In this study we compare different methods of computing the path loss (and therefore the signal strength) in a typical urban area for low-altitude UAS flights. The most accurate path loss estimates arise from ray tracing algorithms, which can take many hours to days just to cover a small part of a city up to an altitude of 400 feet. The fastest algorithms are statistical, which encode the signal strength as a single equation with appropriate parameters and can be solved within ten microseconds, but such models have accuracy and physical limits. Machine Learning algorithms can be applied in two different ways: first by training the ML to directly compute the signal using the ray tracing output as the true signal strength, and secondly by training the ML to predict the inaccuracy between the statistical models and the ray tracing algorithm, and thus provide a correction to the statistical models. We compared all four methods using ten neighborhoods across three different cities to determine the limits of each method.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Comparing Path Loss Prediction Methods for Low Altitude UAS Flights


    Beteiligte:
    Wieland, Frederick (Autor:in) / Drescher, Zach (Autor:in) / Houser, John (Autor:in)


    Erscheinungsdatum :

    03.10.2021


    Format / Umfang :

    2034958 byte





    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Altitude flights

    Henrickson, H.B. | Engineering Index Backfile | 1931


    Altitude flights

    Hendrickson, H.B. | Engineering Index Backfile | 1929


    Altitude record flights

    De Marolles, R.J. | Engineering Index Backfile | 1936


    Altitude Record Flights

    de Marolles, R.J. | Emerald Group Publishing | 1936


    Flight Path Planning for Small UAV Low Altitude Flights

    Szabolcsi Róbert | DOAJ | 2020

    Freier Zugriff