to internal requirements and OEMs according to specifications. However, OEM specifications are mainly defined by overall noise levels and improvements over the years are generally consisting of reducing these specified levels. Furthermore, some HVAC NVH issues may not be well detected when using regular NVH metrics. This raises concerns about the limitations of the regularly used metrics in ensuring specification compliance and, above all, in depicting a subjective assessment at component level. Throughout a statistical analysis of HVAC systems population, this paper first focuses on the discrepancies between the data provided by traditional NVH metrics and subjective evaluations. Then, a deeper analysis involving psychoacoustic metrics provides a relevance overview of the applied metrics depending on the encountered issues and compared to a subjective assessment. Finally, from a responsibility perspective, these findings raise questions about the relevance of regular metrics used by OEMs and the right way to handle HVAC NVH topics nowadays.

Development of Wearable Embedded Hybrid Powered Energy Sources for Mobile Phone Charging System

Mobile phones are an essential part of our day to day living. The increased reliance on mobile phone devices for communication, information sharing, connectivity and entertainment calls for a greater need to keep this device in a functional condition at all times. The epileptic nature of utility power supply, non-accessibility to power in remote locations and affordability of power due to high tariffs calls for a relative option to maintain connectivity of mobile phones. This research proposes to develop wearable embedded powered energy sources for charging mobile phones as a backup for instant and seamless charging of the phone battery once it drains. Our research addresses the inability of mobile phone users to charge their flatten phone batteries conveniently and seamlessly anytime and anywhere they find themselves. The hybrid powered charging system has the ability to harvest dissipated heat within the body of the phone and heat sources from the environment using an energy harvester, alongside the second energy source provided by the solar panel. The energy harvester will serve as a cooling measure to the device to maintain a stable temperature. The entire embedded device not only serves as an alternative power supply but also a protective covering to the mobile phone.

A Hardware-in-the-Loop Simulator to Optimize Autonomous Sailboat Performance in Real Ocean Conditions

In this work, a hardware-in-the-loop (HIL) simulator is designed to diagnose the behavior of an autonomous sailboat as it navigates between waypoints. At its core, the HIL simulator includes the sailboat pilot on an embedded system. The sensor data input to the embedded system is fed by a navigation simulator that takes into account the different forces on the sailboat due to the wind, waves and current conditions. The HIL simulator is then tested for a navigation route from sea trials published in 2014, and the behavior of the automated pilot is compared to its behavior when the vessel is driven by a crew. As demonstrated, the automated system can outperform the man-operated vessel. The tool is also used to diagnose weaknesses in the sailboat autopilot algorithm that can be improved in the future.

ANN-based State of Charge Estimation of Li-ion Batteries for Embedded Applications

The conventional state of charge (SOC) estimation model has several concerns, such as accuracy and reliability. In order to realize robust SOC estimation for embedded applications, this study focuses on three concerns of the existing SOC estimation model: accuracy, robustness, and practicality. In improving the estimation accuracy and robustness, this study took into account the dynamic of the actual SOC caused by the dynamic charging and discharging process. In practice, the charging and discharging processes have characteristics that must be considered to realize robust SOC estimation. The model-based SOC estimation developed based on the virtual battery model causes difficulties for real-time applications. Additionally, model-based SOC estimation cannot be reliably extrapolated to different battery types. In defining the behavior of various types of batteries, the model-based SOC estimation must be updated. Hence, this study utilized data-driven SOC estimation based on an artificial neural network (ANN) and measurable battery data. The ANN model, which has excellent adaptability to nonlinear systems, is utilized to increase accuracy. Additionally, using measurable battery data such as voltage and current signals, the SOC estimation model is suitable for embedded applications. Results indicate that estimating SOC with the proposed model reduced errors with respect to actual datasets. In order to verify the feasibility of the proposed model, an online estimation was out on the embedded system with the use of C2000 real-time microcontrollers. Results show that the proposed model can be executed in an embedded system using measurable battery data.

A Louver Exhaust Fan Centralized Control System with Predictive Maintenance

In order to reduce the failure rate of the louver exhaust fan on the farm and improve the economic benefit of the farm, a louver exhaust fan centralized control system with predictive maintenance characteristics was proposed. The system design is divided into three parts: the perception layer, the monitoring layer, and the analysis layer. The perception layer and the monitoring layer are based on the embedded system, which collects, transmits, and monitors the louver exhaust fan operation data in real time. The monitoring layer issues alarms and emergency control commands when necessary; the analysis layer is based on Gated Recurrent Unit (GRU) and The fusion network model of Convolutional Neural Networks (CNN) performs deep feature extraction on the operating state of the wind turbine. By calculating the similarity between the model output eigenvectors and the actual eigenvectors, it is possible to judge whether there is any abnormality in the louver exhaust fan operation and achieve the purpose of predictive maintenance. The actual test of the system shows that the collection and monitoring functions of the system operate normally, and the judgment accuracy of abnormal operation status is high, which is of practical significance for reducing the failure rate of louver exhaust fans.

Development of an Integrated Library Management System (ILMS)

Abstract— This document presents the design and development of an integrated library management system based on the needs assessment conducted in response to technology, infrastructure, existing condition, facilities, technology readiness of human resources, automation sufficiency level, and support of technical staff among ninety-eight (98) respondents using the descriptive developmental method. Developing a system based on research is crucial in ensuring the accuracy and reliability of the system. This approach addresses specific problems or challenges by incorporating evidence-based practices and the latest advancements in the field. The findings include the need to mentor the respondents on how to use technology and the influence of others as a factor in using technology. There is a difficulty encountered in navigating schools’ virtual libraries, unavailability of materials, and visiting it is not enticing anymore since information is available from the internet. The rarity of introducing new technology and the absence of 3D printers are common pain points. However, improvements on the library website, email/chat with a librarian, online research guides, online book reservations, and book delivery service were noticed. Research gaps and recommendations from the selected Library Management System were also employed as additional input in the development of the system. A deeper understanding of the Internet of Things, Artificial Intelligence, Big Data, Robotics, and Embedded Systems through training, webinars, and workshops are necessary for the librarian. These technologies could be applied in library operations and transactions. Procurement of the latest equipment which includes 3D printers and the development of integrated library systems are steps to take to improve the library's services and operation. By combining the results of assessments and incorporating new knowledge derived from research gaps and recommendations, provided a stronger foundation for the system. It also provides validation and grounding in scientific principles, ensuring that the system is well-informed and effective in its implementation. Ultimately, developing a system based on research leads to more informed and effective solutions to pain points and challenges.