Road Sensing through a Connected Vehicle Fog Approach for Smart Cities

智能城市的车载道路传感涉及大量物联网设备，这些设备可以产生海量数据。车辆可以作为一种非常有效的城市传感工具，收集和分发关于路况、交通状况、天气、空气质量、车辆状况的数据，仅举几例。据估计，一辆智能汽车上各种传感器收集的数据量约为25 GB/h。在这种环境下，需要对数据进行挖掘、过滤、存储、处理，然后及时分发给目标玩家。然而，在车载网络这样的大规模动态系统中，客观高效地获取信息是一个主要的困难。目前在智能城市中存储和处理物联网数据的解决方案涉及集中式云方法。然而，当应用于车载网络时，集中式方法缺乏实时计算特性，并且不能提供快速反馈响应等限制。另一个挑战是如何以一种比现有机制更智能和更有选择性的方式将道路传感数据传递给智能交通系统(ITS)应用的最终用户。为了克服现有方案的不足，迫切需要一种兼顾实时处理约束、车辆机动性和资源异构性的车载雾计算资源编排方案。因此，本研究将调查、设计、实现和评估算法和工具，以促进通过智能城市互联车雾平台收集、处理和分发道路传感数据。更具体地说，本研究计划将：(A)调查现有的车用雾计算资源管理方案；(B)设计和开发一种新的车用雾计算资源分配方案；(C)研究如何将上下文数据整合到拟议的道路传感平台中；(D)设计和开发智能和选择性的数据传输系统；(E)通过模拟使用最相关的基准来评估拟议的技术；(F)设计和实施智能交通系统应用程序，例如创新的湿滑道路预警系统，以展示拟议的道路感知平台。人们高度期待这项研究的结果将促进ICT公司、政府机构、市政当局和其他利益攸关方强烈关注的创新进展。因此，拟议的道路感知平台是及时的，绝对可以提供一个有效的工具来收集和分发城市数据，并帮助在智能城市环境中进行交通和资源管理。这项工作将使服务提供商能够开发新的应用程序，这些应用程序将利用拟议的数据分发平台，并为加拿大人提供智能和定制的服务，以改善他们的通勤和安全，减少碳排放，并为创新解决方案铺平道路。

Vehicular road sensing for smart cities involves a large number of Internet of Things (IoT) devices that can generate a massive volume of data. Vehicles can be the "eyes and ears" as a very efficient urban sensing tool, collecting and distributing data about road condition, traffic status, weather, air quality, vehicle status, just to mention a few. The amount of data collected by the various sensors in a smart vehicle has been estimated to be around 25 GB/h. In this kind of environment, data needs to be mined, filtered, stored, processed, and then disseminated to target players in a timely fashion. However, one of the major difficulties in large scale and dynamic systems such as vehicular networks lies in accessing information objectively and efficiently. Current solutions for storing and processing IoT data in smart cities involve centralized cloud approaches. However, centralized methods lack real-time computation properties and fail to provide quick feedback responses, among other limitations when applied to vehicular networks. Another challenge lies in how to deliver road sensing data to intelligent transportation system (ITS) application's end users in a more intelligent and selective way than existing mechanisms. In order to overcome the shortcomings of existing schemes, a vehicular fog computing resource orchestration scheme is highly needed while considering real-time processing constraints, vehicles mobility, and resource heterogeneity. Therefore, this research will investigate, design, implement and evaluate algorithms and tools to facilitate the collection, processing, and distribution of road sensing data through a connected vehicle fog platform for smart cities. More specifically, this research program will: (a) investigate the existing vehicular fog computing resource management schemes; (b) Design and develop a novel vehicular fog computing resource allocation scheme; (c) Study how contextual data can be integrated into the proposed road sensing platform; (d) Design and develop a smart and selective data delivery system; (e) Evaluate the proposed technologies using the most relevant benchmarks through simulations; (f) Design and implement an ITS application, e.g., an innovative slippery road warning system, to showcase the proposed road sensing platform. It is highly expected that the outcomes of this research will foster innovative advances of strong interest to ICT companies, government agencies, municipalities, and other stakeholders. Thus, the proposed road sensing platform is timely and can definitely provide an efficient tool to harvest and distribute urban data and aid in traffic and resource management in a smart city environment. This work will enable service providers to develop novel applications that will tap into the proposed data distribution platform and provide Canadians with smart and tailored services to improve their commutes and safety, reduce carbon emissions, and pave the way for innovative solutions.