Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Samineni Peddakrishna has a total of 40 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Hierarchical Traffic State Fusion for Infrastructure Assisted Vehicle Speed Regulation of Autonomous Electric Vehicles

    by Zuber Basha Shaik, Samineni Peddakrishna, Sreedhar Kollem

    Infrastructure-assisted autonomous driving has emerged as a promising approach for enhancing the situational awareness and operational efficiency of Autonomous Electric Vehicles (AEVs) in structured environments. However, existing studies often address traffic perception, wireless communication, and vehicle control as separate components, with limited emphasis on an integrated perception–communication–control framework for traffic-aware speed regulation. This paper presents a Hierarchical Traffic State Fusion Vehicle-to-Infrastructure (HTSF-V2I) framework that integrates multi-camera roadside traffic perception, edge artificial intelligence, long-range LoRa communication, onboard LiDAR sensing, and traffic-state-dependent vehicle speed regulation using a lightweight embedded architecture. Three roadside cameras equipped with an optimized lightweight YOLOv8-based detector monitor campus traffic and estimate local traffic states. A deterministic HTSF module combines the multi-camera traffic states into a unified infrastructure-level traffic advisory, which is transmitted to the AEV through LoRa-based V2I communication. Onboard LiDAR provides local obstacle verification and overrides the infrastructure-derived propulsion command when an obstacle is detected within the defined safety threshold. For smart-campus evaluation, a dataset comprising 5,914 manually annotated traffic images was developed for training and evaluating the traffic perception module. The standard YOLOv8n baseline achieved an mAP@50 of 88.3% at an input resolution of $$640\times 640$$ , whereas the deployed $$320\times 320$$ CampusLite configuration achieved an mAP@50 of 70.3%, providing a practical trade-off between detection accuracy and computational efficiency for embedded deployment. Experimental evaluation demonstrates multi-camera traffic perception, infrastructure-level traffic-state estimation, end-to-end LoRa-based traffic-advisory transmission, and vehicle propulsion-command selection using infrastructure-derived traffic information with local obstacle supervision. The results demonstrate the feasibility of integrating infrastructure perception, deterministic traffic-state fusion, wireless communication, and onboard safety intelligence within a lightweight framework for low-speed AEV operation in smart-campus environments. The proposed HTSF-V2I framework provides a computationally practical approach for infrastructure-assisted, traffic-aware speed regulation on resource-constrained embedded platforms.

  • RNN-GRU hybrid deep learning framework for heart disease prediction

    by Nadikatla Chandrasekhar, Peddakrishna Samineni, Sreedhar Kollem

    This research presents a novel hybrid deep learning approach for heart disease prediction, incorporating a Recurrent Neural Network (RNN) with multiple Gated Recurrent Units (GRU) and the Adam optimizer. The performance of the proposed model is evaluated on the IEEE Dataport heart disease dataset, achieving an accuracy of 92%. The primary focus of this research is to investigate the impact of data pre-processing, particularly outlier detection, on the accuracy of the hybrid deep learning model. The comparative analysis demonstrates that without outlier detection, the model achieves an accuracy of 89%, while incorporating outlier detection results in an increase in accuracy of 3%, reaching 92%. The study involves implementing a hybrid RNN-GRU model on a pre-processed dataset. Research tracks the accuracy and loss for each epoch throughout the training and testing phases. The results highlight the usefulness of the model in predicting heart disease and stress the importance of data pre-processing in improving its performance.

  • Open Access

    TurboCBAM-Integrated YOLOv8n Framework for Real-Time Traffic Sign Detection in Embedded Autonomous Vehicles

    by Thotapalli Vinayakumar, Samineni Peddakrishna

    Autonomous vehicles require fast and accurate perception under varying environmental conditions, but most high-performance detectors are too heavy for edge devices. A lightweight yet robust model is needed to ensure safety and real-time operation in practical scenarios. This work proposes to develop a lightweight traffic-sign detection system suitable for real-time resource constrained embedded platforms. The objective of this study is to enhance the YOLOv8n architecture through an effective attention mechanism and validate its embedded performance on a real autonomous vehicle platform. To achieve this, a custom traffic-sign dataset was created and used to train an improved YOLOv8n model integrated with the proposed Turbo Convolutional Block Attention Module (TurboCBAM), and its robustness was further evaluated across diverse datasets. The TurboCBAM model achieved an mAP@50 of 0.977, with inference time increasing from 5.5 ms in YOLOv8n to 6.2 ms. During real-time embedded testing, it maintained a stable performance of 10 to 12 frames per second (FPS). To validate its robustness and generalization capability, the model was tested on external datasets such as TT100K and CCTSDB2021. It achieved mAP@50 scores of 0.860 and 0.784, demonstrating reasonable performance under domain shifts, with some degradation due to dataset differences. These cross-dataset results show that the model maintains reliable detection across diverse environments and performs effectively beyond the conditions seen during custom dataset training.

  • Open Access

    Phase-Controlled Bidirectional Circularly Polarized Dual 4-Port SIW MIMO Antenna with Enhanced Isolation for Sub-6 GHz Vehicular Communications

    by Kamepalli Dharani, M. Sujatha, Samineni Peddakrishna, Jayendra Kumar

    This paper presents a dual four-port circularly polarized (CP) MIMO antenna based on substrate integrated waveguide (SIW) technology for sub-6 GHz applications. The design consists of two identical four-port SIW-based CP-MIMO antennas arranged in a mirror-symmetric configuration with an air gap of 15 mm. Each antenna employs four symmetrically arranged cross-shaped SIW patches excited by coaxial probes. Bidirectional radiation is achieved by applying a 180° phase difference between corresponding ports of the mirror symmetric configuration, referred to as the Backward-Radiating Unit (BRU) and the Forward-Radiating Unit (FRU). The bidirectional radiation mechanism is supported by array-factor-based theoretical modelling, which explains the constructive and destructive interference under phase-controlled excitation. To ensure high isolation and stable polarization performance, the antenna design incorporates defected ground structures, inter-element decoupling strips, and vertical metallic vias. Simulations indicate an operating band from 5.1 to 5.4 GHz. Measurements show a −10 dB bandwidth from 5.25 to 5.55 GHz, with the frequency shift attributed to fabrication tolerances and measurement uncertainties. The antenna achieves inter-port isolation better than −15 dB. A 3 dB axial-ratio bandwidth is maintained across the operating band. Measured axial-ratio values remain below 3 dB from 5.25 to 5.55 GHz, while simulations predict a corresponding range from 5.1 to 5.4 GHz. The proposed configuration achieves a peak gain exceeding 4 dBi and maintains an envelope correlation coefficient below 0.05. These results confirm its suitability for CP-MIMO systems with controlled spatial coverage. With a physical size of 0.733λ0 × 0.733λ0 per array, the proposed antenna is well-suited for vehicular and space-constrained wireless systems requiring bidirectional CP-MIMO coverage.


  • Open Access

    RFID Antenna Deployment Model and Anti-Collision Algorithms to Enhance Vehicle Toll System Efficiency

    by Krishna Mazumder, Anumoy Ghosh, Priyanka Singh, Samineni Peddakrishna, Jayendra Kumar

    This paper proposes a standard applications of Radio Frequency Identification (RFID) system to manage vehicles at electronic toll gate system without compromising the vehicle speed. The existing electronic toll gate system has low latency as each car needs to wait in a queue for passing the gate. Installation of reader antenna at each lane makes the system expensive. We propose a two-ray ground model for using minimal reader antenna to define required antenna height and downtilt angle to cover the total gate area. We estimate that reader antenna at a height of 15 m and 45.4° downtilit angle are required for optimal signal reception from the car at a distance of 35–38 m. Modified Pulse Protocol algorithm is suggested for mitigating reader collision and increase throughput and utilization of the system by 14.71% and 15.2% respectively. The probability distribution function of the contention window is modified to triangular function. The M-ary Trimming Q-tree protocol is suggested to resolve the tag collision by trimming the idle nodes from the query tree, which reduces identification time. The total time slot decreases with higher elements, and the total throughput also increases to 98%. The “no-lane” strategy established a fast, efficient and cost effective vehicle toll collection system.


  • Open Access

    Heart abnormality classification using ECG and PCG recordings with novel PJM-DJRNN

    by Nadikatla Chandrasekhar, Sujatha Canavoy Narahari, Sreedhar Kollem, Samineni Peddakrishna, Archana Penchala, Babji Prasad Chapa


  • Heartfelt AI: Analyzing Emotions Through ECG Signals for Affective Computing

    by Nadikatla Chandrasekhar, Modugu Krishnaiah, Samineni Peddakrishna, Sai Prajith Kancharla, Nikhil Gummadavelly, Uday Nuvvula