=Paper=
{{Paper
|id=Vol-421/paper-8
|storemode=property
|title=Authorization Control in Business Collaboration
|pdfUrl=https://ceur-ws.org/Vol-421/paper8.pdf
|volume=Vol-421
|dblpUrl=https://dblp.org/rec/conf/icsoc/He08
}}
==Authorization Control in Business Collaboration==
https://ceur-ws.org/Vol-421/paper8.pdf
Authorization Control in Business Collaboration
Daisy Daiqin He
Supervised by Jian Yang
Department of Computing, Macquarie University,
North Ryde, NSW 2109 Australia
daiqin,jian@ics.mq.edu.au
Abstract. Authorization control has been well studied for years, and
there are quite a few theories and techniques available for handling ac-
cess control for a single or a centralized system. However unique and
challenging security issues concerning business collaboration in the con-
text of service oriented computing (SOC) have arisen due to the dynamic
and loosely coupling nature of the environment in which business collab-
oration is conducted. In this paper, we discuss different authorization
control issues in business collaboration and present an overview to our
proposed PD-AC framework, which we believe it has laid a good foun-
dation for future work in the area of policy consistency checking, policy
negotiation, and security policy enforcement in business collaboration.
Key words: access control, security policy integration, collaboration
1 Introduction
Web services and Service Oriented Computing (SOC) provides infrastructural
support for cross-organization collaboration in distributed environments. How-
ever security concerns become one of the main barriers that prevent widespread
adoption of this new technology. Each organization or business unit has its own
interest and security polices for defining who has access right for specific services
and how services can be used. In web services environment with complex cross-
organization collaborations, Different security challenges will arise with different
number of participants involved in collaboration.
Access control is enforced in a single organization by using pre-defined au-
thorization control polices. Common authorization control practices include re-
quester credential verifications, role assignments and access decision makings.
In collaborative business world, a service can be accessed by a party which can
pass it to other parties. We shall use some examples in health care to illustrate
some issues.
Suppose a patient granted access right to a General Practitioner (GP) in a
medical centre on patient’s health record. Since the physician is a member of
a research institute, he could also let researchers in this institute access to this
health record based on the security policy of the medical centre. However the
patient may not want the GP give access right to anyone on his health record
unless there is an emergency. How can we use security policies to control the
way in which information or service is propagated between organizations?
Problem can also arise from service composition. For example, a medical
center allows the patients who hold an OSHC (overseas student health cover) to
book appointments on line for general inquiry and ultrasound exam based on its
policy. A radiology institute wants to collaborate with the medical center and
accepts on line bookings from the medical center. However, OSHC is not accepted
by the radiology institute’s policy. It is challenge to decide which authorization
control policy they should follow if these two organizations collaborated in the
presence of this policy conflict.
Moreover, organizations collaborate with each other in various ways. Be-
fore organizations engage in collaboration, their authorization policies need to
be analyzed to decide the possibility of collaboration under the authorization
constraints defined by each individual party. Therefore, we need to evaluate
consistencies of access policies of different organizations. Intuitively, the concept
of ’access policy consistency’ is referred as that the access policies of different
organizations are conflict free, for the same service. And organizations are able
to collaborate in the intended way securely in terms of access control policies.
To address these complex security policy issues, we need a framework that
can analyze, evaluate and integrate security policies if necessary for collaboration
purpose, based on which negotiation can be guided and security integrity can
be enforced. Security control in business collaboration should take individual
organization’s access policy into account, as well as the type of the collaboration
which is referred as collaboration pattern in the paper.
Access control issues in signle organization or signle domain have been well
studied [5, 3, 2]. Access control in collaborative environment has just started to
attract the attention of the research community [1], but little attention have been
given to consistency study between access control policies of different collabora-
tion participants, particularly in the context of Web Services. Furthermore, these
studies only focused on providing solutions to some aspects of security issues in
terms of: security policy specification, access control in distributed environment,
and access decision making. What is missing is a comprehensive analysis of: (1)
what security requirements really are in the context of business collaboration;
(2) security policies can be specified; (3) how security policy can be verified,
evaluated, and integrated for the purpose of business collaboration. No feasible
mechanisms can be developed for policy negotiation and enforcement without
this analysis.
The rest of paper is organized as follows. We first explain the basic elements
included in the security policy and relationship between these elements in Section
2. In Section 3 we identify and model different types of collaborations. In Section
4 we model and discuss security controls for different types of collaboration
in health care environment. We propose our framework in Section 5. Formal
definition and of policies and rules are presented in Section 6. Related work is
discussed in Section 7. In Section 8, we give some concluding remarks and outline
future research directions.
2 Related Work
A number of studies concentrated on authorization architecture [7, 4]. Author in
[7] suggested a brokered architecture to build composite Web services according
to the specified security constraints. They used security matchmaker to find right
collaboration partners who have compatible security policies, which is similar
to our research. However, it did not address the issue that inconsistencies and
conflicts exist between security policies of prospective partner.
A few of studies have touched policy level security issues, which focused on
identifying different security requirements and proposed specifications for these
requirement. Trust-Serv[11] modeled access control processes in web services us-
ing state machine and provides lifecycle management for policies. Ws-AC [6]
provides an adaptive system that is capable of asking users to refine their re-
quests to comply with security policies. Again, all of them are concerned security
policies in a single organization and none of them addressed policy problems in
collaborative environment.
There are few papers on Web service authorization control in the collabora-
tive environment. We are aware of the work presented by [1], which presented a
framework for managing authorization policies for Web service compositions. [12]
proposed an approach to security policy integration and conflict reconciliation.
But they neglected the fact that different types of collaboration affect the way
the collaboration policy is developed as well as the requirements on collaborative
partner’s authorization policy. An evaluation on collaborative partner’s access
policy has to be carried out before the collaboration be established. Our work is
to fill in this gap. We believe this is the first step toward conflicts detection and
suitable collaboration partners discovery.
In summary, none of these studies went deep into different types of cross-
organization collaboration, which could raise different requirements on access
control policy of prospective collaborative partners. Our goal is thus to provide
a framework that could identify types of cross-organization collaborations in
this context; define collaboration requirements in terms of security policy; gen-
erate policy integration rules. Most importantly our work is in the context of
business collaboration which involves multiple organizations rather than simple
interaction between individual requester and single service provider.
In this paper we present some of our initial results in modeling security
requirements and integrating security policies for business collaboration. We be-
lieve this study is the first step towards achieving an understanding of a secured
of business collaboration in terms of authorization control.
3 The PD-AC Framework
The core function of our PD-AC (Policy Driven Authorization Control) frame-
work is realized by a Policy Evaluation Engine, which is associated with every
organization that provides services in business collaboration, see Figure 1. Policy
Evaluation Engine is used to analyze the nature of collaboration, make access
decision, and finally generate the collaborative security policies. The proposed
framework consists of two components: request mediator and policy evaluation
engine. Upon receiving a request for a service, the request mediator firstly iden-
tifies the type of requested service, which is a process that identify whether the
requested service is a collaborative service. If it is a simple service that does not
involve any collaboration, the mediator will perform normal access control func-
tions: looking up database, check which role in the database has the requested
privilege; compare requester’s credential with security requirement; make access
decision.
However, if it is a collaboration request (prospective collaboration partner),
the request mediator will pass the request to the Policy Evaluation Engine to
perform the following functions:
– Identify requested collaboration type;
– Evaluate requester’s authorizaton policy according to policy requirements
for requested collaboration type;
– Make collaboration decision;
– Generate collaborative authorization policies for the collaborated service if
the requester if acceptable (from previous step);
In the following subsections, we will discuss different types of collaboration
and policy evaluation engine in more detail.
3.1 Business Collaboration Patterns
Business collaborations consist of complex relationships and interactions among
organizations. Authorization policies of all participate organizations need be
carefully considered and evaluated. Our analyze shows that different collabo-
ration types affects the requirements on collaborative partner’s authorization
policy. We have conclude four different ways of collaboration between organiza-
tions and provided simple examples in Health Care domain [8].
• Simple Access (SA): it depicts the most simplest ‘request service - provide
service’ scenario that involves two organizations.
• Composite Services. The Composite Service we discuss here is referring to
the service that is based on the integration of multi-service providers. Two
different cases are identified in service composition:
1. Composite service with agent (CSWA): Multiple numbers of service
providers provide their services through an centralized agent, i.e. health
insurance company and health service providers.
2. Joined service without an agent (JSOA): Two organizations in-
volving in a peer-to-peer collaboration and provide a joined service by
integrating their business processes or integrating part of their business
processes together to form a new service directly without any agent.
• Service Outsourcing (SO): As the result of globalization, Outsourcing
and offshore Outsourcing has become a popular trend in many industries,
SO depicts collaboration relationship between outsourcer and outsourcees.
• Service Propagation (SP): it depicts collaborations that involving multi-
ple organizations and ‘forward’ privilege could be passed from one organiza-
tion to another organization.
3.2 Policy Evluation Engine
Authorization policy of prospective partners are compared and evaluated in Pol-
icy Evaluation Engine to determine the suitability for requested collaboration
pattern. Before we can compare and evaluate policies from different organiza-
tions, we need to understand all the necessary elements and their relationships for
a generic authorization policy. Therefore, an authorization policy model is pro-
posed to specify authorization policy in an individual organization. We base our
policy model on Role-Based Access Control (RBAC)[9] and encoded the model
in Description Logic. The main entities in the model are roles, credentials, privi-
leges, obligations and provisions[8]. Policies from two different organizations can
be compared by combining them into a single model.
Three categorizes of inconsistencies have been discussed in our work: role,
credential and privilege inconsistencies[10]. Each category consists of several in-
consistency types. We use a Description Logic reasoner (an automated proof
engine) to analyse the inconsistencies in policies. We encode the inconsistency
tests as concepts and relations in our model. Individual policies expressed us-
ing the model can then be combined and tested. Given a combined policy, with
the roles and privileges of the two organisations suitably related, a reasoner will
prove that the tests are either satisfiable or unsatisfiable and these results can be
analysed to check if they satisfy the requirements for the particular collabora-
tion. Since the tests are part of the general model, so they can be expressed once,
proven to encode the required meaning and used to testing any two policies.
We have identified several cross-organization collaboration patterns, differ-
ent collaboration pattern can result in different requirements for authorization
policies of prospect collaborative partners. The requirements we discussed in our
work are basic requirements that must be satisfied by the prospect partner to be
considered for requested collaboration. Policy requirements for different collabo-
ration patterns are analyzed. Inconsistencies are discussed for each collaboration
pattern[10]. Depends on the collaboration pattern, some of the inconsistencies
are acceptable, some of them needs further negotiation and some of inconsisten-
cies lead to reject.
4 Conclusion
In this paper, we proposed an authorization conrol framework for business col-
laboration. Our analyze shows that different ways of collaboration could affects
the requirements on authorization policy of collaborative partner. The proposed
framework use a policy evaluation engine to analyze collaboration suitability
of prospective partners for requested collaboration pattern from authorization
policy perspective. In our previous work, we have concluded different business
collaboration patterns and discussed different requirements for the prospective
partner’s authorization policies in the collaboration. An description logic based
authorization policy model has been proposed to specify the authorization pol-
icy of an individual organization. Inconsistencies between authorization policies
from different collaboration participants are identified and classified based on the
model, based on which the collaboration possibility are analyzed. In the future
we intend to extend this work to incorporate the following:
– Context constraints that affect access control.
– Inconsistencies that caused by role hierarchies and separation of duty.
– Collaboration access control policy requriements from business transaction
and process perspective.
References
1. Rouached, M., Godart, C.: Reasoning about Events to Specify Authorization Poli-
cies for Web Services Composition. In: 2007 International Conference on Web Ser-
vices, IEEE Press, Salt Lake City (2007)
2. Srivatsa, M., Iyengar, A., Mikalsen, T., Rouvellou, I., Yin, J.: An access control
system for web service compositions. In: 2007 International Conference on Web
Services, IEEE Press, Salt Lake City (2007)
3. Kagal, L., Paolucci, M., Srinivasan, N., Sycara, K., Denker, G.: Authorization and
Privacy for Semantic Web Services. IEEE Intelligent Systems. 19, 50-56 (2004)
4. Ziebermayr, T., Probst, S.: Web Service Authorization Framework. In: 2007 IEEE
International Conference on Web Services, pp. 614-621. IEEE press, San Diego
(2004)
5. Sirer, E. G., Wang, K.: An access control language for web services. In: 2002
SACMAT, pp.23-30. (2002)
6. Bertino, E., Squicciarini, A. C., Mevi, D.: A Fine-Grained Access Control Model
for Web Services. In: IEEE International Conference on Services Computing, pp.
33-40. IEEE press, Shanghai (2004)
7. Carminati, B., Ferrari, E., Hung, P. C. K.: Security Conscious Web Service Com-
position. In: IEEE International Conference on Web Services, pp. 489-496. IEEE
press, Chicago (2006)
8. He, D. D., Yang, J.: Security Policy Specification and Integration in Business Col-
laboration. In: 2007 IEEE International Conference on Services Computing (SCC
2007), pp. 20-27. IEEE press, Salt Lake City (2007)
9. Sandhu R. S., et al.: Role-Based Access Control Models. IEEE Computer. 29, 38-47
(1996)
10. He, D. D., Yang, J.: Identify Authorization Control Requirement in Business Col-
laboration. In: IEEE Service Oriented COmputing (2) 2008 (SCC 2008). IEEE,
2008
11. Skogsrud, H., Benatallah, B., Casati, F.: Trust-Serv: Model-Driven Lifecycle Man-
agement of Trust Negotiation Policies for Web Services. In: 13th World Wide Web
Conf (WWW 2004). ACM Press, New York (2004)
12. Yau, S. S., Chen, Z: Security Policy Integration and Conflict Reconciliation for
Collaborations among Organizations in Ubiquitous Computing Environments. In:
UIC’08, pp. 3-19. (2008)