Leveraging Fog Compute Ecosystems for Robust Connectivity in MaaS Technologies
Fog Computing for Secure MaaS Implementation
Fog Computing: The pressure on transportation systems in urban and rural areas is rising as a result of rising consumer demand. Furthermore, governments and private sector transportation providers are facing new challenges as a result of population growth, rising urbanisation, and rising customer expectations with travel time and overcrowding still increasing on buses, trains, trams, and ferries in most cities despite increased investment. Additionally, many modes of transportation, including road and rail, are showing signs of stress as travel times become more unpredictable due to unforeseen events and constant maintenance needs.
In response to these difficulties, a new wave of innovation known as Mobility as a Service (MaaS) which involves the provision of holistic mobility experiences, is evolving. MaaS predominantly focuses on:
1.) combining the planning of travel, the ordering of services, and the payment process;
2.) offering customers customised travel options based on the most recent service and disruption information;
3.) combining transportation services into practical and affordable ‘mobility plans’ that are customer-specific; and
4.) supplying brand-new, creative transportation services for first and last-mile connectivity to close gaps in the transportation network.
In a decentralised computing environment known as fog computing, data, computing, storage, and applications are distributed between the data source and the cloud. These are all considered necessary for the efficient implementation of MaaS, though it is also important from security and compliance perspectives since MaaS relies heavily on connected infrastructure involving the Internet of Things (IoT), Vehicle to X (V2X) connectivity, and data management.
Innovations in Edge Cloud Ecosystems for MaaS
Researchers in the Department of Computer Science and Engineering at the American University of Sharjah, United Arab Emirates (UAE) conducted a set of experiments to determine the most efficient architectural design for e-vehicles within a MaaS ecosystem within a community setting. The proposed architecture consisted of four main layers: 1.) the physical layer, 2.) the communication layer, 3.) the control and management layer, and 4.) the interface, operation continuity, and security layer. All layers were used to examine the adequacy of the proposed architectures using the Low Power Wide Area Network (LoRaWAN) technology for fog computing in smart mobility sharing with respect to latency, scalability, and energy consumption.
When compared to cloud computing, the results demonstrated that fog computing for smart mobility was able to execute between 1,000 and 10,000 requests simultaneously with an average execution time of hundreds of milliseconds. Additionally, the modular design of the system enabled the deployment of components across different nodes with the option to provision additional servers during runtime so that components can scale when the number of requests was significant. Additionally, the researchers measured the power consumption of the e-bike shared mobility prototype with and without the LoRaWAN end node. It was found that LoRaWAN uses very little power from the e-bike battery when operating normally.
In order to investigate the value of real-time transport demand data in the context of developing MaaS networks using fog computing techniques, Cisco and the University of New South Wales (UNSW), Australia joined forces. In terms of waiting times at transit hubs, such as bus stops, and overall journey times (which include waiting times and trip times from point A to point B), the research team recorded and measured the transportation customer experience from many journeys.
Fog Computing for Enhancing Real-Time Control in MaaS Ecosystems
The researchers utilized the Cisco Kinetic IoT Data Fabric platform to quickly assemble and process real-time transport demand metrics by leveraging a variety of data sources and capture techniques. To understand the impact of dynamic demand response on journey times and overall customer experience, collaborative research at UNSW’s Research Centre for Integrated Transport Innovation (rCITI) and School of Electrical Engineering and Telecommunications (EET) calculated customer experience metrics via the Cisco Kinetic platform and modelled various MaaS scenarios.
Researchers at EURECOM, France have investigated the advantages of fog computing and how they can offer consumer-centric IoT services such as MaaS.
The researchers presented an architecture for connected vehicles where they deployed the fog platform at Roadside Units (RSUs) and Machine to Machine (M2M) gateways. Its architectural characteristics enabled consumer-centric services such as M2M data analytics and semantic web technologies; thus, allowing the deployment of IoT services and the management of connected vehicles. Some of the current work that is ongoing is looking into extending the platform for data management and repository, application and service management, and crowdsourcing of vehicular data.
In order to support real-time control applications for mobility services, researchers from the Beijing University of Technology in China have proposed a novel fog-based distributed network architecture related to MaaS. The researchers considered the cloud, distributed fog, and edge smart devices as three logical layers. They investigated important system variables such as the sampling period, message delay, message dropout, and noisy channel in relation to communication and control systems. The researchers enriched the network resources with fog computing to enable real-time control.
Vehicular Computing Resource Migration in Smart Cities
The researchers conducted a case study of connected cruise control of connected autonomous vehicles to demonstrate the viability of the suggested architecture in the context of vehicular networks within MaaS ecosystems. They identified the need for improvements to real-time control to achieve consistent traffic flow and enhance performance. They also found that reducing data transmission latency was more favorable. In response, the researchers proposed a two-step control scheme for achieving optimal connected cruise control:
1.) They used a backward recursion approach at the offline step to derive the optimal control gain, and
2.) At the online step, they calculated the optimal control strategy in real-time based on the current system states and local cached information.
Shanghai Jiao Tong University researchers in China have developed a geographical migration model for vehicular computing resources based on vehicle mobility in smart cities with fog computing capabilities. The vehicle as a service framework increases the flexibility of conventional cloud computing architecture by fully utilising the unbalance and randomness of vehicular computing resources. The researchers proposed a reward system that influences the choice of vehicle path through resource pricing to balance resource needs and geographically distributed computing resources. The simulation outcomes showed that the suggested scheme’s benefits and effectiveness were substantial. The researchers applied a resource pricing-based incentive scheme to fully utilize the mobility of vehicles and change the distribution of fog computing resources in smart cities. The simulation results demonstrated that the resource pricing scheme attracted vehicles to fog nodes where there was an urgent need for resources.
The Future of Fog Compute Ecosystems for MaaS
The IoT and MaaS ecosystems are likely to be significantly impacted by fog computing. According to forecasts, the fog computing market, which was valued at US$22.3 million in 2017, will significantly increase and reach US$203.5 million by the end of 2023. The demand for connected devices, real-time computing, and IoT interconnectivity is what is causing the acceleration of the fog market.
Fog computing will unavoidably play a significant role in the upcoming waves of smart city development and MaaS ecosystems. As technology advances towards an era where computers must be able to behave like people, the ability of computers to make decisions will become increasingly important for smart city and MaaS ecosystems that want to remain cutting edge.
The emergence of fog computing doesn’t necessarily mean the end of cloud based normalised IoT networks. Since centralised data is almost always easier to access, there are more opportunities for public and private sectors to create open data agreements. This might result in more sophisticated innovation and digital creativity. The uses for data stored in the cloud are also expanding at the same time.