EVALUATING MOBILE AND GRID SERVICES SOFTWARE TECHNOLOGIES IN MISSION CRITICAL APPLICATIONS

Vincenzo Sabbatino, Emiliano Pandolfi

SELEX-Sistemi Integrati1

Piermaurizio Di Placido

AISoftw@re Technologies and Solutions2

Pierpaolo LoretiUniversità degli Studi di Roma Tor Vergata3

Keywords: Homeland Protection, Mobile Computing, SOA, Middleware, Ad Hoc Network

AbstractIn a mission critical scenario multiple organizations can be involved to tackle an emergency coming from natural or non conventional disaster. However only forces that are superior in terms of information gathering and sharing are able to coordinate their action in order to outperform a threat.In this paper we argue that the adoption of a Network Centric approach rather than Platform Centric is the key to open legacy information systems to the collaboration and the key for evolving them toward a new virtual unique information system made by “Grid of Services”. Moreover when terrestrial telecommunication infrastructures are damaged or severely impaired, alternative and flexible networking arrangements supported by Mobile Computing and Ad-Hoc networks, become critically important to ensure ongoing and effective data provisioning, coordination of emergency response and relief efforts. Those objectives can be achieved exploiting and coordinating two new emerging technologies: Service Oriented Architectures (SOA) and the mobile environment. The first one can be used for integrating heterogeneous systems, the second can be adopted to deploy applications on small devices that can nimbly and flexibly move across the disaster territory, sharing real-time context data with the crisis headquarters. This paper introduces the Homeland Protection context, provides a survey of the main software technology mentioned above and presents a case study where a typical emergency scenario has been taken into account.

IntroductionTimely, accurate information easily accessed and capable of being shared across different stakeholders is fundamental to the decision making capability of those tasked with the homeland security mission. But without the real-time ability to quickly visualize activity

1 SELEX Sistemi Integrati SpAVia Tiburtina Km 12.400 00131 Rome {vsabbatino, epandolfi}@selex-si.com2 AISoftw@re Technologies and Solutions S.p.A.Via Cristoforo Colombo 456, 00175 Rome pdiplacido@ais.it3 Università degli Studi di Roma Tor VergataVia del Politecnico 1 , 00133 Rome pierpaolo.loreti@uniroma2.it

patterns, map locations, and understand the multi-layered geospatial context of emergency situations, homeland security could be not efficient enough. Providing decision makers with a common operational picture enables them to have a better and shared understanding of the situation when making critical decisions is needed. Forces that are superior in terms of information gathering and sharing infact are envisioned to be more likely to outperform a threat. The purpose of this work is to investigate emerging software technology to increase their chances of making right and timely decisions. In this paper, we argue that using Service Oriented Architecture (SOA) and Ad-Hoc network infrastructures can be a good start at solving the problem of modelling a Shared Operation Picture (SOP). This paper is organised as follows: we define the problem domain and the main issues related to systems currently adopted. We also present requirements and needs to move from centralised systems to distributed ones through a Network Centric (NC) approach. Then we introduce some of the main software technologies enabling the NC capabilities and finally a case study showing results of a networked distributed system.

Problem DomainMany military and homeland security operations today need to have a high degree of shared situational awareness and accommodate for distributed collaborative working in order to improve the response to major disasters or non-conventional attacks such as terrorism, and to carry out large-scale military operations.Today’s decision-support systems are highly centralized, with central processing of information and message exchange between sites. Connections are static, and have to be administrated by humans. Such an architecture has several weaknesses: for instance, it is not responsive enough to future users’ needs as Ad-Hoc networking based on geography.To enable each actor in the future mission space to obtain customized and detailed information about the current situation, a more dynamic and decentralized system is needed. More ever deployed task force or rescue units are coordinated by heterogeneous information systems with different policies and strategies simply because every stakeholder involves its procedures, communication mechanism and culture.It’s becoming mandatory therefore the need of make interoperable those systems, in order to achieve the best result during a coordinated action among different agencies, without alter the structure of the information system of every single department.The goal then is to evolve from a centralized approach to a distributed one, in which mobile objects equipped with wireless networking and computers can cooperate within a specific geographical area inside emergency and critical missions.The evolution from 'platform centric computing' to 'network-centric computing' has been largely enabled by recent key developments in information technology.Some of the most important developments in information technology include SOA and Mobile Middleware that will be introduced in the next paragraph.

Distributed Architectures and Mobile TechnologiesSOA and Grid of Services

The SOA model is an innovative perspective that achieves the loose coupling among interactive software agents. In this architecture a Service is a mechanism to enable access to a set of one or more capabilities [1] representing a unit of work implemented by a service provider in order to publish generic information to a service consumer. Both provider and consumer agree on a contract that establishes the conditions of the service, how to invoke it and what is expected from this invocation.This kind of architecture defines a way of invocation in which the consumer can be not human but, on the contrary, can be a “web application” (agent) or a set of applications that achieve the capability to interrogate the huge implicit knowledge of the web.

In fact using SOA architecture is achieved the location transparency of the services thanks to an indexing mechanism that discovers the location of the services at run-time allowing the late binding of the invocation, because a service is invoked only when is needed [2].The loosely coupling nature of the SOA achieved by the interface of the service, described in a well defined contract, isolates the back end implementation giving the opportunity to compose existing services in new kind of services: the consumers of a service will see ever a front end made by a single point of access that wraps the original nature of the legacy systems. In this way it is possible to tail the offering to several kind of consumers providing of a Grid of Services that is not Platform Centric anymore, but distributed on the network.Another benefit of SOA can be perceived during the phases of the analysis, design and development because every development team can be antonomous and distributed like the service implementation: the only constrain is the “contract” that a provider has to attend to.This kind of architecture promotes an easy strategy of coordination between needs of the multiple civil and military agencies, in order to expose services and information to the new Homeland Protection emerging department. In our case study we have implemented our SOA architecture using Web Services [3] application based on Simple Object Access Protocol (SOAP) messages that are processed by an engine executed as an embedded Hyper Text Transfer Protocol (HTTP) server and associated with a Java Servlet engine.The chosen architecture guarantees in the interaction among the Service providers and the Service consumers by means of a standard protocol based on the Extensible Markup Language (XML) making easier the interoperability and the distribution.

Mobility and Middleware for a Mobile Environment

In the latest times among the emerging technologies, a significant innovation has been introduced by an ever more wide proliferation of powerful devices of small dimensions (miniaturization phenomenon): notebook, tablets, palms or smartphones. Such kind of devices is able to support applications running on real operative systems and to support wireless connectivity.Developing software – middleware or applications – for mobile devices has to taking in account that mobile applications are unable to have permanent connectivity due moving across a physical space and this needs a non trivial software solution: applications running on mobile devices must deal with the complexity coming from the continuous shifting of the mobile computing environment.Mobile computing in this case it can be considered the enhanced evolution of the Distributed Computing model: in fact represents a scenario where architectures move from a well defined and fixed topology towards more autonomous and independent configurations.The Mobile Middleware is a new area of research and has been seen by various kinds of approaches based on several communication models (RPC, Queue, RMI, QoS): one of the most promising models is the “Tuple Space Paradigm” because it supports the coordination model approach.The Coordination Computing is an original approach to the communication; this new perspective, inspired by the work on Linda and Generative Communication [4], is a model that allows independence and decoupling among processes: in the message passing communication model for instance, business logic is fully immersed in the communication primitive context. Coordination, on the contrary, realizes a strong decoupling among the semantic definition of software component and interaction mechanisms.It has been observed [5] that software development in mobile environment can be tackled successfully by exploiting a coordination perspective: mobility and the related complexity can be managed using a kind of communication as more as possible anonymous and independent. It has also pointed out [6] that the decoupling between behaviour and communication, fostered by coordination, enables one to separate the applicative behaviour of components from the continuously changing context in which they are immersed.

As is said in [7], “Linda In a Mobile Environment” (LIME) redefines the Linda concepts of a globally accessible Tuple Space into a transient sharing Tuple Space, carried by individual mobile units connected among them with a wireless connection. This “abstraction” gives the mobile units the illusion that their “local” Tuple Space Interface (ITS) was plenty of all the tuple published in the entire federated context space, where federated means that it is distributed among all the mobile units are involved.In fact LIME is a fully distributed middleware for a shared space that is not established upon a dedicated centralizated server as happens for JavaSpaces architectures, but is realized when mobile units “engage” each other.

Ad-Hoc Network on Wireless LAN

The kind of network infrastructure that is possible to use in critical situations is a significant topic. The use of new network techonologies such as “Ad-Hoc” networks can heads to achieve small “infrastructure overriding” supporting action coodination where there is a communication infrastructure failures. Ad-hoc networks in fact is a research field involving Accademical Community, analyzing the ability to have applications using transitory network in mobile environment.In fact during emergencies, when terrestrial telecommunication infrastructures are damaged or severely impaired, alternative and flexible networking arrangements become critically important to ensure ongoing and effective coordination of emergency response and relief efforts. Thus, there is a recognized need for wireless Ad-Hoc communications including high capacity wireless technologies, in order to provide common telecommunication services or networking support [9].A Mobile Ad-Hoc Network (MANET), as defined by IETF in [10], "is an autonomous system of mobile routers (and associated hosts) connected by wireless links -- the union of which form an arbitrary graph." (OMISSIS) "Such a network may operate in a standalone fashion, or may be connected to the larger Internet."Thus, a MANET is a network architecture that can be rapidly deployed without relying on pre-existing fixed network infrastructure. The nodes of a MANET (routers) can dynamically join and leave the network, frequently, often without warning, and possibly without disruption to other nodes’ communication. Finally, the nodes in the network can be highly mobile, thus rapidly changing the network topology and the presence or absence of links. Mobility patterns and radio propagation conditions that vary with time and position can produce intermittent and sporadic connectivity between adjacent nodes and consequently the network topology can rapidly change. As a consequence, a MANET requires a large node density to provide effectively communication services [11]. A novel vision to overcome this problem is presented in [12]. The “Mesh network” architecture considers multi-hop Ad-Hoc networks as a flexible extension of wireless or wired infrastructure networks. Mesh networks are built on a mix of fixed and mobile nodes interconnected via wireless links to form a multi-hop Ad-Hoc network. Basically, the mesh network approach introduces the concept of a wireless backbone to reduce the node density requirement. Moreover this backbone usually relays on long range radio technologies and can be auto-configurable, thus resembling the typical scenarios encountered during emergences. The relief teams can use Ad-Hoc networks technologies that can be supported by short range radio communication such as WiFi while an extended coverage can be obtained with deployable transceiver (in the camps or over the vehicles) using for example the forthcoming WiMAX and WiBro standard [13].

A Case StudyDuring the 2005, an internal research project has started with the name SYSCO GRID (A Grid for Cooperating Systems). Purpose of this initiative is to focus the Network Centric approach through the use of emerging and enabling software technologies.In order to define a reference software architecture, a case study has been considered; through the case study we integrated existing applications and middleware using SOA and Mobile Systems within an Ad-Hoc networks infrastructure.The use case has been conceived taking a harbour as reference place, near of which an emergency situation occurred. The emergency needs to be faced with different means and operators (i.e Police, Fireman, Costal Guard,..) each of them coordinated by their own system. A Vessel Traffic System (VTS) is installed in the port and VTS operators monitor vessel traffic as seen by a radar sensor. An Air Traffic Control (ATC) system is also capable to feed the scenario detecting and monitoring air traffic. Finally ground vehicles need to be tracked in order to plan and coordinate a correct intervention. Purpose of this case study is to provide each actor involved in the emergency with a Common Operational Picture (Figure 1) showing on a map the position and identity of each object.

Air Traffi c Control

Air Traffi c Control

VesselTraffi c Control

VesselTraffi c Control

GroundTraffi c Control

GroundTraffi c Control

Common Operational

Picture

Common Operational

Picture

Air Traffi c Control

Air Traffi c Control

VesselTraffi c Control

VesselTraffi c Control

GroundTraffi c Control

GroundTraffi c Control

Common Operational

Picture

Common Operational

Picture

Figure 1 Common Operational Picture

Starting from this scenario, we set up a prototype (Figure 2) in which a simulator generates tracks as seen and formatted by ATC and VTS systems. Track data are collected by MIDAS (Middleware Services) a middleware developed in Selex SI, which elaborates data as follows:

converts tracks in XML format distributes XML data through a Data Distribution Service (DDS) to any operator

console connected to the Local Area Network (LAN) exports traffic, data and map services as Web Services pushes traffic information in a Tuple Space shared by mobile terminals

Web services are published in a UDDI (Universal Description Discovery and Integration) Server to be accessed by any web application. Each mobile terminal is supposed equipped with a Global Positioning System (GPS) device and a Mobile Middleware based GUI application able to manage the shared context. Identity, position and personal data are then inserted into the Tuple Space to be shared in a coordinated environment.

Web Services

Clients

SQL

Web Services

Clients

Clients

ATCATC

Operator Consoles

VTSVTS

MI DAS BUSMI DAS BUS

LimeJ XTA

Web ServicesWeb Services

ClientsClients

SQL

Web ServicesWeb Services

ClientsClients

ClientsClients

ATCATC

Operator Consoles

VTSVTS

MI DAS BUSMI DAS BUS

LimeJ XTA

Figure 2 SYSCO GRID Prototype

The SYSCO GRID prototype is based on a set of software tools which can be subdivided in two classes: SOA and Mobile Infrastructure.On the Service Side we set up SOA exporting some legacy services through MIDAS and exposing them with a three-tier infrastructure:

Data Layer

This layer is responsible of storage and manage data and it is based on MySQL database [14]. MySQL is a widely used and fast SQL database server. It is a client/server implementation that consists of a server daemon and many different client programs/libraries. We used MySQL to store static data, for example data regarding description of the units (ships, aircraft,…) in of the scenario.

Application Layer

The middle tier is based on the Apache and Tomcat [15]. Apache is one of the most a popular Web server while Tomcat is a Java servlet engine and Java Server Pages processor which can run integrated with Apache. We used Tomcat with the purpose to have a powerful engine perfectly working with Axis [16] that is an open source platform to develop Web Services and service oriented applications in Java language. Along with Axis, we also used another open source tool named gSOAP [17], that is a Web Services development toolkit that offers an XML to C/C++ language binding to ease the development of SOAP/XML Web services in C++ environments.

Presentation layer

The presentation layer have been based on thin and fat client: the thin one is a simple Web Browser Application able to access web pages and invoking transparently the the legacy systems fucntionality through Web Services.On the other side the fat client is represented by the Human Console Interface (HCI) named GIOTTO that is a sophisticated front end application built by Selex-SI. It is used by the operators of the traffic control (Air o Vessel) to manage information related traffic scenarios.

On the Mobile side has been used the Mobile Middleware based on LIME and JXTA:

LIME [18] “is a middleware designed to assist in the rapid development of dependable mobile applications over both wired and ad hoc networks. Mobile agents reside on mobile hosts and all communication takes place via transiently shared tuple spaces distributed across the mobile hosts. At the application level, both agents and hosts perceive movement as a sudden change of context. The set of tuples accessible by a particular agent residing on a given host is altered transparently in response to changes in the connectivity pattern among the mobile hosts” [8]. In the SYSCO GRID prototype a tuple is represented by a set of information such as identity and position. Tuples are also displayed via java GUI in a suitable map representing the surrounding emergency area.

JXTA technology [19] is a set of open protocols that allow any connected device on the network ranging from cell phones and wireless PDAs to PCs and servers to communicate and collaborate in a P2P manner. JXTA peers create a virtual network where any peer can interact with other peers and resources directly even when some of the peers and resources are behind firewalls and NATs or are on different network transports. A Chat example provided by JXTA distribution has been integrated in the prototype to enable a text shell shared among peers.

References [1] OASIS (2006) Reference Model of a Service Oriented Architecture. Commettee Draft 1.0pp. 9 – 10.

[2] Alonso, G. Casati, Kuno, H. Machiraju, M. (2004). Web Services. In RPC and Related Middleware, pp. 35 – 45. Heidelberg, Germany. Springer ISBN 3-540-44008-9

[3] Alonso, G. Casati, Kuno, H. Machiraju, M. (2004). Web Services. In Web Services and their Approach to Distributed Computing, pp. 124 – 149. Heidelberg, Germany. Springer ISBN 3-540-44008-9

[4] Gelernter, D. (1985). ACM Trans. on Programming Languages Systems. Generative Communication in Linda, vol. 7, No.1 pp. 80-112.

[5] Roman, G.-C. Murphy, A.L. and Picco, G.P. (2000). In A. Omicini, F. Zambonelli, M. Klusch, and R. Tolksdorf (Eds). Coordination of Internet Agents: Models, Technologies, and Applications. Coordination and Mobility, pp. 254–273. Heidelberg, Germany. Springer ISBN 3-540-41613-7

[6] Murphy, A. L. and Picco, G. P. (2004). Coordination Models and Languages, Springer. In R. De Nicola, G. Ferrari and Greg Meredith (Eds). Using Coordination Middleware for Location-Aware Computing: A Lime Case Study, pp 1. Heidelberg, Germany. Springer ISBN 3-540-21044-X

[7] Murphy, A.L. Picco, G. P. and Roman, G.-C. (2001). in Proc. of the 21st International Conference on Distributed Computing Systems. LIME: A Coordination Middleware Supporting Mobility of Hosts and Agents pp. 524-533

[8] Picco, G. P. Murphy, A.L. Roman, G.P. (2001). In Proc. of the 21st International Conference on Software Engineering. Lime: Linda Meets Mobility, pp. 368-377, Los Angeles ACM Press FEMA Report .

[9] Federal Emergency Management Agency (2001). Information Technology Architecture Version 2.0. The Road to e-FEMA, Vol. 2.

[10] Baker, F. (2002). An outsider's view of MANET, Internet Engineering Task Force document.

[11] Detti, A. Loreti, C. Loreti, P. (2004).In Communications. Effectiveness of overlay multicasting in mobile ad-hoc network. IEEE International Conference Volume 7,  20-24 and pp. 3891 - 3895.

[12] Bruno, R. Conti, M. Gregori, E. (2005). In Communications Magazine. Mesh networks: commodity multihop ad hoc networks. IEEE Volume 43,  Issue 3,  pp. 123 – 131.

[13] Ghosh, A. Wolter, D.R. Andrews, J.G. Chen, R. (2005). In Communications Magazine. Broadband wireless access with WiMax/802.16: current performance benchmarks and future potential. IEEE Volume 43,  Issue 2, pp. 129 – 136.

[14] Project MySQL: http://www.mysql.com

[15] Project APACHE: http://www.apache.org/; Project Tomcat: http://tomcat.apache.org/

[16] Project AXIS: http://ws.apache.org/axis2/

[17] Project gSOAP: http://www.cs.fsu.edu/~engelen/soap.html

[18] Project LIME: http://sourceforge.net/projects/lime

[19] Project JXTA: http://www.jxta.org

ConclusionsIn our work we have evaluated emerging technologies enabling a Common Operational Picture in critical environment where multiple agencies and resources are involved.This evaluation has taken place recreating a possible scenario adopting as case study the target to integrate different “legacy” and heterogeneous control traffic systems: Vessel Traffic System (VTS) and Air Traffic Control (ATC).Using Web Services technology we have implemented a Service Oriented Architecture in order to access functionality of the two systems also with a simple Web Browser. On the other hand we have published information about tracks, coming from VTS and ATC, in the LIME federated and distributed Tuple Space in order to create a “virtual space” of shared data, that can be accessed by every mobile device moving in the physical space supporting wireless connectivity based on Ad-Hoc network. Generally speaking those prototypes has given the feedback of how new technologies are able to support new scenarios make working together heterogeneous systems in extreme conditions and when traditional infrastructures can be partially or totally damaged.

Authors BiographyDr. Vincenzo Sabbatino received his doctor degree in electronic engineering from the II University of Rome- Tor Vergata,  Italy. In 1992 he joined Alenia Difesa (now part of Selex Sistemi Integrati, a leading European industry implementing complex integrated systems) where he is now responsible of the Software Architectures Unit at the Engineering Department. He has been involved in several projects related to design and implementation of advanced functionalities for the Vessel/Air Traffic Control Systems (VTS and ATC). He has had the technical responsibility of R&D projects in the European Community frame and he is now participating in projects in the software areas with particular regards to distributed systems, software architectures and middleware.

Dr. Emiliano Pandolfi received the degree in Electronic Engineering from the I University of Rome - La Sapienza, Italy. During 2002 he joined Alenia Marconi Systems (now part of Selex Sistemi Integrati, a leading European industry implementing complex integrated systems) where he is working in Rome in the Software Architecture Unit. He has participated in SOC1-Malesya project where he was responsible of the Software Architecture and has realized an high performance distributed database. Then he has participated in the realization of Middleware for the Combat Management System (CMS) of the first Italian Aircraft Carries "Cavour". He actually is working in the Sysco Grid Project in the areas of the Service Oriented Architecture and in the Mobile Middleware.

Dr. Piermaurizio Di Placido received the degree in Philosophy from the II University of Rome - Tor Vergata, Italy. During the 1994 joined Auselda AED Group where participated in modelling and designing large integration systems based on Object Oriented Technology for Public Administration. In 1997 has joined in Artificial Intelligence Software (AIS) where has been responsable of the Component Competence Center and involved in a large number of projects of system integration, essentially oriented to Distributed Object Technology mostly based on CORBA Standard. In the last two years has worked as senior consultant in Selex Sistemi Integrati in the area of CORBA Based distributed middleware. In the recent times has been involved in the Mobile Middleware Technology participating to the Virtual Immersive COMmunication (VICOM) Project, co-financied from the Italian Ministry of Education, Universities and Research (MIUR) and Italian National Research Council (CNR).

Dr. Pierpaolo Loreti received the Dr. Ing. degree in electronic engineering (cum laude) and the Ph.D. degree in telecommunications and microelectronics from the University of Rome "Tor Vergata", Rome, Italy, in 1998 and 2002, respectively.Since 1998, he has been collaborating with the Department of Electronic Engineering, the University of Rome Tor Vergata, Rome, Italy. He has worked on several European and National Project such as IST-SUITED and IST-ACCORD. In the 2000, he was with the University of California, Los Angeles, where he worked on Internet application via satellite constellations. In 2002, he worked with the "Consorzio Radiolabs" on technology for the SOHO environment. From September 2002 to September 2005, he was a Researcher of "Consorzio Nazionale Interuniversitario per le Telecomunicazioni" (CNIT) working on the Virtual Immersive Communication (VICOM) project powered by Italian ministry of University. Currently, Loreti coordinates the VICOM project research activities of the University of Rome Tor Vergata. His current interests include wireless solutions for SOHO environment, ad-hoc networks, context aware communication, and ambient intelligence.