Vol. 23 · No. 1

Building complex systems requires a specific kind of modularity as well as incremental development. Here we introduce an architecture where basic modules are reactive agents and the data exchange among them is based on the so-called stigmergic communication through space. In this way we have connected ideas coming from the multi-agent systems and the coordination languages on the one side and ideas of the behavior-based systems on the other side. We demonstrate that this architecture manifests several interesting features which are useful for engineering of real-time systems and modeling of biological creatures or their parts. We advocate for so-called purely reactive agents which are stateless entities usually taken as too weak building blocks of systems. However, their features enable us to use a special method of incremental development (so-called subsumption method).

In this paper, we study the Dominating Set problem with measure functions, which is extended from the general Dominating Set problem. We study the correspondnig problems on complexity, approximation and inapproximability for Dominating Set problem with measure functions. In addition, we extend our results to the weighted graphs.

The method presented in this paper uses a generic C-language written simulation model of an embedded distributed computer system aimed for a safety-critical control application. The considered system is built using Time Triggered Architecture (TTA) concepts. The aim of the presented simulation method is to evaluate the system capability to tolerate a chosen category of faults. The model, being written in ANSI-C, is portable and machine-independent. Its structure is modular and flexible, so that the system to be studied and the experiment setting can easily be changed. The functionality of this model is demonstrated on a set of fault injection experiments aimed mainly to evaluate the correctness of the Time Triggered Protocol (TTP/C) that implements the abstract concepts of TTA. These experiments were done within the EU/IST project Fault Injection for Time triggered architecture (FIT).

Security in computer networks is a very complex task especially if it is required to separate a corporate network from public Internet or to divide a company's intranet into multiple zones with different security requirements. The network security policy that describes these security requirements is primarily presented in a high-level form. Also, the security policy is enforced using some low-level security mechanisms, mainly firewall technology. One of the main difficulties faced by the network administrator is how to translate the high-level policy description to the low-level firewall rule-base. This paper presents Role-Based Network Security (RBNS) model that can be used as an intermediary level between high-level policy form and low-level firewall rule-base. We use the Role-Based Access Control (RBAC) model as a framework for our proposed RBNS model. The main concept of RBNS model is that network services are assigned to roles and hosts are made members of appropriate roles thereby acquiring the roles' network services. Also, the paper presents a compilation algorithm that can be used to automatically generate the low-level firewall rule-base from the RBNS intermediary-level. The paper presents a proposed verification algorithm to prove that the high-level policy and the translated low-level firewall rule-base are equivalent. Based on the RBNS model, we design and implement a firewall management toolkit. The paper demonstrates in brief the toolkit's capabilities through an example, thus showing that the using of this model separates the high-level security policy from the underlying enforcement mechanism. This separation offers easier management and debugging of low-level firewall rule-base at an appropriate level of abstraction.