Research

Research Topics

Service Computing: Resilient Service-oriented Architectures, Service Design, Service Selection, Service Discovery, adhoc and late binding composition, middleware (ESB), service driven processes

Security and Web Services: Security-by-design (Risk Management, Security Services), Security Policies (WS-Policy), end-to-end security (Federated Identity Management, SSO), …

Connected Devices: reusable component-based connected devices, self-adaptation, validation and verification, Hardware/Software design, verifiable composition and assembly strategies, Machine-to-Machine (M2M) middleware;

IoT Big Data Analytics: infrastructures and languages to collect, storage, perform built-in analytics and simultaneously access connected devices at a large scale;

Security and Connected devices: User-centric security in trustless environments, Blockchain-based identity management, smart contract-based access controls, software Moving Target Defense, …

Research Description

The premise of the Internet of Things is to connect almost everything from cyber and physical worlds to create innovative services. Without loss of generality, smart services are IT-enabled services consisting of connected devices and enabling by Big Data Analytics to deliver new facilities and allow intelligent behaviors. Services cover an entire gamut of business models, including opportunities in smart cities, living labs, Internet of services, battle and cyber-space services, e-learning, e-health, etc. In this context, scalability and security are pillars of the Internet of Things and successful smart services.

My current research work primarily contributes to the design and delivery of “smart and secured services” driven by connected devices and analytical supports. To this end, I focus on challenges related to the unconventional characteristics of the Internet of Things. These unconventional characteristics are due to:

1) the burst of diverse protocols (i.e., LoRa, ZegBee, BlueTooth, MQTT, CoAP, XAMPP, …);

2) lacks of standards (e.g., Machine-to-Machine middlewares and IoT architectures, …);

3) hardware’s limited resources (constrained computation and storage capabilities (Arduino, Raspberry, ARM microcontrollers, …), and;

4) mobility (i.e., wearable devices, mobile phones, drones, robots…).

The IoT unconventional characteristics lead me to investigate a new software engineering approach appropriate for smart services. Based on exprtise in service computing and inspired by the object-oriented paradigm, my research strategy seeks to design and implement a cyber-physical component-based model for connected devices aware of limited resources, self-adaptable to contextual changes, and enabled with security capabilities in trustless and decentralized environments. Our component model aims at encapsulating physical and cyber properties and intends to be interoperable, reusable and secure. In addition, high-volumes, heterogeneous and real-time streaming of data generated by connected objects require an adequate chain of Big Data analytical models and manipulation languages to infer facts from distributed devices at a large scale and provide prespective decision supports. My goal is to refine the component-based connected device model at different levels ranging from hardware, middleware, software architecture, to security and data analytics in order to bring the premise of the Internet of Things into a reality. Based on my previous contributions on service computing, my current research topics cover (but limited to) the following:

  1. Interoperable and Reusable Connected Devices: How to buid modular and reusable component-based device model, proactive self-adaptation, validation and verification, Hardware/Software design, verifiable composition and assembly strategies, Machine-to-Machine (M2M) middleware?
  2. Self-Adaptable Connected Devices: How to adapt connected devices to in response to changes and abnormal behavior detection?
  3. IoT Big Data Management and Built-in Analytics: How to design and deploy infrastructures and languages to collect, storage, perform built-in analytics and simultaneously access connected devices at a large scale?
  4. Security and Connected devices: How to build radically a new identity for the Internet of Things, and its corresponding access control and authentication mechanisms based on blockchains and smart contract?

Application Domains: Smart Services (Smart Home, Smart City,…), IT-enabled Services, Wearable Sensors, e-health, Industry 4.0, Battle and Space Management, Crisis Management Systems, …

.

Doctoral Thesis Supervision

Xiaoyang ZHU                                                                                                                                     Septembre 2015 – now

Thesis: Blockchain-based Identity and Access Management for the Internet of Things

Arthur GATOUILLAT                                                                                                                                October 2015 – now

Thesis: Designing Smart services with connected devices and stream event analytics and their application to wearable biomedical connected devices

Maroun ABI ASSAF                                                                                                                                   February 2014-now
Thesis: Integrating streams of Artifacts in the Internet of Things

Pascal Bou Nassar, October 2007 – December 21 2012:

Title: “Managing security in dynamic service infrastructures from a risk management perspective”

Problem: The lack of security awareness in SOA reference models and design methods leads to the development of Web services without effectively supporting security goals and/or efficiently handling potential threats and attacks at design time and runtime. In addition, any Web service security policy that does not take into accounts technological, business and organizational aspects would be incomplete and unstable. The security risk management has been extensively covered in closed environments and stable information system architectures. By such, traditional information systems’ security risk management methods are not well adapted to deal with interoperable and dynamic SOAs in distributed and open environments.

Main contributions: Pascal Bou Nassar proposed a SOA design method integrating risks management for end-to-end information security driven by business needs, service lifecycle and infrastructure assets. His design method relied on the integration of risk management process into the SOA design process in order to evaluate security risks at each stage of the SOA lifecycle (e.g., design, monitoring, and execution). The design method consisted of nine steps covering service identification and specification as well as risk assessment and mitigation at design time. He also covered the runtime with security annotations attached to each Web service profile and security brokers that predict security vulnerabilities by checking daily updated security databases.
Funding: Agence Universitaire de la Francophonie
Current position: Assistant Professor at Université de Saint Joseph Beirut, Lebanon

 

Ahlem Zayati, October 2006 – October 9 2012:
Title: “Mise en œuvre des architectures orientées services pour les systèmes d’information industriels”
Problem: Lean Manufacturing strategies allow enterprises to meet multiple constraints such as cost reducing or just in time response to customized products by focusing on business and production processes that generate value and reducing wastefulness. Though, industrial information systems include often legacy systems (ERP, MES, PLM, and SCM) developed according to specific business needs and consequently, they cause information redundancy, heterogeneity and inconsistency due to the abscense of standards ensuring interoperability.

Main contributions: Ahlem Zayati proposed an extended service-oriented infrastructure to improve industrial information system agility and interoperability in the context of lean strategies. She extended the industrial services proposed by Layth Sliman to reorganize business and production activities from tangible and intangible resources and designed a Lean Enterprise Service Bus, extended by a business semantic layer. The Lean ESB includes four modules; the mediation module improves information interoperability among legacy systems, the dynamic choreography module builds production processes in accordance with customer demands based on industrial service compositions, the intelligent routing module organizes workshop resources and reorganizes processes in a pull flow strategy, and the governance module monitors production performance and quality.
Current position: IT services Consultant at Solutec, Paris

 

Peng Yong, October 2008 – February 2012:
Title: “Modelling and designing IT-enabled service systems driven by requirements and collaboration”
Problem: IT-enabled services cover services facilitated by ICT and achieved through collaboration among service stakeholders to satisfy customer needs, improve service quality, and increase customer satisfaction in social centered and distributed environments. IT-enabled services have received an abundant attention only from a technological perspective as they are often reduced to suitable e-services or Web services. This is a narrow context in which e-services or Web services only focus on user-software interactions to process information and automate tasks. Nevertheless, understanding, analyzing, designing and implementing IT-enabled services require service-modeling frameworks, which cannot be enured by Enterprise Architecture Frameworks.

Main contributions: Yong Peng developed a service architectural framework for designing IT-enabled service systems and supporting their implementation with SOA-aided infrastructures. He focused on designing and building IT-enabled services driven by customer requirements and collaboration among service actors. He developed a collaborative service design method based on a service system reference model with multiple views to develop and bundles IT-enabled services. He also proposed an end-to-end requirement model to capture customer needs at every step in the service design lifecycle.
Funding: Chinese Research Council
Current position: Researcher in Beijing, China

 

Layth Sleiman,              September 2006 – 27 November 2009:
Title: “Collaborative Business and Urbanisation of Enterprise Information Systems”
Problem: Collaborative businesses involve several partners that exchange resources and information through Web services for example, and share common goals such as proposing product-service offers. By such, collaborative processes require agile, reactive and interoperable information systems. Urbanisation in computer engineering is a modelling appraoch that ensures reactivity and evolution of information systems in response to changes. It applies rules to decompose information systems into functional organizational structures with autonomous and self-managed modules, including blocs each of which inludes components. By such, production processes, which are often built across different modules, become rigid and connot support on-demand collaborative processes. In the collaboration context, information security is crucial. When building collaborative processes, exchanged information and computational infrastructure require on-the-fly security configurations despite that partners many have different security policies, requirements and security schemes, which is commonly known as the security interoperability problem.

Main contributions: Within the Inter-Prod project, Layth Sliman proposed a new urbanization modeling approach to build resource-driven organizations of information systems and allow agile reconfigurations of production processes to support collaboration (Production-as-a-Service). In his urbanization approach, he decomposed the production system into production objects defined in terms of logical clusters of interrelated resources (tools, machines, raw material or competencies) and activities. Appropriate computational services, called industrial services, ensure the manipulation of production objects and help in building on-demand collaborative production processes. Processes are designed from the ground up based on the production object model to meet requirements of rapidly evolving businesses collaboration. Regarding the security interoperability problem, a technical solution based a federated identify-based management systems was adapted to PEtALS ESB in the SemEUse project to ensure an end-to-end security enforcement in collaborative environment.
Current position: Associate Professor, École d’ingénieur dans les technologies de l’information (EFREI), France
Funding: Research Contract (SemEUse Project and Inter-prod)

 

Alida Esper, October 2006 – September 1st 2010:
Title: “Intégration des approches SOA et orientées objets pour modéliser une orchestration cohérente de services”
Problem: In the context of service-oriented architectures, the requirement that Web services should be loosely coupled renders them reusable in different orchestrations. In addition, most of Web service discovery and selection approaches consider services independent of each other regardless their success or failure in previous orchestrations. This assumption makes Web services invariants at design-time and run-time and renders them insensitive regarding previous compositions. In social-based environments, users may set up social networks to collaborate and benefit from their social networks to tag, rate or mash up Web services. Nevertheless, implicit relationships between Web services remain unexploited at design time and during Web service discovery and compositions.

Main contributions: Alida Esper developed a hypergraph of services based on SOAs and implicit relationships between software components. These implicit relationships can be established and used at design time or runtime in the context of Web service composition. For example, the inheritance relationship specializes or generalizes a Web service, and thus creates different execution contexts (i.e., versions) whereas the aggregation relationship controls the propagation of contextual information in collaborative processes based on Web services. Recommendation and substitution are hence useful to respectively enrich compositions and ensure robustness orchestration in case of failure. Relationships among software compoenents are adopted and have influenced our work on the Social-based Web service discovery and the ad-hoc Web service composition in the Wenbin Li’s thesis.
Current position: Assistant Professor at Baath University, Syria
Funding: Grant from the Government of Syria.

 

Mayyad Jaber, October 2005 – 19 February 2009:
Title: “Distributed Information Systems for Supply Chain Management”
Problem: Information sharing and exchange are essential activities to interconnect partners’ information systems and evaluate the sypply chain overall performance. Nevertheless, supply chains should also be extended to establish collaboration processes and manage common activities. However, technical interoperability and information security remain challenging tasks.

Main contributions: Within the context of the COPILOTES project, Mayyad Jabber proposed a global workflow infrastructure to design shared collaborative business processes that communicate with partners’ private processes by means of Web services. Based on SOA-based integration services, the global workflow is built as a distributed system enhanced with a distributed and hierarchical transactional mechanisms handle collaboration management of resources accessed by shared collaborative processes.
Current position: Assistant Professor at Baath University, Syria
Funding: Grant from the Government of Syria

 

Wenbin LI, October 2010 – September 2013:
Title: “Bundling IT-enabled Services within the Internet of Services ”.
Problem: In dynamic and social-based environments, building IT-enabled service systems requires adaptative and dynamic SOA-aided infrastructure to support service processes and bundle IT-enabled services.

Main contributions: Wenbin Li works on a high-level structured requirements language to express IT-enabled service requirements and translates them into set of rules, describing structural, contextual constraints and dependency between Web services. He also developed a dynamic and ad-hoc Web service composition based on these rules.
Funding: Chinese Research Council