Past Temporal Logic Research Articles

Hypothetical Temporal Reasoning in Databases

In this paper we integrate a history–encoding based methodology for checking dynamic database integrity constraints into a situation-calculus based specification of relational database updates. By doing this, we are able to: (1) Answer queries about a whole hypothetical evolution of a database, without having to update the entire database and keep all the information associated to the generated states, (2) State and prove dynamic integrity constraints as static integrity constraints, (3). Transform history dependent preconditions for updates into local preconditions. The methodology presented here is based on the introduction of operators of predicate past temporal logic as macros into the specifications, written in the situation calculus, of the dynamics of a database. Temporal subformulas of a query are treated as auxiliary views with the corresponding specification of their dynamics. An implementation of hypothetical temporal query answering is presented.

Temporal triggers in active databases

In this paper we propose two languages, called Future Temporal Logic (FTL) and Past Temporal Logic (PTL), for specifying temporal triggers. Some examples of trigger conditions that can be specified in our language are the following: "The value of a certain attribute increases by more than 10% in 10 minutes," "A tuple that satisfies a certain predicate is added to the database at least 10 minutes before another tuple, satisfying a different condition, is added to the database." Such triggers are important for monitor and control applications. In addition to the languages, we present algorithms for processing the trigger conditions specified in these languages, namely, procedures for determining when the trigger conditions are satisfied. These methods can be added as a "temporal" component to an existing database management systems. A preliminary prototype of the temporal component that uses the FTL language has been built on top of Sybase running on SUN workstations.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Efficient checking of temporal integrity constraints using bounded history encoding

We present an efficient implementation method for temporal integrity constraints formulated in Past Temporal Logic. Although the constraints can refer to past states of the database, their checking does not require that the entire database history be stored. Instead, every database state is extended with auxiliary relations that contain the historical information necessary for checking constraints. Auxiliary relations can be implemented as materialized relational views.

Temporal conditions and integrity constraints in active database systems

In this paper, we present a unified formalism, based on Past Temporal Logic, for specifying conditions and events in the rules for active database system. This language permits specification of many time varying properties of database systems. It also permits specification of temporal aggregates. We present an efficient incremental algorithm for detecting conditions specified in this language. The given algorithm, for a subclass of the logic, was implemented on top of Sybase.

Implementing temporal integrity constraints using an active DBMS

The paper proposes a general architecture for implementing temporal integrity constraints by compiling them into a set of active DBMS rules. The modularity of the design allows easy adaptation to different environments. Both differences in the specification languages and in the target rule systems can be easily accommodated. The advantages of this architecture are demonstrated on a particular temporal constraint compiler. This compiler allows automatic translation of integrity constraints formulated in Past Temporal Logic into rules of an active DBMS (in the current version of the compiler two active DBMS are supported: Starburst and INGRES). During the compilation the set of constraints is checked for the safe evaluation property. The result is a set of SQL statements that includes all the necessary rules needed for enforcing the original constraints. The rules are optimized to reduce the space overhead introduced by the integrity checking mechanism. There is no need for an additional runtime constraint monitor. When the rules are activated, all updates to the database that violate any of the constraints are automatically rejected (i.e., the corresponding transaction is aborted). In addition to straightforward implementation, this approach offers a clean separation of application programs and the integrity checking code.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>