Mobile Digital Rights Management
`
`Zheng Yan
`Nokia Research Center
`zheng.z.yan@nokia.com
`
`Abstract
`
`This paper presents a technical overview of current state in Mobile Digital Rights
`Management (MDRM). Main aspects, such as a DRM system’s requirements and
`architectures are studied. MDRM technologies, such as rights definition languages,
`cryptography and digital watermarking are discussed. The paper also analyzes the
`limitations and extra requirements for developing Mobile DRM systems, classifies
`MDRM based on content types, and proposes MDRM use case models and a MDRM
`terminal structure. Further more, important issues are discussed regarding the success
`of MDRM challenge.
`
`1
`
`Introduction
`
`With the rapid growth of the Internet communications, the Internet has become one of
`the most efficient distribution channels of digital contents for commerce. At present this
`channel is being extended to mobile area. Certainly, It is ideal to distribute all divers of
`digital information via networks to consumers’ desk-top devices or portable devices. But
`digital contents, if not protected and managed, can be easily copied, altered, defaced, and
`distributed to a large number of recipients. Digital Rights Management (DRM), which per-
`mits the smooth, secure, trusted movement of digital works from creators and publishers to
`sellers and consumers, as well as among consumers, is needed for addressing this problem.
`
`In the future, encrypted credit cards, micro-payments, and digital cash will be established
`in mobile devices. Commerce with digital contents will become a suitable area for both
`electronic and mobile domains. Mobile DRM (MDRM), the base-bone of future mobile
`media commerce is the first issue should be addressed. The Mobile DRM is a set of actions,
`procedures, policies, product properties, and tools that an entity uses to manage its rights in
`digital contents according to requirements over mobile networks. This paper aims to give
`an overview of the current state of the MDRM, to analyzes requirements and to discusses
`technologies, use case models and challenges for developing the MDRM.
`
`1
`
`Smart Mobile Technologies LLC, Exhibit 2015
`Page 1 of 16
`
`

`

`HUTTML2001 T-110.501SeminaronNetworkSecurity
`
`2 State of the art
`
`2.1 Basic requirements
`
`The Digital Rights Management concerns techniques, processes, procedures and algo-
`rithms related to establishing a trusted computing environment, and trusted infrastructure
`for the secure preparation, transmission, and prevention of misuse and/or consumption of
`protected digital contents.
`
`General requirements are proposed in an IETF draft on a Digital Rights Trading System
`[1]. In this draft, a digital-right is defined as "a digital representation of the right to claim
`the services or goods". This definition limits digital-rights for claiming services or goods,
`does not contain usage rights for controlling content’s consuming. Therefore, this proposal
`cannot be applied to a DRM system that ensures content integrity, secures copyright, con-
`trols content usage and manages rights acquisition, specification, as well as granting. But
`it is a good reference for proposing basic requirements of a DRM system.
`
`1. From scalability point of view, "it MUST handle diverse types of rights issued by
`different issuers".
`
`2. From system security point of view, "it MUST prevent illegal acts" on both rights and
`contents. For the rights, it MUST prevent them from alternation, forgery, duplicate-
`redemption, reproduction, and repudiation, and SHOULD ensure privacy. For the
`contents, it MUST protect their integrity, prevent illegal copy, and make sure the
`contents are used correctly according to the consumer’s rights, as well as provide
`trust manageability. Because different customer has different preference, privacy
`may not be a mandatory requirement.
`
`3. From business point of view, "it MUST be practical in terms of scalability, simplicity,
`implementation / operation cost and efficiency".
`
`2.2 System architecture
`
`Fig. 1 illustrates a lifecycle of digital rights. Typically, there are four stages:
`
`1. Package stage: The operators of this stage are authors or content providers who
`conduct the following
`• Create rights protection requirements
`• Specify digital rights management policies
`• Specify conditions fee, time, access
`• Specify tracking requirements
`
`2. Sell/protect stage: The handlers of this stage are service providers who do the fol-
`lowing works
`• Define pricing
`
`2
`
`Smart Mobile Technologies LLC, Exhibit 2015
`Page 2 of 16
`
`

`

`Smart Mobile Technologies LLC, Exhibit 2015
`Page 3 of 16
`
`

`

`HUTTML2001 T-110.501SeminaronNetworkSecurity
`
`The rights are managed by a trusted party (a secure server) based on accounts. Any pro-
`cessing of the rights is handled by sending a request to an account manager through a
`network. Generally, on-line verification is needed in order to prevent duplicated rights
`redemption. However, this type of system is expensive because accounts have to be main-
`tained for each service provider and for each user. Therefore, it is hard to support system
`scalability. Additionally, account-based systems have been designed to protect accounts
`from malicious users but provide less protection from malicious managers. Therefore, the
`trust policy of these systems is imbalanced. Some Internet coupon systems, such as Cool-
`Savings and ClipACoupon, use this architecture. And this technology is generally used for
`developing server-based mobile advertisement systems.
`
`Distributed rights management:
`
`There are two approaches to realize it. One is using a tamper-resistant device, like a smart
`card or a Personal Trusted Device (PTD). In the tamper-resistant device based system, dig-
`ital rights are stored in a trusted device and circulated among devices. The tamper-resistant
`device can protect digital-right from both malicious users and malicious service providers.
`Thus, this kind of system seems to have a bright future especially in the application area
`of tickets and coupons since one smart card or PTD can store and manage diverse tickets
`without the cost of maintaining rights centrally. However, these systems create several
`issues that are hard to overcome, i.e. who should be responsible for issuing a smart card
`or a PTD if it is shared by multiple applications, how to achieve high performance given
`the memory and CPU constraints of the small devices. Moreover, the business issue with
`smart cards or PTDs is that the devices for smart card or PTD verification are not very
`common especially as user terminals such as PC.
`
`In general, PTD-based solution is suitable for such applications as eTicket, eCoupon, eLi-
`cense, etc.
`
`The other approach is using a self-protecting container, which is the key element in In-
`terTrust’s commerce platform [11]. The secured container DigiBox enables the associa-
`tion of rules and controls via cryptographic means with information content, to specify the
`types of content usage permitted and the consequences of usage. Containers are manipu-
`lated by using a trusted rights protection application in order to make the protected content
`available according to its associated access control rules. Payment is generally conducted
`when a consumer wants to open the container (pay when use, or download first pay later).
`Similar functionality is provided by IBM’s "cryptolope" container [12, 14]. The secure
`container allows rights management components to be integrated with content in highly
`flexible and configurable control structures. This approach enables true super-distribution
`and can support virtually any network topology and any number of participants, including
`distributors, re-distributors, information retailers, corporate content users, and consumers.
`But it requires pervasive deployment of tamper-resistant hardware devices to perform se-
`cure processing of protected contents. Container technology is playing an increasingly
`important role as a building block for sophisticated digital rights management system.
`
`Container-based architecture is achieving leadership in the digital content distribution, e.g.,
`eBook, eMusic, eImage, and software, etc.
`
`Semi-distributed rights management:
`
`4
`
`Smart Mobile Technologies LLC, Exhibit 2015
`Page 4 of 16
`
`

`

`HUTTML2001 T-110.501SeminaronNetworkSecurity
`
`This architecture tries to combine the advantages of the above two ways and overcome their
`disadvantages. In [13], a proposed scheme uses a ticket-account server to manage user’s
`rights. The personal rights are not managed by the service provider, but the user himself
`or someone delegated to manage the account. A smart card is used only for authentica-
`tion. This approach aims to reduce the account management cost and avoid bottlenecks
`caused by smart cards. Payment consideration during the rights circulation is ignored in
`this scheme. Therefore, it is difficult to practice payment for rights transference between
`users if using this scheme.
`
`3 Technologies
`
`This part introduces technologies for achieving mobile digital rights management.
`
`3.1 DRM languages
`
`A good digital-right representation is necessary in the MDRM system. The representations
`could be different. There are several candidates available, which are from different sources.
`
`Digital rights expressed by relational database [2]
`
`Jams Barker and his colleagues at Case Wstern Reserve University (CWRU) worked out
`a database representation, defined as a set of relational database tables and their interpre-
`tation. More than 10 basic tables are used to describe the right-properties, together with a
`large number of administrative, logging and support tables. The advantage of this method
`is the values in columns of the tables are not restricted by software, but rather by admin-
`istrators’ entries in the support tables. This permits tailoring to any installation’s needs
`together with validity checking of permission table entries. It is convenient to achieve se-
`mantics, syntax and security requirements by making use of database technologies. But
`it is inefficient if table relationships become complicated. And it is only suitable for cen-
`tralized rights management architecture. Some digital libraries support this digital rights
`expression.
`
`Xerox’s DPRL (Digital Property Rights Language) [3]
`
`Xerox’s DPRL (Digital Property Rights Language) is a language that can be used to specify
`rights for digital contents. It provides a mechanism in which different terms and conditions
`related to access, fee and time can be specified and enforced for the different operations on
`digital documents, such as view, print, and copy. Rights specifications are represented as
`statements in DPRL. Different rights can be specified for different parts of a digital work
`using a work specification. Within a work specification, different sets of rights applicable
`to this work are specified. Rights can be grouped into named-groups called "rights groups".
`Each right within a rights group is associated with a set of conditions. Conditions can be of
`different types: fee to be paid, time to use, type of access, type of watermark, type of device
`on which the operation can be performed, and so on. It also allows different categories of
`rights, such as transfer rights, render rights, derivative-work rights, file-management rights
`and configuration rights.
`
`5
`
`Smart Mobile Technologies LLC, Exhibit 2015
`Page 5 of 16
`
`

`

`HUTTML2001 T-110.501SeminaronNetworkSecurity
`
`XrML (eXtensible rights Markup Language) [4]
`
`Originating from DPRL, XrML addresses some DPRL unsolved issues, such as integrity,
`authentication of entities, and extends it by adding a set of structural and semantic tags
`suitable for specifying metadata of XrML documents, validating integrity of XrML doc-
`uments as well as of digital contents, and authoring relatively simple XrML documents
`(such as licenses). In addition, XrML adds support for specifying conditions for usage
`locations and tracking, and simplifies some DPRL document elements and their attributes.
`XrML is driving the standard for digital rights management. This XML-based language is
`deployed for expressing the agreement between the content/service provider and informa-
`tion consumer. Therefore, it is more suitable for centralized digital rights management that
`requires complicated digital rights expression. It has advantage if the centralized DRM
`server deploys a database that supports XML format. But due to its complication, this
`DRM language is not suitable for supporting rights management at the terminal, such as a
`mobile device.
`
`ContentGuard has developed and contributed XrML as an open specification licensed on
`a royalty-free basis to unify the Digital Rights Management industry and encourage inter-
`operability at an early stage. XrML is supported by such companies as Adobe Systems,
`Xerox, Hewlett-Packard, Microsoft, Preview Systems and Time Warner.
`
`ODRL (Open Digital Rights Language) [5]
`
`ODRL is another XML-based digital rights expression language. It is a vocabulary for
`representing terms and conditions over digital contents, which include constraints, per-
`missions, obligations and agreements. The ODRL has no license requirements and it is
`available in the spirit of "open source" software. This policy will attract many new DRM
`vendors’ support.
`
`Compared with XrML, ODRL is very simple. It is focused on concrete rights expression.
`ODRL can be used within trusted or untrusted systems. However, it does not determine the
`capabilities nor requirements of any trusted services. For example, it does not contain any
`information related to content protection key and rights issuer’s digital signature, which are
`needed in the real system. It is extensible for supporting rights management at the mobile
`terminal. And obviously, it is not a suitable candidate for centralized DRM system.
`
`XMCL (Extensible Media Commerce Language) [6]
`
`XMCL is an open XML-based language for media commerce and rights management.
`XMCL is an interchange format that describes usage rules applied to multi-media con-
`tents. It is designed to communicate these rules in an implementation independent manner
`for interchange between business systems and DRM implementations responsible for en-
`forcing the rules described in the language. This language tries to satisfy the requirements
`of media commerce and provides definitions of client information, digital rights expres-
`sion and authentication information. But the rights description is simpler compared with
`ODRL and XrML. It is very suitable for the distributed DRM system, in which digital
`rights is parsed at the terminal to control the content’s consuming.
`
`The above three XML based DRM languages are designed from different points of view
`and considerations. Which one is selected for a MDRM system should be based on the
`system’s requirements and architecture.
`
`6
`
`Smart Mobile Technologies LLC, Exhibit 2015
`Page 6 of 16
`
`

`

`HUTTML2001 T-110.501SeminaronNetworkSecurity
`
`3.2 Cryptographic solution
`
`Cryptographic theory is mature today. Many DRM or MDRM systems are built upon
`cryptographic blocks. We summarize them in the following.
`
`Digital signatures
`
`In MDRM systems, digital signature is often used for non-repudiation rights issuing. The
`digital rights should be digitally signed by the issuer. Therefore, the content user could
`verify the right correctness. And the signature is also a proof of rights purchase.
`
`One-way hash functions
`
`Hash functions are used for integrity checking. Cooperating with digital signature, the hash
`code of the issued rights and encrypted content is signed by the issuer’s private key. The
`rights user verifies the integrity by decrypting the signature and comparing the hash code
`with the re-computing hash code. XrML and XMCL support message digest and digital
`signature.
`
`Symmetric and asymmetric encryption
`
`In some MDRM systems, contents are protected by symmetric-key encryption. While the
`content encryption key is encrypted using asymmetric-key encryption, therefore, only the
`user with the correct private key can decrypt the content key, and then consume the content.
`XCML supports this container based access control.
`
`Traitor tracing (broadcast encryption)
`
`This is an interesting branch emerged into cryptography. The scheme addresses the case
`when an authority broadcasts some valuable information and it is required that only legiti-
`mated clients should be able to decrypt the information. While the schemes make serious
`assumption about the real life models they work in, they also propose quite efficient ways
`to trace down the traitor who have constructed new decryption. Broadcast encryption can
`be used to establish trust between users and service provider, as well as between peer
`devices. It has some beneficial effects, including key revocation, and circumvent device
`denial, besides the main effect of device authentication.
`
`3.3 Digital watermarking
`
`In the MDRM system, usage rights (such as "copy once", "copy never", and "copy no
`more", etc.) are required to tightly bind with the contents using either logical binding or
`physical binding. Traditional cryptographic approaches separate contents from rights by
`using a safe container and expressing container’s key and digital rights with a DRM lan-
`guage. But they suffer from one important drawback: they do not permanently associate
`cryptographic information with the contents. Thus, cryptography alone cannot make guar-
`antees about the redistribution or alteration of content after it has initially passed through
`the cryptosystem. Digital watermarking is a good solution to provide these extended guar-
`antees about digital content usage. The technique modifies the information itself so that it
`is feasible later to detect pirate copies. In the MDRM system, the digital watermark’s func-
`tion is extended for enabling copy protection, rights management and forensic tracking. It
`
`7
`
`Smart Mobile Technologies LLC, Exhibit 2015
`Page 7 of 16
`
`

`

`HUTTML2001 T-110.501SeminaronNetworkSecurity
`
`not only contains copyright information such as creator, content provider, content ID, etc.,
`but also carries usage rights attached to the content. At the user’s PTD, the watermark is
`detected and the content will be processed according to the detected usage rights.
`
`Digital watermarking technologies have been researched and developed not only for static
`images, but also for audio, video, text, as well as software. Fingerprinting is a branch of
`watermarking for the traitor catching.
`
`But the current digital watermarking technology is not mature enough for commercial
`usage. Based on our study and research, there is no standardized watermark solution that
`can be deployed in real applications. The big issues are:
`
`• Watermark’s robustness is not satisfied. Unlike cryptographic algorithm, current
`watermark cannot sustain most normal attacks.
`• Current academic research work has not considered the industrial requirements.
`
`Pure watermarking DRM solution may not secure enough for satisfying the commercial
`requirements.
`
`3.4 Combination
`
`As can be seen from the above, both cryptographic and watermarking technologies should
`be combined and deployed in the MDRM systems.
`
`3.5 Vendors
`
`We compare some of the existing DRM systems as Table 1.
`
`It seems that the best way to do the DRM is to control both the software and the hardware.
`Machine-readable "tags" in the software are then used to represent particular rights that the
`hardware can interpret. When a piece of content is loaded into a trusted device, it checks
`the associated digital rules and acts accordingly. From this point of view, general-purpose,
`but highly untrustworthy device - personal computer may not a good hardware candidate,
`but personal trusted mobile device will be a suitable one.
`
`In what follows, we will discuss limitations and extra requirements for developing Mobile
`DRM based on portable small devices, such as mobile phones and communicators. Fur-
`ther more, we will present and analyze Mobile DRM use cases, propose MDRM terminal
`structure and summarize basic issues that should be considered.
`
`4 Mobile DRM
`
`The demand for various mobile media services is increasing rapidly. Ideally, people hope
`to do everything they can do via the Internet. Portability and mobility provide users great
`convenience, and at the same time attract venders to introduce m-Commerce features into
`
`8
`
`Smart Mobile Technologies LLC, Exhibit 2015
`Page 8 of 16
`
`

`

`HUTTML2001 T-110.501SeminaronNetworkSecurity
`
`System
`
`Architecture DRDL Circulation
`Model
`
`Copy Preven-
`tion/Dection
`
`InterTrust
`
`Distributed
`
`–
`
`EncrypTix
`
`Centralized
`
`–
`
`RightsMarket Centralized DB
`based
`–
`
`PublishOne
`
`Centralized
`
`DigitalOwl
`Centralized
`–
`ContentGuard Centralized XrML
`
`Pay-when-
`use,
`support
`superdistributed
`Pay-before-use
`
`Pay-when-use
`
`Pay-when-use
`
`Pay-when-use
`Pay-when-use,
`not
`support
`superdistributed
`
`Sys-
`
`Distributed
`
`–
`
`–
`
`Wave
`tems
`
`FlexTicket
`
`Semi-
`distributed
`
`XML,
`RDF
`
`Pay-before-use,
`support
`rights
`transfer
`
`Container-
`based
`
`Use authenti-
`cated terminal
`PKL, plug-in
`trusted tool
`Container-
`based
`–
`Integrity, au-
`thority
`
`de-
`Trusted
`vice,
`support
`privacy
`Secure token
`in the tamper-
`proof devices
`
`DRM
`Technolo-
`gies
`DES, RSA,
`etc.
`
`–
`
`–
`
`–
`
`–
`Digital
`signature,
`Hash, Wa-
`termarking
`–
`
`Digital
`signature,
`hash, etc
`
`Table 1: Comparison of different DRM systems
`
`the mobile world. Mobile Digital Rights Management, the base-bone of digital contents
`related mobile services, is the first issue to address.
`
`4.1 Limitations
`
`Some technologies, such as WAP Identity Module (WIM), ECC (Elliptic Curve Cryptogra-
`phy) chip-embedding, and mobile payment solution, etc., have been and will be introduced
`for DRM based m-Commerce. However, we also confront difficulties. The mobile devices
`are small, use low-bandwidth communication technologies, have limited processing power
`and memory, and use batteries with limited life spans. Thus they cannot accommodate
`most strong encryption technologies, which are computationally intensive. What is more,
`the connection speed is very slow and the transaction performance is too bad for many
`people to accept.
`
`Limited device hardware restricts the embedded software. It cannot support large code
`library to fulfill various functions as personal computers can. And in most cases, the es-
`sential code for security cannot be implemented in the small mobile devices. Hardware
`security protection is expected with low cost.
`
`9
`
`Smart Mobile Technologies LLC, Exhibit 2015
`Page 9 of 16
`
`

`

`HUTTML2001 T-110.501SeminaronNetworkSecurity
`
`4.2 Extra requirements
`
`Apart from the requirements listed in 2.1, special requirements are raised due to the above
`limitations:
`
`• User interface design should be satisfied with the small display.
`• System usability should be accepted by potential users, and the basic clue is simpli-
`fying the applications, but providing attractive services.
`• Balance the client and server computation for achieving better performance.
`
`Other extra important requirements are:
`
`• Terminals shall have an environment or features for decrypting secure containers and
`forcing the terminal work following the digital rights.
`• The MDRM architecture shall offer profit for both the service provider and the con-
`tent owner. This is the main power of MDRM development and evolution.
`
`4.3 Classifications
`
`Mobile Digital Rights Management is a big concept that covers various digital data rights
`management. Based on the type of contents, it can be classified into three groups.
`
`• Rich MDRM: The content managed by the MDRM system is rich media, such as
`video, e-books, which can only be consumed by high-end mobile devices, like Nokia
`communicators. Both cryptographic and watermarking technologies are needed for
`protecting the contents and controlling the usage.
`• Light MDRM: The content managed by the MDRM system is light media, such
`as ring tones, images, music, which can be consumed by medium-end or low-end
`mobile devices, such as mobile phones, whose platform is close. Cryptographic
`protection may not be necessary. Watermarking can be used. The device handles
`enforced usage.
`• Minimal MDRM: No digital contents attached. The digital-right itself claims the
`holder’s rights to be served. The typical examples are e-Tickets and e-Coupon. PTD-
`based system is more suitable for the minimal MDRM. The mobile digital rights are
`saved in the secure mobile wallet.
`
`4.4 Use Cases
`
`Use cases are very important for analyzing and developing a new service. In this part,
`we provide three types of Mobile DRM use scenarios, which, we think, are main MDRM
`services in the future.
`
`10
`
`Smart Mobile Technologies LLC, Exhibit 2015
`Page 10 of 16
`
`

`

`Smart Mobile Technologies LLC, Exhibit 2015
`Page 11 of 16
`
`

`

`Smart Mobile Technologies LLC, Exhibit 2015
`Page 12 of 16
`
`

`

`Smart Mobile Technologies LLC, Exhibit 2015
`Page 13 of 16
`
`

`

`Smart Mobile Technologies LLC, Exhibit 2015
`Page 14 of 16
`
`

`

`Smart Mobile Technologies LLC, Exhibit 2015
`Page 15 of 16
`
`

`

`HUTTML2001 T-110.501SeminaronNetworkSecurity
`
`[9] Fujimura K. and Nakajima Y. General-purpose Digital Ticket Framework.
`USENIX Workshop on Electronic Commerce, pp. 177-186, August 1998
`
`3rd
`
`[10] Fujimura K., et. al. ML Ticket: Generalized Digital Ticket Definition Language. The
`W3C Signed XML Workshop, Apr. 1999
`
`[11] Sibert O. DigiBox: A Self-Protecting Container for Information Commerce. 1st
`USENIX Workshop on electronic Commerce, July 1995
`
`[12] Gladney H.M. and Lotspiech J.B. Safeguarding Digital Library Contents and Users:
`Assuring Convenient Security and Data Quality. D-Lib Magazine, May 1997
`
`[13] Matsuyama K. and Fujimura K. Distributed Digital-Ticket Management for Rights
`Trading System. 1st ACM Conferences on Electronic Commerce, Nov. 1999
`
`[14] Auerbach J. S, Creation and Distribution of Cryptographic Envelope. Patent number:
`5673316. Sept. 1997
`
`16
`
`Smart Mobile Technologies LLC, Exhibit 2015
`Page 16 of 16
`
`