Surender Redhu

• Charan, Kamal V.; Redhu, Surender; Soni, Gunjan & Singh, Maneesh (2025). Application of a Large Language Model (LLM) for Failure Modes and Effects Analysis (FMEA). In Singh, Maneesh; Soni, Gunjan; Sinha, Jyoti; Ball, Andrew D.; Gu, Fengshou; Ouyang, Huajiang & Featherston, Carol (Ed.), Proceedings of the UNIfied Conference of DAMAS, IncoME VIII and TEPEN Conferences. Springer Nature Switzerland. ISSN 9783031959639. p. 619–634. doi: 10.1007/978-3-031-95963-9_43. Full text in Research Archive Show summary

  During its operational lifespan, equipment used in chemical, mechanical, and metallurgical process plants is subjected to various degradation mechanisms, including corrosion, erosion, fatigue, and deformation. These factors can severely im-pact equipment integrity, increasing the likelihood of failure. While complete elimination of failures is impractical, an effective asset management program focuses on maximizing reliability and availability through well-structured inspection and maintenance strategies. However, developing and implementing such pro-grams is complex and costly. Risk-based approaches, such as Risk-Based Inspection (RBI) and Reliability-Centered Maintenance (RCM), offer systematic methods to optimize resource allocation and minimize accident risks. Failure Modes and Effects Analysis (FMEA) is a widely used technique for risk assessment. Primarily qualitative, FMEA can incorporate quantitative elements by estimating failure likelihood and consequences. While benefiting from human expertise, traditional FMEA is often criticized for being time-intensive, prone to human error, and inherently static. Recent advancements in Generative AI have opened avenues for automating the FMEA process. A current project is developing a framework that combines open-source Large Language Models (LLMs) with Retrieval-Augmented Generation (RAG) to automate the creation of FMEA/FMECA tables using only pre-vetted, credible documents. At the proof-of-concept (PoC) stage, the project has shown promising results in automating failure analysis. This paper introduces the proposed frame-work and presents initial findings.

• Cheng, Xu; Ilieva, Iliana; Bremdal, Bernt Arild; Redhu, Surender & Ottesen, Stig (2023). A Novel Spatial-Temporal Deep Neural Network for Electricity Price Forecasting. In Nichele, Stefano; Aamodt, Jonas Moræus; Misra, Sanjay & Mölder, Anicka (Ed.), 2023 3rd International Conference on Applied Artificial Intelligence (ICAPAI). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9798350328929. doi: 10.1109/ICAPAI58366.2023.10193970. Full text in Research Archive
• Redhu, Surender; Singh, Aditya & Bremdal, Bernt Arild (2023). Edge Computing for Improving Energy Management in Smart Homes. In 2023, CIRED (Eds.), 27th International Conference on Electricity Distribution - CIRED 2023. Institution of Engineering and Technology (IET). ISSN 9781839538551. p. 3498–3502. doi: 10.1049/icp.2023.0815. Full text in Research Archive
• Redhu, Surender & Bremdal, Bernt Arild (2023). Day-Ahead Zonal Electricity Price Forecasting Using 1d-Lstm With Neighbouring Zones Data, 2023 IEEE Belgrade PowerTech. IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781665487788. p. 1–6. doi: 10.1109/PowerTech55446.2023.10202771. Full text in Research Archive
• Redhu, Surender; Raja, Rameez; Bremdal, Bernt Arild & Nygård, Heidi S. (2023). Cognitive Data Fusion for Improving Flexibility in Smart Homes. In 2023, CIRED (Eds.), 27th International Conference on Electricity Distribution - CIRED 2023. Institution of Engineering and Technology (IET). ISSN 9781839538551. p. 2848–2852. doi: 10.1049/icp.2023.1029. Full text in Research Archive
• Kushwaha, Deepali; Redhu, Surender; Brinton, Christopher G. & Hegde, Rajesh M. (2023). Optimal Device Selection in Federated Learning for Resource-Constrained Edge Networks. IEEE Internet of Things Journal. ISSN 2327-4662. 10(12), p. 10845–10856. doi: 10.1109/JIOT.2023.3243082. Full text in Research Archive
• Singh, Aditya; Redhu, Surender & Hedge, Rajesh M. (2022). UAV Altitude Optimization for Efficient Energy Harvesting in IoT Networks, 22 National Conference on Communications (NCC). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781665451369. p. 350–355. doi: 10.1109/NCC55593.2022.9806741. Full text in Research Archive Show summary

  Energy harvesting plays a crucial role in improving the operational lifetime of an IoT network. Moreover, the source and receiver separation dominates the amount of energy har-vested in rechargeable IoT networks. In recent, Unmanned Aerial Vehicles (UAVs) have been utilized as RF energy transmitters for energy harvesting IoT networks. In this work, a method is proposed for optimizing the altitude of UAV s for energy-efficient charging of the IoT nodes. The proposed method maximizes the energy usage efficiency of the UAV over the IoT network subject to altitude and energy harvesting constraints. The proposed maximization problem is first transformed into an equivalent convex optimization problem using the First-Order Taylor Series Approximation. A heuristic algorithm based on the sequential convex programming approach is employed to obtain the optimal UAV altitude above the IoT network. The accuracy of the obtained results is evaluated analytically by computing the global optimum of the optimization problem via monotonic transformations. The results motivate the usage of UAV-aided energy harvesting in self-sustaining IoT networks.

• Dutt, Micheal; Redhu, Surender; Goodwin, Morten & Omlin, Christian Walter Peter (2022). Sleep Stage Identification based on Single-Channel EEG Signals using 1-D Convolutional Autoencoders, 2022 IEEE International Conference on E-health Networking, Application & Services (HealthCom). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781665480161. p. 94–99. doi: 10.1109/HealthCom54947.2022.9982775. Full text in Research Archive Show summary

  Automatic sleep stage classification can play a vital role when measuring sleep quality and diagnosing different sleep-related ailments. Several automated sleep stage identification algorithms have been proposed using various physiological signals. However, most of these methods use hand-crafted features or multiple Electroencephalography (EEG) signals. This work proposes a one-dimensional convolutional autoencoder (1D-CAE) based on a single-channel EEG signal for sleep stage identification. A total of five 1-D CAEs models are implemented, and each model is trained to reconstructs a specific sleep stage with the lowest reconstruction error, thus enabling the sleep stage identification based on this error. Furthermore, the proposed approach is evaluated on the Sleep EDF expanded datasets and achieved an overall classification accuracy of 87.2% using a single-channel EEG FPz-Cz signal. Also, our approach demonstrated the highest sleep stage identification accuracy compared with the recent algorithms, especially for sleep stage N1, a short period that transitions between sleep stages during a sleep cycle.

• Singh, Aditya; Redhu, Surender & Hedge, Rajesh M (2022). Energy-Efficient UAV Trajectory Planning in Rechargeable IoT Networks, 2022 IEEE International Conference on Signal Processing and Communications (SPCOM). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781665482509. doi: 10.1109/SPCOM55316.2022.9840770. Full text in Research Archive Show summary

  Maintaining adequate energy in low-powered Internet of Things (IoT) nodes is crucial for developing several applications like smart homes, autonomous industries, etc. In this context, energy harvesting plays an essential role in improving the operational lifetime of the IoT nodes. Unmanned Aerial Vehicles (UAVs) have become a feasible option for reaching out to the low-powered IoT nodes in remote areas and recharging them by acting as efficient energy transmitter units. However, ensuring a sustainable and regular supply of power to these IoT nodes mainly depends on the trajectories of UAVs. In this context, the UAV trajectory optimization problem is first formulated. Subsequently, an energy-efficient UAV route planning algorithm (UAV-RPA) is proposed to generate the UAV trajectory to recharge the IoT nodes. The proposed algorithm minimizes the UAV-travel time by selecting an optimal sequence of IoT nodes such that the UAV trajectory length is minimized. Moreover, extensive simulations are also conducted under various network scenarios to evaluate the performance of the route planning algorithm. It is observed that the proposed UAV-RPA generates a minimal length UAV trajectory over an IoT network when compared to other UAV trajectory generation algorithms. Also, the average residual energy per IoT node in the network is also improved. This, in turn, improves the operational lifetime of self-sustaining UAV-powered IoT networks.

• Sironja, Poorva; Redhu, Surender; Hegde, Rajesh M. & Beferull-Lozano, Baltasar (2022). Cluster-based 3D-Beam Power Management in Resource-Constrained Wireless Networks, 2022 14th International Conference on COMmunication Systems & NETworkS (COMSNETS). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781665421041. p. 590–594. doi: 10.1109/COMSNETS53615.2022.9668343. Full text in Research Archive Show summary

  Abstract—Various applications of wireless 5G networks demand high throughput for improving the Quality of Service. Power management of the beams plays a key role in improving the throughput in 5G-mmWave networks. However, resource constraints make beam generation a challenging issue. In this work, cluster-based 3D beam power management schemes are proposed for improving the throughput in resource-Constrained wireless networks. Firstly, the network users are divided into a number of groups. 3D beam patterns are generated from the transmitter to maximize the throughput of users based on clustering information. The proposed cluster-based beam management scheme considers a limited amount of power resources at the transmitter. Available transmit power is distributed optimally to every cluster in the network. The performance is improved further by identifying the outliers in the clusters. The tradeoff between the transmit power and users served decides the threshold on outlier detection. The performance of proposed cluster-based 3D beam patterns is evaluated under different network scenarios using numeric simulations. The proposed method shows significant improvements in terms of throughput, network coverage, average signal-to-noise ratio (SNR), and transmit power consumption. The performance of the proposed methods is also compared with the alternative schemes under power-constrained scenarios.

• Kumar, Sarang; Elnourani, Mohamed; Beferull-Lozano, Baltasar & Redhu, Surender (2022). A Deep Learning-Based Channel Aware Single Step Signal Detection in Downlink Multi-User NOMA, 2022 IEEE 96th Vehicular Technology Conference (VTC2022-Fall). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781665454681. doi: 10.1109/VTC2022-Fall57202.2022.10013001. Full text in Research Archive
• Dutt, Micheal; Redhu, Surender; Goodwin, Morten & Omlin, Christian Walter Peter (2022). SleepXAI: An explainable deep learning approach for multi-class sleep stage identification. Applied Intelligence - The International Journal of Research on Intelligent Systems for Real Life Complex Problems. ISSN 0924-669X. doi: 10.1007/s10489-022-04357-8. Full text in Research Archive Show summary

  Extensive research has been conducted on the automatic classification of sleep stages utilizing deep neural networks and other neurophysiological markers. However, for sleep specialists to employ models as an assistive solution, it is necessary to comprehend how the models arrive at a particular outcome, necessitating the explainability of these models. This work proposes an explainable unified CNN-CRF approach (SleepXAI) for multi-class sleep stage classification designed explicitly for univariate time-series signals using modified gradient-weighted class activation mapping (Grad-CAM). The proposed approach significantly increases the overall accuracy of sleep stage classification while demonstrating the explainability of the multi-class labeling of univariate EEG signals, highlighting the parts of the signals emphasized most in predicting sleep stages. We extensively evaluated our approach to the sleep-EDF dataset, and it demonstrates the highest overall accuracy of 86.8% in identifying five sleep stage classes. More importantly, we achieved the highest accuracy when classifying the crucial sleep stage N1 with the lowest number of instances, outperforming the state-of-the-art machine learning approaches by 16.3%. These results motivate us to adopt the proposed approach in clinical practice as an aid to sleep experts.

• Singh, Aditya; Redhu, Surender & Hegde, Rajesh M. (2021). Context-Aware RF-Energy Harvesting for IoT Networks, 2021 IEEE 7th World Forum on Internet of Things (WF-IoT). IEEE Press. ISSN 9781665444316. doi: 10.1109/WF-IoT51360.2021.9595055. Full text in Research Archive
• Tripathi, Shraddha; Redhu, Surender & Hedge, Rajesh M (2021). Clustering-Assisted 3D Beamforming for Throughput Maximization in mmWave Networks. IEEE International Conference on Communications. ISSN 1550-3607. doi: 10.1109/ICCWorkshops50388.2021.9473596. Full text in Research Archive
• Redhu, Surender; Singh, Amrendra P.; Hedge, Rajesh M & Beferull-Lozano, Baltasar (2021). A Cluster based Sensor-Selection Scheme for Energy-Efficient Agriculture Sensor Networks, 2021 IEEE 18th Annual Consumer Communications & Networking Conference (CCNC). IEEE (Institute of Electrical and Electronics Engineers). ISSN 9781728197944. doi: 10.1109/CCNC49032.2021.9369595. Full text in Research Archive

View all works in NVA