Faculty Publications

Permanent URI for this communityhttps://idr.nitk.ac.in/handle/123456789/18736

Publications by NITK Faculty

Browse

Has files: NoSubject: search.filters.subject.Convolutional Neural Network(CNN)

Search Results

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Blockchain Based Artificial Intelligence of Things (AIoT) for Wildlife Monitoring
    (Springer Science and Business Media Deutschland GmbH, 2024) Madhusudhan, R.; Pravisha, P.
    Climate change poses a significant threat to wild animals and their habitats, increasing the chance of human-wildlife conflict. Traditional camera-based imaging systems are centralized and require operators to install the camera and monitor the video recording constantly. However, manually processing the massive number of images and videos gathered from camera traps is expensive and time-consuming. In this article, we will develop a framework for wildlife monitoring systems that make use of Artificial Intelligence of Things (AIoT), the Interplanetary File System (IPFS), and blockchain. A wildlife camera that uses AIoT to detect wild animal movement in real-time gathers the dynamic properties of animals. Cloud computing solutions are impractical for critical data management in wildlife monitoring due to their high latency and constant internet connectivity requirements. IPFS is a distributed file system that offers efficient data storage, distribution, and persistence, enabling offline-centric paradigms. In our framework, IPFS is used for permanent data storage, and the hash value of data is stored on a private blockchain. The data from multiple forest zones is stored on a consortium blockchain. A simulation is carried out using CNN and a method to improve the scalability of the framework is presented. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.
  • No Thumbnail Available
    Item
    Deep Learning Approach for Wireless Signal and Modulation Classification
    (Institute of Electrical and Electronics Engineers Inc., 2021) B C, B.; Deshmukh, A.; Rupa, M.V.; Sirigina, R.P.; Vankayala, S.K.; Narasimhadhan, A.V.
    This paper aims to classify signal and modulation classes of a given wireless signal with high accuracy using a model having a low number of parameters. We propose an end-to-end method to classify a wireless signal based on its signal and modulation type using a CNN-based architecture. The proposed architecture is similar to that of the LeNet-5. Firstly, we implement signal and modulation classification using a decision tree, followed by a random forest algorithm, classic examples of machine learning(ML) based algorithms. Since our dataset is a time series, we also implement using RNN-LSTM based model for the classification. The proposed model has fewer parameters than that of the CNN-based, RNN-LSTM based architectures. Moreover, it achieves better accuracy for a wide range of signal-to-noise ratios than a decision tree, random forest, RNN-LSTM based classifiers. © 2021 IEEE.