Francis Xavier Engineering College, Tirunelveli, is an Autonomous institution located in the town of Tirunelveli in the state of Tamil Nadu. Tirunelveli is often referred as the 'Oxford of south India' due to the larger number of educational institutions present. The Francis Xavier Engineering College popularly known as FX Engineering College, was established in the year 2000..
The Human Activity Recognition (HAR) from the sensor component has developed the primary part for several real-world scenarios, such as healthcare and disease diagnosis. The conventional techniques have some drawbacks in terms of harmonizing accuracy and speed. Additionally, the relevant methods have not provided a solution for addressing the unfair data in dissimilar activities of HAR, even though it has the main concern for producing enhanced performance. This paper proposes the hybrid technique of Hierarchical Entropy-based GRU-CNN-AdaBoost Framework for performing the classification process of human activities. The proposed framework has the segmentation process with the sliding window concept. The dispersion entropy of several frequency units is produced by the feature vector group. Finally, the hybrid AdaBoost procedure is used to classify the human activities. The Hierarchical entropy computation has the processing of raw signals, frequency components, and time–frequency representations. The proposed model is used to extract the features for capturing temporal patterns in sensor data. The hybrid model produces a robust classification framework by learning spatial–temporal dependencies through a multi-class AdaBoost classifier. Comprehensive evaluation on the datasets of the KU-HAR dataset and USC-HAD dataset demonstrates that the proposed framework produces improved performance and maintains statistically significant enhancements (p < 0.0001) over relevant techniques of GE-EnsemCNN-HAR, DeepConvLSTM, and ICGNet.
Concrete is the primary and most often utilized structural material in civil engineering. Prompt assessment of concrete strength is crucial for ensuring structural integrity and reducing construction delays, therefore preventing potential structural failures. This preliminary assessment guarantees concrete structures support loads throughout their operational lifespan and during construction. A major problem in the construction sector is the precise assessment of concrete strength and detection of possible damage without resorting to destructive testing. Traditional methods frequently necessitate labor processes and may be unfeasible for real-time monitoring. To address this challenge, IoT-based monitoring systems with Polyvinylidene Fluoride Film (PVDF) sensors offer an effective solution for damage detection and ongoing strength assessment at concrete structures. This research employed a polyvinylidene fluoride film sensor, utilizing surface-bonding method to affix sensor to cylindrical specimens. Trial phase lasted four weeks, incorporating assessments on 5th, 10th, and 15th days to detect any structural damage and evaluate required strength levels. This investigation confirmed that the findings achieved by PVDF-based wireless sensor were both dependable and practical. The correlation coefficient values are examined to confirm the relationship between data from IoT-based testing and compressive strength. All results are displayed graphically, demonstrating that this non-destructive method can precisely forecast concrete strength and detect structural problems. This work distinctly contributed by verifying the application of PVDF sensors for continuous, in-situ monitoring of concrete, offering an innovative method for early damage detection and assessing the structural integrity of the structure.
The growing use of wireless sensor network (WSN) has resulted in the growth of demand for efficient, reliable, and secure communication infrastructures. Nonetheless, limitations, including limited energy resources, dynamic topology, and vulnerability to malicious attacks, tend to reduce network performance and reliability. Hence, the proposed research work presents a new Greylag Goose Random Forest-Based Flooding Protocol (GGRFFP) to enhance communication performance in WSNs. The developed hybrid model incorporates Greylag Goose Optimization (GGO) for energy-aware and adaptive cluster head (CH) selection, Random Forest (RF) for identifying malicious nodes, and an enhanced Flooding Protocol (FP) for secure and efficient data transmission. The model was implemented in the NS3 simulation platform, and the experimental results verify that GGRFFP significantly increases energy efficiency, network lifetime (NLT), packet delivery ratio (PDR), throughput, and security accuracy compared to classical routing protocols. The system achieved the maximum detection accuracy (DA), the lowest false alarm rate (FAR), and reductions in energy consumption (EC) and routing overhead (ROH). These results demonstrate that the protocol efficiently reduces redundant transmissions, improves secure communication, and prolongs the operational longevity of wireless sensor network. In the end, GGRFFP offers a scalable, innovative routing paradigm for future industrial and smart-environment applications.
The Internet of Things (IoT) refers to a system of interconnected computing devices, sensors, and supporting infrastructure. Attacks from Distributed Denial of Service (DDoS) and insufficient resources are common issues for this network. Security and access control might be enhanced by integrating the IoT with Software-Defined Networking (SDN). A method for detecting DDoS attacks in Wireless Sensor Networks (WISNE) using machine learning (ML) is discussed in this article. The WISNE-SDN IoT controllers could make use of this technique. In a testbed environment that mimics DDoS attack traffic, the WISNE-SDN controller may gather network events into a pre-processed dataset. For tasks like packet sorting and attack detection, the framework employs some ML algorithms, such as K-Nearest Neighbor (KNN), XGBoost (XGB), and Naive Bayes (NB). Accuracy levels of 97% for KNN, 100% for XGB. The suggested approach improves the security and stability of IoT networks in SDN-IoT environments by making them more resistant to DDoS attacks.
A multi-layer (ML) Ti/TiO2 photoanode was developed to advance the operational functionality of dye-sensitized solar cells (DSSCs). The architecture consists of a sputtered Ti interlayer, a dense TiO2 blocking layer, and multiple spin-coated TiO2 nanoparticle layers to improve structural integrity, light harvesting, and charge transport. XRD analysis confirmed phase-pure anatase TiO2 with improved crystallinity in the ML film, while SEM images revealed a crack-free, well-defined layered structure. Optical studies demonstrate a red shift in the absorption and a reduced band gap of 3.1 eV compared to 3.3 eV for single- and double-layer films, along with defect-induced visible absorption. Photoluminescence quenching indicated suppressed electron–hole recombination. The ML-based DSSC demonstrated a power conversion efficiency of 5.54
