Off-campus Michigan Tech users: To download campus access theses or dissertations, please use the following button to log in with your Michigan Tech ID and password: log in to proxy server
Non-Michigan Tech users: Please talk to your librarian about requesting this thesis or dissertation through interlibrary loan.
Date of Award
Campus Access Dissertation
Doctor of Philosophy in Computer Science (PhD)
Administrative Home Department
Department of Computer Science
Committee Member 1
Committee Member 2
Laura E. Brown
The new era of the Internet of Things is promoting the evolution of self-driving vehicles into connected and autonomous vehicles (CAVs), which enable modern intelligent transportation systems. CAVs provide connectivity for vehicles by integrating the capabilities of new generation wireless technologies such as dedicated short-range communication (DSRC) and cellular networks (e.g., LTE, 5G, and potential future 6G developments). The success of CAVs and their applications is dependent on the performance of their underlying networks, known as vehicular networks, which can be compromised due to the traffic dynamics and limited transmission range of communication devices. This dissertation focuses on the optimization of message coverage and dissemination that improves the overall performance of vehicular networks.
The coverage of message, from a performance perspective, can be studied in terms of dissemination distance, the size of inter-vehicle networks, and the delay of message dissemination. We study the message coverage in infrastructure-based vehicular networks by focusing on the problem of optimal utilization of roadside units (RSUs) in urban environments. Specifically, we develop efficient algorithms, which by considering the driving dynamics and delay constraint of applications find the candidate sites for RSUs deployment to achieve the maximum spatiotemporal message coverage and minimum delay of message dissemination. We also present a disseminator selection algorithm for infrastructure-based environments, which aims at improving message forwarding in the presence of RSUs.
Effective clustering is crucial to mitigate routing scalability and reliability issues in vehicular networks. Clustering provides a hierarchical network architecture as a result of grouping the vehicles. We propose an adaptive clustering scheme to maximize the stability of clusters. To that end, we define a novel concept, named stability degree of vehicles, by taking into consideration the driving dynamics over a prediction horizon. We then formulate the clustering problem as an optimization problem, which is used within a rolling horizon framework in the cluster formation process. Our scheme also provides insight into the optimally of clustering algorithms.
We present a heterogeneous vehicular network architecture, which supports DSRC- and Device-to-Device-based communications for vehicle-to-vehicle to improve routing reliability. We prove that the optimal clustering problem is NP-hard. Therefore, we develop a heuristic algorithm, which by taking into consideration the stability degree of vehicles over the prediction horizon, finds a near-optimal solution. We propose a hybrid routing protocol based on our adaptive clustering scheme to achieve a high packet delivery ratio and low delay. Our hybrid routing protocol is composed of generic proactive and reactive routing methods and is utilized in our heterogeneous vehicular network architecture.
Jalooli, Ali, "ENABLING TECHNOLOGIES FOR INTERNET OF THINGS: OPTIMIZED NETWORKING FOR CONNECTED AND AUTONOMOUS VEHICLES", Campus Access Dissertation, Michigan Technological University, 2020.