MURAL - Maynooth University Research Archive Library



    Differential Privacy and the l1 Sensitivity of Positive Linear Observers


    McGlinchey, Aisling and Mason, Oliver (2017) Differential Privacy and the l1 Sensitivity of Positive Linear Observers. IFAC-PapersOnLine, 50 (1). pp. 3111-3116. ISSN 2405-8963

    [img]
    Preview
    Download (543kB) | Preview


    Share your research

    Twitter Facebook LinkedIn GooglePlus Email more...



    Add this article to your Mendeley library


    Abstract

    We consider the design of differentially private observers for positive linear systems in discrete time. In particular, we first provide a general bound for the l1 sensitivity of the map defined by a Luenberger observer for a linear time invariant (LTI) system and show that this can be tight as well as describing how it relates to previous work. We then define an optimisation problem for minimising this bound for positive linear observers and provide a characterisation of optimal solutions for systems with a single measured output. Finally we show how the addition of Laplacian noise can violate positivity, even for the optimally designed positive observer and discuss directions for future work.

    Item Type: Article
    Keywords: Positive systems; linear time-invariant systems; observers for linear systems; differential privacy; optimization;
    Academic Unit: Faculty of Science and Engineering > Mathematics and Statistics
    Faculty of Science and Engineering > Research Institutes > Hamilton Institute
    Item ID: 12006
    Identification Number: https://doi.org/10.1016/j.ifacol.2017.08.317
    Depositing User: Oliver Mason
    Date Deposited: 06 Dec 2019 12:06
    Journal or Publication Title: IFAC-PapersOnLine
    Publisher: Elsevier
    Refereed: Yes
    URI:
    Use Licence: This item is available under a Creative Commons Attribution Non Commercial Share Alike Licence (CC BY-NC-SA). Details of this licence are available here

    Repository Staff Only(login required)

    View Item Item control page

    Downloads

    Downloads per month over past year

    Origin of downloads