Title context-based personal data protection in smart city

Размещено на http://www.allbest.ru/

National Research University Higher School of Economics

Faculty of Business and Management

Department of Innovation and Business in Information Technologies

MASTER THESIS

title context-based personal data protection in smart city

Student Wang Xizi

Group MBD 161

Supervisor YEVGENI KOUCHERYARY

Submission date 2018/04/25

Moscow 2018



Content

Abstract

Chapter 1. Introduction

1.1 Research Background, Objectives and Significance

1.2 Research Content and Methods

1.3 Main Contributions

Chapter 2. Literature Review

2.1 Research on Concept and Connotation of Smart City

2.2 Research on Services and Techniques of Smart City

2.3 Research on Privacy Issue in Smart City

Chapter 3. Smart City

3.1 Characteristics of Smart City

3.2 Technical Architecture of Smart City

3.3 Supporting Techniques in Smart City

3.4 Services in Smart City

Chapter 4. Personal Data Security in Smart City

4.1 Privacy Concerns in Smart City

4.2 Types of Attacks in Smart City

4.3 Personal Data Classification in Smart City

4.3.1 Importance of Personal Data Classification

4.3.2 Types of Personal Data Involved in Smart City

4.3.3 Personal Data Categories and Samples

Chapter 5. A Classified Personal Data Protection Architecture

5.1 Overall Description

5.2 Hierarchical Elaboration

5.3 Building Blocks for Security Mechanism and Security Service Based on PETs

5.4 Evaluation

Chapter 6. Conclusion

Reference

Acknowledgement



Abstract

Smart city has been a popular research focus on sustainable urban development in recent years. With the development of information technology, network technology and communication technology, the concept of "smart city" has been introduced to solve the increasingly serious problem of “urban disease". Smart city is the advanced stage of the urban informatization development. In smart cities, while citizens enjoy many immediate and comprehensive modern services, their privacy is at the same time threatened: The various network systems in smart cities are vulnerable to attack; Centralized management of the vast personal data can easily lead to "Big Brother” effect. Therefore, the problem of privacy protection in smart city is becoming an urgent issue which attracts more concerns. As a new protection strategy, data classification has been applied in some security systems already. It can help data controller to easily locate sensitive data and deploy different protection levels accordingly. This paper proposes a context-based personal data protection architecture based on the technical architecture of smart city, which provides an integrated system covers from data collection to data fusion in services. The operational process and involved building blocks will also be illustrated in this paper.

Keywords: smart city, privacy issue, personal data classification, data protection



Chapter 1. Introduction

1.1 Research Background, Objectives and Significance

Research Background

First of all, the smart city has become a new direction for the development of cities in the world. In recent years, because it cannot form the sustainable development system of the city, the traditional urban development model is faced with many pressures from the aspects of economy, society and environment. At the same time, the continuous progress of technology and society makes the development demand of cities constantly diversifying, and the difficulty of urban management is also increasing. Environmental destruction, energy shortage and traffic congestion will bring huge obstacles and challenges to urban development. In order to solve the problems encountered in the development of the traditional cities, the city must seek new direction of development. From 2008 to 2009, IBM put forward two ideas of smart earth and its extended concept smart city, hoping to establish a new direction of urban development. IBM hopes that by use of the new technology of networking, cloud computing and other changes in the traditional development model of the city, to establish a way that can help the city to realize more convenient services to benefit citizens. Through smart management of all aspects of the city, smart city will effectively realize the sustainable development and solve all kinds of problems that appear in the process of urbanization [54]. Since 2009, the concept of smart city has been widely accepted. North America, Western Europe, Asia Pacific, Australia and other countries and regions have begun to conduct researches and generalization of smart city. Smart city has become the new direction of city development of all countries in the world.

Secondly, the smart city has become a new hot topic in academic field. Since smart city has become the new direction of urban development recently, the relevant academic research for smart city is extremely necessary and urgent. Further research on related theory will help countries to better the construction of smart city, to promote smart city to the greatest extent, giving full play to its own advantages to achieve the purpose of convenience, also to solve the problems in the process of urbanization. This means that a country's in-depth study of smart city will bring greater progress and development and provide a better quality of life for the domestic people. The above potential advantages of smart city encourage the enthusiasm of all countries to study on it. Governments have increased investment in smart city related research projects to varying degrees. At the same time, scholars' attention to smart city has increased linearly, and the amount of research on smart city has increased year by year. The combination of the above factors leads to the smart city becoming a new hot topic in the academic field.

Thirdly, with the development of society and technology, personal information security has always been a focus in social and academic circles for many years. In terms of how to define the safety of personal information and how to protect personal information security, scholars from different circles put forward different opinions and researches from the aspects of law, technology and management.

Fourthly, although the smart city and personal information security are the current research focuses, but scholars seldom study these two words in combination. The existing research is mainly related to separate study of smart city, or a separate study on relevant personal security problems. This has led to a large gap in the comprehensive study of these two keywords. Under the premise of smart city becoming the new direction of urban development and the new hot spot of academic research, it is very necessary to study personal information security for smart city. On the one hand, it will help to explore the relationship between smart city and personal information security and its impact on personal information security. This kind of research will put forward meaningful advocacy for the future development of smart city. On the other hand, it will help to analyze whether the smart city in the construction process is perfect or not. What are the effects of the smart city on personal information at present? Is there any adverse effect? What is the reason for it if it does have a negative impact? What is the solution to it? The above problems are very worthy of discussion.

Research Objectives

At present, there are relatively few studies on the combination of smart city and personal information security in academic filed. Especially, there is no research on context-based personal data classification and protection that applied in smart city architecture. The purpose of this thesis includes five points:

1. Elaborating the concepts of smart city and personal data classification.

2. Analyzing the relationship between smart city and personal information security, and the challenges on personal information security in smart city.

3. Illustrating the framework of smart city and the supporting techniques, analyzing how personal data is involved in smart city services.

4. Proposing the architecture of classified personal data protection and explaining the context-based personal data classification workflow.

5. Putting forward the classified protection countermeasures and privacy building blocks within the proposed architecture.

Research Significance

Theoretical Significance

On the basis of a certain amount of papers and researches related to smart city and privacy protection, this paper studies the concept and content of smart city and the security of personal information, analyze architecture of smart city and how personal data is collected and processed, used within the architecture, realizing the purpose of combining the two hot topics of smart city and personal data protection.

This paper will further study the challenges and attack types of data protection in smart city. This will help to improve the existing problems of the smart city in the current construction. The analysis and conclusions of the study will also be beneficial to the theoretical research on personal data protection in smart city in the future.

Practical Significance

Based on the analysis of personal data security in smart city context, this paper will be possible to resolve the adverse effects of the construction of smart city on personal information security, and to guarantee personal data security in smart city construction at present.

By putting forward an integrated architecture on context-based data classification and protection on the basis of a three-dimensional security space and smart city architecture, this paper may provide basis of design and deployment of information security system in smart city.

1.2 Research Content and Methods

Research Content

The main content of this paper is the research of personal data security in smart city. One of the main features of this paper is that, the security system is regarded as a three-dimensional space where X-axis refers to security mechanism, Y-axis is the existing function layers in technical architecture of smart city based on Open System Interconnect reference model, and Z-axis is security services corresponding to the security mechanism. A combined research on smart city and personal data protection is carried out on this basis and a context-based personal data classification workflow is built inside. In summary, this paper is divided into six chapters. The contents of this study are as follows:

The first chapter is an overall introduction. It contains the research background, objectives and significance both theoretically and practically. In addition, this chapter covers summarizations of research content, and research method.

The second chapter is a comprehensive literature review towards related studies on smart city and personal data protection. This chapter is developed from different research focus of smart city, and the trend of research is extracted from data in Web of Science(www.webofscience.com).

The third chapter is to elaborate smart city from various perspectives. It contains four sub-chapters. They are: characteristics of smart city, technical architecture of smart city, supporting technologies of smart city and main services of smart city. This chapter aims to provide an overall illustration of smart city especially on technical aspect, which is the basis of proposed personal data protection architecture proposed in this paper.

The fourth chapter discusses the personal data protection in smart city. It analyzes the privacy concerns in smart city as well as the main types of attacks. After that, the concept of personal data classification is introduced and it answers following questions: what personal data classification is; why it is important and necessary; what types of personal data are involved in smart city services and how to classify them into categories. A matrix sample of the output is also provided to better understating the differences among categories.

The fifth chapter is all about the proposed architecture. This chapter first gives an overall introduction on the framework of proposed context-based personal data protection architecture based on smart city architecture, then elaborates each layer from a three-dimensional perspective, which is the operational principle, the security mechanism and the corresponding security services in each layer. The workflow of context-based personal data classification scheme is also covered. After that, we provide building block analysis of data protection mechanism and services in smart city based on Privacy Enhancing Technologies (PETs).

The sixth chapter is the last summary of the main body. There is the conclusion about the application of proposed architecture. It also points out the limitation of this research and the direction of future study.

Research Methods

In general, this paper is a qualitative research and according to the theme of this article, the research methods used are as follows:

Literature Research. This research method is based on two keywords: smart city and personal data protection, and collects relevant literatures published worldwide. so as to fully understand the history and status of the research on the theme and summarize the research results of the two keywords.

System Analysis Method. This research will systematically study the topic of the thesis. This research method is helpful to understand the development pattern, the development principle and the development pattern of research subject. The use of this kind of research method will be advantageous to the research and analysis of the personal data protection and its actual relationship in the process of smart city development. Especially in this paper, it is significantly beneficial to summarize the influence of the smart city to the personal data protection within the technical architecture.

Design-demonstration Method. This method is used to construct and evaluate the feasibility of new design. The proposition and explanation of the classified personal data protection architecture in smart city are mainly accomplished with the help pf this method.

1.3 Main Contributions

In this paper, we systematically focus on the personal data protection in complex smart city. The main contributions are:

Presentation of taxonomies of representative smart city services; the enabling technologies; types of personal data involved in smart city services; types of attacks.

Systematical elaboration on smart city technical architecture, and for each layer, identify possible privacy threats.

Presentation of different personal data categories matrix and the estimation flowchart.

Proposition and demonstration of classified personal data protection architecture and the context-based personal classification workflow, followed by the related building blocks analysis based on PETs.



Chapter 2. Literature Review

The study on smart city has emerged at the end of last century. After decades of efforts and researches, the theoretical research of smart city has been more mature. There are also attempts and breakthroughs in practice which spreading all over the world.

Fig. 1 Research Volume on Smart City (2013~2017)

When collect statistics of smart city related literatures, “TI=Smart* Cit* OR TIЈЅIntelligent Cit*” was used as search expression, and 2868 results were obtained. Most of the papers appeared in and after 2013. The total volume of paper released from 2013 to 2017 is 2773, exact amounts are shown in Fig 1 above. The growth rate began to increase in the past two years. Then the key words of these papers were exported to count word frequency. After eliminating domain words, these researches can be summed up into three themes.



2.1 Research on Concept and Connotation of Smart City

The research on the concept and connotation of smart city is mainly consisted of two different angles. One is to describe its construction target, and the other is developed around its supporting elements such as infrastructure construction, modern communication technology and so on.

Researches on smart city objectives are focused on positive convenience for citizens. R.G.Hollands (2008) pointed out that the definition of smart city is complex and difficult to standardize, but a smart city often links technological information revolution to economic, political and sociocultural changes[20]. P. Lombardi, S. Giordano and H. Farouh et al. (2012) put forward that smart cities attach importance to human and social capital based on modern information and communication technology to support urban economic growth and build wealth creation platform so as to improve the quality of people's life [42]. Angelidou (2014) put forward the concept of smart city, which is to enhance the development and prosperity of urban agglomeration by using human, collective and technological capital [35]. These studies point out the positive effects of smart city from different perspectives.

In terms of pluralistic supporting elements in smart city, researches mainly focus on infrastructure construction, modern communication technology utilization, innovation ability training and citizen quality improvement. In 2008, IBM defined smart city as a city that could make full use of the material infrastructure, IT infrastructure, social infrastructure, and commercial infrastructure, to produce collective wisdom. IBM believes that the key lies in the deep fusion of virtual network and above four aspects of infrastructures to realize the wisdom of the deep integration between the mutual connection and the real-time data[9]. In addition, J. R.Gil-Garcia, T. A. Pardo and T. Nam(2015), in their publication, described that smart city is a fairly complex system that contains many aspects of vision[9]. Moreover, R. Khatoun, S. Zeadally (2016) also shared the same opinion [48]. These studies have proposed a variety of supporting elements for building a smart city, among which the information technology has been given the greatest importance.

2.2 Research on Services and Techniques of Smart City

Smart city is a type of urban morphology based on the technology of Internet of things, cloud computing, artificial intelligence and so on. Jiong Jin, Jayavardhana Gubbi, Slaven Marusic (2014) et al. presented a framework for the realization of smart city through the Internet of Things which compasses the complete urban information system, from the sensory level and networking support structure through to data management and cloud-based integration of respective systems and services [25]. R.Wenge, C Dave, X Zhang(2014) et al. released a technology guide for implementation and design challenges that may exist in smart city, proposed a new architecture from the perspective of the data that underpins all the functionality of smart cities[51].

Apart from technique analysis, researches on smart city services has exploded in past two years. D. Menniti, N. Sorrentino, and A .Pinnarelli (2014), studied the coordination of smart grid in the virtual energy filed in future smart city, and they believed new information and communication technology will enable the existing resources to be managed more effectively in the next stage of smart city [13]. Schlingensiepen J, Mehmood R, and Nemtanu FC (2015) analyzed the framework of automatic transport system in smart city and it was considered as a core function in the city. Also, they designed framework for automatic transport system that could provide users with personalized mobile services in the intelligent environment [27]. I. A. T Hashem, V. Chang, N. B.Anuar et al.(2016) discussed the feasibility of applying big data to the construction of smart citiy, suggesting that big data can obtain potentially valuable clues from a large number of data via different data sources and be helpful for city's decision-making system[61].

Besides, there are studies of smart retirement (such as the AmIE Project) using citizen's information system to relieve the worsening of population aging. To better alleviate the problem of urban parking jam, there are designs and solutions about smart parking to improve the efficiency of parking operation and optimize parking space usage. With the gradual improvement of the smart city prototype and the renewal of related technology, people start to focus on analyzing these specific urban services constructions.

2.3 Research on Privacy Issue in Smart City

In recent years, as smart city has become more mature and universals, issues related to privacy and security have come into limelight. Li, Dai, Ming, and Qiu (2015) identify the over-collection of data as a severe security risk, especially when it comes to the sensitive data that people hold on their smart phones [64]. Martinez-Balleste, Perez-Martinez, and Solanas (2013) similarly fear for the privacy of citizens in smart cities, especially when it comes to protecting information about their identity, the kind of information they look for, their location, energy usage and possessions [7]. Privacy scholars offered solutions to privacy risks in smart cities focus on particular technological solutions, such as cloud computing analyzed by Kahn, Pervez, Ghafoor (2014) [68], privacy enhancing technologies studied by Rebollo Monedero, Bartoli, Hernбndez Serrano et al. (2014) [2]. Wright and De Hert (2011) turned to privacy impact assessments as a tool to identify whether a specific technology or applications involves a privacy risk and how this can be mitigated [14].

The good environment of smart city information security will benefit from long-term research and so far, the researches regarding to personal data protection in smart city can be mainly divided into three categories:

Situation analysis on information security in smart city

By investigating the status of information security in smart City, Silva and Guimaraes (2014) discussed the new challenges of information security in smart city context and concludes nine information security issues [56]. Cilliers and Flowerday (2015) used the quantitative analysis method to study the relationship between the citizen's privacy, information security and the credibility of the crowdsourcing system in the current smart city, the results show that the information security of smart city has positive correlation with the credibility of the public-packet system. Privacy in crowdsourcing projects in smart cities can be mitigated by enhancing the credibility of the system [30]. AS Elmaghraby and MM Losavio (2014) examined the privacy and security challenges in smart city based on an interacted model [55]. Kuan Zhang, Jianbing Ni, Kan Yang et al. (2017) published a paper about security and privacy challenges in smart city applications within the smart city architecture [29].

(2) Factor analysis on information security in smart city

In their another research, Silva and Guimaraes (2014) summarized the information security issues that have an impact on smart cities, expounds the consequences of these security issues in terms of confidentiality, completeness and usability. The new concept “Interoperable security” was proposed to summarize the most prominent elements of information security in smart cities [57]. Paul Wang, Amjad Ali and William Kelly (2015) analyzed the data security and threat factor modeling in smart city and proposed solutions that can be used to reduce threat factors [44]. S Ijaz, MA Shah, A Khan et al. (2016) investigated the security concerns in smart city and provided a categorization that based on several factors such as governance, socioeconomic and technological factors [58]. Cilliers and Flowerday (2017), in another cooperated paper, studied the decision-making process of the public reporting information security events in the smart City IVR system, the results show that the information security factors in the smart city's information system, such as confidentiality, integrality and usability, have a prominent influence on the public risk perception, and the most important is the integrity factor of the system [31].

(3) Strategy Analysis on Information Security in Smart City

A Bartoli, J Hernбndez Serrano et al. (2011) released an overall analysis on open issue of privacy and security, it is observed that privacy can be achieved i) by imposing high security requirements onto the used technology to avoid third party abuses; and ii) by decoupling technical smart city data streams from the personal one to avoid abuse of data by insiders [2]. HC Pцhls, V Angelakis, S Suppan et al. (2014) analysis the key technologies applied in RERUM, which is a European Union project that aims to develop a framework which will allow IoT applications to consider security and privacy strategy early in their design phase [10]. S Chakrabarty and DW Engels (2016) presented a secure IoT architecture for smart city that contains four IoT architecture blocks, which enables the mitigation of cyber-attacks from the IoT nodes [52]. There are also researches of smart city privacy based on cloud. For instance, Z Khan, Z Pervez, and A Ghafoor (2014) proposed a framework that enables end-to-end security and privacy features for trustable data acquisition, transmission, processing and legitimate service provisioning based on cloud [68].

Due to advanced level of economic development and system perfection, smart cities in developed countries have better information security environment. Early in 2002, the United States passed the Federal Information Security Management Act (FISMA), establishing a perfect standard for city information security through a number of standardized organizations. In 2013, the American National Standards Institute, with the standard development units, the information security enterprises and the academic experts, launched a discussion on the information security of smart city, and issued a series of information security management standards, such as “AAMI TIR57:2016”, “ANSI/NEMA SB” etc. Meanwhile, the US has established a comprehensive infrastructure with information security monitoring and early warning services, and implemented classified monitoring for information system security and Cyberspace Security. On the other hand, the National Institute of Standards and Technology (National Institute of Standards and Technology, NIST) has issued a number of guidelines on information security awareness education and training, focusing on the "safety awareness" and "security basic culture" of citizens' information security. And advocated the establishment of a flexible and skilled quick response team to deal with information security incidents. Other developed countries, such as Japan and Britain, are also more prominent in the information security work of smart cities. They have formulated a perfect information security monitoring system, strengthen the laws and regulations to protect the personal information of citizens, improve the awareness of information security, trained special personnel in the field of information security, and actively participated in the field of international information security in the same time.

Besides, there are also certain number of researches regarding to the smart city factor analysis from diverse angle of views, such as objective perspective, driving force perspective and utility perspective. As the most creative urban morphology, smart city has become a strategic choice for global cities to seek development. Now there are more than 200 smart city projects in the world. At present, the proportion of smart city construction is also increasing gradually in the related research of smart city.



Chapter 3. Smart City

Smart city is a new urban form of innovation and new information technology development and growth in the social innovation environment. It is the product of the integration of traditional urban civilization and new knowledge and technology. The smart city will greatly change the way of urban management and life and ensure the sustainable development of the city. The construction and development of smart city is a new driving force for economic growth and an important turning point in the transformation of economic development mode. It is also an important way to improve the quality of urbanization and alleviate the "big city disease".

3.1 Characteristics of Smart City

Smart city is equipped with various technologies and characteristics. Generally, the Smart city has the following characteristics:

(1) Perception and interconnection. Through the intelligent sensor, the physical space of the city is fully perceived, and the real-time monitoring of the operation system in the city is carried out. The intelligent sensor network is connected through Internet, to realize the further storage, management and use of the data.

(2) More concentrated on infrastructure construction. Through the integration of Internet, Internet of things technology, satellite sensor network, and the integration of multi-source and heterogeneous data, all of these constitute the information infrastructure of smart city and ensure the healthy development of smart city from the source [43].

(3) Mobility and openness. With the universal application of the ubiquitous network and handheld terminal, mobile becomes an important feature of the smart city. Widely used wireless technology in smart city, such as the electronic tag of the perception layer of the Internet of things, is sometimes limited by cost, which leads to weak password mechanism. Therefore, the internal data of the electronic tag is easy to be cracked [23]. At the same time, many institutions construct the institutional network on the public network through VPN and other ways. This kind of openness leads to increased security risks.

(4) High penetration and increased privacy risk. The scale of Internet of Things and wireless broadband network has been greatly increased, and the constraint of time and location about people's use of network has been broken. Smart applications attract people to actively participate in the creation and release of information, or in certain network operation and other links. In this situation, penetration levels of the smart city to human society is greatly improved. At the same time, the construction of smart city is people-oriented, involving privacy data, including personal basic information, personal preference, personal location, personal behavior data, etc. High penetration has caused a dramatic increase in the risk of personal privacy protection.

3.2 Technical Architecture of Smart City

Because of the wide scope of smart city, the definition of smart city has not been standardized yet, therefore the proposed architecture of smart city may diverse accordingly. There are already many proposed smart city architectures based on either Internet of Things or cloud. This chapter will use the most commonly accepted technical architecture of smart city and elaborate elements and building blocks of each layer into details.



Fig. 2 Sample of Technical Architecture of Smart City

The technical architecture of smart city is shown above. The architecture periphery are: on the top is the serving objects, including the public, the enterprise user and the government manager. Different access channels will focus on serving objects and integrate together to achieve multi-channel access; at the bottom is the peripheral natural environment, which is the data acquisition source of the whole architecture.

The core of the technical architecture of smart city is, from the perspective of the overall construction of urban informatization, puts forward four levels and two supporting systems. The upper level of the horizontal elements is dependent on its lower level, and the vertical support system has constraints on the four horizontal layers [1] [21] [34].

Physical Perceptual Layer

In the general framework of smart city technology system, the physical sensing layer is responsible for sensing and collecting all aspects of the data in the urban environment, processing and automatically controlling the collected information, and directing the data to the right location through the communication module. It provides intelligent perception and execution ability to the environment, and realizes the identification, information collection, monitoring and control of infrastructure, environment, equipment and personnel in the city based on IoT technology.

Fig. 3 Constitution Elements in Physical Perceptual Layer

The commonly used sensing units are RFID, IP Cam, mobile, GPS, remote sensor, radar and other sensing devices which can be equipped to monitor the outer environment and activities. At present, there are many mature, stable and widely used wireless communication technologies that can be applied in network transmission of perceptual layer, such as 3G/4G, Wi-Fi, Zigbee, 6LoWPAN, Bluetooth, Ethernet and so on. Choosing the right network can ensure the efficiency and accuracy of information interaction between devices, since different networks have diverse characteristics which are appropriate to be deployed in different circumstances. Afterward, the collected massive data will be connected to different access gateways.

Network Communication Layer

The main goal of the network communication layer of smart city is to build a universal, shared, convenient and high-speed network communication infrastructure, which can provide the basis for the flow and sharing of city data. It is responsible for transmitting, routing and distributing the perception data in smart city.



Fig 4. Heterogeneous Communication Network in Smart City

This layer focuses on the Internet, telecommunications networks, radio and television networks, and the integration of three networks (e.g., mobile Internet), therefore to equips the smart city with large capacity, high bandwidth network coverage by the help of GSM/GPRS, WLAN, 3G, 4G TD-LTE technologies. The 4G, represented by TD-LTE, has the capacity of ultra-high wireless bandwidth. And the forthcoming 5G, which will be deployed very soon, would dramatically improve the network performance in smart city.

Apart from communication technologies, private network technologies (WSN, Social Network, etc.), transmission control technology (real-time connection, Multi-source Aggregation technology, etc.) also make contributions in this layer.

It is worth noting that, in addition to networks shown on the above graph, vehicular ad hoc networks, a special class of Mobile Ad Hoc Networks(MANETs), can also be formed via vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communications. It allows online monitoring of travel time, original destination route selection behavior, air pollutants and noise emissions, as well as prediction of possible accidents [26].

Data Layer

The data platform is usually cloud-based because the massive volume of urban data is nearly impossible to store in physical disks when taking the investment into consideration.

Fig 5. Multi-tiered Framework in Data Layer

There are two main tasks in this layer. One is data storage and maintenance, and another is data computation and vitalization for service integration. Both of them are the core in whole architecture. The city's data resources include urban basic data such as population-related data, geographical data, Internet data or specific industrial data. In order to realize efficient and effective operation in smart city, these datasets shall be stored and maintained separately.

This layer is also responsible for data analytics and vitalization. The analytics modules include: prediction modules, description modules and decision-making modules. These modules enable the categorization and aggregation of massive urban data, and achieves data activation through data association, data evolution, data mining and data maintenance technology, providing active data to service integration.

After data being vitalized, this layer further encapsulates data into smart city services and provides reusable and flexible deployment capabilities for the development of upper-level applications, which includes cloud platform, visualization and simulation, public data engine and other platforms and services. This layer is in the upper middle of the architecture, which has an important role to link up.

Application Layer

Through the integration of data and services, related applications can be carried in smart application layer. The application layer is the foundation to realize and derive value from the data generated from the city. The city's stakeholders can use collected data to analyze, take decision and modify it for specific purpose, and the role of a smart city in the application layer is more like specifier than application developer.

Fig 6. Services Types and Top Users

The end of intelligent application is the business requirement of the specific field of data, which is integrated processing of all kinds of perceptive information in time: intelligent analysis, auxiliary statistics, prediction, simulation and so on. Smart city application server running on smart city big data center, and has a website, mobile phone/tablet APP for public access, to provide more refined and smart services for government, enterprises and citizens.

Security System

The city's security and privacy are at great risk since various types of attacks may happen in each different layer. For instance, the sensors or other scanning devices in perceptual layer might be hacked to produce false data, or, the third parties or authorized parties may have the access to data which they should not have. Meanwhile, people and devices are getting connected more closely at a larger scale, which means the construction of smart city needs a refined information security system, in order to improve the safety and control level of urban basic information network, core information system, providing reliable information security environment for the city. From a technical point of view, a perfect security system shall provide all kinds of necessary authenticity, confidentiality, integrity and reliability protection, which are involved in every horizontal layer mentioned previously. And in each horizontal layer, the protection strategies shall consist of physical security, system security, network security and data security [66].

More extendedly, a security system in smart city can be designed as a three-dimensional architecture, in which X-axis refers to security strategies, Y-axis is based on the OSI (Open System Interconnect) model and Z-axis represents different security services that can be deployed and implemented according to strategies. In specific, the security mechanism outlines the major security concerns in each level, and the OSI reference model is the base on which the network protection can be carried out in all aspects and levels. And the security services cover full set of security supports from all levels of the network. Alternatively, the OSI reference model, in which seven-layers framework of network interconnection is defined, could also be replaced with the four layers in the technical architecture of smart city. The three-dimensional security system can expand gradually with the expansion of the network, and the scope of security is gradually enriched.

The smart city security mechanism includes public infrastructure security, smart application security and smart application platform security, data access security, network transmission security, smart city operation security, smart management security, authorization and audit security, and information security detection system.

Smart city security services include peer authentication services, encryption services, data integrity services, data source authentication services, and so on.

Security technology in smart city includes identity authentication technology, encryption technology, digital signature technology, access control technology, data integrity technology, authentication technology, data mining and analysis technology.

Operation Management system

Operation management system is mainly responsible for normal operation in various types of hardware, software, data center and the necessary parameter configuration, state monitoring, logging, abnormal warning, troubleshooting and repair, including configuration management, event management, fault management, log management etc. The whole system involves various horizontal levels to ensure the overall construction management and long-term operation of the smart city. The design and maintenance of operation management system of smart city is an important guarantee for the smooth development of city construction.

3.3 Supporting Techniques in Smart City

Internet of Things

Technologies of Internet of Things is the use of advanced intelligent sensing technology and equipment, perception, analysis of certain behavior in physical world, then make use of the network transmission technology to transmit these data. Through computing, processing and information mining, we can complete information communication between people and objects, or among objects. With the support of Internet of things, we can manage and control the physical world scientifically and make effective decisions. In a simple way, the Internet of things technology is the further derivation and expansion of communication network and Internet technology. The structural framework of the Internet of things can be divided into three parts in general: the perception layer, the network layer and the application layer. Specifically, the perception layer part is mainly to analyze the behavior of the physical world; The network layer is responsible for data transmission of network layer; The application layer is to use information to calculate and process related data and control the application of various industries effectively. The most critical technologies in the application of Internet of things are RFID and sensor fusion technology, recognition and environment awareness technology, and Internet of things interaction and control technology.

The application of Internet of things technology has been reflected in every aspect of smart city, providing good support for the construction and development of smart city information system. For example, in agriculture, people use the technology of Internet of things to complete the irrigation of farmland, the monitoring of crop growth and the tracing of crop circulation. In business, the most common application of IoT technology is the vending machine, POS terminal etc. In addition, the Internet of things technology has been involved in the fields of transportation, electricity, medical, education, home and environmental protection [60]. The sensor assembly equipment to the sensor equipment using the embedded technology. Through networking technology, the sensors penetrate the city infrastructure in various aspects and areas, provide all kinds of information and data and decision analysis services for city construction, and realize the city's monitoring and early warning, real-time monitoring and management, to provide data support for the construction of smart city top-level design.

However, there are still some problems in the process of its application, mainly due to the isolation phenomenon between different regions and different industries. At present, it is not yet possible to form an integrated cooperation platform. The main reason for this problem, except for the different economic conditions, is that there is no uniform standard for all systems. On this basis, cloud computing is an effective solution to the storage and calculation of a large number of data.

Cloud Computing Technology

With the acceleration of economic and social development, a lot of data and information have been generated. Especially after the universal application of Internet technology and IoT technology, the requirements for information processing, cost calculation and intelligent applications are constantly improving. Under this trend of development, cloud computing technology came into being and has been developed rapidly. Cloud computing is a new application technology, which has a new understanding of the traditional computing method and business operation mode. Cloud computing technology, through virtualization, distributed storage and parallel computing, makes use of broadband network technology to form plug and play(PNP) processing mode, so as to complete the needs in processing different service forms. Moreover, using cloud computing technology, smart city intelligent service control system will better accomplish all kinds of service tasks, including data processing in different industries of the city, controlling of industrial production activities and precepting of environmental information. Through the use of the cloud computing system, the utilization of resources is effectively improved, and the requirements of the fusion of user terminal and cloud computing technology are reduced. At present, the user can use the simple terminal system of the embedded chip to complete the complicated computing process [67].

Big Data Technology

In the construction of digital and smart city, a large number of data resources will be produced across regions, across sectors and across departments. As time goes on, the magnitude of storage is increasing, and whether there is a useful information in these data resources has become a problem for people to think about. The application of big data technologies in smart city are involved in data collection, data storage, data processing, data mining, data visualization etc. Among them, data mining enables people to analyze each data and find its rules or extract valuable information from a large number of data [6]. In the process of data mining, it is necessary to preprocess the data, such as collection, integration, specification, cleaning and transformation, then carry out data mining, model evaluation, knowledge representation and other mining implementation processes, to complete the analysis of association, clustering, classification, anomaly, specific group and evolution, finally realizes the application of big data in data visualization, data storage, data retrieval and data intelligence etc.

Traditional machine learning and data mining perform inefficient when dealing with massive and heterogeneous smart city data, which needs to be improved through distributed computing framework. According to the real-time requirement, data processing and mining can be divided into offline processing and online processing. The processing of off-line data is usually stored in distributed databases or distributed file systems, and can be analyzed through distributed computing technologies like Apache, Mahout, etc. For data that needs to be processed online, such as stream data. The method used is very different here since the primary requirement is real-time. The memory database has fast access speed, and can be combined with the data processing engine to improve the processing speed. For example, ComplexEventProcessing (CEP) is a data processing engine to capture the various simple events or event streams of different sources. According to the predefined event model, the relationship between massive events or more meaningful events can be found in real time and efficiently, and the decision basis is then provided.

Besides, other technologies of big data include artificial intelligence technology, data specification technology, statistical analysis technology, database technology, big data storage and management technology are also widely applied in the construction of smart city.

3.4 Services in Smart City

At present, there are already certain quantity of well-applied services regarding to smart city construction, such as smart travel, smart education, smart medical, smart environment etc. Various of smart applications would be developed and designed in the near future. Some representative services would be listed and described as follow [12].

Smart Bank

Smart bank is a kind of bank management model which is in the advanced stage of intelligence. It is characterized by the effective use of new technologies, the effective allocation of banks to their resources, and the real-time monitoring and management of customer needs, providing personalized help for changing customers' needs. The way for the development of smart banks is to improve and upgrade the traditional banks. By combining traditional services with information technology, the services provided by smart banks will be more intelligent.

2) Smart Logistics

Smart logistics is a modern integrated logistic system based on the existing theoretical research, IBM's explanation and other related contents. The technological support of smart logistics is information technology, which will ensure smart logistics to achieve comprehensive information analysis and management in a series of environments such as packaging, transportation, storage, loading and unloading, distribution to households. At the same time, it also ensures that the smart logistics system will carry out the function adjustment in time to realize the standard, modern and intelligent service of the logistics industry.

3) Smart Healthcare

Smart healthcare is a smart project that combines life science theory with modern communication technology. Compared to traditional medical treatment, smart healthcare can ensure patients get timely consultation and service related to medical treatment. Its service model is based on patient data, and can be divided into three stages: data acquisition, knowledge discovery and remote service. Among them, data acquisition is completed by the Internet of things, knowledge discovery relies on big data processing technology, and remote service is provided by cloud service and portable intelligent medical terminal.

Smart healthcare includes the efficient and effective provision of health care. For example, smart medical centers can combine patient health records from multiple sources and thus improve medical care. Smart healthcare can also rely on data from connected medical devices and wearables. Lastly, smart healthcare aims to empower patients by granting them access to their own health records and information about their conditions.

...