Faculty Publications

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

Publications by NITK Faculty

Browse

Subject: search.filters.subject.Data aggregation

Search Results

Now showing 1 - 5 of 5
  • No Thumbnail Available
    Item
    Motion tracking and data aggregation in distributed sensor networks using mobile agents
    (2005) Shivakumar, S.; Vinaykumar, K.
    In this paper a distributed, energy aware and collaborative approach for motion tracking and data aggregation in distributed sensor networks is proposed based on mobile agents. Proposed approach is simulated using network simulator (NS2) and the results are compared with that of the client-server model approach based on energy consumed for tracking and data aggregation. Results show that mobile agent based performs very much better than that of client server approach.
  • No Thumbnail Available
    Item
    Data Aggregation of Tweets and Topic Modelling Based on the Twitter Dataset
    (Association for Computing Machinery, 2021) Srinivasan, V.; Chandrasekaran, K.
    Twitter is one of the most popular online social networks. It has a relatively simple data model and an intuitive API to access Twitter data. This makes it easy to collect social data and analyse the patterns of online behaviour. Twitter has an impactful presence among politicians, entrepreneurs, news agencies, public figures, and this makes it a crucial playground for social discussion. The topics discussed on Twitter often lead to or are the cause of social events. Therefore, a lot of information can be inferred from Twitter data. This can be used by NGOs, government agencies or policymakers to develop meaningful understanding and respond to the emerging trends. In this project, I will discuss a method to aggregate tweets related to Elon Musk and Tesla from Twitter servers using the Twitter API in the form of a web crawler. The data obtained from the web crawler will be combined with a ready-made dataset containing similar information, and the datasets will be merged together. After collecting relevant tweet information, I will perform topic modelling using Latent Dirichlet Allocation (LDA) on his tweets to find out the most common topics tweeted by Elon Musk. © 2021 ACM.
  • No Thumbnail Available
    Item
    Data Format Heterogeneity in IoT-Based Ambient Assisted Living: A Survey
    (Springer Science and Business Media Deutschland GmbH, 2023) Sandeep, M.; Khatri, S.; Chandavarkar, B.R.
    Ambient Assisted Living (AAL) has become a significant component of the lives of the elderly in the present decade, allowing them to live independently by assisting their daily activities with automation. Different sensors from various manufacturers with proprietary data formats to detect environmental changes and monitor a person’s health metrics. These data formats are the root cause of the data Heterogeneity issue in AAL and, in turn, contribute to data interoperability challenges. In this paper, we have presented a survey on currently available state-of-the-art solutions to address data heterogeneity challenges in AAL and made a comparative study of suggested methods to overcome the data interoperability. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
  • No Thumbnail Available
    Item
    Data aggregation using compressive sensing for improved network lifetime in large scale wireless sensor networks
    (Serials Publications serialspublications@vsnl.net, 2016) Puneeth, D.; Ruthwik, R.; Kulkarni, M.
    In large scale Wireless Sensor Networks(WSN's) the amount of data generated is enormous. The data has to be processed efficiently before it reaches the Base Station (BS) by using an efficient routing algorithm as well as data aggregation methods. The nodes in WSN's are randomly deployed, the data emerging from these nodes are highly correlated either spatially or temporally. The data aggregation scheme should employ simple encoding since the sensor nodes are battery operated. The proposed method discusses about a data aggregation scheme using Compressive Sensing(CS) technique which makes use of correlation among the sensor nodes. Our primary focus is to increase the lifetime of the overall network. The underlying protocols used are Low-energy adaptive clustering hierarchy (LEACH) and Multi-threshold adaptive range clustering (M-TRAC). We have computed several network parameters for different network configuration. The reconstruction algorithm is sufficiently robust against noise. The reconstruction of the data is done using greedy method and L1 norm regularization. The implementation of the algorithm is done using the real data-set from Intel Lab. Simulation results validate the data aggregation scheme guarantees data accuracy and doubles the network lifetime. © 2016 International Science Press.
  • No Thumbnail Available
    Item
    Trust Based Data Gathering in Wireless Sensor Network
    (Springer New York LLC barbara.b.bertram@gsk.com, 2019) Karthik, N.; Ananthanarayana, V.S.
    Wireless sensor nodes have been successfully employed in various pervasive applications. In all pervasive applications, a gathering of sensor data from the environment is the main operation held in a sensor network, where sink node or base station gathers all generated data to do data analysis and decision making. The data generated by the sensor node in the pervasive environment should be transmitted to the sink node for data analysis and decision making. We strongly conceive that each process from perceiving the environment to decision making, demands trust based process to ease and ensure the trustworthy data exchange among trustworthy nodes such as trust-based data collection, trust-based data aggregation, trust-based data reconstruction and trust-based data analysis for decision making. In this work, we propose a Trust-based Data Gathering which focus on trust-based data collection, data aggregation, and data reconstruction to show that the absence of trust in a sensor-driven pervasive environment could affect the normal functionality of an application. Experimental results show that the proposed method achieves better performance in detecting data faults, malicious nodes and demonstrates that the absence of trust based process in data collection, data aggregation, and data reconstruction in harsh environment consumes more energy and delay for handling untrustworthy data, untrustworthy node and affects the normal functionality of the application. © 2019, Springer Science+Business Media, LLC, part of Springer Nature.