Embedded WiSeNts - Project FP6-004400

The main objectives of the studies carried out in the Embedded WiSeNts Coordination Action are:

• In-depth analysis of the current state-of-the-art in Cooperating Embedded Systems and Wireless Sensor Networks.
• Identify open issues and trends in the field.

In the scope of the project, a Cooperating Object (CO) is defined as a collection of sensors, actuators, controllers or other COs that communicate with each other and are able to achieve, more or less autonomously, a common goal. The inclusion of other cooperating objects, as part of CO itself indicates that these objects can combine their sensors, controllers and actuators in a hierarchical way and are, therefore, able to create arbitrarily complex structures.

Five studies and a whitepaper on have been performed:

• Study 3.1.1: Applications and applications scenarios.
• Study 3.1.2: Paradigms for algorithms and interactions.
• Study 3.1.3: Vertical Systems Functions.
• Study 3.1.4: System Architectures and Programming Models.
• Study 3.2: Visions for Innovative Applications.

The following figure points out some relations between the Studies.

It should be noted that the first study deals with the applications that can be understood today and have been implemented or could be implemented in the next ten years. In addition, an analysis of visionary applications is being conducted in Embedded WiSeNts. The studies and the Visionary Applications will be the basis of the Research Road Map that will provide guidelines to achieve the full vision of cooperating objects and wireless sensor networks.

This document provides an initial overview of cooperating object (CO) applications and application scenarios that can be readily understood today. The main objective of the study is to identify relevant state of the art, projects and activities in the CO domain. For this purpose, both European projects and other projects outside Europe are considered. Some application scenarios that enable us to better understand the area of CO in the wide sense of the term from two different perspectives: socio-economic and application-type aspects have been identified and developed.

The applications and scenarios take into account the state of the art of current data-centric approaches (wireless sensor networks). In data-centric approaches efficient management of data is the major concern whereas service-centric approaches are mostly concerned with the definition of the interface or API in order to provide functionality for the user. The wide spectrum of potential applications indicates that the constraints for a CO application domain may be much different from another CO application domain. Therefore, first, general application characteristics for all domains are identified. Then CO applications have been categorized. CO applications can be classified in many different ways as each application has common features with others. In this study, sectors which have a social and economic impact in the society are used as the basis of classification. The classification is then used as a basis for the analysis of common characteristics in the field of action. Some application scenarios from different domains have been chosen and they are given along with their typical parameters, requirements, roles, traffic, threads, legal/economic issues that best characterize the research performed in the field of cooperating objects in the wide sense of the term. Considering the current trends, sectoral areas which can benefit from cooperating objects are defined as follows: control and automation, healthcare, environmental monitoring, security&surveillance, logistics, home&office, transportation, tourism and education&training.

In this study, you can also find:

• General characteristics of data-centric CO applications (Section 3).
• A survey of state-of-the-art CO projects in the order of evolvement (Section 4).
• Classification of CO applications and projects into sectoral areas (Section 5).
• The object symbol set used for the functional description of CO scenarios (Section 6).
• Detailed sample scenarios from control, surveillance, monitoring and transportation domains along with their typical parameters, requirements, roles, traffic, threads, legal/economic issues that best characterize the research performed in the field. (Section 7).
• A summary of the outcome of the study which later be used as a measurement for application domain's importance and will act as a means of weighting conflicting requirements (Section 8).

One of the primary targets of the Embedded WiSeNTs consurtium is fostering the development of systems based on cooperating-objects in the near future. To achieve such a goal, however, it is first necessary to understand what is already in place, what is ongoing and what is still missing in this area.

This study was conceived to provide an up-to-date overview of the fundamental design paradigms, algorithms and interaction patterns that enable the realization of systems based on Cooperating Objects (COs).  In order to cope with the large heterogeneity of CO-systems, the literature has been divided in four Thematic Areas, namely Wireless Sensor Networks for Environmental Monitoring, Wireless Sensor Networks with Mobile Nodes, Autonomous Robotic Teams and Inter Vehicular Networks. Such thematic areas differ for characteristics and requirements and, we believe, are representative for a number of possible application scenarios. 

For each thematic area, the study provides a rather comprehensive survey of the most interesting algorithms proposed in the literature for the following aspects: Medium Access Control, Routing, localization, Data Aggregation & Data Fusion, Time synchronization, Navigation, Objects coordination & Cooperation.  The algorithms and paradigms are, hence, classified according to the requirements that have been identified by Study 3.1.1. In this way, the study permits to clearly identify commonalities and differences among the different design paradigms adopted in each specific system, thus revealing the most promising research trends and the research gaps that should be covered in the next future.

To summarize, the study includes:

• a survey of the most relevant algorithms in the field of 
• Wireless Sensor Networks for Environmental Monitoring (Section 5);
• Wireless Sensor Networks with Mobile Nodes (Section 6);
• Autonomous Robotic Teams (Section 7);
• Inter Vehicular Networks (Section 8);
• a classification of the algorithms according to the application requirements (Section 9);
• a discussion concerning the research gaps and the most interesting research trends in the area (Section 10).

This document complements other three studies in identifying the relevant state of the art in the context of distributed cooperating objects (COs) and wireless sensor networks (WSNs). The most important results of these surveys will guide the preparation of a roadmap in this research domain.

The set of characteristics exhibited in CO applications are more diverse than the ones found in applications of traditional wireless and wired networks. Study 3.1.1 (Applications and Application Scenarios) examined this set of requirements in detail covering among other issues the location and context of COs, security, privacy and trust, system scalability and reliability, to list a few.

Critical factors impact the architectural and protocol design of such applications. Organising software and hardware components of COs into a framework that can cope with the inherent complexity of the overall system will be an important exercise for application developers.

The current operating systems proposed for WSNs and COs cannot offer all the required functionality to the these applications. The main goal of this study is to discuss the roles and effects of vertical system functions (VFs), which are defined in this document as the functionality that addresses the needs of applications in specific domains and in some cases a VF also offers minimal essential functionality that is missing from available real-time operating systems.

We revisited and reviewed the most relevant application requirements and discuss the most suitable VFs to address these needs. The document is organised as follows:

• Definition of vertical system functions in the context of cooperating objects (Section 3).
• Discussion of the characteristics and requirements of the CO application studied in WP3.1.1 - Applications and Application Scenarios (Section 4).
• Different types of VFs to address these application needs are then discussed (Section 4). The VFs discussed are:
  • Context and Location Management.
  • Data Consistency.
  • Communication functionality.
  • Security, Privacy and Trust.
  • Distributed Storage and Data Search.
  • Data Aggregation.
  • Resource Management.
  • Time Synchronisation.
• The document concludes with a summary of trends and open issues identified (Section 5). Briefly, the most urgent issues that deserve attention of the research community are the distributed accurate location of COs and system support for node mobility and sensor heterogeneity. The software frameworks require consolidation through standardised APIs and high-level descriptive languages. In practical terms, there is a need for deployment and configuration of sensors in real application scenarios.

This study provides a survey about the current state of the art of programming models and system architecture for Cooperating Objects and motivates their importance for a successful development of these technologies. Section 2 provides a brief introduction to the topics and motivates the need of designing suitable programming abstractions for Cooperating Objects. In Section 3, the most relevant existing programming abstractions are surveyed and classified. The main reason for the development of these abstractions is to allow a programmer to design applications in terms of global goals and to specify interactions between high level entities (such as agents or roles), instead of explicit managing the cooperation between individual sensors, devices or services. For example, the database abstraction allows to consider a whole sensors network as a logical database, and performing network-wide queries over the set of sensors. The various paradigm are surveyed and a set of criterions allowing to easily review their strengths and weaknesses are presented.

Section 4 presents the existing system architectures for Cooperating Objects at two different levels: first, system architectures of individual nodes, which includes the structure of the operating system running at node level and its facilities; second, system architectures supporting the cooperation of different nodes, such as communication models.

Finally, in section 5, the document points out some of the limitations of current approaches, and proposes some research perspectives. In particular, programming paradigms should provide more support to ease of programming, heterogeneity, as well as scalability issues. Regarding system architectures, real-time aspects, which are not currently well addressed, will become increasingly important for Cooperating Objects. Dynamic maintenance (such as code deployment and runtime update support) is another important issue to address in future systems. At last, an effort is required to better integrate the various paradigms and systems into a unified framework.



Top

Study 3.2: Visions for Innovative Applications


This work is licensed under a Creative Commons Attribution-ShareAlike 2.5 License.

Study leader: UCAM-CL
Contributors: TUB, UCAM-CL, SICS, UT, YTU, AICIA, INRIA.

Whereas the "application study" discusses concrete applications that can be well understood and characterised today, this whitepaper explores visions for application areas that could potentially be realised once a wide-ranging technology of cooperating objects becomes available. The focus here has a 10 year time horizon and is aimed at stimulating exploratory application studies as opposed to the analytical emphasis of the other WiSeNts studies. To achieve this goal, we elicited application visions from three sources:

1. Researchers at the WiSeNts partner institutions and our industrial partners.
2. A small scale competition held during the 2005 Dagstuhl Summer School on wireless sensor networks.
3. A large scale public Sentient Future Competition (SFC) organised by the WiSeNts personnel with substantial prizes generously donated by Deutsche Telekom Laboratories. The competition received 79 entries which were evaluated by more than 25 reviewers.

• Immediate is used for those scenarios ready to be picked up in industrial development. As expected, a few entries (only three out of 33) were within this timescale.
• Middle-term refers to research activities that could start now but with planned development in 5 years. 18 of the entries in this document were classified in this group.
• Long-term visions account for the cases for which any research should begin after 5 years, mostly because of unavailability of the required technology and infrastructure. 12 of the visions are believed to be of long-term.