Encyclopedia of Wireless Networks

Living Edition
| Editors: Xuemin (Sherman) Shen, Xiaodong Lin, Kuan Zhang

Architecture and Data Management for Smart Community Information Platform

  • Hiroaki NishiEmail author
Living reference work entry
DOI: https://doi.org/10.1007/978-3-319-32903-1_249-1



Smart Infrastructure: A new style of infrastructure composed of conventional infrastructure and information and communication technologies (ICTs). Smart infrastructure refers to an efficient and highly functional infrastructure accomplished by ICT-based monitoring, control, and management.

Smart Community: An integration of smart infrastructures implemented in a certain region.


A smart community is an integration of smart infrastructures implemented in a certain region. Each smart infrastructure may provide dedicated merits. However, the essential merit of a smart community is community big data, which covers all infrastructures. To pursue the effective use of the community, big data is the key to the success of smart community projects. Therefore, data infrastructure is a necessity for a smart community. The information platform considering the safe data exchange and privacy preservation is described as a case study in Saitama City, Japan.

Smart Community

A smart city is an integration of smart infrastructures that are composed of conventional infrastructure and information and communication technologies (ICTs). Smart infrastructure refers to an efficient and highly functional infrastructure accomplished by ICT-based monitoring, control, and management (Nishi 2018). For example, a smart grid is composed of the interaction between the power grid and ICTs, and it accomplishes high-functioning grid operation and effective electricity usage. Smart transportation achieves automated drive and fare systems in logistics, cars, and road systems, which is also referred to as an intelligent transportation system (ITS). Smart agriculture improves the product value by controlling the growth of crops and the total efficiency of farm work. A smart government provides administrative services using the Internet to improve usability and operational efficiency. A smart city is an implementation of several smart infrastructures in a city. Similarly, a smart town and a smart island are implemented in a town and on an island, respectively. A smart community refers to a similar concept and is unrelated to the target region. Moreover, it is meaningless when these smart infrastructures are independently implemented in the target region. A smart community provides new services by integrating and linking information of different smart infrastructures. An example of this information integration is the efficient charge/discharge management for electric vehicles, combining data from an electric vehicle and the electric power system with traffic information. The intensive introduction of smart infrastructures and strong information linkage enable the provision of more advanced services to communities. The penetration of Internet of Things (IoT) has created a large amount of data, and the success of the smart community project depends on the effective use of the data. The data management for a smart community is indispensable to provide attractive smart community services.

A smart community is described in the Smart Communities Guidebook (1997) by the State University of San Diego as “a geographical area ranging in size from neighborhood to a multicounty region whose residents, organizations, and governing institutions are using information technology to transform their region in significant ways. Co-operation among government, industry, educators, and the citizenry, instead of individual groups acting in isolation, is preferred. The technological enhancements undertaken as part of this effort should result in fundamental, rather than incremental, changes.” The ICT created a paradigm shift in infrastructures, and a significant amount of data processing and network transactions was generated. The data processing is primarily achieved in cloud services. However, the location of cloud causes obstacles to some services. The communication latency may cause a serious problem to hard real-time control applications. Open Fog (OpenFog Consortium 2018; Bonomi et al. 2012; Yi et al. 2015) proposed the placement of services closer to the terminal devices than the cloud for improving the efficiency of providing the services. Therefore, Fog would improve the service latency and improve the service distribution in the networks. Edge computing is used as a similar technology to Open Fog. Edge computing provides a computing environment at the edge of the Internet. This also means that the locations of the computing resources are closer to the IoT nodes than to the cloud. Data freshness is indispensable for some services, such as the ancillary service of power grids, and the automated car drive service. The study regarding Fog and Edge computing has become active as the discussions of new smart community services become popular. For data management, FIWARE (https://www.fiware.org) (Ferreira et al. 2017) provides several types of data management API for smart community services. FIWARE provides various types of APIs to manage the smart community data; this shows the importance of data management, especially the multiple perspectives for the data management of smart city services. Herein, the smart community information platform from the perspective of its data management is focused and described.

Smart Community Data Specifications and Requirements

The requirements of smart city services are diverse. When providing smart city services, data specifications are required to provide the service. Several points for data specifications and requirements are used and described as follows.

Immediate data handling: Regarding time constraints, for example, ancillary services in a smart grid do not legally allow a delay of over 1 Hz. The IEC 61000-4 standard determines that the control delay has to be smaller than 10-ms intervals. Moreover, the IEC 61850-9-2 standard determines the tolerance of the delay to be less than 1 μs. Thus, it is almost impossible to provide ancillary services as cloud services. However, it is necessary to accommodate new services that handle hard real-time system requirements, such as haptic communication for telemedicine and tactile sensing, which are sensitive to delay because they are required to transmit action–reaction force feedbacks to convey tactile sensations. This application also permits a 10-ms delay for maintaining stabilization in the control of its feedback control system (Anderson and Spong 1989). Cloud services cannot meet this need owing to their comparatively large communication delay.

Flexible data handling: Many protocols are proposed for communicating with IoT nodes. When focusing only on smart community protocols, especially the application layer, several examples can be provided: IEEE1888, IEEE1451, HTTP, MQTT, SEP2.0, Bluetooth (such as health thermometer profile, Bluetooth Low Energy (BLE)), etc. In the application layer protocols, the primary purpose is to define the data formats. Therefore, server applications are required to support different protocols for receiving data from IoT devices, which differ in the types of application layer protocol. Moreover, new protocols and new data expression rules are successively developed to support the emerging smart community services. However, it is better to continually use conventional IoT devices that do not support new protocols because the replacement of all old IoT devices is costly. Moreover, it is typical that IoT terminals, which have similar sensors and functions, are installed redundantly at the same place owing to different protocols required to handle the data, and this poses a significant problem in the implementation of smart community services. It is important to design a smart community information infrastructure to address this problem.

Data cycle for improving QoL and QoS: For providing smart community services, it is essential to generate, store, process, and analyze the data, as well as foster the data to useful services, use the data through the services, change someone’s behavior or control something using the data, and anonymize the data for its secondary use. Finally, the result of the behavior or control is measured by the sensors, and it creates the data again. This data cycle is shown in Fig. 1. This cycle enables the human behavior and machine control to be improved. Therefore, it improves the quality of life (QoL), such as the improvement in comfort or wellbeing, and the quality of service (QoS), such as the system efficiency or functionality. Thus, the smart city data infrastructure has to maintain the effective data cycle. The similar cycle is also discussed in ambient intelligence from a different perspective.
Fig. 1

Smart community data cycle

Data encapsulation: In the series of data processes, most parts of the process are achieved in the cloud, as shown in Fig. 2a. The use of cloud has the merit of low management cost. However, to use the cloud means to use a centralized system. Therefore, cloud services may cause a single point of failure and suffer from an advanced persistent threat. What types of countermeasures can be provided for it? It is strategically correct to prohibit gatherings and distribute people when the danger of terrorism is predicted. However, despite the repeated incidents of cyber terrorism, private data continue to be centrally managed in the cloud. It is also natural for people to resist the revelation of personal information by moving these data from the cloud to their residential area. However, the service providers only consider their points of view and would force users to observe their service provision policy: “personal information must be managed in the cloud for providing services.” This situation is not desirable in the provision of smart community services. Private data should be encapsulated locally, and this encapsulation is effective from the perspective of protecting several types of attacks or threats, as shown in Fig. 2b. The cloud should provide a global service using abstract and unified information.
Fig. 2

(a) Conventional smart city data services. (b) Encapsulated smart city data services

Data hierarchy: IoT devices send measured data to the cloud. On their way to the cloud, the data may pass a gateway at a smart house or smart building, or switches and routers at the internet providers, and finally, it will arrive at the service application programs in the cloud. As explained in data encapsulation, the data should be processed at an appropriate location when required by the service. As shown in Fig. 3, the switches and routers as well as the gateways can become computational resources referred as edge and fog computing. The selection of the appropriate computational resources requires appropriate indexes, and the typical indexes are shown in the figure. When the services are provided in a gateway, it focuses on narrow space service provisioning and short-range communications, e.g., an indoor environment, and the amount of calculation cost and computing platform can be small because the required data amount is small. Anonymity is a new and important index, as given in data encapsulation. The gateway is often connected with IoT devices and handles the raw data generated by the devices. Contrastingly, the services provided in the cloud can use anonymized data, and vice versa, as explained in data encapsulation. The Edge or Fog computing environment provides a balanced environment between the gateway and cloud. By using these indexes and the differentiated environment, data can be stored at an appropriate network location and processed at an appropriate network location. Moreover, the needs of data processing chains on the Internet are increasing. This chain of data processing is also discussed as service function chaining in network function virtualization (NFV) (Trajkovska et al. 2017). This data chain is important in providing services using the data. Figure 4 shows the mapping of promising smart community services according to the indexes. For example, HVAC control is provided in a building and may use private information. Electricity exchange service is provided in the city area, and the price can be defined using generalized information. Additionally, the service applications have to select the best location.
Fig. 3

Smart community service hierarchy

Fig. 4

Mapping of smart community services considering service locations (Nishi 2018)

IoT data security: The risk of IoT terminals being hacked has increased in the recent years. Hence, the maintenance of security levels must be ensured. However, most IoT terminals have low computing and power capacities. Therefore, it is difficult to facilitate better resistance to cyber-attacks using complicated protocols to add new functionalities. Moreover, introducing additional security software, or new protocols, is undesirable from the viewpoint of power consumption and system cost. Thus, it is necessary to design an information infrastructure to address these problems. Namely, the network system should provide new security services for IoT devices and locally stored data.

Private data handling: The protection of private information is important in smart community services because the data generated by residents for receiving smart community services could have privacy constraints. One of the protection methods is the aforementioned local data encapsulation. However, it is desirable to share the data for secondary use. This secondary use of data is the most promising service in the future smart community. Anonymization technology is key for the safe sharing of private data. If the data are perfectly anonymized and any type of private data is not extracted from the anonymized data, it is safe but is not useful for smart community services. The balance between difficulty in revealing private information and the usefulness in providing services is important. For achieving the balance at the highest level, the appropriate anonymization method must be designed and selected for each data service. Medical records, smart meters, locations, and other data will have their respective suitable anonymization methods. The development of an anonymization method is indispensable. Moreover, watermarking technology for anonymized data is indispensable (Nakamura et al. 2017). In data service, the information of user rights must be managed, such as who generates the data, who uses the data, and what the purpose is. Watermarking technology can include the information in the anonymized data. The watermark in anonymized data can prevent data leakage by tracing the data leakage and protect the user rights holders.

Data ownership: Who possesses the data? Although this is a simple question, it is sometimes difficult to answer because the ownership and stakeholders of data are easy to entangle and handover. Moreover, the national speculations have worsened this situation. The General Data Protection Regulation (GDPR) was issued, and it provided a new stage of data management from the perspective of data privacy. In consideration of these situations, flexible data ownership management is required. Regarding the style of service provision, vendors have to manage the consumers; vendors draft contracts with consumers for providing services and provide reasonable prices for the services, in which the vendors aim to maximize both their benefit and the consumer satisfaction. In this contract, the vendors request the disclosure of private data. This is called consumer relationship management (CRM). Contrastingly, new trend requires an alternative relationship management between vendors and consumers such that the consumer can manage the vendors’ services. This is called vender relationship management (VRM). A user may think that a certain service is useful and worth providing the user’s private data to the service vendor for a more efficient service. Meanwhile, the user may think that the service is enough to use as a trial and not worth providing the private information. In this case, the data management system will anonymize the data to preserve privacy. For attaining the flexible service relationship management between vendors and users, vendor and consumer relationship management (VCRM) is proposed (Niwa and Nishi 2017). Figure 5 shows the structure of VCRM. Both the vendor and consumer manage the database, which stores the information of the service provider or who the data user. A relationship is established when a contract of providing and receiving a service is engaged. The relationship is managed by its dedicated database, and the database has various information, such as the method and level of data anonymization, target area of service, location to be processed, service provider, service user, expiration date, constraints for providing services, required computational power, required latency, and amount of data. VCRM manages these data and verifies the integrity with real use.
Fig. 5

Vendor and consumer relationship management

Smart Community Information Platform (SCIP)

The fusion of Fog/Edge, gateway, and cloud are appropriate as a desired future direction. However, their current shortfalls must be compensated. Therefore, an information mechanism called authorized stream contents analysis (ASCA) was proposed (Nishi 2018). ASCA is a mixed mechanism of software and hardware for supporting service provision. It supplies a method to process the information flowing through a communication network on intermediate communication devices. ASCA reconstructs the TCP stream on the device, decodes the SSL using the key shared with the cloud service, decodes Chunk and Gzip, executes string matching and the extraction function using regular expression rules, and subsequently sends the analyzed result of the stream to the dedicated service process. ASCA can modify the stream contents directly, under the constraint of the buffer size. When using 128 G of memory, it can analyze more than two million TCP streams simultaneously without the limitation of the TCP stream length by employing a dedicated context-switch technology. ASCA on a general enterprise server can provide 20 Gbps of processing performance using hardware accelerators, such as DPDK for the accelerating network throughput using userland zero-copy communication, HyperScan on the Intel Xeon Processor for accelerating the string matching function, and Intel QuickAssist Technology for accelerating the throughput of the encryption, decryption, compression, and decompression.

A service application using ASCA on a Fog/Edge terminal can process the network stream and provide dedicated services at an intermediate location on the route to the target in the cloud. This does not require any modification of the IoT terminals; the destination IP address of the data stream can maintain its original IP address of the target in the cloud. Moreover, ASCA provides the functionality to modify the stream contents, enabling the direct removal or anonymization of private information at the intermediate nodes. NEGI (Takagiwa and Nishi 2015) is an original library created for ASCA. It is designed with no acceleration and can therefore be utilized in any Linux-based environment, including an ARM-based embedded platform. The Intel Xeon platform can achieve 1 Gbps of streaming. DooR is a hardware accelerated library of the ASCA, and it achieves a 20-Gbps stream process. The proposed smart community information platform (SCIP) uses NEGI or DooR as basic libraries for stream processing. On these ASCA basic libraries, the user and service provider can design their services as a Docker container (Miura et al. 2017). Docker is a virtualized environment for executing application programs, and it reduces the cost of launching and terminating application software, compared with the conventional virtual machine environment. Because Docker provides a virtualized environment isolated from the host machine, the zero-copy architecture including the DPDK is not available for the application software as a Docker container. Therefore, Docker’s shared memory option is used to communicate with a DPDK-supported NIC via shared memory to cope with this problem.

ASCA can handle the aforementioned smart community data specifications and requirements. ASCA provides a transparent environment as network nodes. Because ASCA can monitor or modify a network stream at any point, it enables immediate data handling using zero-copy communication and hardware accelerators. It can also offer flexible data handling because it can provide services at the intermediate nodes on the Internet and change a communication protocol transparently. When ASCA is used as a basic library in the smart community data platform, it can be the key device for rotating the data cycle for improving the QoL and QoS. All the given functions in Fig. 1 can be achieved or measured using service applications with ASCA. Transparent data encapsulation can be achieved by designing the data firewall of the data anonymization application on the ASCA. Moreover, the hierarchical design of the ASCA-based encapsulation enables the data hierarchy. The firewall function and antivirus function on ASCA-supported devices can provide IoT data security. By implementing the appropriate applications on the ASCA, private data can be handled and the data can be anonymized. VCRM can be an application of ASCA. Therefore, either the user or the service provider can define and modify the relationship at any time. This feature adopts the opt-in and opt-out of data registration. The default value is given as the opt-in. The on-demand modification of the relationship achieves the opt-out.

Another important point is the mobility of the service applications. The initial allocation to an appropriate location and the subsequent execution of a service should ideally occur automatically, and be migrated as necessary, without the need for an explicit migration request. Executing this effectively requires managing several resources and activities, including memory, storage, CPU, communication, task allocation, and distribution, to manage the Docker containers with service applications. Thus, a mechanism for resource management that performs functions similar to the basic functions of an operating system is necessary. This resource management system is called the SCIP OS. Some basic functions of the SCIP OS resemble those of the orchestrator for Docker containers. However, the orchestrator only considers the management of the Docker containers, whereas the SCIP OS focuses on the service applications more broadly from the perspectives of service feasibility, IoT feasibility, and future feasibility. Moreover, ASCA enabled wireless station can be a center of local data manager in the wireless environment. It gathers data from sensor nodes and offers a variety of benefits to local services in the reachable range of its wireless signals.

System Demonstration

The data anonymization and watermark insertion application of UDCMi, a smart town project in Misono Town, Saitama City, Japan, was demonstrated at the Global City Team Challenge EXPO (https://pages.nist.gov/GCTC/). This application uses the smart metering of smart houses in UDCMi. As a smart community service, a lifestyle recommendation service for eco-life is provided as a nudge service. In this demonstration scenario, the smart meter sends the data to the cloud, and the nudge report was automatically generated using machine-learning technology. En route to the cloud, the data are also captured, anonymized, and watermarked at the gateway by ASCA. The extension of the function of IoT devices was proven, and the effectiveness of the nudge report using anonymized data was compared with the other report using raw data.


Data-oriented smart community services are indispensable for the sustainable advancement of smart cities. The proposed smart community information platform is a system considering the given data specification and requirements. The complexity of data handling at the network infrastructure will be increased according to the development of a society structure. However, an information platform maximizing the flexible data management can solve various regional problems, including urbanization problems. The improvement of QoL and QoS, i.e., the ultimate goal of a smart community, is not achieved by a single metric but by an interoperable data approach including the residents’ behavior change and machinery control that enables hopeful societies to be established. Further developments of the architecture and data management of the smart community information platform are expected.




This work was partially supported by MEXT/JSPS KAKENHI Grant (B) Numbers JP17H01739, and also by the Technology Foundation of the R&D project, “Design of Information and Communication Platform for Future Smart Community Services” by the Ministry of Internal Affairs and Communications of Japan.


  1. Anderson RJ, Spong MW (1989) Bilateral control of teleoperators with time delay. IEEE Trans Autom Control 34(5):494–501. https://doi.org/10.1109/ICSMC.1988.754257
  2. Bonomi F, Milito R, Zhu J, Addepalli S (2012) Fog computing and its role in the internet of things. In: Proceedings of the first edition of the MCC workshop on mobile cloud computing. ACM, New York, pp 13–16. https://doi.org/10.1145/2342509.2342513
  3. Ferreira D, Corista P, Gião J, Ghimire S, Sarraipa J, Jardim-Gonçalves R (2017) Towards smart agriculture using FIWARE enablers. In: International conference on engineering, technology and innovation (ICE/ITMC), Madeira Island, pp 1544–1551. https://doi.org/10.1109/ICE.2017.8280066
  4. Miura T, Wijekoon JL, Prageeth S, Nishi H (2017) Novel infrastructure with common API using docker for scaling the degree of platforms for smart community services. In: International conference on industrial informatics, Emden, pp 474–479. https://doi.org/10.1109/INDIN.2017.8104818
  5. Nakamura Y, Nakatsuka Y, Nishi H (2017) Novel Method to Watermark Anonymized Data for Data Publishing. IEICE Trans Inf Syst E100-D(8):1671–1679CrossRefGoogle Scholar
  6. Nishi H (2018) Information and communication platform for providing smart community services system implementation and use case in Saitama City In: Proceedings of the 2018 IEEE international conference on industrial technology (ICIT), Lyon, pp 1375–1380. ISBN: 978-1-5386-4053-1/18/. https://doi.org/10.1109/ICIT.2018.8352380
  7. Niwa A, Nishi H (2017) An information platform for smart communities realizing data usage authentication and secure data sharing. In: Fifth international symposium on computing and networking, Aomori, pp 119–125. https://doi.org/10.1109/CANDAR.2017.73
  8. OpenFog Consortium. https://www.openfogconsortium.org/#fog-computing. Accessed 7 July 2018
  9. Smart Communities Guidebook: Building smart communities, how California’s communities can thrive in the digital age. International Center for Communications, College of Professional Studies and Fine Arts, San Diego State University, San Diego, 1997Google Scholar
  10. Takagiwa K, Nishi H (2015) Local trend detection from network traffic using topic model and network router. In: ICOMP’15 – The 2015 international conference on internet computing and big data, Las Vegas, pp 53–59Google Scholar
  11. Trajkovska I, Kourtis M-A, Sakkas C, Baudinot D, Silva J, Harsh P, Xylouris G, Bohnert TM, Koumaras H (2017) SDN-based service function chaining mechanism and service prototype umplementation in NFV scenario. J Comput Stand Interfaces Arche 54(P4):247–265 . Elsevier Science Publishers. https://doi.org/10.1016/j.csi.2017.01.002CrossRefGoogle Scholar
  12. Yi S, Li C, Li Q (2015) A survey of fog computing: concepts, applications and issues. In: Proceedings of the 2015 workshop on mobile big data. ACM, New York, pp 37–42. https://doi.org/10.1145/2757384.2757397

Copyright information

© Springer Nature Switzerland AG 2018

Authors and Affiliations

  1. 1.Department of System Design EngineeringKeio UniversityYokohamaJapan

Section editors and affiliations

  • Vincent Wong
    • 1
  1. 1.Department of Electrical and Computer EngineeringThe University of British ColumbiaVancouverCanada