Conference Papers

Permanent URI for this collectionhttps://idr.nitk.ac.in/handle/123456789/28506

Browse

Filters

2017 - 201932020 - 20232

Settings

Author: search.filters.author.Jain, V.Has files: No

Search Results

Now showing 1 - 5 of 5
  • No Thumbnail Available
    Item
    Profile Matching of Online Social Network with Aadhaar Unique Identification Number
    (Institute of Electrical and Electronics Engineers Inc., 2017) Gautam, B.; Jain, V.; Jain, S.; Annappa, B.
    Matching user's profile over multiple Online Social Network (OSN) brings many new insights. Considering the fact of user's disambiguation over different OSN, Existing user profile matching are highly computationally expensive. In this paper, we propose a novel e-identity Architecture using Infrastructure as a service (IaaS) of Cloud to reduce the computational cost of the profile matching algorithm. Profile equivalent through proposed architecture are based on the public attributes available in overall social network and lastly, profiles are mapped to the Aadhaar Unique Identification Number (UID). © 2016 IEEE.
  • No Thumbnail Available
    Item
    Proportional rate reduction for ns-3 TCP
    (Association for Computing Machinery acmhelp@acm.org, 2018) Mittal, V.; Jain, V.; Tahiliani, M.P.
    The implementation of Transmission Control Protocol (TCP) in the Linux kernel has kept pace with its ongoing research. However, it is not the case with the implementations of TCP in network simulators. This limitation has been a major hindrance in thoroughly evaluating the performance of new TCP extensions, since carrying out real-time experimental evaluations is a non-trivial task. Recently, there have been significant efforts to align the TCP implementation in ns-3 to that of Linux. Nonetheless, there still exist several features that the implementation of TCP in ns-3 lacks. In this work, the TCP implementation in ns-3 has been extended to support Proportional Rate Reduction (PRR) algorithm. PRR is a de facto loss recovery algorithm used since Linux kernel 3.2. This paper describes the implementation and validation of PRR algorithm in ns-3 and also highlights the need to modularize the implementation of loss recovery algorithms in ns-3. © 2018 Copyright held by the owner/author(s).
  • No Thumbnail Available
    Item
    Design and implementation of TCP BBR in ns-3
    (Association for Computing Machinery acmhelp@acm.org, 2018) Jain, V.; Mittal, V.; Tahiliani, M.P.
    Bottleneck Bandwidth and Round-trip propagation time (BBR) is a congestion based congestion control algorithm recently proposed by Google. Although it can be deployed with any transport protocol that supports data delivery acknowledgement, BBR is presently implemented alongside TCP (known as TCP BBR) in Linux kernel since 4.9 and is the default congestion control used in Google Cloud Platform. However, to the best of our knowledge, TCP BBR is not yet supported in popular network simulators such as ns-3. This limitation is a major hindrance in thoroughly studying the benefits of TCP BBR since carrying out large-scale and real-time experimental evaluations is a non-trivial task. In this paper, we discuss the design and implementation of a new model for TCP BBR in ns-3. We validate the proposed model by performing different sets of simulations to ensure that the model in ns-3 exhibits key characteristics of TCP BBR. © 2018 Copyright held by the owner/author(s).
  • No Thumbnail Available
    Item
    Data Center TCP in ns-3: Implementation, Validation and Evaluation
    (Association for Computing Machinery, 2020) Jain, V.; Henderson, T.R.; Shravya, K.S.; Tahiliani, M.P.
    Data Center TCP (DCTCP) is a standard congestion control scheme used to provide high burst tolerance, low latency and high throughput in Data Center Networks (DCNs). It uses in-network feedback obtained through Explicit Congestion Notification (ECN) to scale the congestion window (cwnd). Recently, there have been significant efforts to extend DCTCP to provide low latency transport (e.g., TCP Prague) in a Wide Area Network, but with congestion control and in-network queuing mechanisms that are backward compatible with legacy implementations. This paper presents the implementation and validation of a new ns-3 model for DCTCP which can be used to implement its extensions, such as TCP Prague. The proposed model is verified and validated by comparing it to the Linux model of DCTCP by using ns-3 Direct Code Execution (DCE), a framework to run ns-3 simulations using Linux network stack. The results obtained from this comparison show that the ns-3 model and Linux model of DCTCP exhibit similar characteristics. © 2020 ACM.
  • No Thumbnail Available
    Item
    Optimizing Super-Resolution Generative Adversarial Networks
    (Springer Science and Business Media Deutschland GmbH, 2023) Jain, V.; Annappa, B.; Dodia, S.
    Image super-resolution is an ill-posed problem because many possible high-resolution solutions exist for a single low resolution (LR) image. There are traditional methods to solve this problem, they are fast and straightforward, but they fail when the scale factor is high or there is noise in the data. With the development of machine learning algorithms, their application in this field is studied, and they perform better than traditional methods. Many Convolutional Neural Networks (CNNs) and Generative Adversarial Networks (GANs) have been developed for this problem. The Super-Resolution Generative Adversarial Networks (SRGAN) have proved to be significant in this area. Although the SRGAN produces good results with 4 upscaling, it has some shortcomings. This paper proposes an improved version of SRGAN with reduced computational complexity and training time. The proposed model achieved an PPSNR of 29.72 and SSIM value of 0.86. The proposed work outperforms most of the recently developed systems. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.