Chukka Rajasekhar
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Chukka Rajasekhar has a total of 38 co-authors. Here are the top co-authors by number of shared publications:
- Dasari Srinivasa Rao
GMR Institute of Technology
4 shared publications D. Srinivasa Rao
Dept. of ECE, GMPJT, Rajam, A.P, India
2 shared publicationsG. Sasibhushana Rao
2 shared publicationsK. Vijay
Dept of ECE, GITAM University, Visakhapatnam, A.P, India
2 shared publications- V. B. S. Srilatha Indira Dutt
Department of EECE, GITAM Deemed to be University, Visakhapatnam, India
2 shared publications A. Chitti Babu
1 shared publication
Latest Publications
- Open Access
- 9 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 37827-37844
LapGraph-CutFS: A Deep Learning-Based Framework for Multi-Class Classification of Maize Leaf Pests, Diseases, and Healthy States
Monitoring crop health is a critical component of sustainable agriculture, enabling the early detection and effective management of pests and diseases. We propose LapGraph-CutFS, a graph-theoretic feature selection framework designed for multi-class classification of maize leaf states, encompassing three insect pests (fall armyworm, grasshopper, and leaf beetle), one viral disease (maize streak virus), and a healthy control class. The framework constructs a Laplacian graph of the feature space and applies graph-cut optimization to preserve discriminative properties while eliminating redundancy, thereby improving the class-wise separability. LapGraph-CutFS integrates a Support Vector Machine (SVM) classifier and is benchmarked against baseline deep learning models (ResNet-50, Inception V3) and optimization-based approaches, including the Weighted Grey Wolf Optimizer (WGOA) and Particle Swarm Optimization (PSO). The proposed model was validated on a maize dataset, achieving 99.02% accuracy, 99.28% balanced accuracy, and a Cohen’s kappa score of 0.9869. It significantly outperformed both the baseline models and the optimization methods reported in the literature. Further validation on a brain tumor dataset demonstrated the strong generalization of the framework capability across distinct application domains. In summary, the results establish LapGraph-CutFS as a robust and scalable framework for addressing classification challenges in both the agricultural and biomedical imaging contexts.
- 9 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 37827-37844
Open AccessLapGraph-CutFS: A Deep Learning-Based Framework for Multi-Class Classification of Maize Leaf Pests, Diseases, and Healthy States
Monitoring crop health is a critical component of sustainable agriculture, enabling the early detection and effective management of pests and diseases. We propose LapGraph-CutFS, a graph-theoretic feature selection framework designed for multi-class classification of maize leaf states, encompassing three insect pests (fall armyworm, grasshopper, and leaf beetle), one viral disease (maize streak virus), and a healthy control class. The framework constructs a Laplacian graph of the feature space and applies graph-cut optimization to preserve discriminative properties while eliminating redundancy, thereby improving the class-wise separability. LapGraph-CutFS integrates a Support Vector Machine (SVM) classifier and is benchmarked against baseline deep learning models (ResNet-50, Inception V3) and optimization-based approaches, including the Weighted Grey Wolf Optimizer (WGOA) and Particle Swarm Optimization (PSO). The proposed model was validated on a maize dataset, achieving 99.02% accuracy, 99.28% balanced accuracy, and a Cohen’s kappa score of 0.9869. It significantly outperformed both the baseline models and the optimization methods reported in the literature. Further validation on a brain tumor dataset demonstrated the strong generalization of the framework capability across distinct application domains. In summary, the results establish LapGraph-CutFS as a robust and scalable framework for addressing classification challenges in both the agricultural and biomedical imaging contexts.
- Open Access
- 1 March 2026
- Published by Institute of Advanced Engineering and Science in International Journal of Applied Power Engineering (IJAPE)
- Vol. 15 (1) , 298-307
Feature transformation with ensemble learning for power grid stability in sustainable energy and industry systems
Power grids today operate under unpredictable and rapidly changing conditions, making reliable stability prediction increasingly important. This study evaluates two hybrid learning frameworks that integrate deep feature transformation with ensemble classification. In the first framework, an autoencoder (AE) is used for feature encoding before classification with extreme gradient boosting (XGBoost), while the second applies a TabTransformer (TT) followed by the same classifier. For comparison, conventional ensemble models, including random forest and standalone LightGBM, are also assessed. The models are tested on a large public dataset using stratified cross-validation and standard performance metrics. Results show that the AE-XGBoost hybrid achieves the highest performance, with a test accuracy of 97.73% and an F1-score of 0.98 for both stable and unstable states. LightGBM also performs strongly, offering consistent accuracy (95.8%) and good interpretability. In contrast, TT-XGBoost, despite its architectural novelty, achieves lower accuracy (89.4%) and struggles with unstable states. These findings highlight that model effectiveness depends not only on architectural complexity but also on the synergy between feature transformation and classification. The results provide practical insights for building dependable, confidence-aware predictive systems to support smart grid decision-making.
- 1 March 2026
- Published by Institute of Advanced Engineering and Science in International Journal of Applied Power Engineering (IJAPE)
- Vol. 15 (1) , 298-307
Open AccessFeature transformation with ensemble learning for power grid stability in sustainable energy and industry systems
Power grids today operate under unpredictable and rapidly changing conditions, making reliable stability prediction increasingly important. This study evaluates two hybrid learning frameworks that integrate deep feature transformation with ensemble classification. In the first framework, an autoencoder (AE) is used for feature encoding before classification with extreme gradient boosting (XGBoost), while the second applies a TabTransformer (TT) followed by the same classifier. For comparison, conventional ensemble models, including random forest and standalone LightGBM, are also assessed. The models are tested on a large public dataset using stratified cross-validation and standard performance metrics. Results show that the AE-XGBoost hybrid achieves the highest performance, with a test accuracy of 97.73% and an F1-score of 0.98 for both stable and unstable states. LightGBM also performs strongly, offering consistent accuracy (95.8%) and good interpretability. In contrast, TT-XGBoost, despite its architectural novelty, achieves lower accuracy (89.4%) and struggles with unstable states. These findings highlight that model effectiveness depends not only on architectural complexity but also on the synergy between feature transformation and classification. The results provide practical insights for building dependable, confidence-aware predictive systems to support smart grid decision-making.
- 1 December 2024
- Published by Technoscience Publications in Nature Environment and Pollution Technology
- Vol. 23 (4) , 2483-2487
Sustainability and Environmental Impact of Mining and Maintaining Cryptocurrencies: A Review
Cryptocurrency has seen an increased popularity with the introduction of Bitcoins. It has been adapted in several countries and has become an alternate solution to conventional currency. Despite its benefits, some controversies surround the manufacturing of bitcoins. While all the countries are moving to sustainability development and global warming control, Bitcoin production has raised several concerns about environmental pollution and sustainability. The increased carbon emissions and high electrical consumption have accompanied the popularity of cryptocurrency. Hence, there is an immediate need to reduce the carbon footprint and electricity consumption caused by human cryptocurrency for a sustainable future. This study presents the current scenario and trends of worldwide cryptocurrency growth and discusses the environmental impact of cryptocurrency mining. It explores crypto mining worldwide and provides a qualitative review. Further, this article highlights the need to take necessary measures to control cryptocurrency circulation.
- 1 December 2024
- Published by Technoscience Publications in Nature Environment and Pollution Technology
- Vol. 23 (4) , 2483-2487
Sustainability and Environmental Impact of Mining and Maintaining Cryptocurrencies: A Review
Cryptocurrency has seen an increased popularity with the introduction of Bitcoins. It has been adapted in several countries and has become an alternate solution to conventional currency. Despite its benefits, some controversies surround the manufacturing of bitcoins. While all the countries are moving to sustainability development and global warming control, Bitcoin production has raised several concerns about environmental pollution and sustainability. The increased carbon emissions and high electrical consumption have accompanied the popularity of cryptocurrency. Hence, there is an immediate need to reduce the carbon footprint and electricity consumption caused by human cryptocurrency for a sustainable future. This study presents the current scenario and trends of worldwide cryptocurrency growth and discusses the environmental impact of cryptocurrency mining. It explores crypto mining worldwide and provides a qualitative review. Further, this article highlights the need to take necessary measures to control cryptocurrency circulation.
- 8 August 2024
- Published by Springer Nature
- p. 17-28
Performance Comparison of QoS Aware Power Allocation and Optimization Techniques for the Small-Cell 5G Networks
- 8 August 2024
- Published by Springer Nature
- p. 17-28
Performance Comparison of QoS Aware Power Allocation and Optimization Techniques for the Small-Cell 5G Networks
- Open Access
- 1 January 2021
- Published by KEAI Publishing in International Journal of Intelligent Networks
- Vol. 2, 42-45
Weighted GDoP for improved position accuracy using NavIC and GPS hybrid constellation over Indian sub-continent
- 1 January 2021
- Published by KEAI Publishing in International Journal of Intelligent Networks
- Vol. 2, 42-45
Open AccessWeighted GDoP for improved position accuracy using NavIC and GPS hybrid constellation over Indian sub-continent
- 9 November 2015
- Published by Springer Nature in Wireless Personal Communications
- Vol. 88 (2) , 385-393
Investigation of Best Satellite–Receiver Geometry to Improve Positioning Accuracy Using GPS and IRNSS Combined Constellation over Hyderabad Region
- 9 November 2015
- Published by Springer Nature in Wireless Personal Communications
- Vol. 88 (2) , 385-393
Investigation of Best Satellite–Receiver Geometry to Improve Positioning Accuracy Using GPS and IRNSS Combined Constellation over Hyderabad Region
- 1 April 2015
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1185-1189
Role of optimum satellite-receiver geometry in the accuracy of GPS Navigation solution - an investigation
The field of Civil Aviation sector in India is going to be revolutionized by the Indian Regional Navigation Satellite System, which is regional navigation system of the Indian sub-continent. The accuracy of the Navigation solution of any satellite based navigation system is affected by several factors such as delay due to refraction in atmospheric layers, clock drift errors, errors due to multipath and satellite-receiver geometry. The navigation solution accuracy of any regional navigation system can be improved by augmenting the regional system with a Global satellite based Navigation system such as Global Positioning System, which is the popular and completely operational Global Navigation Satellite System In order to augment the IRNSS with GPS, initially it is necessary to investigate the techniques to improve the navigation solution accuracy of the GPS. Even though the delay errors due to atmospheric layers and multipath are mitigated, the accuracy of the GPS navigation solution is affected by satellite-receiver geometry, which represents the geometrical locations of the Navigation satellites seen by the GNSS receiver. As the satellite-receiver geometry is measured by Geometric Dilution of Precision (GDOP), monitoring of GDOP is necessary to determine the navigation solution accuracy. In this paper, the dual frequency GPS receiver data of the IGS Station, NGRI, Hyderabad (Lat./Long: 17.407430/78.526620) is collected using the SOPAC to carry out the required investigation of the navigation solution accuracy with DOP, by selecting four optimum satellites. This work can be extended by considering the IRNSS satellite positions along with GPS satellite locations.
- 1 April 2015
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1185-1189
Role of optimum satellite-receiver geometry in the accuracy of GPS Navigation solution - an investigation
The field of Civil Aviation sector in India is going to be revolutionized by the Indian Regional Navigation Satellite System, which is regional navigation system of the Indian sub-continent. The accuracy of the Navigation solution of any satellite based navigation system is affected by several factors such as delay due to refraction in atmospheric layers, clock drift errors, errors due to multipath and satellite-receiver geometry. The navigation solution accuracy of any regional navigation system can be improved by augmenting the regional system with a Global satellite based Navigation system such as Global Positioning System, which is the popular and completely operational Global Navigation Satellite System In order to augment the IRNSS with GPS, initially it is necessary to investigate the techniques to improve the navigation solution accuracy of the GPS. Even though the delay errors due to atmospheric layers and multipath are mitigated, the accuracy of the GPS navigation solution is affected by satellite-receiver geometry, which represents the geometrical locations of the Navigation satellites seen by the GNSS receiver. As the satellite-receiver geometry is measured by Geometric Dilution of Precision (GDOP), monitoring of GDOP is necessary to determine the navigation solution accuracy. In this paper, the dual frequency GPS receiver data of the IGS Station, NGRI, Hyderabad (Lat./Long: 17.407430/78.526620) is collected using the SOPAC to carry out the required investigation of the navigation solution accuracy with DOP, by selecting four optimum satellites. This work can be extended by considering the IRNSS satellite positions along with GPS satellite locations.
- 1 February 2014
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
PAPR reduction performance in OFDM systems using channel coding techniques
Orthogonal frequency division multiplexing (OFDM) is the most popular modulation technique in modern wireless communication systems involving high data rates. However, due to the large number of sub carriers used, OFDM systems have a large dynamic signal range with a very high peak-to-average-power ratio (PAPR) which makes OFDM sensitive to nonlinear effects. In this paper, a comparative study of PAPR reduction performance is studied using various channel coding techniques for the OFDM signals. Here, for performance analysis Hamming codes, Cyclic codes are considered. It is shown that each of the channel coding techniques has a different PAPR reduction performance and the coding technique with better PAPR performance is identified.
- 1 February 2014
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
PAPR reduction performance in OFDM systems using channel coding techniques
Orthogonal frequency division multiplexing (OFDM) is the most popular modulation technique in modern wireless communication systems involving high data rates. However, due to the large number of sub carriers used, OFDM systems have a large dynamic signal range with a very high peak-to-average-power ratio (PAPR) which makes OFDM sensitive to nonlinear effects. In this paper, a comparative study of PAPR reduction performance is studied using various channel coding techniques for the OFDM signals. Here, for performance analysis Hamming codes, Cyclic codes are considered. It is shown that each of the channel coding techniques has a different PAPR reduction performance and the coding technique with better PAPR performance is identified.
- 11 January 2013
- Published by Springer Nature
- p. 217-225
Channel Estimation and Equalization for Time-Varying OFDM System Using Kalman Filter
- 11 January 2013
- Published by Springer Nature
- p. 217-225
Channel Estimation and Equalization for Time-Varying OFDM System Using Kalman Filter