Pratibha Lanka
@gvpcdpgc.edu.in
RESEARCH, TEACHING, or OTHER INTERESTS
Artificial Intelligence
Scopus Publications
- Unsupervised Detection of Groundwater Quality Anomalies via Autoencoder-LOF Ensemble
T.M.N. Vamsi, Saaketh Choudarapu, Pratibha Lanka, G. Kamal, DNVSLS Indira, J. N.V.R. Swarup Kumar
Proceedings of the 4th International Conference on Intelligent Computing Information and Control Systems Icoiics 2025, 2025
An unsupervised pipeline is introduced for detecting groundwater quality anomalies by combining an autoencoder (AE) with Local Outlier Factor (LOF) in the AE latent space. The pipeline targets multi-parameter hydrochemistry and operates without labels. Robustness is improved through deterministic training and a lightweight ensemble that averages LOF percentile ranks across nearby neighborhood sizes with MC‒dropout latent draws; an optional consensus with AE reconstruction error further suppresses spurious flags. Evaluation is conducted on three consecutive post-monsoon datasets from Telangana State, India (2018‒2020), comprising village-level measurements of pH, EC, TDS, major ions, hardness, SAR, and geospatial attributes. In the absence of ground-truth labels, performance is assessed using policy-aligned proxies: Stability Index under bootstrap resampling, Regulatory Violation Enrichment against BIS/WHO limits (EC, TDS, F, NO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</inf>), and separation in latent space (silhouette). The AE‒LOF ensemble attains near-perfect violation enrichment (≈1.00) and the highest separation (silhouette ≈0.70), with competitive stability after ensembling. Visualizations (latent projections, parameter distributions, geospatial maps) and case studies confirm salinity-, nitrate-, and mixed salinity-hardness anomalies that are actionable for water management. The approach is scalable, tunable via a single percentile threshold, and readily implemented in Python/Colab. The resulting anomaly watchlist supports prioritization of inspections and targeted interventions and is applicable to other environmental sensing contexts. - Learning processes for the internet of autonomous vehicles and intelligent transportation systems
T. Mohana Naga Vamsi, L. Pratibha
Wireless Ad Hoc and Sensor Networks Architecture Protocols and Applications, 2024
This chapter explores the learning processes within the Internet of Vehicles (IoV), intelligent transportation systems (ITS), and autonomous vehicles (AVs), emphasizing the essential role of machine learning and autonomous system architecture in advancing these technologies. It begins by detailing the application of machine learning in ITS, IoV, and AVs, highlighting its forms—supervised, unsupervised, and reinforcement learning—and their impact on traffic management, route optimization, and autonomous navigation through real-world examples. The significance of sensor networks in data collection for decision-making and safety is discussed, along with the essential elements of autonomous system architecture that enable vehicle autonomy, such as sensor fusion and safety mechanisms, illustrating the synergy of machine learning in autonomous decision-making. Further, the chapter addresses machine learning models suited for IoV, ITS, and AV applications, from traffic forecasting to self-driving technologies, underscoring their adaptability and effectiveness. It provides model training with IoV data, emphasizing methods for validating and enhancing model robustness to ensure successful deployment. Additionally, the narrative explains the roles of deep learning in image recognition, reinforcement learning in improving driving experiences, and federated learning in maintaining privacy in decentralized training. It acknowledges the challenges faced by these technologies, including security, technological constraints, and regulatory issues, while also highlighting their collective evolution toward more intelligent, adaptable, and efficient transportation systems. Conclusively, the transformative potential of integrating machine learning and autonomous system architecture in future transportation technologies is underscored, illustrating a vision for innovation-led, improved transportation ecosystems. - Improving Text-Driven Image Synthesis: Diffusion Models for Photorealistic Outcomes
T.M.N. Vamsi, J. N.V.R. Swarup Kumar, I.S. Siva Rao, Pratibha Lanka
International Journal of Computing, 2024
In recent developments, there has been a noteworthy demonstration of the effectiveness of generating high-quality images of diffusion models. This success is further enhanced when these models are combined with a technique that allows for a strategic balance between image diversity and fidelity. Addressing the challenge of text-conditional image synthesis, we extensively explore the utility of diffusion models along with two distinct guiding approaches: CLIP (Contrastive Language–Image Pretraining) guidance and classifier-free guidance. Through a comprehensive analysis, we uncover intriguing insights. The classifier-free guidance method consistently emerges as a standout performer, producing images with remarkable photorealism. This method showed a PSNR of 183.66 dB and an SSIM of 99.99%, indicating efficient photorealism and structural similarity to ground reality images. It presents a unique approach that combines diffusion models with classifier-free guidance for text-conditional image synthesis, focusing on photorealism and alignment with captions. Therefore, it can be useful for human evaluators to proficiently maintain both visual realism and associated captions. - An IoT Based Smart Water Contamination Monitoring System
M N Vamsi Thalatam, Pratibha Lanka, J.N.V.R. Swarup Kumar
Proceedings of the 2023 International Conference on Intelligent Systems for Communication Iot and Security Iciscois 2023, 2023
In the current era of urbanization, there is a need for effective monitoring and alert the concerned water body maintenance agencies to control the contamination levels in the residential water supply. Because of the increasing population and exponential growth of pollution, this task of quality water supply becomes challenging. At present, the concerned water quality maintenance agencies rely on traditional techniques. In this method, they are collecting samples of suspected water from the tanks and dams. Then send these samples to concerned laboratories for testing the level of contamination in the sample. This is not only a time-consuming process but also a costly affair. Also, with the existing method, dissemination of analyzed real-time data to the concerned stakeholders for taking necessary timely action has become a cumbersome job. To overcome all the issues this novel system is proposed. In this paper, a real-time water contamination monitoring system is developed using Internet of Things (IoT) and Embedded Systems. This system can able to estimate the water quality in residential homes as well water body tanks. The system consists of an Arduino Uno, ESP 8266, Analog to Digital Converter, pH sensor, turbidity sensor, temperature sensor, and other electronic components. With this setup, the water contamination can be estimated by measuring the pH value, turbidity of water and temperature of the water. The measured values at the edge level are then pushed to ThingSpeak cloud server via the ESP8266 wifi module. The accumulated data is analyzed in the cloud and from which the results are accessed via an interfacing device. The results obtained can help determine the quality of the water sample to be tested. The system has achieved the intelligence of data processing and networking of information transfer for automated water quality monitoring. It is distinguished by the advantages of efficiency, accuracy, and frugal use of labour and material resources. - An IoT Based Smart Garbage Monitoring and Disposal Support System
T. M. N. Vamsi, G. Kalyan Chakravarthi, Pratibha Lanka, B. Divakar
Proceedings 5th International Conference on Computing Methodologies and Communication Iccmc 2021, 2021
Today globally people designed different systems based on Internet of Things (IoT) principles and made human with smart living. In this article, an important application is proposed for our daily usage and is named as Smart Garbage Monitoring and Disposal Support System (SGMDSS). Since the advent of mobile communication and the spread of the Internet anywhere and anytime, there is a lot of scopes to control the devices from remote places. Solid garbage is one of the prior concerns today globally because improper execution of garbage removal from the villages, towns, and cities affects the people living there and it is the ultimate cause for new diseases. So, proper garbage management by using technology is a major concern of the hour. In the traditional existing system, the monitoring and disposal of garbage are done by humans and it is cumbersome, therefore supporting the traditional system with the flavor of IoT, then monitoring of garbage bins will be easy. This helps the people who are involved in the traditional garbage collection system and it will be the best feasible solution. The SGMDSS is a very innovative information management control system that helps the metros, cities, and villages hygiene and clean with a better garbage disposal. This system uses an advanced approach in which waste monitoring and disposal support are automated. SGMDSS monitors the garbage bins located at different locations and notifies about the level of garbage accumulated in the garbage bins through an android mobile application to the cleaning personnel for disposal and provides the shortest path to the garbage bin location that is almost filled. This information is also sent to the webpage and the entire data is stored and accessed through the cloud. Also, an alert message is sent to the worker. - Defense against frequency analysis in elliptic curve cryptography using k-means clustering
Bh Padma, D Chandravathi, Lanka Pratibha
Proceedings IEEE 2021 International Conference on Computing Communication and Intelligent Systems Icccis 2021, 2021
Elliptic Curve Cryptography (ECC) is a revolution in asymmetric key cryptography which is based on the hardness of discrete logarithms. ECC offers lightweight encryption as it presents equal security for smaller keys, and reduces processing overhead. But asymmetric schemes are vulnerable to several cryptographic attacks such as plaintext attacks, known cipher text attacks etc. Frequency analysis is a type of cipher text attack which is a passive traffic analysis scenario, where an opponent studies the frequency or occurrence of single letter or groups of letters in a cipher text to predict the plain text part. Block cipher modes are not used in asymmetric key encryption because encrypting many blocks with an asymmetric scheme is literally slow and CBC propagates transmission errors. Therefore, in this research we present a new approach to defence against frequency analysis in ECC using K-Means clustering to defence against Frequency Analysis. In this proposed methodology, security of ECC against frequency analysis is achieved by clustering the points of the curve and selecting different cluster for encoding a text each time it is encrypted. This technique destroys the regularities in the cipher text and thereby guards against cipher text attacks. - IoT Applications and Recent Advances
T. Mohana Naga Vamsi, Pratibha Lanka
Cyber Physical Iot and Autonomous Systems in Industry 4 0, 2021
During this current decade, the most familiar technology that is creating a standard in the world is the Internet of Things (IoT). The IoT is the inter-networking of physical things, mobile systems, buildings, and other objects which are embedded with electronic processors, sensors, actuators, and network connectivity devices that enable these things to transfer data between them via the Internet. The IoT is an interdisciplinary research domain that involves computer science and electronic engineering principles. The major research focus is on the design and development of hardware and software systems that will make IoT-enabled systems intelligent and secure. This is possible by incorporating artificial intelligence, machine learning, and Internet concepts while developing intelligent IoT systems. Since research in IoT applications is the amalgamation of embedded electronics, sensor technology, networking, cloud computing, artificial intelligence, data analytics, and other technologies, it will become a good area for interdisciplinary research for both academics and industrial people. At the outset there are a lot of new methodologies and findings that have evolved with this IoT research, and these outcomes give better solutions in the areas of smart environments, IoT services, intelligent systems, smart health, smart homes, smart agriculture, networking protocols, communication protocols, and autonomous industrial applications. Hence the research direction in IoT applications is definitely a challenge for academic researchers and industrial developers. The major challenges are in identification technology, choosing an IoT architecture technology, and establishing networks and communication technologies for the specific research problem. In this chapter the authors attempt to depict all such challenges by focusing on these IoT-enabled topics and throw light on these issues for new researchers who are entering this domain. - An Embedded System Design For Guiding Visually Impaired Personnel
T. M. N. Vamsi, G. Kalyan Chakravarthi, T. Pratibha
2019 International Conference on Recent Advances in Energy Efficient Computing and Communication Icraecc 2019, 2019
One of the most important senses in human life is vision. But there are few people who lack this ability of sensing things via vision and are considered as visually impaired people. These people face many challenges and hurdles when they are moving in unknown public places and are usually accompanied by a sighted person. In this paper, a simple and novel approach has been proposed and developed an automated guiding system for the visually impaired personnel. The whole idea of our project is to ensure that the visually impaired people travel independently with the help of a guiding system. The current paper focuses on developing an embedded system for guiding visually impaired which helps them travel without the support of others and is also comfortable to use. This proposed idea has been undertaken with the sense of responsibility towards the society and the sole aim remains to enhance the life of the underprivileged by making the wonders of technology accessible to them in their everyday life in order to promote a better living.
RECENT SCHOLAR PUBLICATIONS
- Unsupervised Detection of Groundwater Quality Anomalies via Autoencoder-LOF Ensemble
TMN Vamsi, S Choudarapu, P Lanka, G Kamal, D Indira, JS Kumar
2025 International Conference on Intelligent Computing, Information and … , 2025
2025 - Improving Text-Driven Image Synthesis: Diffusion Models for Photorealistic Outcomes
TMN VAMSI, JS KUMAR, ISS RAO, P LANKA
2024 - An IoT based smart water contamination monitoring system
MNV Thalatam, P Lanka, JS Kumar
2023 International Conference on Intelligent Systems for Communication, IoT … , 2023
2023
Citations: 17 - IoT Applications and Recent Advances
TMN Vamsi, P Lanka
Cyber-Physical, IoT, and Autonomous Systems in Industry 4.0, 199-220 , 2021
2021 - An IoT based smart garbage monitoring and disposal support system
TMN Vamsi, GK Chakravarthi, P Lanka, B Divakar
2021 5th International Conference on Computing Methodologies and … , 2021
2021
Citations: 23 - Defense against frequency analysis in elliptic curve cryptography using k-means clustering
B Padma, D Chandravathi, L Pratibha
2021 International Conference on Computing, Communication, and Intelligent … , 2021
2021
Citations: 2
MOST CITED SCHOLAR PUBLICATIONS
- An IoT based smart garbage monitoring and disposal support system
TMN Vamsi, GK Chakravarthi, P Lanka, B Divakar
2021 5th International Conference on Computing Methodologies and … , 2021
2021
Citations: 23 - An IoT based smart water contamination monitoring system
MNV Thalatam, P Lanka, JS Kumar
2023 International Conference on Intelligent Systems for Communication, IoT … , 2023
2023
Citations: 17 - Defense against frequency analysis in elliptic curve cryptography using k-means clustering
B Padma, D Chandravathi, L Pratibha
2021 International Conference on Computing, Communication, and Intelligent … , 2021
2021
Citations: 2 - Unsupervised Detection of Groundwater Quality Anomalies via Autoencoder-LOF Ensemble
TMN Vamsi, S Choudarapu, P Lanka, G Kamal, D Indira, JS Kumar
2025 International Conference on Intelligent Computing, Information and … , 2025
2025 - Improving Text-Driven Image Synthesis: Diffusion Models for Photorealistic Outcomes
TMN VAMSI, JS KUMAR, ISS RAO, P LANKA
2024 - IoT Applications and Recent Advances
TMN Vamsi, P Lanka
Cyber-Physical, IoT, and Autonomous Systems in Industry 4.0, 199-220 , 2021
2021
