30 October 2019 in IEEE Transactions on Vehicular Technology
The next frontier for automotive revolution will be connected vehicles and the Internet of Vehicles that will transform a vehicle into a much more than a smartphone on wheels. The fifth-generation (5G) networks are expected to meet various communication requirements for vehicles. This will enable the creation of a large market of services with the exploitation of the generated data from in-vehicle sensors. The cloudification of vehicular network resources through Software Defined Networking (SDN) is a new network paradigm. A critical part of SDN is the transport of control plane. Therefore, it is crucial for SDN-based On-Board Unit to keep a robust connection with the SDN controller. In the literature, several works propose SDN-based architecture vehicular communication. However, nearly all performance evaluations are based on theory or simulation results. This paper proposes a design of a complete Software Defined Vehicular Network prototype. At first, the SDN-based backbone is tested in real hardware composed of OpenFlow switches. Next, the SDN-based Radio Access is tested based on WiFi Access Points that support Click Modular Router and OpenvSwitch/OpenFlow. Finally, a Single Board Computer is used as On-Board Unit (OBU) in which are implemented OpenFlow switch functionalities. Several SDN controllers are used to implement routing algorithms to transport vehicles control plan through the backbone, to process information received from vehicles to predict topology and compute routing path for V2V and V2I communication, and finally to manage mobility schemes. Communication quality is evaluated by measuring throughput, delay, processing time, handoff latency and packet loss.
27 April 2020 in IEEE Access
The fifth-generation (5G) networks are expected to meet various communication requirements for vehicles. C-V2X, introduced in LTE V2X in Release 14, is designed to provide ultra-high reliability and ultra-low latency performance required by the most demanding V2X applications. In the literature, research interests are primarily focused on safety-critical applications in a dynamic environment. Therefore, in most communication models, both safety and non-safety critical applications operate through the same radio access technology. This is the case of both C-V2X Direct Communication and IEEE 802.11p. However, in an urban environment characterized by high traffic density, the availability of resources can be problematic. In that case, it would be best to propose new communication strategies because different use cases will have different sets of requirements. In this paper, we propose to increase the capacity of C-V2X Direct Communication by introducing WiFi Direct as a second connection alternative. Indeed, several works have shown that WiFi offloading can alleviate the congestion of cellular networks. Thus, an SDN-based P2P Group Formation is proposed by extending OpenFlow to manage the WiFi Direct control plane. This solution also allows establishing multi-hop communication, something that is not possible in the standard version of WiFi Direct. The performance evaluation of the P2P Group Formation procedure is proposed via simulations in an urban environment. The results show that our proposed procedure performs better compared to those proposed in the literature. To demonstrate the implementation feasibility of the proposed solution in real hardware, we also performed prototyping.
18 April 2019 in IEEE Sensors Journal
The automotive sensor technology is advancing rapidly and is actually a vital part of any modern automobile design. These sensors generate a large quantity of data that are exploited to implement intelligent functions in the vehicle (improvement of driving-efficiency, safety, convenience’). 5G Next generation networks are expected to meet various communication requirements for vehicles. This will enable the creation of a large market of services with the exploitation of the generated data from in-vehicle sensors. However, with the dynamic nature of vehicular networks and the heterogeneity of invehicle sensors data, it is a major challenge for applications/services to meet the 4A vision (Anywhere, Anytime, by Anyone and Anything). In this paper, we propose a Sensing as a Service (S2aaS) model based on the novel architecture for the collection and the transformation of in-vehicle sensors data. Sensor Semantic Network ontology is employed to effectively model heterogeneous sensors data and sensors proprieties. This ontology allows the deployment of vehicle sensing resource discovery service, ranking, and selection of sensing data based on user and application requirements. The proposition of caching and distributed decision-making systems allow the reduction of network overhead and latency compared to others works in literature. Then a practical example of the use of the S2aaS is presented. Android automotive architecture with the proposed API and App to support S2aaS is presented at the end of the paper.
27 November 2023 in Bulletin of Social Informatics Theory and Application
In Senegal, stock recovery and fish classification are based on manual data collection, and the fish caught by the fishery are not often declared. What's more, data collection suffers from a lack of tools for monitoring and counting fish caught at fishing docks. Researchers have carried out studies on the fishery in Senegal, but data collection is almost non-existent. Moreover, there is no local fisheries database or automatic detection and counting algorithm. In this paper, a semantic segmentation algorithm is proposed using intelligent systems for the collection of fishery catches, for the formation of the local database. The data are collected by taking images of fish at the Soumbédioune fishing wharf in Senegal, and are completed with the Fishbase database. They were applied to the algorithm and resulted in a segmented dataset with masks. This constitutes our local database. The database is used with YOLO v8. The latter is very important for detecting images with bounding boxes in order to train the model. The results obtained are very promising for the proposed automatic poison detection and counting model. For example, the recall-confidence scores translate into bounding box performance with scores ranging from 0.01 to 0.75, confirming the performance of this model with bounding boxes