Abstract: Mobile agent technology makes it possible to reduce network traffic, overcome network latencies and enhance robustness and fault-tolerant capabilities of distributed applications. However, it is sometimes difficult or even impossible to take full advantage of these technical benefits because of the lack of an appropriate infrastructure for overcoming problems related to connectivity (e.g. access through firewalls), security, location transparency, and use of proprietary tools. This paper discusses these problems and introduces the requirements for various infrastructure components and their implementation with the aim of enhancing the practicality and accelerating the deployment of mobile agents.

Abstract: Network management is gaining increasing importance due to the pervasiveness of computer networks. Neverless, mainstream approaches to network management are presently limited by centralized management strategies and poor flexibility-a consequence of their rigid client-server architecture. In this paper we analyze how to overcome these problems by new design paradigms and technologies encompassing the capability to relocate dynamically the components of a distributed application. We evaluate the opportunities offered by this approach and provide feasibility considerations, also discussing a few interim architectural solutions adopted in our on-going implementation work.

Keyword: code mobility, mobile agents, network management

Abstract: Active networks are receiving increasing attention due to their promises of great flexibility in tailoring services to applications. This capability stems from the exploitation of network devices whose behavior can be changed dynamically by applications, possibly using technologies and architectures originally conceived for mobile code systems. Notwithstanding the promises of active networks, real-world applications that clearly benefit by them are still missing. We describe the design of a videoconference system conceived expressly for operation over active networks. The goal of this activity is to pinpoint the benefits that mobile code and active networks bring in this application domain and to provide insights for the exploitation of these concepts in other application domains.

Abstract: Driven by the question of how to identify potential communication partners and the need for well-suited communication schemes in agent-based systems, the authors discuss two communication concepts: sessions and global event management. Sessions establish either actively or passively a context for inter-agent interactions. Communication partners are addressed by globally unique agent identifiers or via badges. Communication in sessions is based on RPC or message mechanisms. Global event management addresses the need for anonymous communication. Event managers are employed as a synchronization means within agent groups. Based on this approach, they introduce synchronization objects-active components that offer various synchronization services. The presented model is finally mapped onto OMG event services.

Abstract: Orphan detection in distributed systems is a well researched field for which many solutions exist. These solutions exploit well defined parent-child relationships given in distributed systems. But they are not applicable in mobile agent systems, since no similar natural relationship between agents exist. Thus new protocols have to be developed. In this paper one such protocol for controlling mobile mobile agents and for orphan detection is presented. The `shadow' approach presented in this paper uses the idea of a placeholder (shadow) which is assigned by the agent system to each new agent. This defines an artificial relationship between agents and shadow. The shadow records the location of all dependent agents. Removing the root shadow implies that all dependent agents are declared orphan and eventually be terminated. We introduce agent proxies that create a path from shadow to every agent. In an extension of the basic protocol we additionally allow the shadow to be mobile. The shadow approach can be used for termination of groups of agents even if the exact location of each single agent is not known.

Abstract: The technical enforcement of intellectual property (IP) rights often conflicts with the ability to use the IP. This is especially true when the IP is data, which may easily be copied while it is being accessed. As the electronic commerce of data becomes more widespread, traditional approaches will prove increasingly problematic. In this paper, we show that the mobile agent architecture is an ideal solution to this dilemma: by providing full access to the data but charging for the transmission of results back to the user-results-based billing-we resolve the access vs. protection conflict. We define new requirements for agent frameworks to implement results-based billing: "data-aware accounting" and "data-tight sandboxing", which, along with common requirements such as authentication, authorization, agent self-monitoring and efficiency, provide the mechanisms by which database owners can effectively grant users access to their intellectual property.

Abstract: Very large networks with thousands of applications and millions of users pose serious problems to the current traditional technology. Moreover, if synchronised client behaviour exists then the limitations are strongly felt. Mobile agent systems can be a suitable technology to answer to these requirements if some aspects of the architecture are carefully designed. This paper proposes a cooperative mobile agent system with a very dynamic and scalable trading service. The system allows applications to deploy servers onto the network to respond to demand making them self-configurable. Clients can also use mobile agents with performance gains. Sets of simulations were performed to study the scalability of the overall system.

Abstract: Mobile code technology is gaining growing importance, for example for electronic commerce applications. For the widespread use of mobile agents, many security aspects have to be seriously considered, and security problems have to be solved to convince potential users of this technology. So far, most work concerning security in the area of mobile code has been done to protect hosts from malicious agents. However, in the more recent literature, approaches have been discussed which lead to different levels of security for the mobile agent against attacks by dishonest hosts. A central problem consists in the integrity of computation: in order to profit from mobile agent technology, techniques have to be used which guarantee the correctness of the results returned by a mobile agent to its originator. In this paper, we explain a general approach to cope with the integrity problem by supplementing computation results with very short proofs of correctness which can be checked, a posteriori, by the originator of the mobile code to verify whether the result is reliable or not.

Abstract: This paper focuses on agents and their interplay in manufacturing on two levels: one is the production plant control level employing agents that cope with malfunctions of production units in order to keep the production flow up and running. The other is the production plant configuration level incorporating agents for the investigation of improved production plant topological layouts in an anytime fashion. The interplay of agents from both levels represents a new approach in production plant configuration, taking into account the highly dynamic flavour of the scenario imposed by the liability to machine failures. We present experimental results and experiences with our implementation of the lower level and propose an elaborate architecture of the upper level.

Keyword: agents, manufacturing, scheduling, configuration

Abstract: In this paper we present the basic design philosophy of the Bond agent system, the multi-plane agent model and the component-based architecture implementing the model. We discuss several applications of Bond agents: resource discovery, an adaptive video service, a workflow management system, a system of agents for remote monitoring of web servers, and a network of PDE solvers.

Keyword: agents, resource discovery, workflow, remote monitoring, scientific applications

Abstract: JavaSeal is a secure mobile agent kernel that provides a small set of abstractions for constructing agent applications. This paper describes the design of these abstractions and their implementation. We address the limitations of the Java security model that had to be overcome, and then present a medium-sized e-commerce application that runs over JavaSeal.

Abstract: Existing mobile agent systems are often constructed with a focus on intelligence and autonomy issues. We have approached mobility from a different direction. The area of distributed systems research is quite mature, and has developed mechanisms for implementing a "sea of objects" abstraction. We have used this as our starting point, and have added to this the ability for objects to move from host to host, whilst maintaining location-transparent references to each other. This provides a powerful and straightforward programming paradigm which embraces programming language semantics such as strong typing, method invocation and encapsulation. We have built a mobile object workbench on top of a flexible Java middleware platform, which can be used as a the basis for a mobile agent system. In this paper, we examine the philosophy and design of the mobile object workbench and describe how this is being extended to provide a security framework oriented towards agents.

Abstract: The paper surveys several coordination models for mobile agent applications and outlines the advantages of uncoupled coordination models based on reactive blackboards. On this base, the paper presents the design and the implementation of the MARS system, a coordination tool for Java-based mobile agents. MARS defines Linda-like tuple spaces that can be programmed to react with specific actions to the accesses made by mobile agents.

Abstract: A particular challenging area where agent technology is increasingly applied is the Communication Networks field. As networks become increasingly complex, hard to manage and control the ideal of a distributed, intelligent network management and control system is becoming more and more of a necessity. This paper concentrates on the allocation of service demands spanning network domains owned and controlled by distinct operators. In particular, the focus is on an agent-based solution that has been defined in order to allow automatic intra-domain resource allocation and inter-domain negotiations between peer operators. Self-interested agents interact in order to define consistent end-to-end routes that satisfy Quality of Service requirements, network resources availability and utility's maximisation.

Keyword: agent negotiation, agent interaction, agent communication

Abstract: GMD FOKUS, IBM and The Open Group have jointly submitted to the OMG a specification on the Mobile Agent Facility for evaluation and adoption. The paper gives an overview of the technical requirements of the OMG Request for Proposal, the design goals of the joint submission, the architecture of the proposed Mobile Agent Facility, and the salient features of its interfaces.

Abstract: Intelligent agents have been shown to be a good approach to addressing the issues of mobile computing. However, before the approach can be commercially viable, a set of management capabilities that support the controls of intelligent agents in a mobile environment need to be in place. Since controls can only be applied after the target agent is located, an effective agent search algorithm is an indispensable part of the management functions. The authors propose a new algorithm, the highest probability first algorithm, for locating the target agent. The approach makes use of the execution time information to reduce cost and network traffic. The execution time of the agent on a server is assumed to be binomial distributed and therefore is more realistic.

Abstract: To realize its promise of providing scaleable and optimal use of network resources, mobile agent technology must be integrated with non-mobile architectures (e.g., client-server) while taking into account the specific needs of applications. The paper introduces a general characterization of mobile agent applications that provides a framework for reasoning about such optimization issues. This framework is refined to enable analysis of alternative mobility policies, and the development of a new strategic mobility policy that uses the application characterization to intelligently decide on mobility. The analysis shows that the extreme cases of `always mobile' and `always stationary' are sub-optimal with respect to strategic mobility. Strategic mobility has been implemented in a software prototype called Mobile-AgentX and successfully tested in an application of concurrent engineering of micro-electro-mechanical systems (MEMS).

Abstract: Jocaml is a system for mobile agents built inside the Objective-Caml language. Jocaml eases the development of concurrent, distributed and mobile agent based applications, by expressing useful distribution abstractions using a small set of simple but powerful primitives taken from the Join-Calculus[5]. The system provides total transparency for migration, application states (after migration, all threads resume their execution in the state before migration), communications (communication channels with other agents are kept during migration) and composition (sub-agents migrate with their parent agent). Other features of the Jocaml system are mobile objects with transparent remote method invocation, distributed garbage collection, failure detection and execution efficiency. Jocaml has already been used in several applications, such as a mobile editor, some distributed games and a distributed implementation of Ambients[4]. This paper describes the Jocaml programming model and language, its current implementation and some interesting applications.

Abstract: Jackal is a Java-based tool for communication using the KQML agent communication language. Some features that make it extremely valuable to agent development are its conversation management facilities, flexible, blackboard style interface and ease of integration. Jackal has been developed in support of an investigation of the use of agents in enterprise-wide integration of planning and execution for manufacturing. This paper describes Jackal at a surface and design level, and demonstrates its use in a multi-agent system that supports intelligent integration of enterprise planning and execution.

Abstract: In a de-regulated market the distribution utilities will compete with added value for the customer in addition to the delivery of energy. We describe a system consisting of a collection of software agents that monitor and control an office building. It uses the existing power lines for communication between the agents and the electrical devices of the building, such as sensors and actuators for lights, heating, and ventilation. The objectives are both energy saving and increasing customer satisfaction through value added services. Results of qualitative simulations and quantitative analysis based on thermodynamical modeling of an office building and its staff using four different approaches for controlling the building indicate that significant energy savings, up to 40 per cent, can be achieved by using the agent-based approach. The evaluation also shows that customer satisfaction can be increased in most situations. In fact, this approach makes it possible to control the trade-off between energy saving and customer satisfaction (and actually increase both in comparison with current approaches).

Keyword: multi-agent systenms

Abstract: An open problem in mobile agent research is the interaction between agents in different `worlds'. The authors survey various approaches to agent interaction and discuss the interoperation between Telescript and ffMAIN agents through the use of the hypertext transfer protocol (HTTP) and the shared information space provided by ffMAIN.

Abstract: Mobile Maude is a mobile agent language extending the rewriting logic language Maude and supporting mobile computation. Mobile Maude uses reflection to obtain a simple and general declarative mobile language design and makes possible strong assurances of mobile agent behavior. The two key notions are \emph{processes} and \emph{mobile objects}. Processes are located computational environments where mobile objects can reside. Mobile objects have their own code, can move between different processes in different locations, and can communicate asynchronously with each other by means of messages. Mobile Maude's key novel characteristics include: (1) reflection as a way of endowing mobile objects with ``higher-order'' capabilities; (2) object-orientation and asynchronous message passing; (3) a high-performance implementation of the underlying Maude basis; (4) a simple semantics without loss in the expressive power of application code; and (5) security mechanisms supporting authentication, secure message passing, and secure object mobility. Mobile Maude has been specified and prototyped in Maude. Here we present the Mobile Maude language for the first time, and illustrate its use in applications by means of Milner's cell-phone example. We also discuss security and implementation issues.

Keyword: mobile agent, programming language, mobile object, mobile code, object oriented, security

Abstract: This paper describes a guide system and the software architecture of an autonomous interactive robot based on a multi-agent system. A robot navigation system is developed so that the robot can guide people through halls in various types of exhibitions. An infrared location system is implemented on hallway ceilings, and thus the environment is part of a sensor-distributed robot system. Real-world guide agent plays the role of "ROBOT-MEDIA" to integrate information between the information space of the mobile computer and the physical space of the exhibits in order to guide visitors in physical space. This research aims to develop a cooperative adaptive systems as two-way communication among space, media and human beings to construct transparent knowledge boundaries between real space and virtual space generated from computer data is using shared-space technology to create distributed intelligence on shared space in order to manage the communication and guide in a laboratory. The experimental results of using an autonomous mobile robot equipped with a compass verify this proposed software architecture.

Keyword: a multi-agent system, autonomous mobile robot, mobile assistant, real-world guide agent

Abstract: We describe a way to save and restore the state of a running Java program. We achieve this on the language level, without modifying the Java virtual machine, by instrumenting the programmer's original code with a preprocessor. The automatically inserted code saves the runtime information when the program requests state saving and re-establishes the program's runtime state on restart. The current preprocessor prototype is used in a mobile agent scenario to offer transparent agent migration for Java-based mobile agents, but could generally be used to save and re-establish the execution state of any Java program.

Abstract: Mobile agents add a new communication paradigm to traditional network communication mechanisms. In contrast to the classical mechanisms like remote programming, RPC, or client-server systems, mobile agents have specific advantages when used in a heterogeneous networking environment such as the World Wide Web. So far, the pervasiveness of publicly available mobile agent platforms is not given. Offering a seamless integration of mobile agents into the widespread and well-accepted WWW environment is crucial for the success of mobile agents. One of the growing fields of interest in the Web is the area of electronic commerce. Mobile Web-commerce agents could play a prominent role in future electronic commerce scenarios, if the malicious host problem could be solved. Our paper describes the integration of mobile agents into the Web and the use of Java cards to allow a mobile agent to store and transport data securely. This should promote the usage of mobile agents for electronic commerce purposes.

Keyword: mobile agents, electronic commerce, trusted computing base, WWW, security, smartcard, java card

Abstract: We present a programming model and system support for the development of self-monitoring distributed applications, which sense changes in their networked environment and react by dynamically relocating their components. The monitoring service provides two unique capabilities. First, it enables to perform application-level monitoring, as opposed to only conventional system-level monitoring, without interfering with the basic application logic. Second, it enables dynamic relocation of the monitoring components, in addition to the migration of the monitored components, again, without requiring changes inside application components. The monitoring service and programming model were implemented as a subsystem of FarGo, a programming environment for dynamically-relocatable distributed application. In addition to a programming language interface, relocation can be programmed using a high-level script language, and manually controlled using a graphical tool that tracks component relocations.

Keyword: agent-based monitoring, programming model, mobile code

Abstract: The opportunities offered by the Internet are encouraging research aimed at the creation of a computational infrastructure that exploits the wide spread communication infrastructure. The mobile computation paradigm is a proposal to build a computational infrastructure that goes beyond the well-known client-server paradigm and increases dynamicity and flexibility. Despite the promising first steps, there is still confusion on the role of paradigms and technology in the development on applications based on the mobile computation paradigm. We present a case study in which we develop serveral versions of an application using different paradigms and different technologies in order to show when these concepts come into play and which are their relationships.

Keyword: mobile code, design paradigms, case study

Abstract: Mobile agents, programs that move within a system performing a set of tasks, are an active field of research. The focus of current research, however, is on the development of execution platforms and applications for mobile agents and not on methodologies for building agents. Creating mobile agents can be tedious and susceptible to errors. We propose a framework where the agent is composed using a well-defined set of categories of software components. Building systems from software components has already proven useful in the context of large software systems, increasing the productivity of the development process and the reliability of the resulting system by reusing proven components. We claim that the same holds true for the construction of mobile agents for network and systems management as well as for other domains. We have designed and implemented an agent construction toolkit (the AgentBean Development Kit-ADK) to demonstrate the usability and flexibility of this approach.

Keyword: Java, Mobility, Systems, Tools, Software Components, Network Management, Agent Construction

Abstract: Wireless communication with mobile computing devices is known to be problematic. It is very different in character from conventional communication over wired networks. Since many distributed applications make assumptions about network characteristics, they may not be used in a hostile mobile environment. The authors are proposing a new kind of messaging system which incorporates adaptive behaviour into the messages themselves. They call these `smart messages', and implement them using mobile agents. They are transported between machines via agent airports. The metaphor used is of a message being delivered by a courier (mobile agent), through agent airports, on a potentially unresolved route. The `intelligence' is in the messages (couriers in our metaphor) themselves rather than in the network. The approach taken expands on self-routing capabilities of current mobile agent systems such as Aglets or Telescript. They aim to provide structured support for handling the particular problems associated with wireless communications. These include very limited, variable and asymmetric bandwidth, frequent and prolonged disconnections, geographical mobility and high usage costs. They argue that this offers an efficient, adaptable and robust solution to many of the problems associated with this hostile communications environment.

Abstract: In this paper, we present experiences from building several mobile agent-based distributed applications using the agent system TACOMA (Tromsoe And COrnell Moving Agents). Our hope is to demonstrate mobile agent applicability potential through some real and concrete examples. We conclude that mobile agents, even if they simplify remote installation of software, basically complement other structuring techniques in distributed applications.

Abstract: Cooperating merchants establish a distributed marketplace under the auspices of an independent market authority. Each merchant's server is equipped with a trusted device, a smart card for example, provided by the market authority. The market authority plays the role of a trusted third party for the customer as well as for the merchants. This paper describes protocols that prevent the malicious alteration of the data collected by visiting mobile agents roaming through the marketplace without being detectable by subsequent servers or by the owner of the agent upon its return. Another protocol makes the trusted device a secure execution platform for routines provided by the agent owner.

Keyword: agents, e-commerce, security, mobile agent

Abstract: When mobile agents do comparison shopping for their owners, they are subject to attacks of malicious hosts executing the agents. We present a family of protocols that protect the computation results established by free-roaming mobile agents. Our protocols enable the owner of the agent to detect upon its return whether a visited host has maliciously altered the state of the agent, thus providing forward integrity and truncation resilience. In an environment without public-key infrastructure, the protocols are based only on a secret hash chain. With a public-key infrastructure, the protocols also guarantee non-repudiability.

Abstract: Systems comprised of multiple interacting mobile agents provide an alternate network computing paradigm that integrates remote data access, message exchange and migration; which up until now have largely been considered independently. This paper focuses on basic one-to-one agent interactions, or paradigms, which can be used as building blocks; allowing larger system characteristics and performance to be understood in terms of their combination. This paper defines three basic agent paradigms and presents associated performance models. The paradigms are evaluated quantitatively in terms of network traffic, overall processing time and size of memory used, in the context of a distributed DB system developed using the Bee-gent Agent Framework. Comparison of the results and models illustrates the performance trade-off for each paradigm, which are not represented in the models, and some implementation issues of agent frameworks.

Keyword: agents, agent interaction

Abstract: Role theory [6] deals with collaboration and coordination; roles have also been applied to distributed systems management [31] and to agent and robot systems [2, 43]. However, it has been difficult to realize these representations in an automated or semi-automated system, due to the lack of adequate formalism and corresponding abstractions in software. Role models are relatively new concepts in object-oriented software engineering that emphasize patterns of interaction and therefore rectify this situation for software analysis and design. This paper provides examples of agent role models and explains how role modeling can be used to facilitate agent system analysis and design. We also discuss role model implementations based on two approaches: the Role Object pattern and aspect-oriented programming (AOP).

Abstract: As various Internet services, e-commerce, and information systems permeate our lives, their continual availability becomes a dominant issue. But continuing software evolution requires system reconfiguration. Running systems must upgrade their components or change their configuration parameters. In addition, Internet services often need to serve thousands or millions of users. This scenario raises three conflicting issues: availability, configurability, and scalability.

We propose the use of mobile reconfiguration agents for the efficient, secure, and scalable dynamic reconfiguration of Internet systems. We describe a CORBA object-oriented framework that supports dynamic reconfiguration and allows customization to different kinds of computing environments ranging from PDAs and embedded systems with limited resources to powerful workstations.

Keyword: mobile agent, CORBA, distributed system

Abstract: With the advances of wireless communication, mobile computing will become ubiquitous. This will lead to adaptive computing, where electronic services support the user anytime, anywhere and in every form. Adaptive services must take the user requirements, the current geolocation, the situation he or she is facing, the available network connection and the associated usage costs into account. This very much relies on flexible and adaptive system mechanisms in the network. This paper presents the results of the ACTS AMASE project which has adapted a mobile agent system for the use in wireless networks to support mobile users. We concentrate on issues related to making mobile agent-based services aware to the context of the user. We examine (a) a mechanism that allows agents to adapt their behavior to the current situation in the wireless network, and (b) a system mechanism that automatically adapts agent execution to the given context. Finally we present an overview about the current Europe-wide AMASE test bed and the realized application.

Abstract: Mobile agents are a new paradigm for distributed computing that is especially well suited for mobile computing over global wireless networks. This paper describes the approach taken in the ACTS (Advanced Communication Technologies and Services) OnTheMove project to integrate a mobile agent system into the Mobile Application Support Environment (MASE), a middleware for mobile computing. In this project, an existing mobile agent system was adapted for the requirements of mobile computing. We present the changes that had to be made to the agent system to adapt it to wireless communication. We also present some of the application areas where a mobile agent system is suitable for mobile communication. We describe an agent-based pre-fetcher application where an agent operates disconnected from the user on the fixed network and prepares Web pages for the anticipated next connection of the user using the quality-of-service trading functions available in MASE.

Abstract: Third-generation mobile networks will be characterized by service variety and multi-provider scenarios, requiring new concepts for service control and location management. Mobile agents seem to be appropriate for service customization and user localization. However, it is very unclear whether or not their migration costs lead to an overload of the underlying signaling network. This paper answers the question of whether the use of mobile agents in telecommunication systems makes sense at all, by analytically comparing a conventional stationary concept with mobile agents. These analyses are based on various call and movement behavior patterns of mobile customers, and on measurement results achieved in the mobile agent system JAE (Java Agent Environment).

Abstract: There is considerable interest in developing runtime infrastructures for programs that can migrate from one host to another. Mobile programs are appealing because they support efficient utilization of network resources and extensibility of information servers. This paper presents a scheduling scheme for allocating resources to a mix of real-time and non real-time mobile programs. Within this framework, both mobile programs and hosts can specify constraints on how CPU should be allocated. On the basis of the constraints, the scheme constructs a scheduling graph on which it applies several scheduling algorithms. In case of conflicts between mobile program and host specified constraints, the schemes implements a policy that resolves the conflicts in favor of the host. The resulting scheduling scheme is adaptive, flexible, and enforces both program and host specified constraints.

Keyword: Mobile programs, CPU allocation, Java virtual machine, Security

Abstract: This paper presents a mobile agent approach that aims at satisfying the following requirements of open Internet-based electronic service markets: the mobile agent system should be usable by any Internet user without a need for specifically configurated non-standard software tools. It should reduce costs in mobile computing environments and therefore enhance overall efficiency. It should suit well to an electronic service market where local services are commerically offered and business transactions predominate the interaction between customers and suppliers.

As a part of the project OSM (Open Service Model), mobile agents are built on top of two well-established technologies: CORBA and Java. The first is used as a conceptual framework for interoperability, the latter as the programming environment. Since Java does not provide the necessary persistency of execution state, the concept of OSM service profiles is used to embed Java classes and to transfer a coarse-grained execution context in a secure and efficient manner.

Abstract: Communication between mobile agents and their users is an interesting but largely unresearched topic with important applications. We investigate the various types of interaction required and propose a versatile, easy-to-implement and secure WWW-based method to allow mobile agents to interact with arbitrary users.

Abstract: MASIF is a standard for mobile agent systems which has been adopted as an OMG technology. It is an early attempt to standardize an area of industry that, even though popular in the recent past, still has not caught on. In its short history MASIF has raised interest in industry and academia. There are already a number of projects pursuing MASIF reference implementation. MASIF addresses the interfaces between agent systems, not between agent applications and the agent system. Even though the former seem to be more relevant for application developers, it is the latter that impact interoperability between different agent systems. This paper describes two sets of interfaces that constitute MASIF: MAFAgentSystem and MAFFinder (the acronym MAF is used for historical reasons). MASIF extensively addresses security. The paper provides a brief description of MASIF and its interfaces, data types and data structures.

Abstract: Hive is a distributed agents platform, a decentralized system for building applications by networking local system resources. This paper presents the architecture of Hive, concentrating on the idea of an "ecology of distributed agents" and its implementation in a practical Java based system. Hive provides ad-hoc agent interaction, ontologies of agent capabilities, mobile agents, and a graphical interface to the distributed system. We are applying Hive to the problems of networking

"Things That Think," putting computation and communication in everyday places such as your shoes, your kitchen, or your own body. TTT shares the challenges and potentials of ubiquitous computing and embedded network applications. We have found that the flexibility of a distributed agents architecture is well suited for this application domain, enabling us to easily build applications and to reconfigure our systems on the fly. Hive enables us to make our environment and network more alive.

Keyword: agents, mobile agents, distributed objects, ubiquitous computing, Java, networks, embedded systems

Abstract: The provision of a reliable communication infrastructure for mobile agents is still an open research issue. The challenge to reliability we address in this work does not come from the possibility of faults, but rather from the mere presence of mobility, which slightly complicates the problem of ensuring the delivery of information even in a fault-free network. For instance, the asynchronous nature of message passing and agent migration may cause situations where messages forever chase a mobile agent that moves frequently from one host to another. Current solutions rely on conventional technologies that either do not provide a solution for the aforementioned problem, because they were not designed with mobility in mind, or enforce continuous connectivity with the message source, which in many cases defeats the very purpose of using mobile agents.

In this paper, we propose an algorithm that guarantees delivery to highly mobile agents using a technique similar to a distributed snapshot. A number of enhancements to this basic idea are discussed, which limit the scope of message delivery by allowing dynamic creation of the connectivity graph. Notably, the very structure of our algorithm makes it amenable not only to guarantee message delivery to a given mobile agent, but also to provide multicast communication to a group of agents-another open problem in research on mobile agents. After presenting our algorithm and its properties, we discuss its implementability by analyzing the requirements on the underlying mobile agent platform, and argue about its applicability.

Keyword: mobile agents, communication mechanism, message delivery, multicast

Abstract: Quality of service (QoS) is receiving strong attention due to its key role in distributed multimedia computing. QoS is not a new concept, being usually employed in the domain of computer networks to specify a set of parameters typically assigned to transport connections. As a rule, QoS is established through negotiation between service users and providers. The process of negotiation is simple if the resources are managed by a centralized entity (e.g. an operating system) or by a set of homogeneous entities such as a network protocol. Unfortunately, in distributed multimedia applications the negotiation and management of resources is a difficult task since resources are very diversified, dispersed and maintained by heterogeneous entities. To cope with diversity and distribution, an agent-based approach for QoS negotiation and management in open distributed environment is proposed in this paper.

Abstract: Straightforward approaches to team coordination with the expressive power of finite state automata are doomed to fail under a wide range of heterogeneity due to the combinatorial explosion of states. In this paper we propose a coordination scheme based on operational semantics, which allows an extremely compact and modular way of specifying soccer-robot team behaviors. The capabilities of our approach are demonstrated on two examples, which, though just being simple demo implementations, perform very well in a simulator tournament.

Keyword: agent coordination

Abstract: Mobile agents have been advocated to support electronic commerce over the Internet. While being a promising paradigm, many intricate problems need to be solved to make this vision reality. The problem of \emph{fair exchange} between two agents is one such fundamental problem. Informally speaking, this means to exchange two electronic items in such a way that neither agent suffers a disadvantage. We study the problem of fair exchange in the mobile agent paradigm. We show that while existing protocols for fair exchange can be substantially simplified in the context of mobile agents, there are still many problems related to security which remain difficult to solve. We propose three increasingly flexible solutions to the fair exchange problem and show how to implement them using existing agent technology. The basis for ensuring the security properties of fair exchange is a tamper-proof hardware device called a trusted processing environment.

Keyword: mobile agent, e-commerce, security

Abstract: This paper proposes a multi-agent system comprising of mobile agents communicating with a fixed infrastructure of agent servers. This script-based multi-agent system enables easy access to Internet services and combines remote execution to support mobile users and migration of agents within the agent server network. Such an environment offers a higher degree of flexibility and efficiency to the user by performing much of the work on the server located in the fixed network. The focus of this paper lies on the description of the agent server architecture and the agent definition methods.

Keyword: mobile agent, agent server, multi-agent system, agent services

Abstract: We describe a platform for the portable and secure execution of mobile agents written in various interpreted languages on top of a common run-time core. Agents may migrate at any point in their execution, fully preserving their state, and may exchange messages with other agents. One system may contain many virtual places, each establishing a domain of logically related services under a common security policy governing all agents at this place. Agents are equipped with allowances limiting their resource assets, both globally per agent lifetime and locally per place. We discuss aspects of this architecture and report about ongoing work.

Keyword: migration, multi-language, interpreter, Tcl, C, byte code, Java, persistence, authentication, security domain

Abstract: General Dynamics Information Systems (GDIS), in cooperation with the U.S. Navy, has been applying agent-based technology to intelligently retrieve distributed data and developing personalized tools to assimilate and manage the knowledge inherent in that data. Our initial hypothesis was that by filtering information based on user need, we could greatly decrease the amount of remote information transferred and increase the value of information locally available to the user. In evaluating our hypothesis, we found that the user's information needs were implicit; that is not codified. By explicitly capturing the information requirements, we could repeat our initial experimentation and extend our solution to a generalized set of knowledge management problems. As a driving scenario for this research we worked with 3rd Fleet Staff to formalize and expedite the process of gathering and organizing information for their daily situation briefs. Our experience includes a refined list of knowledge management issues and lessons learned in applying agent-based technology.

Keyword: agent coordination, knowledge management, information dissemination, agent system integration

Abstract: Although a thorough evaluation of mobile code technology does not exist yet, some studies already evidenced that the powerful (and often heavyweight) abstractions and mechanisms proposed so far are not always flexible enough to fully exploit the benefits of migrating code. muCODE is a new mobile code toolkit designed to be flexible, extensible and lightweight. Its small set of abstractions and mechanisms can be used directly by the programmer or composed in higher-level abstractions-mobile agents included. This paper discusses the fundamental concepts and features of muCODE, together with its rationale and motivation.

Abstract: Operators of telecommunications networks dedicate significant human and capital resources to managing their networks. Current network management approaches, however, do not scale well in dynamically evolving and expanding networks, cannot easily be customized for (or by) human users, and involve high overhead costs.

We desire to collect management-related data across a changing domain of networked components, and to periodically compute aggregated statistics, or infer events, based on that data. These aggregated results must be generated in a way that is minimally "intrusive" into other network operations, yet are produced with the desired periodicity. To do this, the aggregation process must be adaptive to changing environmental conditions. Such requirements suggest that these aggregation operations must be distributed throughout the network, rather than centralized at a single platform.

We describe a conceptual model of an "aggregation network" of adaptive mobile agents for this purpose, and describe the operation of a testbed for demonstrating simple aggregation network concepts using mobile-agent technology.

Keyword: network monitoring, service monitoring, network management, adaptation, aggregation, telecommunications, intrusive, dataflow, distributed, Aglets, CORBA, SNMP

Abstract: In this paper we deal with the study of the actual convenience of using the agent programming paradigm for accessing distributed service. We try to point out the benefits of such a communication paradigm, by providing an analytical study of its basic features in comparison with client-server approach and remote evaluation. The results that we have obtained show how agents must not always be considered the only solution to any communication issue, since in several cases their use might even reveal a drawback. In this paper we present several models of non-Markovian Petri nets, which have been solved through the WebSPN tool, and we provide a close comparison between the agents technique, the client-server and the remote evaluation. We also focus our attention on providing some practical remarks, which can help the developer during the design, in order to select the communication paradigm which best suits to the features of the application that has to be developed.

Keyword: Client-Server, Remote Evaluation, Agent technology, Petri nets

Abstract: Mobile code is slowly gaining acceptance but it is still not clear where it is really useful. If not used judiciously it may incur greater complexity of programming and degradation of performances. The goal of this paper is to show that mobile code is particularly well suited as a glue for the composition of immobile services, where flexibility and extensibility are necessary. To support our claim we describe two services and one application that have been programmed with mobile code in the context of active networking. We study the impact on the flexibility, complexity and performances of the resulting systems. We observe positive effects on flexibility and complexity and acceptable performance penalties.

Keyword: Mobile code, Active network, Service configuration, Event propagation

Abstract: In order to maintain their performance in a dynamic environment, agents may be required to modify their learning behavior during run-time. If an agent utilizes a rule-based system for learning, new rules may be easily communicated to the agent in order to modify the way in which it learns. However, if an agent utilizes a connectionist-based system for learning, the way in which the agent learns typically remains static. This is due, in part, to a lack of research in communicating sub-symbolic information between agents.

In this paper, we present a framework for communicating neural network knowledge between agents in order to modify an agent's learning and pattern classification behavior. This framework is applied to a simulated aerial reconnaissance system in order to show how the communication of neural network knowledge can help maintain the performance of agents tasked with recognizing images of mobile military objects.

Abstract: The abstractions and protocol mechanisms that form the basis for inter-agent communications can significantly impact the overall design and effectiveness of Mobile Agent systems. We present the design and performance analysis of a reliable communication mechanism for Mobile Agent systems. Our protocols are presented in the context of a Mobile Agent system called AGNI. We have developed AGNI communication mechanisms that offer reliable peer-to-peer communications, and that are integrated with our agent location tracking infrastructure to enable efficient, failure-resistant networking among highly mobile systems. We have analyzed the design parameters of our protocols using an in-situ simulation approach with validation through measurement of our prototype implementation in real distributed systems. Our system assumptions are simple and general enough to make our results applicable to other Agent systems that may adopt our protocols and/or design principles.

Keyword: mobile agent communication, mobile agent message passing, distributed reconfigurable scripting, mobile sockets

Abstract: CarPAcities provides a highly dynamic, easy to use, cost effective and safe approach to car sharing. It is designed to be tightly integrated into today蘳 mobile network infrastructure. The underlying model as well as an agent-based prototype implementation employing the Jini technology for service management are described throughout this paper. Simulation results underpin the effectiveness of CarPAcities.

Keyword: agent application

Abstract: Network management comprises of monitoring and control of a network. Ideally a network management system should be able to provide the facility of managing the network from any site and from any system and network. However the present network management systems are centralized and tied to particular systems. We introduce the unique concept of a mobile network manager (MNM) to provide network administrators with a system and location independent network manager which will be useful under a variety of circumstances. In order to enable the MNM mobile agents have been used. The mobile agents are furnished by the mobile agent environment MAGENTA (Mobile AGENT environment for distributed Applications) which provides autonomous, reactive, proactive and communicative mobile agents. MAGENTA has been designed and implemented to support mobile user aware applications. The mobile agents are used not only to facilitate the disconnected mode of functioning of the MNM but also to decentralize network management functionalities. The applicability and usefulness of mobile agents in implementing the functionalities of MNM has been compared with the client-server mechanism.

Abstract: This paper proposes a Java bytecode transformation algorithm for realizing transparent thread migration in a portable and efficient manner. In contrast to previous studies, our approach does not need extended virtual machines nor source code of target programs. The whole state of stack frames is saved, and then restored at a remote site. To accomplish this goal, a type system for Java bytecode is used to correctly determine valid frame variables and valid entries in the operand stack. A target program is transformed based on the type information into a form so that it can perform transparent thread migration. We have also measured execution efficiency of transformed programs and growth in bytecode size, and obtained better results compared to previous studies.

Keyword: Java, bytecode transformation, binary transformation, mobile agent, thread migration

Abstract: In this paper we present practical experiences gathered from the employment of two popular Java-based mobile-agent platforms, IBM's Aglets and Mitsubishi's Concordia. We present some basic distributed computing models and describe their adaptation to the mobile-agent paradigm. Upon these models we develop a set of frameworks for distributed database access over the World-Wide Web, using IBM's Aglets and Mitsubishi's Concordia platforms. We compare the two platforms both quantitatively and qualitatively. For the quantitative comparison, we propose, employ, and validate an approach to evaluate and analyze mobile-agent framework performance. For the qualitative assessment, we present our observations about the programmability and robustness of, and mobility provided by, the two platforms.

Abstract: MobiDoc is a framework for building mobile compound documents, where the compound document can be dynamically composed of mobile agents and can migrate itself over a network as a whole, with all its embedded agents. The key of this framework is that it builds a hierarchical mobile agent system that enables multiple mobile agents to be combined into a single mobile agent. The framework also provides several added-value mechanisms for visually manipulating components embedded in a compound document and for sharing a window on the screen among the components. This paper will describe the MobiDoc framework and its first implementation, currently using Java as implementation language as well as component development language, and then illustrate several interesting applications to demonstrate the utility and flexibility of this framework.

Keyword: mobile agent, software component, compound document

Abstract: There is an increasing desire to use constellations of autonomous spacecraft working together to accomplish complex mission objectives. Multiple, highly autonomous, satellite systems are envisioned because they are capable of higher performance, lower cost, better fault tolerance, reconfigurability and upgradability. This paper presents an architecture and multi-agent design and simulation environment that will enable agent-based multi-satellite systems to fulfill their complex mission objectives, termed TeamAgent. Its application is shown for TechSat21, a U.S. Air Force mission designed to explore the benefits of distributed satellite systems. Required spacecraft functions, software agents, and multi-agent organisations are described for the TechSat21 mission, as well as their implementation. Agent-based simulations of TechSat21 case studies show the autonomous operation and how TeamAgent can be used to evaluate and compare multi agent-based organisations.

Keyword: multi-agent application

Abstract: Traffic telematics applications, such as road traffic management systems, operate in an extremely dynamic mobile computing environment. The mobile agent paradigm then becomes a promising alternative to the conventional client/server approach. In this article, we evaluate the application of mobile agent technology in the area of vehicular traffic management and introduce a general application partitioning model which facilitates the combination of asynchronous and autonomous operation, data filtering and scheduling in a user-specific manner. A sample application and performance results are presented.

Abstract: The paper presents a service-oriented platform for the development and execution of distributed applications based on contracting stationary and migrating services. Services are seen as active objects build on top of middleware using OMG/CORBA and added features. Customized services add to the middleware the ability to handle transparently application start-up and distribution according to load-balancing and inverse caching application demand. Services can be considered of any kind ranging from scientific specialized processing to data archiving juke-boxes. An application on system management in scientific experimental environment drives the work on some aspects of the architecture.

Keyword: mobile agents, service-oriented architecture, agents distribution, distributed processing, ORB, load-balancing

Abstract: Gives an overall overview of the AgentSpace framework, a next-generation Java mobile agent system developed on top of the ObjectSpace Voyager system. We first introduce the notion of dynamic and distributed agent-based applications and argue that the AgentSpace features are suitable to support them. The AgentSpace novelties include: flexible and dynamic association between agents, security policies and users; transparency of agent location through the use of views; and an easy and clean way to create agents through the use of abstract classes and method factories.

Abstract: The authors continue arguing that persistence is a fundamental requirement to support the development of next-generation agent-based applications. After a general overview of mobility and persistence to clarify the main issues discussed in the paper, they propose a tentative list of facilities that should be supported by persistent mobile agent systems. The main contribution of the paper is a survey of existing persistent and mobile agent systems that includes a comparison based on how well (or badly) they support the proposed list of facilities.

Abstract: This paper presents a fault-tolerance protocol for mobile agent executions that tolerates long-term failures of agencies. If the agency where an agent execution is being performed fails for a long-time, the execution can be recovered and continue at another agency. This is not only important for avoiding a mobile agent execution to become blocked, but it also contributes for enforcing the autonomy of organizations in an open environment emitting mobile agents to execute applications that cross the boundary of autonomous organizations. The protocol presented in this paper is based on mobile agent replication. Our protocol differs from previous work mainly in the sense that an agent can execute more than a single atomic transaction at an agency; it integrates distributed storage of recovery information; and it supports partial recovery of the activity carried out at an agency. The motivation of this work is on building a support for the execution of open nested transactions with a set of mobile agents.

Abstract: The paper describes a model of agent-based transactions, i.e., distributed transactions which are realized with the use of mobile agents, for an environment of heterogeneous and autonomous systems. Combining an asynchronous operation mode (provided by agent mobility) with the enforcement of complex control flows with transactional semantics, agent-based transactions represent an adequate base concept for reliable processing in future distributed systems, such as virtual enterprises and electronic markets. Among other properties, agent-based transactions provide a very suitable support for transaction processing in massively distributed environments; support high adaptivity to dynamically changing environments; and are adequate to support processing requirements of mobile devices. The authors consider only transactions which are implemented by a single mobile agent. The presented transaction model is being developed in the context of the MAGNA mobile agent project, under development at GMD Fokus and Technical University of Berlin.

Keyword: distributed transaction processing, mobile agents, mobile agent platforms, workflows, extended transaction models

Abstract: Implementing an application as a mobile agent may improve the application's functionality and performance, but may have a detrimental effect on overall system performance. In this paper we consider the effect of moving an application from a client to a file server (as an agent), both on the application and the server. Under what circumstances does application performance improve, and does it come at the expense of other (non-mobile) background applications using the same server? We use a trace-driven simulation to measure the effect of mobile code, allowing system parameters such as the size of the server memory and server speed relative to client speed to be varied. We found that several factors influence the benefit of mobile agents. Server memory does not appear to be a significant problem; relatively small server caches have a high hit rate even when shared with mobile agents. The relative CPU performance of the client and server has a bigger effect on system performance: mobile agents should not be run on the server if its CPU is a bottleneck.

Abstract: Many real-world tasks can be decomposed into pipelines of sequential operations (where subtasks may themselves be composed of one or more pipelines). JGram is a framework enabling rapid development of such multi-agent systems. Each agent's services are specified in the JGram Description Language (JDL), and automatically converted into Java source code. These services may be invoked synchronously (analogous to function call) or asynchronously (analogous to message passing), in a manner that is transparent to the service's implementation. Complex tasks are created by composing several agent services into hierarchical JGram pipelines in which each agent may dynamically delegate its subtasks to other agents in a recursive manner, and in which errors are handled by a cross-agent exception mechanism. Although JGram agents communicate using Java's Remote Method Invocation (RMI) protocol, the framework provides significant enhancements such as authentication, encrypted channels and dynamic service specification. JGram has been used to develop several real-world agent systems. This paper discusses ARGUS, a visitor identification system that integrates a security camera with face detection, face recognition, and user notification systems to automatically identify regular visitors arriving at the front door of our building.

Abstract: NOMADS is a Java-based agent system that supports strong mobility (i.e., the ability to capture and transfer the full execution state of migrating agents) and safe agent execution (i.e., the ability to control resources consumed by agents, facilitating guarantees of quality of service while protecting against denial of service attacks). The NOMADS environment is composed of two parts: an agent execution environment called Oasis and a new Java-compatible Virtual Machine (VM) called Aroma. The combination of Oasis and the Aroma VM provides key enhancements over today's Java agent environments.

Keyword: mobile agent, Java, thread migration, virtual machine

Abstract: An often claimed benefit of mobile agent technology is the reduction of communication cost. Especially the area of information filtering has been proposed for the application of mobile filter agents. However, an effective coordination of agents, which takes into account the current network conditions, is difficult to achieve.

This contribution analyses the situation that data distributed among various remote data servers has to be examined with mobile filter agents. We present an approach for coordinating the agents' employment, which minimizes communication costs. Validation studies on the possible cost savings for various constellations show that savings up to 90% can be achieved in the face of actual Internet conditions.

Keyword: distributed filtering, mobile code, Internet, communication cost, coordination of mobile agents

Abstract: This paper describes a mobile agent-based approach for supporting coordination of user activities in distributed collaborations. The approach presented here uses XML to specify a collaboration plan in terms of various participants' roles, access rights based on roles, and the coordination actions to be executed when certain events occur. Using this plan an agent-based distributed middleware system provides each user an interface to perform the tasks pertaining to the collaboration. The actions of a user transparently create and dispatch coordination agents to other users. The middleware also enforces the security constraints defined in the collaboration plan. We illustrate our approach with an example system for collaborative authoring implemented using the Ajanta mobile agent system.

Keyword: mobile agent, agent collaboration

Abstract: In this paper, we present a mechanism to capture and reestablish the state of Java threads. We achieve this by extracting a thread's execution state from the application code that is executing in this thread. This thread serialization mechanism is implemented by instrumenting the original application code at the byte code level, without modifying the Java Virtual Machine. We describe this thread serialization technique in the context of middleware support for mobile agent technology. We present a simple execution model for agents that guarantees correct thread migration semantics when moving an agent to another location. Our thread serialization mechanism is however generally applicable in other domains as well, such as load balancing and checkpointing.

Keyword: mobile agent, thread migration, Java

Abstract: Mobile code technology is leading to a new type of "open systems": instead of applying openness to a standardization process one now requires the running systems to become open for foreign code. The question then is how far this technical openness can go for mobile code. The fewer constraints one imposes on hosts running mobile code, the more the benefits of mobile code can be exploited. However, there must necessarily be basic constraints regarding the utilization of resources which are always finite and most of the time will be operated near the saturation point. The author argues in favor of openness even at the level of resource allocation. He links this topic to (open) market models, describes the mechanisms developed so far for communication messengers and shows how they are used to allocate resources in an open way. Finally he presents experimental results of validation runs which help to test these mechanisms.

Keyword: mobile code, communication messagers, open resource allocation, market, computational ecosystems

Abstract: The diversity of research and development work on agent technology has led to a strong distinction between mobile and intelligent agents. This paper presents an architecture aiming at providing a step towards the integration of these two aspects, concretely by providing an approach of dynamically embedding negotiation capabilities into mobile agents. In particular, the requirements for enabling automated negotiations including negotiation protocols and strategies, a plug-in component architecture for realizing such requirements on mobile agents, and the design of negotiation support building blocks as components of this architecture are presented.

Abstract: MESSENGERS is a system that supports the development and use of distributed applications structured as collections of autonomous objects. With self-migrating computations, the main challenge is the extraction and subsequent restoration of the computation's state during migration. This is very difficult when the navigational statement may be placed anywhere in the code and, hence, many systems place the burden of state capture on the application programmer. We describe an intermediate approach, where the use of navigational statements is restricted to the top level of the self-migrating computation. This permits an efficient implementation of a fully transparent state capture and restoration. We demonstrate that this approach is applicable not only to interpreted mobile code but also to compiled self-migrating computations executing entirely in native mode.

Abstract: Large computer systems are expected to have long lifetimes. For this reason, a system should evolve as user and system demands change. Attempts have been made at this in the field of distributed systems where configuration languages have been developed for just this purpose. Multi-agent systems provide a better solution to this problem but lack the development environments that are available for traditional software. The paper presents a generic agent shell and two agent system languages, ADLe (definition language) and ARCLe (reconfiguration language) for use with this shell. It is proposed that they will provide the basis for an environment that fills this gap.

Keyword: re-configuration, multi-agent systems, agent programming

Abstract: Quality of Service (QoS) routing generally requires fast reaction times, tight coupling of interactions between routing systems and mechanisms for ensuring that actions taken throughout the network are coherent. Our previous work showed how an agent based QoS routing approach can benefit significantly from making controller agents mobile and allowing them adapt the information and control distribution in the network over time. This paper discusses how giving mobile agents organisational models can bridge the gap between the need for tight, fast coordination and freedom to move around the network. Furthermore coordination is achieved without imposing any globally or external controls on the mobile agents in the system.

Keyword: mobile agent, agent coordination, quality of service, routing, adaptation

Abstract: We study the distributed infrastructures required for location-independent communication between migrating agents. These infrastructures are problematic: different applications may have very different patterns of migration and communication, and require different performance and robustness properties; algorithms must be designed with these in mind. To study this problem we introduce an agent programming language - Nomadic Pict. It is designed to allow infrastructure algorithms to be expressed as clearly as possible, as translations from a high-level language to a low level. The levels are based on rigorously-defined process calculi, they provide sharp levels of abstraction. In this paper we describe the language and use it to develop an infrastructure for an example application. The language and examples have been implemented; we conclude with a description of the compiler and runtime.

Keyword: Distributed Infrastructure, Location Independence, Communication, Coordination, Languages, Mobility, Systems, Theory

Abstract: Use of the Internet and the World-Wide-Web has become widespread in recent years and mobile agent technology has proliferated at an equally rapid rate. The authors introduce the Concordia infrastructure for the development and management of network-efficient mobile agent applications for accessing information anytime, anywhere, and on any device. Concordia has been implemented in the Java language to ensure platform independence among agent applications. The design goals of Concordia have focused on providing complete coverage of flexible agent mobility, support for agent collaboration, agent persistence, reliable agent transmission, and agent security. Concordia offers a flexible scheme for dynamic invocation of arbitrary method entry points within a common agent application and extends the notion of simple agent interaction with support for agent collaboration, which allows agents to interact, modify external states (e.g., a database), as well as internal agent states. Concordia provides support for agent persistence and recovery and guarantees the transmission of agents across a network. Concordia has also been designed to provide for fairly complete security coverage from the outset. An alpha release of Concordia is available.