Microsoft Technology Licensing, LLC

Moritz Y Becker

Blair Brewster Dillaway

Andrew Donald Gordon

Cedric Fournet

Jason Forrest Mackay

Gregory Darrell Fee

John Michael Leen

Brian LaMacchia

Jason Hogg

Security language constructs may be translated into logic language constructs and vice versa. Logic resolution may be effected using, for example, the logic language constructs. In an example implementation, translation of a security language assertion into at least one logic language rule is described. In another example implementation, translation of a proof graph reflecting a logic language into a proof graph reflecting a security language is described. In yet another example implementation, evaluation of a logic language program using a deterministic algorithm is described.

Security language translations with logic resolution

Techniques for evaluating detectablity of confidential information stored in authorization policies are described. In an example, an authorization policy has a confidential property. The confidential property is defined by whether application of a test probe to the authorization policy results in the grant of access to a resource. A processor automatically determines whether at least one witness policy can be generated that is observationally equivalent to the authorization policy from the perspective of a potential attacker, but the application of the test probe to the witness policy generates an access denial result. In the case that such a witness policy can be generated, an indication that the confidential property cannot be detected using the test probe is output. In the case that such a witness policy cannot be generated, an indication that the confidential property can be detected using the test probe is output.

Evaluating detectability of information in authorization policies

Delegation-based authorization is described. In one example, a reference monitor receives from a first entity a request and a credential statement comprising a delegation of authority over a fact to a further entity. An authorization node then determines whether the further entity consents to provide the fact to the first entity and evaluates the request in accordance with an authorization policy and the credential statement. In another example, an assertion comprising a statement delegating authority over a fact to a further entity is received at an authorization node from a first entity. An authorization policy is then used to determine that the first entity vouches for the fact if each of these conditions are met: i) the first entity consents to import the fact from the further entity, ii) the further entity consents to export the fact to the first entity, and iii) the further entity asserts the fact.

Delegation-based authorization

A security language expresses assertions and authorization queries in a manner that facilitates logic resolution. In an example implementation, assertion syntax and authorization query syntax are described. In another example implementation, checks on the safety of assertions and authorization queries are described. In yet another example implementation, semantics rules are described.

Security language expressions for logic resolution

Software tools assist an access-policy analyst or creator to debug and/or author access policies. An access request contains a query that evaluates to either true or false depending on whether access is to be allowed. Abduction may be used to generate assumptions that, if true, would cause the access request to be true. The tool may perform analysis on the generated assumptions, such as: comparing the assumptions with tokens to detect errors in the tokens or to suggest changes to the tokens that would cause the query to be satisfied, or comparing the assumptions to a meta-policy. The tool may allow an analysis, policy author, or other person to interactively walk through assumptions in order to see the implications of the access policy.

Access policy analysis

Security language constructs may be translated into logic language constructs and vise versa. Logic resolution may be effected using, for example, the logic language constructs. In an example implementation, translation of a security language assertion into at least one logic language rule is described. In another example implementation, translation of a proof graph reflecting a logic language into a proof graph reflecting a security language is described. In yet another example implementation, evaluation of a logic language program using a deterministic algorithm is described.

Access to a resource may be controlled by a policy, such that a request to access the resource is either granted or denied based on what assertions have been made by various principals. To find the assertions that support a grant of access to the resource, a template may be created that defines the nature of assertions that would cause access to succeed. Assertions may be stored in the form of tokens. The template may be used to search an existing token store to find assertions that have been made, and/or to generate assertions that have not been found in the token store and that would satisfy the template. The assertions in the template may be created by performing an abductive reasoning process on an access query.

Delegation in logic-based access control

In an example implementation, a bifurcated security scheme has a first level that does not allow usage of negations and a second level that does permit usage of negations. In another example implementation, an authorization query table maps respective resource-specific operations to respective associated authorization queries. In yet another example implementation, authorization queries are permitted to have negations, but individual assertions are not.

Security authorization queries

Architecture that employs dynamic assertion providers in logic-based security policy languages. The dynamic assertion providers provide a mechanism by which data can be processed in an efficient, goal-directed manner. The application of dynamic assertion providers enables the inferencing over type hierarchies, including hierarchies of personally identifiable information and also for file systems which are hierarchal in nature. When searching for facts to bind against for a given subgoal, dynamic assertion providers are also queried. These objects follow an interface which can implement an arbitrary method to provide facts to the logic engine in a goal-directed, on-demand manner. Hierarchical types can be implemented using a hierarchy provider.

Dynamic assertion providers for logic-based security policy languages

An authorization system includes an authorization node, a storage device and a reference monitor. The authorization node executes an authorization policy, and the storage device stores an authorization state associated with the authorization policy. Requests for access to a secured resource are received at the reference monitor, and the reference monitor queries the authorization node, which uses the authorization policy to determine whether to grant access to the secured resource based on a rule having at least one access condition. The rule, executed as part of the authorization policy on the authorization node, is configured to update all the entries in the authorization state for which an update condition is met.

State-updating authorization

The delegation of rights may be controlled in a number of manners. In an example implementation, a delegation authority assertion is formulated with a delegator principle, a delegatee principal, a verb phrase, a resource, and a delagation-directive verb. In another example implementation, a delegation mechanism involving an assertor, a first principal, and a second principal enables a delegation to be specifically controlled. In yet another example implementation, a chained delegation mechanism enables explicit control of a permitted transitive chaining depth.

Controlling the delegation of rights

Security assertion revocation enables a revocation granularity in a security scheme down to the level of individual assertions. In an example implementation, a security token includes multiple respective assertions that are associated with multiple respective assertion identifiers. More specifically, each individual assertion is associated with at least one individual assertion identifier.

Security assertion revocation

Logical abduction is used to derive the premises that support an access query. In a logic-based access-control system, a query, as to one or more principals' right to access one or more resources, is a statement that can be either true or false. The statement evaluates to true if the principal is allowed to access the resource under the existing set of assertions. Assertions that, if made, would cause the statement to be true can be abduced from the query and from the policy against which the truth of the query is to be judged. The abduced assertions can be used to assist in making the appropriate assertions to cause the query to evaluate to true so that access to the resource can be granted.

Growing community of inventors

Cambridge, United Kingdom

Moritz Y Becker