Implementation of spatially consistent channel models for real-time full stack C-ITS V2X simulations

Forskningsoutput: Kapitel i bok/rapport/Conference proceedingKonferenspaper i proceedingPeer review

Sammanfattning

Vehicle-to-everything (V2X) wireless communication is expected to play a key role in enabling complex Advanced Driving Assistance Applications, including the support of emerging autonomous vehicles. The development of V2X applications, especially safety-related, requires rigorous verification. Field trials, though, have to be supplemented with less expensive and more repeatable methods. Thorough V2X modeling could be an efficient approach to boost V2X development. However, to be helpful, V2X modeling has to be realistic and properly capture the signal propagation in a complex V2X environment. To this end, Geometry Based Stochastic Models (GSCM) have been derived to account for the influence of the environment as well as small yet critical details, where we typically lack information about their physical and geometrical properties. In this paper, we present an implementation of a GSCM in the Unity3D game engine. Due to extensive signal path search in a 3D environment with ray-tracing, the GSCM computation has to be further optimized to enable real-time V2X simulations. Thus, we also propose the approach for GSCM computation optimization and explain the practicalities of its implementation. The results show that optimization can accelerate the simulation framework up to a few thousand times without losing the precision in modeling channel characteristics.

Originalspråkengelska
Titel på värdpublikation55th Asilomar Conference on Signals, Systems and Computers, ACSSC 2021
RedaktörerMichael B. Matthews
FörlagIEEE Computer Society
Sidor67-71
Antal sidor5
ISBN (elektroniskt)9781665458283
DOI
StatusPublished - 2021
Evenemang55th Asilomar Conference on Signals, Systems and Computers, ACSSC 2021 - Virtual, Pacific Grove, USA
Varaktighet: 2021 okt. 312021 nov. 3

Publikationsserier

NamnConference Record - Asilomar Conference on Signals, Systems and Computers
Volym2021-October
ISSN (tryckt)1058-6393

Konferens

Konferens55th Asilomar Conference on Signals, Systems and Computers, ACSSC 2021
Land/TerritoriumUSA
OrtVirtual, Pacific Grove
Period2021/10/312021/11/03

Ämnesklassifikation (UKÄ)

  • Datavetenskap (datalogi)

Fingeravtryck

Utforska forskningsämnen för ”Implementation of spatially consistent channel models for real-time full stack C-ITS V2X simulations”. Tillsammans bildar de ett unikt fingeravtryck.

Citera det här