| Summary: | 博士 === 國立成功大學 === 資訊工程學系碩博士班 === 101 === This thesis proposes a novel approach to clustering the interests of car drivers, increasing the lifetime of interest groups, and increasing the throughput in vehicle-to-vehicle (V2V) environments. It develops an interest ontology of cellular automata (CA) clustering using zone of interest (ZOI) for mobicast communications in vehicular ad hoc network (VANET) environments. The key to the proposed method is to integrate CA clustering with the ontology of users’ interests. This study argues for the use of both an interest profile (ontology) of drivers and information about vehicles to form a group of VANET-related interests. The current study evaluates the performance of the approach by conducting computer simulations. Simulation results reveal the strengths of the proposed CA clustering algorithm in terms of increased group lifetime and increased ZOI throughput for VANETs.
The car society interest group mechanism is focus on how the vehicle join into a interest group through cell (car) evolution and CA mechanism. In this work, we want to investigate the interest group evolution progress. When interest group and car member evolution at different time stamp, how the car member join or leave the interest group.
This thesis proposes a novel cognitive CA approach that can adapt to immediate requirements, spread to use in cross-area car societies, enhance system performance, and decrease traffic congestion problems. We propose a mechanism that operates in a cognitive radio mode for increasing the channel reuse rate and decreasing the consumption of redundant channels. The advantage is a heterogeneous communication interface available through cognitive mechanisms that can recognize different transmission modulation modes. The receiver will get messages through different transmission modulation modes. In this work, we consider vehicles connecting to traffic-congestion computing centers (TCCCs) by vehicle-to-roadside (V2R) communications under a car society. RSUs serve each segment, and we suppose that every car has a navigation device.
We propose an innovative congestion reducing mechanism that can help vehicles get directions with the help of a navigation device after drivers set the origin location and the destination location. This mechanism can calculate the congestion status of the upcoming segment of road. By tracking roadway segments’ status from a point of origin to a destination, our proposed mechanism can handle cross-area car societies. The current study evaluates this approach’s performance by conducting computer simulations. Simulation results reveal the strengths of the proposed CA mechanism in terms of increased lifetime and increased congestion avoidance for urban vehicular networks.
This thesis proposes a novel eight-direction mechanism that can receive location information about nearby vehicles to perform self-analysis for lane-changing activities. In this work, we assume that each vehicle creates 3x3 grids called safety distance fields, and that the central grid is based on the given vehicle’s location. Also assuming one on-board unit (OBU) in each vehicle, this work uses location information of nearby vehicles as input for a CA model that calculates whether or not vehicles in nearby lanes are a safe distance from the vehicle for which the calculations are being performed. The current study evaluates this approach’s performance by conducting computer simulations. Simulation results reveal the strengths of the proposed CA model in terms of increased safety distance and increased collision-avoidance for urban vehicular environments.
|
|---|
