God datakommunikasjon mellom biler (V2V) vil i fremtiden kunne avverge farlige trafikksituasjoner og bedre flyten i tett trafikk.
Artikkelen er mer enn to år gammel, og kan inneholde utdatert informasjon.
Ali Chelli forsvarer sin doktoravhandling “Channel Modelling and System Performance Analysis of Vehicle-to-Vehicle Communication Systems” den 13. august 2013 ved Campus Grimstad. Chelli er tatt opp på doktorgradsprogrammet med spesialisering i IKT ved Fakultet for teknologi og realfag ved Universitetet i Agder.
I avhandlingen analyserer Ali Chelli modeller for hvordan bil-til-bil-kommunikasjon kan fungere i forskjellige trafikksituasjoner, fra tett bytrafikk til motorveier, og simulerer forskjellige måter å gjøre dette på.
Slik beskriver kandidaten selv essensen i avhandlingen:
Channel Modelling and System Performance Analysis of Vehicle-to-Vehicle Communication Systems
Vehicle-to-vehicle (V2V) communications are expected to play an important role in intelligent transportation systems and ad hoc networks. V2V communications have attracted a lot of interest due to their numerous benefits, such as the reduction of the number of road accidents and the improvement of the traffic flow. For the development of future V2V communications, the exact knowledge of the channel statistics is indispensable. The absence of appropriate channel models that capture the propagation conditions along street environments propelled us to derive novel V2V channel models for straight street and for street intersections.
In this dissertation, a geometry-based approach is utilized to derive V2V channel models for various street layouts under different propagation conditions. Stationary narrowband and wideband V2V channel models are developed and their statistical properties are derived. For each of the investigated channel models, a channel simulator is designed such that a good fitting between the statistics of the reference and the simulation model is obtained.
Stationary V2V channel models are well suited for propagation environments which are quasi-static. This is observed if the propagation environment contains few scattering objects and the vehicles move with moderate speed as it is the case in rural environment. Urban traffic congestions can be also assumed to be quasi-static due to the low speed of the vehicles during traffic congestions. However, for dynamic environments, such as highway and urban environments, significant changes in the scattering environment lead to time-variant channel statistics. This fact makes the investigated V2V channel non-stationary. Following this line of thought, non-stationary V2V channel models for straight street and street intersections are proposed.
Beside geometry-based approaches, measurement-based approaches constitute a major method to explore the fading behaviour of V2V channels. In practice, measurements are time and band limited. Under such conditions the classical time-averaging and frequency-averaging methods lead to estimation error. The accurate estimation of the channel temporal autocorrelation function (ACF) and the frequency correlation function (FCF) is important from a system design perspective. To improve the estimation of the temporal ACF and the FCF from time-limited and band-limited signals, respectively, novel estimation techniques are introduced.
In this thesis, we also explore the performance of two hybrid automatic repeat request (HARQ) schemes over V2V channels. The choice of the HARQ scheme in particular is motivated by its adaptive nature and its high spectral efficiency. These features make the HARQ scheme very attractive for V2V communications. Our study of the HARQ scheme revealed that it allows communicating at a rate close to the ergodic capacity after few retransmission rounds.
Kandidaten: Ali Chelli received the B.Sc degree in communications from Ecole Supérieure des Communications de Tunis (SUP'COM), in 2005. He received the M.Sc. degree in information and communication technology from University of Agder, Norway, in 2007. He received the best master project award for the Academic Session 2006-2007. From 2008 to 2012 he was with the University of Agder, Norway, as a PhD research Fellow under the supervision of Professor Matthias Pätzold. He is currently a Postdoctoral Fellow at King Abdullah University of Science and Technology (KAUST), Saudi Arabia.
Prøveforelesning og disputas finner sted i Auditorium A3 041, UiA Campus Grimstad. Instituttleder ved Institutt for IKT, professor Andreas Prinz, leder disputasen
Tid for prøveforelesning: 13. august 2013 kl. 10.00
Oppgitt emne for prøveforelesning: “Propagation mechanisms and current trends in channel measurement methodology”
Tid for disputas 13. august 2013 kl. 12.30
Tittel på avhandling: “Channel Modelling and System Performance Analysis of Vehicle-to-Vehicle Communication Systems”
Les avhandlingen i AURA - Agder University Research Archive, http://brage.bibsys.no/hia/ som er et digitalt arkiv for vitenskapelige artikler, avhandlinger og masteroppgaver fra ansatte og studenter ved Universitetet i Agder.
Førsteopponent: Førsteamanuensis Nils Torbjörn Ekman, NTNU
Annenopponent: Professor Fethi Choubani, High School of Communications, Department of Electronics, Physics and Propagation, Complex of Communications Technologies, Tunisia
Bedømmelseskomitéen har vært ledet av førsteamanuensis Geir Køien, Institutt for IKT, UiA
Veiledere: Professor Matthias Uwe Pätzold (Institutt for IKT) og professor Frank Reichert (Institutt For IKT)