Sahaya Beni Prathiba

Dr. Sahaya Beni Prathiba

Vellore Institution of Technology, Chennai, India
orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
10 August 2021 in IEEE Transactions on Network Science and Engineering
Verified
Journal Article
Closed access
SPSahaya Beni Prathiba
Sahaya Beni Prathiba
GRGunasekaran Raja
Gunasekaran Raja
SASudha Anbalagan
Sudha Anbalagan
KDKapal Dev
Kapal Dev
SGSugeerthi Gurumoorthy
Sugeerthi Gurumoorthy
ASAtshaya P. Sankaran
Atshaya P. Sankaran
Humankind's urbanization and luxuriousneed increase the number of vehicles day by day. Alongside the advancement of technologies, autonomous vehicles have now come into a greater existence. Due to the resource constraints for obtaining a high processing rate in performing the intelligent tasks, Vehicular Edge Computing (VEC) has been introduced. However, a delay occurs in transferring the information due to the traits of automotive tasks. In this paper, a Federated Learning empowered computation Offloading and Resource management (FLOR) is proposed, which offloads and manages the resources with low latency. The FLOR framework operates over heterogeneous networks that include Vehicle-to-Everything (V2X) communication, Dedicated Short Range Communication (DSRC), 5G millimeter-wave (5G-mmWave), and 6G Vehicle-to-Vehicle (6G-V2V) communication. FLOR employs the stochastic network calculus mechanism for computing the upper bound delay of the heterogeneous communication and obtains the probability of offloading mechanism. Upon the detected probability value, the FLOR offloads the computation task to the optimal network. Further, to avoid the increase in resource cost and utilize the available resources efficiently, an effective Radio Resource Management (e-RRM) based on Federated Q-Learning is proposed in FLOR, for allocating and using the available resources optimally. The simulation result shows that the proposed FLOR framework effectively offloads the computation between the available heterogeneous networks with optimal resource allocation by 20.69% than the existing solutions.

26 October 2021 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
SPSahaya Beni Prathiba
Sahaya Beni Prathiba
GRGunasekaran Raja
Gunasekaran Raja
KDKapal Dev
Kapal Dev
Neeraj KumarNeeraj Kumar
Neeraj Kumar
Mohsen GuizaniMohsen Guizani
Mohsen Guizani
The development of Autonomous Vehicles (AVs) envisions the promising technology of future Intelligent Transportation Systems (ITS). However, the complex road structures and increased vehicles cause traffic congestion and road safety, which eventually leads to horrible accidents. Cooperative driving of AVs, a groundbreaking initiative of vehicle platooning, epitomizes the next wave in vehicular technology through minimizing accident risks, transport times, costs, energy, and fuel consumption. However, the traditional machine learning-based platooning approaches fail to regulate the policy with the dynamic feature of AVs. This paper proposes a hybrid Deep Reinforcement learning and Genetic algorithm for Smart-Platooning (DRG-SP) the AVs. The leverage of the deep reinforcement learning mechanism addresses the computational complexity and accommodates the high dynamic platoon environments. Adopting the Genetic Algorithm in Deep Reinforcement learning overcomes the slow convergence problem and offers long-term performance. The simulation results reveal that the Smart-Platooning effectively forms and maintains the platoons by minimizing traffic congestion and fuel consumption.

4 July 2022 in IEEE Transactions on Network Science and Engineering
Verified
Journal Article
Closed access
SPSahaya Beni Prathiba
Sahaya Beni Prathiba
GRGunasekaran Raja
Gunasekaran Raja
SASudha Anbalagan
Sudha Anbalagan
ASArikumar K S
Arikumar K S
SGSugeerthi Gurumoorthy
Sugeerthi Gurumoorthy
KDKapal Dev
Kapal Dev
Autonomous Vehicles (AVs) exchange real-time and seamless data between other AVs and the network, thus revolutionizing the Intelligent Transportation System (ITS). Automated transportation brings numerous benefits to human beings. However, the concerns such as safety, security, and privacy keep rising. In navigation and trajectory planning, the AVs require exchanging sensory information from their own and other AVs. In such cases, when a malicious AV or faulty sensor-equipped AV comes into connectivity, it can have disruptive consequences. This paper proposes a Hybrid Deep Anomaly Detection (HDAD) approach for effective anomaly detection and cyber-attack mitigation in AVs. The Multi-Agent Reinforcement Learning (MARL) algorithm in HDAD approach acts over the 6G network to combat new-age cyber-attacks and provide a swift and accurate anomaly detection mechanism. In conjunction with Maximum Entropy Inverse Reinforcement Learning (MaxEntIRL), the HDAD approach identifies and isolates malicious AVs. It is envisioned that the obtained results prove the effectiveness of HDAD and have an 8.2% higher accuracy rate than the existing systems.

11 November 2021 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
SPSahaya Beni Prathiba
Sahaya Beni Prathiba
GRGunasekaran Raja
Gunasekaran Raja
Neeraj KumarNeeraj Kumar
Neeraj Kumar
The rapid growth in Autonomous Vehicle (AV) technology endeavors increased attention towards road safety in recent days. Particularly, a higher number of road accidents occurs when the AV tries to overtake or change the lane. To cut down the number of accidents and improve traffic reliability, the AV should be capable of making intelligent decisions and communicating those to other AVs. Therefore, in this paper, a Cooperative Collision avoidance scheme for AVs at Overtaking and Lane Changing maneuver (CCAV-OLC) is proposed. The Inverse Reinforcement Learning (IRL) in the CCAV-OLC scheme, processes on the given number of expert demonstrations for automatically acquiring the reward function, and thereby imitating actual human driving strategy and decisions. However, the adaptability of IRL to a high-dimensional AV environment restricts the performance of the CCAV-OLC scheme. To overcome this, the IRL in CCAV-OLC leverages the Gaussian Process (GP) regression model (IRL-GP), which enables data-efficient Bayesian prediction even when the number of demonstrations is very low. After taking intelligent decisions in overtaking and lane changing maneuvers, the AVs cooperatively communicate and exchange the decisions with each other by 6th Generation Vehicle-to-Everything (6G-V2X) communications, which further improves the accuracy and lessens the time taken for making optimal decisions. The experimental results show that the AVs clone the expert's optimal driving strategy and avoid the collisions to a greater extent.

9 December 2021 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
SPSahaya Beni Prathiba
Sahaya Beni Prathiba
GRGunasekaran Raja
Gunasekaran Raja
SASudha Anbalagan
Sudha Anbalagan
SGSugeerthi Gurumoorthy
Sugeerthi Gurumoorthy
NKNeeraj Kumar
Neeraj Kumar
Mohsen GuizaniMohsen Guizani
Mohsen Guizani
The rapid growth of Autonomous Vehicle (AV) technology and the integration of edge computing grasp new challenges along with the ever-increasing mobile internet traffic and services. Tackling such challenges through customized edge computing services is the critical research in 6G Vehicle-to-Everything (6G-V2X) communication. V2X contributes detailed information about the current navigation of vehicles, automatic payments for toll roads, parking fees and other services. With the countless, unique, and personalized service requirements of AVs over computation-intensive applications, exploring the edge resources for the excellent Quality of Service (QoS) provision is the greatest concern. This paper proposes a Federated Learning and edge Cache-assisted Cybertwin (FLCC) framework for personalized service provision in 6G-V2X. Integration of cybertwin in 6G enables the connectivity of the physical system to the digital realm, allowing for adequate instantaneous wireless access. The FLCC jointly considers the edge cooperation and optimizations through the proposed Federated Multi-agent Deep Reinforcement Learning based (FM-DRL) algorithm. The FM-DRL algorithm balances the FLCC’s learning accuracy. It minimizes the time and cost by taking the factors such as cybertwin association, training data batch size, and bandwidth. Finally, caching is performed using the Federated Reinforcement Learning-based Edge Caching (FREC) algorithm to obtain the desired datasets required that train the model for providing personalized 6G-V2X services for the AVs. Numerical studies and simulation results reveal that the proposed system outperforms the baseline learning approaches by 17.6%.

16 December 2022 in Electronics
Verified
Journal Article
Open access
PDF file
ASArikumar K S
Arikumar K S
AKA. Deepak Kumar
A. Deepak Kumar
Thippa Reddy GadekalluThippa Reddy Gadekallu
Thippa Reddy Gadekallu
SPSahaya Beni Prathiba
Sahaya Beni Prathiba
TMTamilarasi Kathirvel Kathirvel Murugan
Tamilarasi Kathirvel Kathirvel Murugan
The rapid development of Autonomous Vehicles (AVs) increases the requirement for the accurate prediction of objects in the vicinity to guarantee safer journeys. For effectively predicting objects, sensors such as Three-Dimensional Light Detection and Ranging (3D LiDAR) and cameras can be used. The 3D LiDAR sensor captures the 3D shape of the object and produces point cloud data that describes the geometrical structure of the object. The LiDAR-only detectors may be subject to false detection or even non-detection over objects located at high distances. The camera sensor captures RGB images with sufficient attributes that describe the distinct identification of the object. The high-resolution images produced by the camera sensor benefit the precise classification of the objects. However, hindrances such as the absence of depth information from the images, unstructured point clouds, and cross modalities affect assertion and boil down the environmental perception. To this end, this paper proposes an object detection mechanism that fuses the data received from the camera sensor and the 3D LiDAR sensor (OD-C3DL). The 3D LiDAR sensor obtains point clouds of the object such as distance, position, and geometric shape. The OD-C3DL employs Convolutional Neural Networks (CNN) for further processing point clouds obtained from the 3D LiDAR sensor and the camera sensor to recognize the objects effectively. The point cloud of the LiDAR is enhanced and fused with the image space on the Regions of Interest (ROI) for easy recognition of the objects. The evaluation results show that the OD-C3DL can provide an average of 89 real-time objects for a frame and reduces the extraction time by a recall rate of 94%. The average processing time is 65ms, which makes the OD-C3DL model incredibly suitable for the AVs perception. Furthermore, OD-C3DL provides mean accuracy for identifying automobiles and pedestrians at a moderate degree of difficulty is higher than that of the previous models at 79.13% and 88.76%.