Graduate School of Engineering and Science

Permanent URI for this collectionhttps://hdl.handle.net/10679/9877

Browse

Browsing by Author "Eldeeb, Hossien Badr Hossien"

Filter results by typing the first few letters
Now showing 1 - 1 of 1
  • Placeholder
    PhD DissertationPublicationMetadata only
    Channel modeling for vehicular visible light communication
    Eldeeb, Hossien Badr Hossien; Uysal, Murat; Uysal, Murat; Demiroğlu, Cenk; Durak, Kadir; Erküçük, S.; Başar, E.; Department of Electrical and Electronics Engineering; Eldeeb, Hossien Badr Hossien
    The demand for vehicular communication systems has increased since they are considered the key enabling technology for Intelligent Transportation Systems (ITSs). Vehicular communications allow information sharing between the vehicles and the infrastructures, which have great potential in improving road safety, traffic flow, and passenger comfort along the road. The resulting vehicular connectivity forms are Vehicle-to-Vehicle (V2V), Vehicle-to-Infrastructure (V2I), Infrastructure-to-Vehicle (I2V), and Vehicle-to-Pedestrian (V2P), commonly referred to as Vehicle-to-Everything (V2X) communication. Most of the research efforts and standardization activities on V2X communication have focused on Radio Frequency (RF) technologies. However, limited RF bands allocated for V2X networks can suffer high interference levels in heavy traffic, and channel congestion might be particularly problematic for delay-sensitive safety functionalities. To address such issues, Visible Light Communication (VLC) has been proposed as an alternative or complementary vehicular access solution to RF-based V2X communications. VLC is based on the principle of modulating the intensity of the Light-Emitting-Diode (LED) and enables the dual use of LED for both illumination and communication purposes. The ubiquitous availability of LED-based streetlights, traffic lights, and automotive exterior lighting positions VLC as a potential wireless connectivity solution for vehicular networks. Vehicular VLC has received increasing attention recently in several aspects such as physical layer design, upper layer network protocols, and integration with RF-based solutions for hybrid systems. As in any other communication system, channel modeling plays a critical role in VLC system design and optimization. Earlier results in the literature have focused on indoor channel modeling. Those results do not apply to vehicular VLC systems since they exhibit inherently different characteristics to indoor counterparts. For example, the ideal Lambertian model, used for indoor LED luminaries, fails to match the illumination characteristics of automotive Headlights (HLs), Taillights (TLs), traffic lights, and streetlights. In addition, the effects of road reflectance, road type, weather conditions, the orientation of the user/vehicle equipment and infrastructures, receiver aperture size, and the sunlight might strongly affect the link performance of vehicular VLC systems. Despite the increasing attention on indoor VLC channel models, there is a lack of realistic vehicular VLC channel models. Motivated by these, we provide in this dissertation a comprehensive channel modeling study for different vehicular VLC links. In the first part of this dissertation, we explain our channel modeling approach, which builds upon advanced non-sequential ray tracing features of OpticStudio software. This approach allows the integration of any realistic light source radiation pattern. It can handle a large number of reflections for better accuracy. Since this is a simulation-based channel modeling methodology, it must be validated, at least once, against the measurements in a real scenario. Therefore, we carry out measurements and simulations for the same cases to validate our channel modeling approach considering both the Line of Sight (LoS) and the Non-Line of Sight (NLoS) cases. Such a realistic channel modeling approach is then adopted to model the V2V VLC channel when two vehicles travel in the same lane and by utilizing the Headlights (HLs) of the source vehicle as wireless transmitters. The most recent proposed channel path loss model was a linear function of transmission distance applicable only for ranges less than 20 meters. Therefore, we propose a new path loss expression that takes the form of a negative exponential function and provides an excellent match to simulation results for large transmission ranges and under different weather conditions. This expression is then utilized to derive the achievable transmission distance for a targeted data rate while satisfying a given value of Bit Error Rate (BER). We then consider V2V based Taillights (TLs) and derive a new path loss model that works for measured TL radiation patterns of different commercial car models. Utilizing the derived path loss model, we further derive a closed-form expression for the maximum transmission distance under the target BER value. Furthermore, the Root-Mean-Square (RMS) delay spread is investigated by considering different V2V density scenarios. In the above points, the effect of both the lateral shift between the two vehicles, the exact geometry of the vehicle's HLs, and the receiver aperture on the path loss model of the V2V system is excluded. Therefore, to address such shortcoming, we develop a closed-form path loss expression for the V2V VLC system as a function of link distance, lateral shift between the two vehicles, weather type, transmitter beam divergence angle, and receiver aperture diameter. While several vehicular VLC efforts investigated the V2V link channels while only a few attentions were paid to V2I/I2V channels, and by considering very idealistic assumptions. Motivated by this, we consider the I2V VLC links where the traffic lights and streetlights are deployed as wireless transmitters. First, we model the I2V based on a commercial traffic light and derive a closed-form expression of the path loss model as a function of both longitudinal and lateral shift distances. Then, we model the I2V system with street light transmitters and derive a closed-form expression for channel path loss as a function of pole spacing, the height of both the lighting pole and the vehicle, the longitudinal and lateral distance between the vehicle and the pole, and the aperture size of the Photodetector (PD). The effect of these transceivers and infrastructure parameters on the system average error rate performance is also investigated considering the mobility of vehicular communication, which makes the path loss no longer deterministic. We further model the reverse channel link, i.e., the V2I system, where the vehicle communicates with an infrastructure pole deploying its HLs as transmitters, and three PDs located within the traffic pole to act as receivers. Based on the Channel Impulse Responses (CIRs), obtained from the ray tracing, we introduce an expression for the achievable capacity considering the effect of propagation environment and the LED non-linear characteristics. In most V2V and I2V VLC works, the most common underlying assumption is using one or two PDs. That might be sufficient for establishing a connection between two vehicles cruising in the same straight lane or between the vehicles and infrastructure with clear LoS. To position VLC as a strong candidate for vehicular connectivity, it is essential to realize multi-directional reception in various deployment scenarios supporting V2V and I2V links. To address this of practical relevance, we investigate the channel modeling of multi-directional coverage for vehicular VLC systems in different road types and traffic scenarios. We quantify the capability of receiving signals in several cases including the V2V connectivity (with HLs and TLs) and the I2V connectivity (with traffic and streetlights). We further quantify the contribution of individual PDs to elaborate on the usage cases of each PD. In the last part of this dissertation, we explore the vehicular VLC as a wireless connectivity solution to enable outdoor broadcasting for public safety systems. We utilize the ubiquitous streetlights as wireless transmitters taking into account the fundamental differences imposed by the outdoor medium and lightning infrastructure. These include the effect of the asymmetrical pattern of streetlights, the orientation of the user equipment, the weather condition, and the solar irradiance. We consider two broadcasting scenarios; VLC broadcasting in the roadway and VLC broadcasting in the sidewalk path, and obtain the received Signal-to-Noise Ratios (SNRs) for all links under the mobility condition.