Dr. Mahalingam College of Engineering and Technology

Federated learning (FL) has been proposed as an effective solution in the context of intrusion detection in IoT networks, where models can be trained collaboratively with the security of raw data protection. In this paper we present a privacy-preserving FL framework based on light weight neural network, differential privacy (DP) and homomorphic encryption (HE). With a dataset of 1,191,264 instances and 47 attributes, the proposed model conducted on the IoT Intrusion Detection Dataset available on Kaggle produces overall accuracy (93.5), precision (94.2), recall (93.4), and the F1-score (94.2), with the detection time of 90-130 ms and no distinction between the attacks, where detection latency was considered in this study. At the attack level the model delivered 94.1 %, 92.5 %, and 93.6 % accuracies on DoS, DDoS, and Mirai respectively, and above 85 % accuracy on Malware and Web-based attacks. DP experiments showed that augmenting the privacy budget parameter 0.5 to 20.0 increased the levels of accuracy by 2.6 % to 94.0 %, and decreased the computational time 150 ms to 121 ms, depicting a compromise between privacy and performance. HE experiments likewise exhibited a negligible accuracy reduction (94.1 % to 93.5 %) between no encryption to complete homomorphic encryption, but required more computation time (120 ms to 200 ms). Devices-level testing demonstrated that the model had >91 % accuracy at the low-end (0.5 GHz CPU, 128 MB memory) and up to 94.5 % accuracy with 110 ms inference time on powerful processors, irrespective of whether or not the sensor was heterogeneous, demonstrating a viable solution to the heterogeneous IT situation. Audit mechanisms further enhanced greater compliance of 0 % to 99 % with minimal reduction in accuracy (< 0.8 %). The results show that privacy-preserving intrusion detection specifically can be performed with real-time intrusion detection, high detection gene, and privacy guarantees in resource-constrained IoT networks.

The increased requirement and use of renewable energy worldwide has compounded the urgency to enhance solar thermal system efficiency, particularly Solar Evacuated Collector Tubes (SECTs), This research aims at enhancing the thermal and exergy efficiency of SECT systems by using progressive ternary nanofluids. Nanofluids of titanium dioxide (TiO2), aluminum oxide (Al2O3), and zinc oxide (ZnO) by 0.15 wt in 1:1:1, 2:1:1 and 3:1:1 combinations were prepared and tested using water as the base fluid. Experiments were conducted at flow rates of 1 L/min and 2 L/min to evaluate heat absorption, thermal efficiency, and exergy efficiency. The results revealed that the ternary nanofluid exhibited superior performance compared to mono-nanofluids: at 2 L/min, the peak heat absorption reached 530 W, which was 56.15 % higher than that of single nanofluids (TiO2: 460 W, Al2O3: 480 W, ZnO: 500 W). The ternary nanofluid achieved an average thermal efficiency of 77 %, outperforming TiO2 (63 %), Al2O3 (70 %), and ZnO (73 %), while the exergy efficiency reached 16.5 %, compared to 6.5 % for TiO2, 8.5 % for Al2O3, and 10.2 % for ZnO. The 1:1:1 nanoparticle ratio demonstrated the most stable enhancement due to balanced thermal conductivity and Brownian motion effects, while higher ratios (2:1:1 and 3:1:1) improved transient heat transfer. The observed improvement in performance is attributed to synergistic interactions among nanoparticles, which enhanced energy transport and reduced thermal resistance. These findings confirm the significant potential of ternary nanofluids to advance SECT technology, contributing to the development of more efficient and sustainable solar thermal energy systems.

Drought prediction is vital for early warnings, agricultural planning and management of water resources. The Standardised Precipitation Index (SPI) provides a quantitative measure of the severity of the drought and its extent, enabling preparedness and mitigation actions for short- and long-term drought events. Being solely dependent on precipitation records having spatio-temporal inconsistencies, conventional methods of SPI estimation suffer several computational challenges. In this study, four deep learning models, namely, Long Short Term Memory (LSTM), Stacked LSTM (S-LSTM), Bi-directional LSTM (BiLSTM), and Stacked BiLSTM (S-BiLSTM), were used to predict SPI values at various time scales (SPI-1, SPI-3, SPI-6, SPI-9, SPI-12, SPI-18, SPI-24, SPI-48, and SPI-yearly) for assessing the meteorological drought of Coimbatore station in India, based on 120 years of rainfall data. The short-term predictions (SPI-1) showed poor consistency (R-2 <= 0.11, RMSE similar to 1.0) with LSTM, though BiLSTM and S-BiLSTM experienced slight improvements. For moderate aggregations (SPI-3 to SPI-9), the S-LSTM resulted in improved performance (R-2 approximate to 0.62 to 0.89). For long-term scales (SPI-12 to SPI-24), S-BiLSTM outperformed with the highest consistency (R-2 > 0.94), indicating improved learning of long-range dependencies. The drought predictions at the annual scale were found to be weak due to the masking effect of critical short-term variabilities, resulting in a near-normal abundance. Bidirectional and stacked variants of LSTM models demonstrated better computational reliability with accurate, stable, and generalizable predictions, particularly for time scales SPI-6 to SPI-24. Deep learning improves drought forecasting using multi-source predictors, but marginal data variability must be addressed to ensure transparency and interpretability.

Passive thermal management of electronic devices employing circular fin–pin heat sinks embedded with phase-change materials (PCMs) that enhance thermal conductivity via mono-/hybrid nanoparticles is the focus of the present investigation. Two categories of nano-enhanced PCM were fabricated [mono-nano-PCM (MPCM) and hybrid nano-PCM (HPCM)] by combining paraffin wax (PFW) with 0.5 and 1.0 mass

This study introduces an innovative photodetector device that integrates an L-shaped top gate with a photosensitive back gate (LTG-PBG-TFET), specifically engineered to detect incident light within the near-infrared (NIR) wavelength range of 750–1050 nm. The LTG-PBG-TFET design leverages the advantages of both the L-shaped top gate and the photosensitive bottom gate to extend the edge tunneling zone at the channel/source (C/S) junction. This structural configuration enhances the photocurrent (Ilight) response, subthreshold swing (SSavg), and turn-ON voltage (Vth) under illumination. Furthermore, the slight elevation near the gate corner minimizes corner effects at the source/channel interface, which in turn improves Ilight/Idark performance as the illumination wavelength (λ) transitions from 1050 to 750 nm. Consequently, notable enhancements in Ilight, SSavg, and the Ilight/Idark ratio were observed, yielding a high spectral sensitivity (Sn) of approximately 54.4 and a signal-to-noise ratio (SNR) of about 82.2 for the proposed LTG-PBG-TFET device. Additionally, incorporating a light exposure window in the back gate region increases the active area for electron-hole pair (EHP) generation, thereby enhancing quantum efficiency (η) and responsivity (R), particularly at longer wavelengths around 1050 nm. Finally, the influence of acceptor–donor trap charges at the semiconductor/oxide interface on the Sn of the device was examined. Results indicate that Sn remains relatively stable as the wavelength (λ) is tuned between 750 and 1050 nm.