International Journal of Engineering & Technology Sciences

The increasing demand for secure and efficient digital signal processing (DSP) in embedded systems has intensified the need for hardware platforms capable of delivering real-time performance without compromising cryptographic robustness. Field-Programmable Gate Arrays (FPGAs) have emerged as a versatile solution, offering fine-grained parallelism, low-latency execution, and reconfigurable logic for integrating both signal processing and cryptographic functions. This review presents a comprehensive synthesis of recent advancements in FPGA-based secure DSP architectures, encompassing classical primitives such as AES and ECC, as well as post-quantum cryptographic algorithms like Kyber, Dilithium, and NTRU. It explores design methodologies ranging from Register Transfer Level (RTL) to High-Level Synthesis (HLS), evaluating trade-offs in power, area, and latency, and detailing hardware-level countermeasures against side-channel attacks. Practical applications are surveyed across domains including healthcare, defense, secure communications, and multimedia processing, with comparative benchmarking on major Xilinx and Intel FPGA platforms. The review identifies key challenges, such as resource constraints, cross-platform portability, and the real-time implementation of cryptographic workloads, and highlights the integration of AI-based threat detection using models like convolutional and graph neural networks. A classification of machine learning applications in secure DSP is provided to contextualize current research directions. Finally, emerging trends are discussed, including post-quantum secure DSP systems, FPGA–AI co-acceleration, secure and reconfigurable hardware architectures, processing-in-memory (PIM), and heterogeneous platforms combining RISC-V SoCs with FPGA fabrics. By bridging cryptographic assurance with signal integrity, this work offers a holistic overview of the secure embedded computing landscape and a roadmap for future innovation at the intersection of signal processing, reconfigurable computing, and hardware-level security.

Digital Signal Processing (DSP) is critical in the development and optimization of modern telecommunication and wireless systems. The advent of complex standards like 5G and AI-native networks has increased the need to revisit classical and new DSP algorithms. This paper presents a systematic review of DSP techniques that include time-domain, frequency-domain, and advanced signal processing methods, as well as new AI-based approaches, such as Deep Reinforcement Learning (DRL), Neural Architecture Search (NAS), Graph Neural Networks (GNN), and Bayesian Optimization (BO). The survey explores basic DSP principles, historical developments, and key performance metrics like latency, throughput, and computational complexity. It compares classical and new algorithms, analysing their merits, limitations, and suitability for real-world applications, such as modulation schemes (QAM, OFDM), channel estimation, noise suppression, MIMO, and beamforming. Additionally, the paper addresses the integration with Software-Defined Radio (SDR), edge computing, and the Internet of Things (IoT), highlighting challenges related to real-time implementation and hardware acceleration using FPGA and ASIC platforms. Key research challenges and future directions are identified, such as the need for scalable, adaptive DSP in dynamic environments, power-efficient hardware implementation of AI models, and the growing importance of optimization in future wireless systems. This review is a comprehensive resource for researchers at the intersection of signal processing, wireless communication, and intelligent systems.

Digital filter design remains a foundational pillar of modern signal processing, facilitating the extraction, enhancement, and suppression of signal components across a broad spectrum of applications, including wireless communication, biomedical imaging, the Internet of Things (IoT), industrial automation, and edge computing. This review comprehensively examines both classical approaches, such as analog filter designs (e.g., Butterworth, Chebyshev, Elliptic) and digital implementations (FIR and IIR), and advanced, optimization-driven techniques that incorporate machine learning, neural networks, reinforcement learning, and quantum-inspired algorithms. Particular emphasis is placed on practical FPGA-based realizations, highlighting their reconfigurable, low-latency architectures tailored for real-time systems. A comparative analysis of FIR and IIR filters is presented in terms of latency, computational complexity, and hardware–software resource trade-offs across CPU, DSP, and FPGA platforms. Furthermore, the study explores adaptive filtering in dynamic, resource-constrained environments, hybrid classical–deep learning filter structures, and secure designs integrating cryptographic methods and machine learning–based Trojan detection. Finally, emerging trends and research challenges are discussed, including reconfigurable and neuromorphic architectures, holistic hardware–algorithm co-design, and seamless integration into heterogeneous high-performance computing platforms, laying the groundwork for the next generation of intelligent, adaptive, and secure signal processing systems.

This study, based on 47 articles, explores cost-effective practices in reinforced concrete projects, highlighting sustainability, innovation, and process optimization. Examples include the use of concrete ties in Brazil, concrete recycling in Japan, and the use of recycled aggregates in Australia, all contributing to reducing environmental impact. Nanotechnology, steel fibers, and carbon-reinforced concrete improve the mechanical properties and durability of concrete. Additionally, resistance to chloride and sulfate attacks is enhanced through FRP polymers and nano-iron. The adoption of technologies like linear programming, predictive models, and BIM optimizes production and project planning. Regarding sustainability, low-carbon cements and the use of recycled materials help reduce the carbon footprint. Finally, the impact of polypropylene fibers on seismic resilience is highlighted, along with the need to update building codes to improve safety in high-rise buildings.

The integration of Artificial Intelligence (AI) and Machine Learning (ML) into structural engineering is transforming traditional methods of designing and analyzing reinforced concrete. These advanced technologies enable the creation of more accurate predictive models and real-time design optimization, reducing error margins and enhancing construction efficiency. This article explores how deep neural networks and supervised and unsupervised learning algorithms are revolutionizing engineers’ approaches to complex problems such as load distribution, material strength, and structural failure estimation. Additionally, it examines case studies where AI and ML have facilitated the design of safer, more sustainable, and cost-effective structures. Automation of repetitive tasks and analysis of large experimental datasets are among the advantages these technologies bring to the civil engineering sector. However, challenges such as integration into traditional processes, the need for specialized training, and ensuring the reliability and transparency of obtained results are also identified. This article concludes that the incorporation of AI and ML in the design and analysis of reinforced concrete promises to revolutionize the future of structural engineering, offering innovative solutions tailored to the demands of sustainability and efficiency in construction.

This work compiles a series of investigations on the use of nanosilica (NS) as an additive in concrete, evaluating its impact on mechanical properties, durability, workability and sustainability. The studies consider the incorporation of NS in various proportions (0.4%-3%) and its combination with other additives such as microsilica and superplasticizers, using methodologies based on international standards (ACI, ASTM). The results show that NS significantly increases compressive strength (up to 56.92%) and flexural strength, improves impermeability and durability against aggressive environments, such as coastal and sulfated areas, by reducing the porosity of concrete. It is also observed that it optimizes workability in fresh mixtures, being ideal for self-compacting concretes. However, proportions above 1.5% can have adverse effects on cohesion and mechanical behavior. In terms of sustainability, the use of NS makes it possible to reduce the amount of cement needed, decreasing the carbon footprint of the material, and is considered economically viable in projects that prioritize durability and high resistance.