Real Time Embedded Systems

The migration from classical public-key cryptography to post-quantum cryptography requires systematic methods for identifying algebraic structures whose computational properties remain secure against both classical and quantum adversaries. Traditional selection is largely based on theoretical hardness assumptions, implementation cost, parameter constraints, and expert judgement. This paper develops a Machine Learning Assisted Algebraic Structure Selection Framework (MLASSF) for classifying candidate algebraic structures as suitable or unsuitable for post-quantum cryptographic use. The framework combines finite group invariants, conjugacy class statistics, character-theoretic indicators, elliptic-curve descriptors, lattice parameters, implementation efficiency, and estimated security categories into a supervised learning pipeline. A reproducible simulated dataset of 5,000 candidate structures was generated to model algebraic and computational features relevant to post-quantum design. Five classifiers were trained and evaluated: logistic regression, support vector machine, random forest, gradient boosting, and artificial neural network. The best-performing model was gradient boosting, with an accuracy of 93.80%, precision of 95.04%, recall of 96.21%, F1-score of 95.62%, and ROC-AUC of 95.29%. Random forest followed closely with 93.60% accuracy and also provided interpretable feature importance, showing that lattice dimension, implementation efficiency, character-degree variance, elliptic-curve rank, and derived length were influential predictors. The results show that machine learning can serve as a decision-support layer for algebraic selection in post-quantum cryptographic research, especially when combined with rigorous mathematical analysis rather than used as a substitute for proof-based security.

Perkembangan Internet of Things (IoT) berbasis WiFi semakin luas digunakan pada sistem
monitoring dan otomasi, namun kualitas koneksi perangkat IoT masih dipengaruhi oleh jarak
terhadap access point. Penelitian ini bertujuan menganalisis kinerja dan stabilitas koneksi perangkat
IoT berbasis WiFi terhadap variasi jarak access point menggunakan parameter Received Signal
Strength Indicator (RSSI), latency, dan packet loss. Metode penelitian menggunakan pendekatan
kuantitatif eksperimental dengan pengujian langsung pada perangkat ESP32 yang terhubung ke
access point pada variasi jarak 1 meter hingga 30 meter. Pengambilan data dilakukan pada kondisi
line of sight (LOS) dan non line of sight (NLOS) dengan interval pengiriman data setiap 2 detik
selama 10 menit di setiap titik pengujian. Hasil penelitian menunjukkan bahwa peningkatan jarak
menyebabkan penurunan kualitas sinyal RSSI, peningkatan latency, serta kenaikan packet loss. Pada
kondisi LOS, koneksi masih stabil hingga jarak 25 meter dengan rata-rata RSSI -71 dBm, latency 86
ms, dan packet loss 2,8%, sedangkan pada kondisi NLOS kualitas koneksi menurun signifikan pada
jarak di atas 15 meter. Unsur kebaruan penelitian ini terletak pada integrasi analisis tiga parameter
jaringan secara simultan pada perangkat IoT ESP32 berbasis WiFi dengan perbandingan kondisi LOS
dan NLOS menggunakan skenario pengiriman data real-time yang mendekati implementasi sistem
IoT aktual. Penelitian ini diharapkan menjadi referensi dalam optimasi penempatan access point pada
implementasi jaringan IoT berbasis WiFi.

Bio-inspired Artificial Intelligence (AI)
architectures have become an innovative paradigm in solving
intricate problem-solving tasks, which in many cases, are
beyond the reach of traditional computational models. Inspired
by natural phenomena like neural plasticity, evolutionary
adaptation, swarm intelligence, and immune system defence
systems, bio-inspired AI systems offer scalable, adaptive and
resilient solutions to real-world problems. In this paper, the
interest is in the design and integration of bio-inspired
architectures to address multidimensional issues in areas such
as optimization and network security, healthcare and
autonomous systems. The proposed architectures have the
benefit of using self-organization, distributed intelligence, and
evolutionary learning to exhibit greater uncertainty, dynamic
environment, and high-dimensional data processing
capabilities. The experimental literature points at their high
level of performance in relation to traditional models of AI by
their adaptability, fault tolerance, and computation efficiency.
The findings of this paper highlight the promise of bio-inspired
AI architecture as a foundational block of future intelligent
systems that can provide viable and understandable responses
to the problems of an ever more complex landscape.

This essay argues that "one's own time" is not a private possession but a socially produced, globally mediated, and ecologically conditioned achievement. The welfare society created historically new forms of socially protected own-time through education, healthcare, care, social security, labour regulation, and public services. Under the conditions of ecological crisis, however, this achievement must be reinterpreted. Modern time has been progressively disembedded from the rhythms of the living world: from solar and seasonal cycles to church bells, mechanical clocks, factory whistles, global supply chains, quarterly capitalism, and financial microtime. Drawing on Marx's distinction between free time and capital's usurpation of the time created for society, Frigga Haug's Four-in-One Perspective, and critical theories of social acceleration and abstract time, the essay proposes that the total time-budget of life must be reorganized around socially necessary labour, care and reproduction, democratic participation, and self-development. Services should be understood as institutions of time-production, and commoning reproduction as the democratic organization of the time required to reproduce life. The task is not merely liberation from work, but the liberation of work for the ecological reproduction of the conditions of life.

Existing autonomous lawn mower systems are exclusively ground-based, leaving a critical gap in the development of aerial cutting platforms capable of navigating complex terrain without physical contact. This study presents the design, fabrication, and validation of an Automated Self-Lifting Lawn Mower (ASLM) — a novel quadrotor unmanned aerial vehicle (UAV) integrating an active rotary cutting mechanism with an AI-driven machine vision system for real-time obstacle and pothole avoidance. The system addresses four identified gaps in current literature: the absence of a formal aerial mowing machine category, missing aerodynamic–blade interaction theory, the lack of empirical performance data for combined hover and cutting operations, and the non-existence of a validated AI-integrated aerial mowing prototype. The proposed ASLM is built on a 450 mm carbon-fibre reinforced polymer (CFRP) quadrotor frame with an all-up weight of ≤ 2.8 kg, operating at hover altitudes of 50–250 mm above ground level. A 200 mm hardened steel rotary blade driven by an independent brushless motor provides the cutting function, while an Intel RealSense D435 depth camera and YOLOv8-nano neural network on an NVIDIA Jetson Nano deliver real-time terrain sensing and obstacle avoidance. A boustrophedon coverage path planning algorithm enables fully autonomous lawn traversal. Mathematical models governing thrust, blade torque, energy partitioning, and machine vision detection confirm design feasibility, with an estimated field endurance of 12–15 minutes and coverage of approximately 200 m² per charge cycle. This research contributes a new machine class, novel aerodynamic–blade interaction theory, and empirical performance benchmarks to the field of aerial autonomous robotics.

Bahasa pemrograman C merupakan salah satu bahasa pemrograman yang banyak digunakan dalam dunia pendidikan maupun industri karena memiliki struktur yang sederhana, cepat, dan efisien. Dalam bidang Teknik Industri, bahasa C dapat digunakan untuk membantu menyelesaikan berbagai perhitungan produksi, optimasi, maupun pengolahan data sederhana. Artikel ini membahas dasar penggunaan bahasa pemrograman C, struktur program, tipe data, penggunaan percabangan dan perulangan, serta penerapannya dalam pembuatan program perhitungan produksi sederhana. Program yang dibuat mampu menghitung total produksi, biaya produksi, dan keuntungan berdasarkan data yang dimasukkan pengguna. Program ini diharapkan dapat membantu mahasiswa memahami penerapan bahasa pemrograman C dalam menyelesaikan permasalahan sederhana di bidang industri. Kata kunci: Bahasa C, Pemrograman, Teknik Industri, Produksi, Algoritma Pendahuluan Perkembangan teknologi informasi memberikan dampak besar terhadap berbagai bidang, termasuk dunia industri. Saat ini hampir seluruh aktivitas industri memanfaatkan komputer untuk membantu proses produksi, pengolahan data, hingga pengambilan keputusan. Oleh

A Mobile Ad-hoc Network (MANET) is a temporary wireless network composed of mobile nodes, in which an infrastructure is absent. If two mobile nodes are within each other's transmission range, they can communicate with each other directly; otherwise, the nodes in between have to forward the packets for them. Several routing protocols have already been proposed for ad hoc networks. This paper suggests an approach to utilize location information obtained by using DLRDP F (Dynamic located Routing and Detection Protocol Far), if the destination is Remote and DLRP L (Dynamic located Routing and detection Protocol Local), if the destination is local. in this routing scheme affected various attacks. In MANETs, it is difficult to identify malicious attacks as the topology of the network dynamically changes. In this Progress we proposed new efficient dynamic detection scheme for attacks also. Using location information to help routing is often proposed as a means to achieve scalability benefits like Packet delivery ratio, Average delay ,packet loss and Qos in large mobile ad-hoc networks. and detect the attacks. We use the network simulator 2 (ns-2) system to conduct the MANET simulations and consider scenarios for location and detection. and on the Tiny based Real time operating systems.

Penulisan kode program yang besar dan kompleks pada bahasa pemrograman C sering kali menghadapi kendala keterbacaan serta tingginya risiko duplikasi data akibat pendekatan monolitik. Penelitian ini bertujuan untuk menganalisis penerapan konsep fungsi dalam membangun arsitektur program yang terstruktur dan modular. Metode yang digunakan adalah analisis komparatif dan implementasi fungsional melalui studi kasus pembuatan program mesin kasir restoran. Hasil pembahasan menunjukkan bahwa penggunaan fungsi mampu memecah program menjadi subprogram yang lebih kecil dan spesifik, sehingga mempermudah proses pelacakan kesalahan (debugging) dan pemeliharaan kode. Selain itu, pemilihan mekanisme pengiriman parameter melalui call by value terbukti efektif dalam menjaga keamanan data asli, sedangkan call by reference dengan pointer memberikan efisiensi tinggi pada manipulasi memori secara langsung. Ketepatan dalam mengatur ruang lingkup variabel melalui kelas penyimpanan lokal, eksternal, dan statis turut mengoptimalkan kinerja program. Kesimpulan dari penelitian ini menegaskan bahwa penerapan fungsi bukan hanya sekadar merapikan visualisasi baris kode, melainkan sebuah instrumen krusial dalam menciptakan arsitektur perangkat lunak yang efisien, adaptif, dan sistematis.

The inline current-voltage flash-test measurement performed on every cell at the end of passivated emitter and rear cell manufacturing lines is the single most information-dense moment in the entire production sequence. In a fraction of a second, the electrical characterization of a completed cell produces a curve containing the integrated signature of every process step upstream of the measurement point, from wafer texturing through firing. Extracting the statistical information embedded in that signature, and using it to attribute efficiency variation to specific process variables, is the analytical activity that separates a factory that merely binning its output from a factory that systematically optimizes its process to shift the efficiency distribution itself. This paper presents a production-validated methodology for inline current-voltage curve analysis and binning strategy optimization, developed and deployed at the ES Foundry 1 gigawatt domestic PERC cell manufacturing facility in Greenwood, South Carolina during the April through June 2025 production window. The analytical methodology decomposes each measured current-voltage curve into seven primary electrical parameters-opencircuit voltage, short-circuit current density, fill factor, maximum power point, series resistance, shunt resistance, and diode ideality factor-and correlates each parameter statistically against a suite of twenty-two upstream process variables drawn from the full process data pipeline. The correlation analysis identifies the three highestleverage process variables and quantifies their fractional contribution to the overall efficiency distribution. A structured optimization program targeting these three variables delivered a shift in the top-bin yield from sixpoint-six percent of production to thirteen-point-seven percent of production over a twelve-week production window, with a corresponding mean efficiency improvement of twenty-three one-hundredths of one percent absolute and a narrowing of the efficiency distribution standard deviation from forty-one one-hundredths to thirtythree one-hundredths of one percent absolute.

In embedded systems, timing and resource utilization are vital aspects, having impacts on the deployed software. In this paper, an extensible formal framework for modeling of resource usage and adaptability of software components is introduced. The formalism explicitly supports modeling resources of various types and in different ways; and the timing requirements that applications have. The models can be used as a basis for quantitative analysis, and in the extension as a basis for system synthesis. An example of usage of the formalism is given.

Real-time national fraud detection platforms-exemplified by Brazil's PIX, India's UPI and Nigeria's NIBSS-embed centralised AI-and-rule-driven monitoring at the settlement layer, yet share three structural limitations: human-in-the-loop investigation bottlenecks, batch retraining latency and weak cross-institution coordination. This paper proposes an architectural successor: an agentic AI fabric in which autonomous agents, with bounded autonomy, persistent memory and observable handoffs, distribute detection, investigation, remediation and learning across the terminal, institutional and national-authority layers.

Biometrics is one of the most used technologies in the field of security due to its reliability and
performance. It is based on several physical human characteristics but the most used technology is the
fingerprint recognition, and since we must carry out an image processing to be able to exploit the data in
each fingerprint we propose in this article an image preprocessing procedure in order to improve its
quality before extracting the necessary information for the comparison phase.

This laboratory report presents the design and
implementation of a light-sensitive alarm system
using LDR (Light Dependent Resistor) sensors.

The circuit operates on the principle of a voltage
divider — when ambient light decreases, the resistance
of the LDR increases, raising the output voltage.
This voltage triggers an NPN transistor (2N2222),
which activates a relay module (HW-482), subsequently
sounding a buzzer.

A secondary LDR sensor independently controls one
color channel of an RGB LED, providing a continuous
visual indication of light intensity.

Keywords: LDR, voltage divider, relay, NPN transistor,
alarm system, light sensor, electronics

Verified by: Physics Teacher Telman Askeraliyev (Azerbaijan, Baku)

Technical Laboratory Report: The Light-Sensitive Switch (LDR+ Voltage Divider) Connector Dilara Rzazada, Nazrin Aliyeva, Ruslan Qasimov, Ulker Aliyeva, Khanverdi Chelebiyev, Nazila Aliyeva, Ali Shukurov, Imran Cherchiyev Verified by: Physics Teacher Azerbaijan Telman Askeraliyev (Fizika muellimi, Azerbaijan, Baku)

The rapid advancement of technology across the globe resulted in an exponential growth of data production. The rate of data production is directly proportional to the volume of data generated and stored. A domain that adheres to the stated fact is transportation, as it heavily relies on real-time data for accurate traffic analysis and prediction. The satisfaction that accompanies the successful completion of the thesis would be incomplete without mentioning the people who made it possible. The constant guidance and encouragement of these people crown all the efforts with success. First and foremost, I am incredibly grateful to my supervisor, Professor Richard O. Sinnott of the Department of Computing and Information Systems at the University of Melbourne, whose expertise was invaluable in formulating the research questions and methodology. His insightful feedback pushed me to sharpen my thinking and brought my work to a higher level. I would also like to thank Mr. Luca Morandini of the Department of Computing and Information Systems at the University of Melbourne for his invaluable advice, technical guidance, continuous support, and patience throughout my project. His immense knowledge and ample experience have encouraged me throughout my academic research and daily life. In addition, I would like to thank Mr. Abdul Rehman Mohammad and Mr. Xiaolin Zhang for their efforts and cooperation when needed during this project. I would like to thank my parents for their wise counsel, tremendous understanding, encouragement, and support all through my life and studies. Finally, I could not have completed this dissertation without the support of my friends, who provided stimulating discussions and happy distractions to rest my mind outside of my research.

In this paper, a new VLSI architecture is provided for the application of quad-input and dual-output Fuzzy Logic Controller (FLC) with maximum seven fuzzy membership functions. Our approach is based on classical three stage implementation process -fuzzification, rule inference and defuzzification cores. An innovative design methodology is proposed by splitting the process between DSP processor and FPGA to implement run time configurable FLC. Since the target application takes a maximum of 4 inputs and 7 membership functions, the rule base comprises of 2401 (7 4 ) rules. It increases the complexity of the overall system. To minimize this effect, rule reduction VLSI architecture is suggested to bind the no of rules to 16 (2 4 ). The Rule inference is designed for seeking the maximum frequency of operation for targeted Virtex 5 LX110T FPGA. The simulation results obtained with Modelsim 6.3g show satisfactory results for all test vectors.

A Fuzzy Logic Controller (FLC) is designed with heuristic knowledge of the plant. There are several application based Fuzzy Logic Controllers in literature. However, application specific controllers lack reusability and are expensive, if not produced in bulk. This paper presents a general purpose and easily configurable fuzzy logic based controller which can work with 4 inputs and generates a control action based on the knowledge base. Expert operator knowledge can also be directly programmed into the knowledge base (also known as rule base) of the designed controller without ceasing the control operation. The ease of file handling and user interactivity of DSP processors is heavily exploited and TMS320C6713 DSP was chosen as the hardware platform. Further this paper also analyses the system performance with reference to the Fuzzy Logic Toolbox of Matlab.

In today's data-centric world, organizations have multi geographic regulatory regimes to manage and complex requirements to timespan privacy and storage policies like GDPR, HIPAA, and regional data protection laws. In this paper propose a new AI-based model for managing data that helps organizations operating in different countries to handle and align their privacy and storage policies in real time imaging using a combination of deep learning techniques called Convolutional Neural Networks (CNNs) and RNN.The framework is able to consume disparate data streams such as user access logs, policy documents, audits, and storage metadata. CNNs are used for capturing spatial dependencies and feature interactions in structured or unstructured inputs event logs, legal texts. The spatial representations are later fed to Recurrent Neural Networks (RNNs), to learn the dynamics of the environment and capture sequential behaviors identifying temporal policy violations, unauthorized data access patterns, or misalignments to some regulation. The proposed model has been realized with 98.76 semi-independent classification accuracy, which is higher than CNN, RNN and GRU alone. Explanation is improved through attention or saliency maps, which can enable legal or compliance teams to trace the reasons behind decisions. For organizations that function within a complicated, multi-territory, cross-border landscape, this AIdriven solution offers scalable, transparent, and efficient policy enforcement. It is also paving the way for future developments in federated learning, edge computing, and multilingual regulation modeling towards the goal of autonomous, compliant, and intelligent data governance.

Teacher education systems are increasingly required to prepare graduates for digitally mediated classrooms, where learning, assessment, and administration rely on interconnected technologies. The integration of Internet of Things (IoT) offers potential for data-driven and adaptive learning environments, but its effectiveness depends on contextual feasibility, institutional capacity, and pedagogical alignment. This theoretical paper develops a Liberia-centred framework for integrating IoT capabilities into teacher education while preserving the central role of pedagogy, mentoring, equity, and

As immersive instruments like the AlloSphere at UCSB evolve, the primary challenge in computational architecture shifts from simple visualization to the real-time synthesis of complex, multi-modal datasets. This paper explores the "Shaping Space as Information" framework through the lens of a Senior AI Architect. By implementing asynchronous data streams and generative spatial logic, we can transform immersive environments from passive display tools into active, forensic research engines where every pixel is a traceable data point.

The study proposes a transition toward a Forensic Computational Pipeline, utilizing Python-driven spatial culling and Generative Adversarial Networks (GANs) to scale information within immersive 3D environments. The goal is to provide a technical blueprint that ensures data integrity while maximizing the structural potential of real-time, large-scale data visualization.

The development of distributed real-time embedded systems presents a significant practical challenge both because of the complexity of distributed computation and because of the need to rapidly assess a wide variety of design alternatives in early stages when requirements are often volatile. Formal methods can address some of these challenges but are often thought to require greater initial investment and longer development cycles than is desirable for the development of non-critical systems in highly competitive markets. In this paper we propose an approach that takes advantage of formal modelling and analysis technology in a lightweight way, making significant use of readily available tools. We describe an incremental approach in which detail is progressively added to abstract system-level specifications of functional and timing properties via intermediate models that express system architecture, concurrency and distribution. The approach is illustrated using a model of a home automation system. The models are expressed using the Vienna Development Method (VDM) and are validated primarily by scenario-based tests.

Real-time operating system (RTOS) is a very useful tool for developing the application on embedded boards with least software development effort. Though number of RTOS products are available in the market, µC/OS-II is a freeware with minimum facility and more popular among the hobbyist, researchers and small embedded system developers. The µC/OS-II supports preemptive scheduling which is not efficient with respect to processor utilization. As a result, this may lead to missing deadline of the task assigned and hence may cause system failure. In this paper, a Rate Monotonic scheduling (RM), which is a better scheduling method when compared to preemptive technique, is implemented on µC/OS-II and its operation in terms of task execution and processor utilization are discussed. This paper presents the RM Analysis (RMA) on µC/OS-II with two different types of hardware: 1) Low end microcontroller i.e., 8051 based system and 2) High end system based on ARM9 is used. For the demonstration of RM Analysis, two tasks, namely, 1) Advanced encryption standard (AES) and 2) Text message display and graphic display are implemented. The software tools Keil IDE, SDCC compiler and Phillips Flash Magic are used for implementation of tasks on 8051 embedded development board. The ARM developers Suite v1.2 and DNW are used for implementation of tasks on ARM9 development board. In addition to the above said tasks, additional tasks like Real time clock interface, graphical LCD interface and UART interface for communication with computer are also implemented. The scaled-version of µC/OS-II with multiple tasks uses 4 kB of flash and 512 bytes of RAM in 8051 board. The entire MUCOS RTOS in ARM9 with multiple tasks uses 29.23 kB of flash and 596.36 kB of RAM. The results obtained indicate optimum utilization of processor with RMA scheduler for realizing low cost software for developing the application on embedded boards with least software development effort.

This note formalizes δ as the deterministic transition function governing admissible state evolution within the IDDA architecture. The model distinguishes execution, holding, blocking, and controlled reset as distinct outcomes of a unified transition rule. A key design property is that blocking is not terminal: inadmissible progression triggers a controlled reset to a designated safe baseline s 0 , preserving operational continuity. The repeated application of δ induces a piecewise-stable state trajectory under admissibility constraints. The result is a compact formal layer connecting admissibility, execution, and recovery.

The rapid evolution of cyber threats-including polymorphic malware, fileless attacks, and software supply chain compromises-has exposed a fundamental limitation in modern cybersecurity: the absence of a persistent identity mechanism for software systems. Traditional approaches such as cryptographic hashing, behavioral analysis, and network-based attribution operate independently and fail to provide a unified lifecycle-based identity framework. This paper introduces the Program Identity Chain (PIC 2.0), a dynamic, multi-layered framework that establishes a continuous and evolving identity for software systems from creation through execution and mutation. Drawing inspiration from biological DNA and industrial traceability systems, PIC 2.0 integrates cryptographic hashing, distributed ledger technologies, hardware-based trust anchors, lineage-based identity tracking, and privacy-preserving verification mechanisms. The proposed architecture formalizes a lifecycle identity pathway that spans code creation, build processes, distribution, execution monitoring, mutation tracking, and forensic reconstruction. By transforming software identity into a cumulative and multi-dimensional construct, PIC 2.0 enhances traceability, resilience, and attribution capabilities. While acknowledging that absolute attribution in cyberspace remains theoretically unattainable, this framework significantly advances the state of cybersecurity by enabling near-continuous identity tracking and forensic reconstruction.

This second part of the report examines the evolution of the Phenomenological Reconstruction of Complex Systems (PRCS) method from binary anomaly detection toward expert-level diagnostic capability. Building upon multiscale analysis of structural invariants in reconstructed phase space, a meta-feature space is constructed wherein streaming clustering and dynamic pattern tracking are performed. The emergence of mechanical defects appears to be associated not merely with an increase in a scalar risk index, but with the emergence of new, quasi-stable dynamical regimesa process we term patternogenesis. By associating identied clusters with computed bearing defect frequencies and analyzing state transition dynamics, physical localization and typication of damage modes become feasible. Results are validated using the publicly available CWRU Bearing Data Center dataset.

This technical report presents the Phenomenological Reconstruction of Complex Systems (PRCS) method as applied to predictive diagnostics of rotating machinery. In contrast to conventional spectral analysis or monitoring of energy-based parameters (RMS, peak factor), the proposed approach relies on the analysis of structural invariants of the reconstructed phase space (J , C, Φ), derived from a one-dimensional vibration time series. This methodology appears to enable the detection of dynamical changes associated with the initiation and progression of mechanical faults, potentially prior to their manifestation in amplitude-based indicators. The report provides a detailed examination of: the mathematical foundation of the method, the procedure for computing the structural risk index R, validation results on the publicly available Bearing Data Center (CWRU) dataset, and engineering considerations for implementation on embedded computing platforms (ARM Cortex-M). Particular attention is given to evidence suggesting robustness of the method to load variationsa property of critical importance for mobile platforms and equipment operating under variable conditions. A central premise of this work is that PRCS is not intended to replace existing emergency protection systems (RMS monitoring, vibration limit checks), but rather to serve as a complementary intelligent interpreter layered atop them, supporting early fault detection and facilitating a transition toward condition-based maintenance strategies.

The widespread use of microservices and event-driven architectures requires reliable tools to manage data structures in motion. A robust foundation for high-throughput data pipelines, Apache Kafka decouples event producers and consumers, which introduces complications regarding schema evolution ability within enterprise integrations. This research paper studies the critical Confluent Schema Registry and serialization formats, namely, Apache Avro, Protocol Buffers, and JSON Schema, for backward and forward compatibility in enterprise integrations. By combining academic literature on compatibility enforcement with industry case studies, the analysis investigates compatibility enforcement levels including transitive modes and their impact on system resilience. Additionally, schema validation in automated continuous integration continuous delivery (CI/CD) pipelines, the use of architectural patterns such as the Transactional Outbox and CQRS, and the governance frameworks needed for scalable data ecosystems will be examined. The results reveal that automatic schema management reduces operational overhead by more than 70% and prevents catastrophic integration failures enabling seamless evolution in heterogeneous environments.

This paper presents the design and simulationbased evaluation of a smart home energy monitoring and management system that integrates IoT-enabled sensing, GSMbased alerting, and automated load control. The proposed framework enables real-time monitoring of electrical and environmental parameters, including voltage, current, temperature, humidity, gas concentration, and human presence. A NodeMCU-based controller coordinates sensor data acquisition, rule-based decision-making, and bidirectional communication via Wi-Fi and GSM networks. Automated load shedding is implemented to disconnect non-essential appliances during peak demand periods, supporting demand-side energy management. The system is validated through Proteus-based simulation, demonstrating reliable operation with an average alert generation time of approximately 1.3 s and a switching latency of 170 ms. Performance evaluation shows an average energy saving of 8.6%, supported by statistical validation across multiple simulation runs. Additional analyses on scalability, stability, and communication reliability confirm the robustness and practical applicability of the proposed approach. The results indicate that the system provides a scalable, low-cost solution for intelligent residential energy management and safety enhancement.

Saat ini, perkembangan teknologi semakin pesat dan keterampilan dalam bidang elektronika serta pemrograman menjadi sangat penting, terutama bagi pelajar, mahasiswa, maupun masyarakat umum yang ingin memahami konsep robotik secara praktis. Buku ini disusun secara sistematis, dimulai dari pengenalan Arduino, pemahaman dasar komponen elektronik, hingga proyek-proyek robotik sederhana yang dapat langsung dipraktikkan. Harapan saya, buku ini bisa menjadi jembatan awal untuk menumbuhkan minat dan kreativitas pembaca dalam mengembangkan teknologi berbasis mikrokontroler. Saya menyadari bahwa buku ini masih jauh dari sempurna. Oleh karena itu, saya sangat terbuka terhadap masukan, kritik, dan saran yang membangun demi penyempurnaan edisi berikutnya. Akhir kata, saya mengucapkan terima kasih kepada semua pihak yang telah memberikan dukungan, baik secara langsung maupun tidak langsung, dalam penyusunan buku ini. Semoga buku ini bermanfaat dan dapat menjadi langkah awal dalam mencetak generasi yang melek teknologi dan inovatif.

The rise of streaming technologies has spurred the emergence of real-time data ingestion and  streaming  pipelines,  which  serve  as  the  foundation  for  building  streaming  data platforms. Real-time data ingestion focuses on collecting, validating, and transforming data  as  it  becomes  available.  A  streaming  pipeline  consists  of  a  series  of  data transformation  steps  applied  to  streams  of  data—effectively  a  set  of  streaming  ETL processes.  Both  concepts  are  wide  in  scope  and  can  be  addressed  along  various dimensions,including  protocol  semantics  and  guarantees;  data  serialization  formats; support  for  schema  evolution  and  data  quality;  pipeline  architectures;  data transformation  and  enrichment  such  as  windowing,  aggregation,  and  complex  event processing; durable storage; and consistent state management. Real-time data processing is acknowledged as a rapidly maturing field of research and engineering practice.The supporting foundations and concepts, along with an outline of open challenges, offer an overview of real-time data ingestion and streaming pipelines. The latter can be seen as  the  foundations  of  streaming  data  platforms—semantically  richer  and  more  fully featured  than  their  streaming  processing  framework  subsystems.  While  firms  such  as Confluent,  Databricks,  and  Streamlio  provide  specialty  products  in  the  commercial space, a  wide variety of open-source technologies are also available for building real-time data ingestion and streaming data platforms.

This research presents a low-cost system for detecting microplastics in water using ultraviolet (UV) light and a photodiode sensor. The system utilizes fluorescence and light scattering properties of microplastic particles. An Arduino Uno processes the sensor data and displays results in real time on a 16x2 LCD. Experimental results show clear differences between clean and contaminated water, demonstrating the feasibility of this method for field-level detection.

This research work introduces a scalable and real-time student attendance monitoring system designed to improve accountability and academic discipline through automation and surveillance. Traditional attendance methods like manual roll calls and RFID are often inefficient and prone to manipulation; in contrast, this system uses Face API JS a JavaScript-based deep learning facial recognition library to detect and identify students and lecturers via live video streams captured through webcams. Lecturers are the only authorized personnel to initiate attendance sessions, while students can access their attendance records through a secure web interface. The system incorporates Socket.IO for real-time, bidirectional communication between clients and the server, allowing for seamless updates and efficient attendance coordination. Administrators can monitor live classroom activities using mobile devices or built-in Raspberry Pi modules. Experimental evaluation across varying lighting and classroom conditions yielded an average face recognition accuracy of 95.7%, with a false acceptance rate of 2.1% and average detection latency under 1.2 seconds, confirming the system's robustness and responsiveness. In conclusion, the proposed solution offers an automated, non-intrusive, and transparent attendance tracking mechanism that significantly enhances institutional oversight, reduces human error, and ensures both student presence and lecturer participation.

Lorentz invariance-the exact symmetry of physical laws under rotations and boosts-is a foundational principle of modern physics, forming the basis of both Special Relativity and the Standard Model of particle physics. Nonetheless, numerous theoretical approaches to quantum gravity, discrete spacetime, and emergent geometry suggest that Lorentz invariance may be an effective, rather than fundamental, symmetry. This article provides a physicsgrounded overview of Lorentz violation: what it means, why it is theoretically motivated, how it is parameterized within effective field theory, how it is tested experimentally, and what its discovery-or continued absence-would imply for our understanding of spacetime, matter, and time.

Traffic is significant issue in our nation, particularly in urban ranges. Aftereffect of this, activity clog issue happens. Crisis vehicle like rescue vehicle, fire unit, squad cars confront bunches of issue to achieve their goal on account of congested driving conditions, coming about loss of human lives. To minimize this issue, we approach new idea named as "Traffic control framework for blockage control and stolen Vehicle location". In this framework activity freedom done by transforming Red flag into Green flag. We demonstrate an idea of what we call "Green wave". Alongside this, we distinguish stolen vehicle by utilizing extremely advantageous RFID innovation. In the event that stolen vehicle is been distinguished, the framework gives ready sign through ringer. Framework sends Message with the assistance of GSM to Police station. In this framework we Use diverse RFID labels for recognizing rescue vehicle and the stolen Vehicle. On the off chance that Red flag is on and IR sensor is initiated, then framework gives ringer alarm to movement police. This is novel framework which encourage great answer for comprehend traffic clog.

Embedded real-time applications are often built from scratch on a trial-and-error basis, which leads to sub-optimal designs with latent errors that are not detectable in early stages of use or deployment and often incurs prolonged time-to-market. A new application framework called Verifiable Embedded Real-Time Application Framework (VERTAF) is proposed for embedded real-time application development, with the aim of reducing design errors and increasing design productivity. VERTAF is an integration of three technologies, namely object-oriented technology, software component technology, and formal verification technology. VERTAF consists of five software components: Implanter, Modeler, Scheduler, Verifier, and Generator. Experiences of using VERTAF show a significant increase in design productivity through design reuse, and a significant decrease in design time and effort through design verification. An example shows a relatively low design effort on the part of the designer using VERTAF.

🚀 AI-Powered Smart Hospital System for Real-Time Healthcare Delivery

This research proposal presents the design of an intelligent hospital management system that integrates telemedicine, real-time communication, and AI-powered health insights into a unified platform.

The proposed system aims to solve common challenges in healthcare such as poor communication, fragmented systems, and delayed medical services by combining video consultations, smart patient management, and predictive analytics.

Built using modern technologies including React, Node.js, Firebase, and AI tools, the system is designed to enhance efficiency, improve patient experience, and support data-driven healthcare decisions.

📌 This work highlights how AI and real-time systems can transform healthcare delivery, especially in developing regions like Uganda.

Architectural decisions are among the earliest made in a software development project. They are also the most costly to fix if, when the software is completed, the architecture is found to be inappropriate for meeting quality objectives. Thus, it is important to be able to assess the impact of architectural decisions on quality objectives such as performance and reliability at the time that they are made. This paper describes PASA, a method for performance assessment of software architectures. It was developed from our experience in conducting performance assessments of software architectures in a variety of application domains including web-based systems, financial applications, and real-time systems. PASA uses the principles and techniques of software performance engineering (SPE) to determine whether an architecture is capable of supporting its performance objectives. The method may be applied to new development to uncover potential problems when they are easier and less expensive to fix. It may also be used when upgrading legacy systems to decide whether to continue to commit resources to the current architecture or migrate to a new one. The method is illustrated with an example drawn from an actual assessment.

In recent times, most hospitals in the world are crowded and facing the problem of increasing numbers of patients, specifically with the spread of infectious viruses such as Covid-19 and other chronic diseases such as cardiac disease and Geriatrics diseases, therefore patients need continuous and accurate emergency care. The heart rhythm disturbances, unsaturated oxygen in the blood, and high-temperature may be a high risk to the life of the patient. Most hospitals and medical quarantines have a shortage of Intensive care units (ICU) rooms. This paper discusses the design and improvement of the Patient monitor device for converting it to a handheld device to be small in size (L145 x W85 x D45 mm), lite Wight (398 gm.), low-cost nearly (98USD), easy-to-use by touch screen, and secure connections to internet with (SSL) protocol [22]. The maximum errors of ECG HR is 1.408% , and SPO2 is 1.075% ,SPO2 HR is 1.408% , temperature is 0.549%, BP systolic is 2.439% , BP Diastolic is 3.448%,The patient can use the device anywhere by himself, in an easy way and safely. This device measures physiological parameters such as (blood pressure by using (MAP) method, the saturation of oxygen in the blood (SPO2) by photoplethysmograph (PPG), the electrical and muscle movement of the heart through 3Lead (ECG), temperature, and finally, Heartbeat rate by different three ways). All data are monitoring on Smart Touch TFT, and storing the Main information of patient for more than thirty days Continuously on SD. Also using (IoT) technology to transfer all patient information in real-time through ESP8266 to Android or Website to follow-up the patient directly and permanently by Doctor or Caretaker, The device gives some important Speech voice sentences and sound alerts to improve the health of the patient. to design and implement the system with low cost, we are using the traditional lowcost sensors, with changing the large bulker analog circuits to intelligent signal conditioning by using (CMRR) with Sallen-Key Active filter and low-cost successive approximation 12-bit ADC MCP3208, with DAC MCP4728 and improving the efficiency by DSP software Specifically FIR and Moving Average Filter. This paper not using any Platforms under (RTOS), but using embedded BASIC language with the wide-range microcontroller PIC18f46k22, to keep the program in small size of memory and low power consumption

High loads frequently lead to throttling of diagnostic logging on mixed-criticality automotive ECUs due to blocking and timing interference created by traditional logger designs. This paper delivers ASIL (Automotive Safety Integrity Level)-Aware Circular Logger (AACL) a FreeRTOS-based framework that incorporates hook-based execution-domain knowledge and ASIL-prioritized buffering to eliminate the possibility of blocking once safetycritical execution paths are defined. A closed-form calibrated deferred-latency upper bound is derived from representative automotive task sets; experimental testing used the FreeRTOS GCC_POSIX port with a Cortex-M4 at 168 MHz reference task model with unique benchmark against a naive logger (directly based on mutexes) will provide evidence hosts but not certified timing (in target systems). Results showed no measured ASILtask blocking for the AACL, subordinate deferred latencies (bounded) with quantifiable tightness bounds, significant throughput/drop-rate robustness game over baseline logger provided, only public FreeRTOS API used in implementation and designed for integration into ISO 26262 timing arguments.

Mental health is essential to overall well-being, influencing physical, psychological, and social health. Despite advancements in healthcare, diagnosing and treating mental health disorders remains challenging. Delayed recognition of symptoms often leads to worsening conditions, emphasizing the need for real-time mental health monitoring solutions. This paper presents an AI-driven mental health monitoring system that continuously assesses emotional well-being. The system integrates a MAX30102 sensor and an ESP32 microcontroller to measure heart rate and oxygen saturation, transmitting realtime data via the Blynk API for continuous analysis. Additionally, Apple Health is utilized to track sleep duration and physical activity, providing a comprehensive evaluation of mental and physical health. By leveraging IoT-based data collection and AIdriven analytics, the system detects stress levels and emotional imbalances through key physiological indicators, including heart rate variability, sleep patterns, oxygen saturation, and physical activity. A random forest regression model achieves a coefficient of determination score of 0.97 and a mean absolute error of 0.17, demonstrating high accuracy in stress prediction. This technology-driven approach enhances mental health monitoring, enabling early detection, personalized intervention, and improved well-being through real-time insights and proactive support.

Distributed vehicular control systems include several safety-critical processes so that reliability aspects are of growing importance, raising concerns about faults affecting them. Observing this fact, this paper presents a combination of aspectoriented concepts to model faults in early-design phases of distributed vehicular control systems. A fault modeling approach in communication protocols as non-functional requirements-NFR is proposed, using aspect-oriented modeling (AOM) with the support of the Real-Time From Requirements to Design using Aspects (RT-FRIDA) framework. Following practical experiments about the effect of electrical fast transients in vehicular control systems, a study was performed specifying NFR associated with these faults. Then, an evaluation with a SIG graph using the softgoal weight method and the NFR framework is presented to validate the proposed approach. The results emphasize that early fault modeling could improve the control system modeling mapping fault behaviors in order to mitigate and diagnosis the fault impact in critical tasks. The approach cover gaps related to fault modeling and allow requirements specification with AOM concepts emphasized in the case study by the correlation among network performance degradation and requirements related to faults. The analysis with the softgoal weight method also provides an alternative view of the impact of fault modeling in vehicular critical real-time systems.

This paper presents the design and development of an alternative, cost-effective automated piece positioning system, specifically tailored for Small and Medium-sized Enterprises (SMEs), which integrates computer vision with EtherCAT-controlled servo motors. The proposed method combines a robust vision system with an AI-enhanced algorithm based on edge detection to precisely identify object contours. This enables a Programmable Logic Controller (PLC) to control the servo motor, adjusting the piece's angle with high accuracy. Experimental results demonstrate the solution's practical viability, achieving a minimal angular oscillation of less than 0.0012°and a promising low image processing time of approximately 20ms, showcasing its potential for enhancing manufacturing efficiency and quality in industrial applications.

including interactive systems, medical support tools, safety appl focus on a single input source, such as text, speech, or facial expressions, relying on one modality often limits accuracy in real voice system processes user text using an NLP MFCC, and determines facial emotions using a CNN trained on expression datasets. By merging outputs generated by the three independent modules, the system produces a more reliable and consistent emotion output. Experimental shows that the multimodel architecture outperforms individual models, demonstrating improved accuracy and stability, especially in situations involving noise, ambiguous text, or partial visibility. This approach highlights the potential of de learning driven multimodal emotion detection to build more smart and responsive intelligent systems. Emotion is a fundamental part of human communication and often carries more meaning than spoken words. In recent years, This is mainly applications. expressions. This approach often reduces accuracy when the input is unclear, incomplete, or affected by noise. Advancements in deep learning have improved emotion detection by allowing models to identify patterns from large amounts of data. comm context, and meaning in written text. Even with these developments, relying on a single type of input is life situations based different information. Text helps identify the user's inten expressions show visual cues related to emotional state. a more accurate and meaningful result compared to using them individua implementation, and evaluation of the proposed multimodel emotion detection system. severa Poria S detection

Good Distribution Practice (GDP) compliance is critical to ensuring the integrity, safety, and efficiency of pharmaceutical supply chains, particularly in developing economies where infrastructural and regulatory challenges persist. This study examines the influence of GDP compliance on pharmaceutical supply chain performance in Nigeria, focusing on four core dimensions: traceability and serialization, handling and packaging practices, storage and warehousing compliance, and quality control and monitoring. Adopting a survey research design, data were collected from 251 employees in logistics, supply chain, and sales/marketing departments of pharmaceutical companies in Nigeria. Descriptive statistics and simple linear regression analyses were employed to test the study hypotheses. The findings reveal that all four dimensions of GDP compliance exert a statistically significant positive effect on pharmaceutical supply chain performance, measured in terms of product availability, delivery reliability, quality integrity, and responsiveness. The study concludes that strengthening GDP compliance is of strategic importance to improving supply chain outcomes and safeguarding public health in Nigeria. It recommends enhanced regulatory enforcement, investment in traceability technologies, improved warehousing infrastructure, and continuous quality monitoring as pathways to achieving sustainable pharmaceutical supply chain performance.

Inventory management remains a critical challenge in modern supply chains, with traditional systems often suffering from limited visibility, manual data entry errors, outdated information, and delayed responsiveness to demand fluctuations. These limitations result in stock discrepancies, excess inventory, stockouts, and inefficient warehouse operations that collectively undermine operational performance and customer satisfaction. This article examines the deployment of Radio Frequency Identification and Internet of Things technologies to achieve real-time stock visibility and transform inventory management practices across warehousing, retail, and distribution environments. RFID tags and IoT sensors enable continuous, automated tracking of inventory items throughout the supply chain, while network connectivity transmits data to middleware platforms that filter, aggregate, and contextualize information streams. Advanced analytics engines process this real-time data to generate actionable insights, predictive alerts, and decision support capabilities that enhance inventory accuracy, optimize replenishment cycles, reduce shrinkage, and improve operational efficiency. The integration of RFID and IoT systems with enterprise resource planning and warehouse management platforms creates unified visibility across organizational boundaries, facilitating coordinated decision-making and responsive supply chain operations. Looking forward, the convergence of realtime inventory visibility with artificial intelligence, machine learning, and predictive analytics promises autonomous inventory optimization, demand-driven replenishment, and intelligent exception management. However, successful implementation requires addressing scalability challenges, interoperability standards, cybersecurity vulnerabilities, privacy concerns, and organizational readiness to fully realize the transformative potential of smart inventory management systems in the context of Industry 4.0.