Marijan Kunstic

The Intelligent Agent-Based Control of Service Processing Capacity

The paper presents new approach to processing capacity protection in the service system with multiple server units. In the integrated service communication networks an important problem is to implement call admission control and routing so as to optimally use the network resources. We assumed reality of several classes of jobs and tried to preserve an amount of processing capacity for high priority jobs that can arrive in a burst. In parallel, current low priority jobs were processed with continuous regulation of servers load. Simulation results showed rapid adaptation and good balancing around the predetermined maximum processing level. Inspiration was found in the paradigm of software agent technology and potential advantages appearing when applied in telecommunication network. Main characteristic of the usage of intelligent agent is the opportunity to permanently transfer adaptation regarding one or more parameters following optimal and requested policy.

Verification of Intelligent Agents with ACTL for Epistemic Reasoning

Verification of multi-agent systems (MAS) is a huge challenge, especially for those systems where security and safety are of major importance. Verification detects faults, defects and drawbacks in an early stage of software development. Here, we give a formal model for verification of MAS by means of model checking technique. We extend the existing action computation tree logic (ACTL) with epistemic operators in order to reason about knowledge properties of MAS. We introduce new operators for manipulation on agents actions with data. We explain their syntax and semantics for our ACTL-er (ACTL for epistemic reasoning), and provide a case study for a MAS system of foraging bees.

The effect of alteration in service environments with distributed intelligent agents

The paper presents some properties of the communication network supported by the intelligent agents. The intelligent agents were placed into network nodes and they were immobile. They were used to regulate the transfer of mobile agents from the network input, through the routing nodes and finally, towards the service processing nodes. These nodes were the programs running on the computers. A new model of distributed and collaborating intelligent agents was designed and presented. Continuous adaptation and agents collaboration was achieved by reinforcement Q-learning. For such a model the results show rapid tendency to reduce state time of mobile agents in the service. The obstructive effects expansion to the other agents when a change of processing capabilities in the region occurs was detected and described.

Self-trained agents optimize communication service by intelligent selection

The paper presents a method using an optimal selection of an agent in a thought client-server environment. The selection criteria are based on continuous learning and monitoring of the agents behavior. The work has been motivated by the different abilities and properties of the agents in the network, particularly when they act in distributed environments. The main idea presented is a permanent transfer adaptation of the requests from a client agent to an optimal service provider agent. Continuous adaptation is achieved by reinforcement Q-learning. Simulation results show that by implementing knowledge into an agents behavior, it is possible, in particular situations, to significantly accelerate the service system.

Load protection model based on intelligent agent regulation

Paralleling rapid advancement in the telecommunication network expansion is necessary for advanced network traffic management surveillance. The increasing number and variety of services being offered by networks have emphasized the demand for optimized load management strategies. The paper deals with regulation of a mobile agent moving toward service processing resource in the part of the agent network. We have constructed the agent architecture for the control of service processing load. The goals of the controlling system were both to protect the processing load and to predict the arrival rate of clients requests. Self-adaptive property is implemented by reinforcement Q-learning. The analysis is based on experimentation through simulations.

SDL Ontology for Specifying Systems Based on Finite State Machines

Specification and Description Language (SDL) is an object-oriented and formal standardized language for the specification of complex, event-driven, real-time and interactive applications involving many concurrent activities that communicate using discrete signals. Using SDL formal model for system specification we bridge the gap between ideas in our minds and the actual implementation of the system. In this paper we propose the ontology for the basic SDL elements. We also propose a formal framework of SDL Markup Language as a medium for translating SDL model to SDL ontology.

Petri Net-Based Simulation and Analysis of the Software Development Process

This paper explores improvements which can be achieved by applying Petri nets to the modeling, simulation and analysis of the software development process. The huge complexity of this process, in conjunction with the demand for rapid reaction to market pressure, hard limits on time and cost, and the fluidity of human resource organization can make it considerably difficult to establish a confident software development process. Simulations of such processes using Petri net models show considerable benefits with respect to real factors such as resource requirements, representation of critical borders, the effects of resource deficit, delays in process phases, etc.

Verification of communication protocols in a multi-agent system

This paper provides a framework for specification and verification of intelligent agent-based systems, with the emphasis on their cognitive and reactive features. We use an epistemic transition system to specify the agents and an epistemic synchronous product to specify the multi-agent system. We verify the system by means of a special action-based logic - ACTLW for Epistemic Reasoning (ACTLW stands for Action Computation Tree Logic with Unless Operator). Using temporal and epistemic operators we create the appropriate formulae to check whether the relevant property holds, or the protocol works within the given conditions. We test our method by the example of security communication protocol called Dining Cryptographers.

Agent-Based Connection Control for Digital Content Service

This paper explores improvements that can be achieved by applying intelligent agent techniques to solve the problem of self-adaptive routing. A potential redundancy has been recognized in telecommunication network configuration, hidden in the routing method between the user access points and service providers. The main idea presented here is perpetual trans fer adaptation for all requests that are sent from a user to a service location over all the network elements. Self-adaptation is based on the continuous monitoring of the available communication channel capacity between the user and the service. The actions are based on perpetually seeking the optimal throughput via the nodes that maximize exploitation of the communication channel. From the users point of view, accumulation and exploration of knowledge concerning throughput properties in the network can optimally utilize redundant capacities thus providing service more rapidly.

Implementation of Epistemic Operators for Model Checking Multi-agent Systems

The problem of multi-agent system (MAS) specification and verification has been introduced in this paper, Epistemic transition system (ETS) represents an agent as the smallest unit in a multi-agent system, while Epistemic synchronous product (ESP) represents the formal model for a multi-agent system. Therefore, a formal framework for epistemic properties of multi-agent systems has been provided. A special extension of Action computation tree logic with unless operator for epistemic reasoning (ACTLW-ER) is used for MAS model checking. Epistemic operators of ACTLW-ER are implemented by symbolic model checking algorithms using binary decision diagrams.