DESIGNING AND SECURING DISCOVERY SERVICES FOR EPCGLOBAL NETWORK IN SUPPLY CHAINS

DESIGNING AND SECURING DISCOVERY SERVICES FOR EPCGLOBAL NETWORK IN SUPPLY CHAINS

ABSTRACT

Internet of Things (IoT) is an emerging, global scale, Internet-based information service architecture enabled by the Radio Frequency Identification (RFID) and sensors technologies for information sharing and corresponding service discoveries. The EPCglobal network is considered as a sub-project under the IoT umbrella, which was created to develop a universal identification system and an open architecture to provide interoperability in a complex multi-vendor scenario.

This universal identification system is based on the allocation of a unique ID—Electronic Product Code (EPC)—to every item. The EPC key is a globally unique identifier that is carried by the RFID tag. As a result, the EPCglobal network is an architecture proposed for enabling the sharing of information about the individually identifiable objects among organizations. As such, the life history of each individual object is shared and linked to an object through the unique EPC key. Since a huge volume of data is involved, a suitable service oriented architecture (SOA), which is called EPCglobal Discovery Service (EPCDS), is needed for locating both the key and additional information associated with it. Although several EPCDS models have been introduced by existing studies, most of them are in the very early development stages and cannot fulfill the design requirements, especially in advanced services, network performance, and security aspects.

This dissertation introduces an advanced EPCDS model, which centers around three main tasks. First, we discuss the architecture of EPCglobal network and the needs for advanced discovery services. We examine the detailed EPCDS design requirements that are classified into three categories, functionality and service, network performance, and security and privacy. We select several typical EPCDS models that are proposed in other studies for detailed discussion, and then summarize and analyze current EPCDS models against the design requirements. The conclusion points out how to take advantage of current EPCDS designs to eliminate the gaps for further exploration.

Next, according to the design requirements and the assessment, we propose an advanced EPCDS design. The enhanced EPCDS network is based on a peer-to-peer infrastructure and a consistent hashing method, adopted for the EPC key assignment function. It enhances the functionalities and services of the system, especially core functions such as publishing and lookup services. Our model can fulfill all the design requirements. In particular, compared to other models, the design is a highly distributed network, which is able to provide both basic and advanced services, as well as superior networking capability, such as reliability, scalability, extensibility, etc. It could mitigate some of the aforementioned shortcomings without completely abandoning the established standard.

Finally, we investigate security threats and vulnerabilities for the RFID-enabled EPCglobal network in detail. We categorize the security issues into several categories, and discuss how to resolve them. To address the challenges posed by complex and dynamic EPC objects and events, we propose a relationship-based federated access control model for EPC Discovery Service, which can be tightly integrated with the proposed EPCDS infrastructure, to fulfill all the security and privacy design requirements. To the best of our knowledge, this is the first study that identifies and examines in detail the causes and impacts of key publishing security threats and interdependent security issues in an EPCDS context.

 

Chapter 1:  Introduction

1.1. Problem Statement

The convergence of the Internet, wireless communications, sensor networks, information processing, and integration technologies has given birth to a new field of study and applications called the Internet of Things (IoT) (Gershenfeld et al., 2004). This paradigm has been widely adopted and used in our daily life, such as in healthcare, transportation, energy consumption, entertainment, the environment, and homes.

The IoT describes a collective Internet-based information service architecture that is deployed globally. In the vision of this concept, all the items are equipped with a Radio Frequency Identification (RFID) or similar function tag to increase transparency and facilitate the information exchange of the products in the global supply networks. Besides the initial application scope, IoT ultimately aims to serve as a backbone for ubiquitous computing, enabling smart environments to easily recognize and identify objects, and retrieve information from the Internet.

As the IoT is growing rapidly, a huge amount of data of real-world objects and events will be available through the association of RFID technologies. RFID is one of the fundamental technologies for automatic identification, data capture, and object tracking. It has been quickly emerging as a popular technology that can be broadly applied in many places. It is able to facilitate the IoT in exchanging information and the management of merchandise and resources in the supply network. As a result, the EPCglobal network was designed for the purpose of standardizing and diffusing the use of RFID and sensor technologies, and is considered as a subproject under the IoT. In this network, the Electronic Product Code (EPC) is a unique identifier (ID) embedded into each RFID tag (EPCglobal, 2010b) that is attached to every item. It aims at developing a universal identification system and an open architecture to provide interoperability in a complex multi-party environment. The greatest promise is to enable the seamless sharing of data and item-level visibility for all the participants across the supply network, and to eventually achieve the blueprint of the IoT that every participant could easily share and access valuable business data in a secure manner.

On the other hand, the EPCglobal network is also constrained by several limitations, for instance, high cost, lack of standardization, security and privacy issues, information sharing issues, the voluminous data (Big Data) issues, and the scalability issues (Michael and McCathie, 2005; Juels, 2006). In order to take the advantage of the EPCglobal network, information from different organizations should and must be shared for facilitating each business to achieve a smart and highly networked environment. EPCglobal is a global scaled network involving vast volumes of data generated by the RFID and sensor devices, and stored in widely distributed and heterogeneous databases. In order to respond to the worldwide query demands from business partners and individual consumers, a discovery service system is needed to provide the relevant services. While EPCglobal is improving people’s daily life, problems such as information security and large-scale data management are critical and should be studied before the EPCglobal systems are deployed. A highly efficient and scalable model for discovery services is the foundation for this ubiquitous computing.

Although the EPCglobal architecture defines a functional component called EPC

Discovery Services (EPCDS), no detailed specification has been developed, and many problems

(e.g., standardization, high cost, limited services provided, and inefficiency) remain to be solved.

In addition, a systematic and explicit design requirement framework is missing. Along with the information sharing benefits generated by the network, there are serious security challenges due to the lack of appropriate protections. Several discovery service models have been proposed to meet different functionality and services requirements (to be reviewed in details in Chapter 3); most models, however, are designed as intermediate brokers for simple lookup services, which cannot fulfill clients’ expectations and needs. Also, they can hardly convince us that they have good network performance, and are able to handle global-scale distributed data. Thus, a comprehensive and detailed discussion and assessment of each model is needed for directing future design and implementation. Furthermore, a new EPCDS design is expected to provide more of the favorite functions and services, improve networking performance, and make the EPCglobal network more secure.

Finally, there is the issue of security and privacy, and in particular, the data within the EPCglobal network, such as information about price, location, volume and flow of products, and relationships between trading partners, etc., is always commercially sensitive and complicated. In the case of missing the proper access control mechanism, this information can be accessed by any parties that include the hostile entities and the competitor organizations. Because of such concerns, certain security countermeasures, such as key publishing control and federated access control, should be integrated with the EPCglobal network to allow network participants to obtain the power of controlling their own data, since the information is decentralized across multiple organizations (Vanalstyne et al., 1995). While the security and privacy of each organization is guaranteed, they can share their information with more confidence, and thus obtain further

benefits.

1.2. Background and Motivation

1.2.1. Internet of Things and RFID Architecture

RFID is a communication and identification technique known at least since the Second World War from friend-or-foe identification systems of military airplanes (Rieback et al., 2006). However, it has been used in more and more other domains since then, e.g., animal or human identification, anti-counterfeiting, and access control payment to global supply chains (Garfinkel and Rosenberg, 2005). Currently, it has been considered as an important underlying technology for smart and ubiquitous computing. The term IoT refers to the networked services that speak about things, rather than services that reside inside the objects themselves (Liu and Albitz, 2006). People are interested in the data and information associated with these things. This valuable information will be available on the internet because of the IoT. It has been noted that the cost of establishing a real IoT is not affordable under current situations; in order to make this dream come true, people have put their hope in a simple chip called the RFID tag, which is mostly externally powered and communicates via radio waves issued by RFID readers (EPCglobal, 2007a). In the recent years the RFID technology has become cheaper and cheaper, and the massive deployment of these tags will most likely be achieved in the near future. They work with the RFID reader, which interrogates all the tags that are close to it via radio waves of a specific frequency. The reader also provides energy to the passive tags. The reader and tags follow an anti-collision protocol to initiate the probe, and then the tags return the data to the reader. Because of the capability of the passive tags, they can only do simple operations and store very limited data, e.g., EPC keys. Besides carrying the identifier data, most of other processes are done by the back-end system and the application layer.  It enables these data to be accessed via the Internet,

e.g., via different companies’ Web services or EPCDS’s query interfaces.

RFID systems can be very complex and may be integrated with many other applications. Traditionally, a simple RFID sub-system includes hardware and software, as shown in Figure 1.1. The hardware is composed of tags, readers, and a back-end system. The software system contains a user-interface, middleware, and database system. An RFID enabled network, such as an EPCglobal network, is constructed by many RFID sub-systems that are deployed in different stages of supply chains.

 

Figure 1.1 RFID Sub-System Architecture

An RFID tag is a small electronic device that consists of a small chip and an antenna to carry and transmit useful information, such as the identity data. The major characteristics of an RFID tag include identifier format, power source, operating frequencies, functionality, and form factor (Karygiannis et al., 2007). According to these characteristics, the RFID tag can be divided into several categories, which greatly differ in many aspects, e.g., the cost, size, performance, and security features. Because of such variety of the RFID tag, it will always need to be customized in order to meet the requirement of certain applications.

Generally, the RFID reader can read the RFID tag, and in some cases write into the tag. It is obvious that the communication cannot happen between different types of RFID tags and RFID readers, unless they are under the same communication protocol. Basically, the RFID tag and reader that are produced by different vendors cannot connect with each other. The prominent international RFID standard is the EPCglobal Class-1 Generation-2 standard, which is essentially equivalent to the ISO/IEC 18000-6C standard.

After the RFID reader extracts the data from the RFID tag, it sends the data to the backend system. The back-end system is responsible for the storage and maintenance of that information. It contains two major components: the middleware and the analytic system. RFID middleware is responsible for preparing the collected data for the analytic systems. For instance, the middleware filters duplicate, incomplete, and erroneous information that it receives from readers. The middleware can immediately transfer the filtered data to the analytic systems, which aggregate and store them for later retrieval. It also can be used to monitor and manage readers. For example, system administrators use middleware to adjust the power output and duty cycle to reduce the number of transaction errors. Analytic systems directly support business processes. They are mainly composed of Web servers and databases, and support each customized business logic. In the EPCglobal context we normally consider the backend system as a database.

1.2.2. EPCglobal Network Architecture

EPCglobal is an activity of the global not-for-profit standards organization, which supports the global adoption of the electronic product code (EPC) and related industry driven standards to enable accurate, immediate, and cost-effective visibility of information throughout the supply chain. The EPCglobal Architecture Framework is a collection of interrelated hardware, software, and data standards, together with shared network services that are operated by EPCglobal, its delegates, and others, all in service of this common goal (EPCglobal, 2010a).

Traditionally, EPCglobal network is composed five major components: EPC tag, RFID reader, Application Level Events (ALE), EPC Information Services (EPCIS), and EPC Object

Name Service (EPCONS) (see Figure 1.2.)

 

Figure 1.2 The EPCglobal Network Architecture

The working process is the following: the EPC tag will capture the events of the physical object exchange among different organizations. The captured raw data will then be processed and delivered to EPCIS by ALE, which was originally called Savant. ALE is essentially a middleware performing the counting, logging, accumulation, and filtering operations on the data. EPCIS contains two major components—the EPCIS repository and EPCIS query interface. The filtered data will be stored in the EPCIS repository. When the users in the EPCglobal network request information about a specific object, a query will be initiated via the EPCIS query interface. The interface will then interact with the upper layer applications, such as the EPCONS and EPCDS, to extract the target resources.

The EPC is a unique identification number that is used to retrieve information associated with the object in the supply chain. The EPC Tag Data Standard (TDS) (EPCglobal, 2010b) defines several encoding URI syntax schema, such as serialized global trade item number

(SGTIN), Serial Shipping Container Code (SSCC), and General Identifier (GID). EPC tags are

potentially the most important class of RFID tags, and constitute the physical embodiment of the EPC to be attached or integrated into supply chain pallets and transporting cases, and possibly to all applicable manufactured single goods in the future. An example of an SGTIN-96 encoded schema—the most popular standard, the serialized global trade identification number (SGTIN)— is given in Figure 1.3.

 

Figure 1.3 EPC Discovery Service Framework

In this SGTIN-96 variant, the EPC includes a header to denote its EPC identity type, e.g., here is SGTIN-96, a filter value for fast logistic decisions, a partition value that indicates the boundary of the next two fields, and a Company Prefix number (CPN) that is a unique identifier of the item’s manufacturer. Furthermore, the manufacturer can assign item reference numbers to classes of objects that belong to it. Within the same class, similar objects can be distinguished by their serial number, which is a fundamental extension compared to the conventional barcode. In our dissertation, we will use the SGTIN-96 coding schema for our examples.

Data exchange between end users in the EPCglobal architecture is facilitated by EPCIS, which encompasses both data exchange interfaces and the data itself. The goal of EPCIS is to

enable disparate applications to leverage EPC event data via information sharing within and across organizations. Ultimately, this sharing aims at enabling participants of the EPCglobal network to gain a shared view of the disposition of EPC bearing objects within a relevant business context (EPCglobal, 2007b). Specifically, it also provides a standard query interface for retrieval of detailed EPC event data that is stored within the EPCIS repository. Much of the data provided by EPCIS consists of events such as observations of the object in particular locations within the premises of a company, as well as actions performed on it, e.g., packing, unpacking, shipping, and receiving. It may also include other attributes and sensor measurements associated with the object or its environment, e.g., price, quality, temperature, and humidity. The event data at the EPCIS level may be much richer than data provided at the ALE level because the EPCIS data model can answer more than the basic questions. Data stored in the EPCIS allows a number of additional descriptive fields to be specified per event and to provide annotations about the additional business context corresponding to each event. In our dissertation, we consider EPCIS as a repository that stores detailed information about an EPC event.

Finding information about a tagged object in an open system is a challenge because the EPC event data could be maintained by any EPCIS repository. To solve this problem, EPCglobal declares a component called ONS, which leverages the existing Internet standards and infrastructure. ONS uses the existing Domain Name System (DNS) for looking up/resolving information about an EPC (EPCglobal, 2008). The only purpose of ONS is to offer a lookup service, which provides no information except the address of the manufacturer’s EPCIS repository. It encodes complicated services in a standard way, which allows the service endpoint to be expressed as a URI (uniform resource identifier). The raw EPC sequence of bits will be resolved and converted, and the query and response formats must adhere to the DNS standards to enable the DNS to find the corresponding EPC data. After resolution of an inquired EPC, the root

ONS retrieves the local ONS and returns the address of relevant EPCIS servers. Figure 1.4 illustrates the ONS looking up process, which contains the following steps (EPCglobal, 2008):

 

Figure 1.4 ONS Looking Up Process

1. A sequence of bits denoting an EPC is read from a 64-bit RFID tag.
2. The tag reader sends that sequence of bits to a local server.
3. The local server converts the bit sequence into the pure identity URI form as defined in EPC TDS (EPCglobal, 2010b).

Example: urn:epc:id:sgtin:0614141.000024.400

4. The local server presents the URI to the local ONS resolver.
5. The resolver converts the URI form into a domain name and issues a DNS query. Example: 000024.0614141.sgtin.id.onsepc.com
6. The DNS infrastructure returns a series of answers that contain URLs that point to one or more services, e.g., an EPCIS Server.
7. The local resolver extracts the URL from the DNS record and presents it back to the local server. Example: http://epc-is.example.com/epc-wsdl.xml
8. The local server contacts the correct EPCIS server found in the URL for the

EPC in question.

In some cases when the EPCIS address associated with the EPC, which is not known in advance due to the fact that it is the first time it appears in the network, or the EPC record is out of date though it is already exists in the network, the local ONS will consult the root ONS (in step 6). The root ONS contains the addresses of the local ONS, which is managed by an EPCglobalassociated organization. When the root ONS receives the query, it will identify the local ONS by recognizing the company prefix part of the DNS encoded EPC URI. Then it delegates the query to the identified local ONS server.

1.2.3 Motivation

Although the ONS can offer simply a lookup service, there is still a need for EPC

Discovery Services (EPCDS) due to following reasons. In the EPCglobal specification documents (EPCglobal, 2010a), it describes the Discovery Services as a mean to locate EPCIS resources in the most general situations arising from multiparty supply chains or product lifecycles in which several different organizations may have relevant data about one EPC, but the identities of those organizations are not known in advance. Besides the functional and security concerns of the ONS that we mentioned above, it is also restricted by some other factors that make the discovery services necessary. We summarize the main differences between the EPCONS and EPCDS in Table 1.1.

The ONS is a lookup service useful for finding the address of the original EPCIS resource of a certain EPC tag. It does not address the issues of discovering the set of EPCIS data sources that may contain information about a particular EPC or a set of EPCs. Therefore, ONS is the authoritative directory of information sources, which like a static pointer to point out an EPC’s head position in the supply chain, i.e., the manufacturer of the product. In the meantime, the EPCDS records all the history locations that an EPC has passed through, and thus it allows users to find the detailed, real-time, as well as the historical information about a certain EPC. The ONS usually stores the item’s class level information—for instance, the item’s ratings, reviews, price, availability, and specifications. However, the discovery service is able to store the serial level links for a specific item, such as traceability, authenticity, warranty registration, ownership, customization, and so on.

Table 1.1 Differences between EPCONS and DS

	Object Naming Services (ONS)	Discovery Services (DS)
Functionality	•  Point to current address of the manufacturer’s EPCIS, not designed for historical traceability information. •  Intended for relatively static records. •  Stores class-level, not serial level, record.	•  Store historical addresses for every host, not only current state. •  Offer dynamic and up to date information. •  Handle serial level information for individual item.
Scalability	 Centralized service, limited by root ONS.	 Global scale distributed architecture.
Robustness	 Less robust, centralized service, limited by root ONS.	 More robust, distributed architecture.
Services	 Only specifies how to query ONS, not how to publish; it is merely a lookup service.	•  Support different query types, one-off and standing. •  Support publishing, multi-parties for one item. •  Support complex service, active pull and push data.
Security	 No authentication or authorization mechanisms.	 Offer mutual access control enforcement for each party.

 

In addition, the ONS is a centralized service that was designed based on the DNS infrastructure. Thus, this infrastructure is not a scalable solution as the number of EPC growing rapidly in the IoT and EPCglobal network. Additionally, since the root ONS is not replaceable, it is not robust enough once the errors and attacks occur. Moreover, from the functionality and service perspective, the ONS specification only indicates how to query, but not how to publish, while the discover services should support both basic services and advanced services. Some studies also indicate that the ONS, from the security perspective, is not a good skeleton for building the discovery service (BRIDGE, 2007a), because the ONS does not support authentication or authorization for queries, while the EPCDS is located at the perfect place to enforce the security mechanism. In conclusion, the EPCONS and EPCDS should co-exist and serve different roles in the EPCglobal architecture.

Furthermore, EPCglobal claims that the root ONS runs by itself, whereas it is actually managed by a particular company based in a certain country. In this scenario, the functionality and reliability of the root ONS can be easily compromised by hostile companies or countries.

Thus, for users in other countries, the ONS will not be a trusted system. Some studies

(Evdokimov et al., 2008) have also indicated the possible attack model towards the traditional ONS architecture, e.g., unilateral ONS blocking and traffic eavesdropping and analysis.

As we can see from the discussion, the ONS is a lookup service for locating the address of the manufacturer of a specific item, which is part of EPCDS’s functions, but has some inherent deficiencies. Therefore, the secured Discovery Service is needed for providing more flexible and comprehensive services.

A number of EPCDS models have been conceived, but most of them are in the very early stage and can hardly fulfill the EPCDS design requirement, especially in the global scaled, highly dynamic real world environment, which involves a large amount of tags, and organizational entities, and machines. The number of tags is in the billions, and thus the data volume associated with the EPC event is extremely large. It is necessary to analyze current preliminary EPCDS models in our dissertation and explore their strengths and weaknesses.  Few works have been done for this purpose, since EPCDS design is a relatively new topic. None of the current model can fully meet all design requirements. The designs can be grouped into three categories. The first group of designs (EPCglobal, 2008; BRIDGE, 2007a; Afilias, 2008; Polytarchos et al., 2008; Khair et al., 2014a; Khair et al., 2014b; Muller et al., 2010) focused more on basic lookup and publishing services, but overlooked advanced services such as standing query and message pushing services. They address neither network performance nor security issues. A second group of designs

(Fabian, 2008; Xu et al., 2011; Dahbi et al., 2013; Paganelli and Parlanti, 2012; Liu et al., 2014; Shrestha et al., 2010) adopted P2P technology to address core network design requirements but did not drive to specific performance requirements such as query accuracy, network independency and extensibility. The third group of designs (Grummt and Müller, 2008; Cantero et al., 2010; Jakkhupan et al., 2010; Fabian et al., 2012; Kerschbaum and Chaves, 2012; Shi et al., 2012a; Shi et al., 2012b; Dahbi et al., 2013; Kywe et al., 2013; Shi et al., 2013; Dahbi et al., 2014; Torvekar and Li, 2014) addressed traditional security issues such as access control, but did not resolve specific EPCDS security issues such as the mutual anonymous authentication, key publishing control, or interdependent security issue.

A comprehensive EPCDS design requirement should be systematically summarized from the extensive literature, and explicitly described. Then, a more sophisticate EPCDS network should be built based upon a peer-to-peer enabled framework, to enable multi-functionality, and support better network performance. Moreover, an integrated security countermeasure is needed to offer mutual anonymous authentication, key publishing security control, and a federated access control mechanism. Our ultimate goal is to satisfy all these design requirements.

1.3. Research Framework and Contributions

1.3.1. Research Framework

Motivated by the above observations and discussions, this dissertation proposes a framework that allows a versatile EPCDS model design and also enhances its networking capability and ensures its security. The framework is shown in Figure 1.5, where the research issues are organized by the design requirements of the EPC Discovery Service. These are grouped into three categories: EPCDS functionality and service, EPCDS network performance and

EPCDS security and privacy.

Category I: EPCDS functionalities and services

• Problem 1: How to provide basic EPCDS functionality and service efficiently and effectively?
• Problem 2: How to provide advanced EPCDS functionality and service?

Category II: EPCDS network performance

• Problem 3: How to ensure the accuracy of EPCDS results?
• Problem 4: How do the EPCDS network performances meet the real world usage environment?

Category III: EPCDS security and privacy

• Problem 5: How to provide the mutual anonymous authentication for all the network participants?
• Problem 6: How to resolve the interdependent security issues?
• Problem 7: How to secure the key publishing procedures?

 

 

Figure 1.5 Research Framework

A number of studies have contributed to the research of EPCDS design, and most of them focus on solving the EPCDS functionality and service issues, especially to provide basic services (problem 1). However, they rarely address the issues relate to the advanced services. Thus, a better solution that can support advanced services (problem 2) is needed. Another group of designs tries to solve part of the network performances, whereas it can hardly fulfill all the requirements such as the query accuracy (problem 3). Most of them focus on improving the network scalability, while they overlook other requirements. Hence, we need a comprehensive EPCDS design to enhance the comprehensive networking performances (problem 4), such as the query accuracy, scalability, availability and reliability, independency and extensibility. Moreover, in the EPCglobal network context, each entity, whether human or machine, needs to be authenticated. Before the entities establish any business partnership and trust, the mutual anonymous authentication is a prerequisite (problem 5). Thereafter, entities are able to exchange their information based on their authentication results. An authorization is enabled by the relationship-based federated access control model (problem 6). In the meantime, the anonymous

authentication brings out the key publishing security issue, which needs to be resolved by the key publishing control mechanism (problem 7).

1.3.2. Contributions of this work

To address the above issues, this dissertation focuses on designing a new EPCDS model that can support all the required features. Figure 1.6 labels the problems and points out the corresponding solutions and chapters in our work. In the following sections, we will detail the two groups of problems and provide an overview of the solutions in this work.

The contributions of this dissertation are many-fold:

• First, this dissertation provides extensive discussion and analysis of selected typical EPCDS design models that were introduced by previous studies.
• Second, the detailed design requirements for the EPC Discovery Service system are discussed in a systematic fashion.
• The existing EPCDS model is discussed and summarized in relation to the

EPCDS design requirements.

• The presentation of a P2P-based EPCDS design, which takes the comprehensive design requirements into account.
• A prototype of a new proposed EPCDS model has been implemented and

evaluated. The prototype presents empirical evidence for the feasibility of core ideas of the new design.

• A security and privacy analysis of RFID and EPCglobal in general is presented for the EPCDS security and privacy model design preparation.
• It is the first study that addresses the resulting key publishing security issue and the interdependent security issue.
• A novel relationship-based federated access control model, which includes the federated mutual anonymous authentication and fine-grained authorization mechanisms has been introduced for the security and privacy purpose.
• An EPCDS source data ontology has been developed, which will facilitate the future works in this specific research area.

 

 

Figure 1.6 Research Problems and Solutions

1.4. Organization of the Dissertation

The rest of the dissertation is organized as follows. In Chapter 2, we review the main research that is related to our work, including background and preliminary technologies that we will adopt in the system design, particularly focusing on the selected typical EPCDS models. In

Chapter 3, we collect the requirements that EPCDS should fulfill, and the current EPCDS models

will be analyzed and compared by using those requirements as a foundation. In Chapter 4, we introduce a new proposed P2P-based EPCDS network model. In Chapter 5, an EPCglobal network security and privacy analysis is presented, and a novel relationship-based federated access control model will be described. Finally, in Chapter 6, we conclude our work with a summary and will discuss remaining issues and future research directions.

DESIGNING AND SECURING DISCOVERY SERVICES FOR EPCGLOBAL NETWORK IN SUPPLY CHAINS

 

Sharing is caring!