Final Report Summary - VISION (Video-oriented UWB-based Intelligent Ubiquitous Sensing) VISION’s overall objective is the development of an innovative wireless infrastructure providing real-time sensing services, with particular emphasis on 3D video, with mobile and context-aware operation. Numerous challenges raise by the limitations of current technology for wireless sensor networks, thus VISION has proposed:- Exploiting the 60 GHz ultra-wide band radios to enable broadband transmissions, miniaturized devices and reduced interference. - Optimizing the design of VISION's sensor network and customizing solutions at several planes, to face with the inherent high resource consumption of broadband wireless necessary for real-time 3D video. VISION has been now completed. The work has developed according to the requirements targeted over specific application scenarios defined at the beginning of the project. We are proud of being far beyond the typical scenarios exploiting video, audio and sensing, e.g. rescue, public security and entertainment, as we proposed also an interesting and challenging scenario promising a great social impact: the communicative impaired children scenario, where VISION represents the key enabling technology. This is one of the cross-disciplinary developments obtained in the framework of VISION, through the collaboration with researchers in the domain of Cognitive Systems. The definition of VISION's architecture offered the context for applying unconventional approaches based on bio-inspired methodologies, coming from the cell life-cycle and the stem cell. A bio-inspired approach has been pursued also in the definition of the new reconfiguration language PROTEUS that aims at building and managing rules to reconfigure software applications using the concept of Virtual Membrane and will be used in VISION to improve the communication exchange among the context and resource aware QoS management and the reconfiguration components.Within the VISION's framework, the 60 GHz radios represent the enabling communication technology. The characteristics of the propagation channel have been understood through the channel modelling based on the large database of experimental channel responses obtained by a measurement campaign in six scenarios (Office, Residential, Outdoor, Desktop, Corridor, Human) over 8 sub-bands of 1.2 GHz bandwidth each, spanning the spectral regions 54-59 GHz and 61-66 GHz. Several path-loss models have been devised, based on least-square regression fits of the distance and frequency power laws into the experimental path loss data for measurement locations within the Corridor, Office, Human and Outdoor scenarios. The channel temporal dispersion for the Office and Outdoor scenarios, i.e. mean excess delay and RMS delay spread, has been analysed and a cluster model for the power-delay profiles and small scale statistics for Office scenario has been devised. The complete channel model for the Residential environment is being currently refined, and the residential data have been used for performance evaluations of modulations and 60 GHz transceiver schemes by semi-analytical evaluations. We proposed the use of Rake receivers to achieve high data rates for wireless home entertainment with acceptable SNR and low Rake’s complexity, applying the most common modulation schemes, i.e. QPSK, 16-QAM and 64-QAM, defined in the existing standards. The promising results obtained in this study place the Rake receiver architecture as a valuable candidate for future 60 GHz WPANs and WLANs transceivers. The radio transceiver design started by the performance evaluation of the standard IEEE 802.11ad over the real experimental channel impulse responses recorded through the VISION’s measurement campaign. We observed a strong dependence of performance on both frequency and distance due to the sole multipath, which calls for an appropriate selection of the best suitable frequency band according to the required service. We proposed and evaluated the impulse radio ultra-wideband signalling operating at 60 GHz. An adaptive 60 GHz transceiver architecture for FFD nodes based on single-carrier and orthogonal frequency division multiplexing modulation blocks complying as close as possible to the 802.11ad requirements has been prototyped. The designed transceiver architecture proposed by UKIM satisfies the needs for multimedia wireless applications operating in frequently changing environment and using battery-operated devices with limited power source imposed by the VISION project goals. The proposed transceiver can be easily, in operation, adjusted to the current channel characteristics as well as the battery state of the networked mobile devices, using several control signals activated by other parts of the system detecting and measuring the proposed thresholds. The control signals enable automatic adjustment of the device operation modes, thus allowing optimal use of device's resources, while achieving transfer of uncompressed video with HD quality, as required by the emergency scenarios.The adaptable and resource-aware WSN of VISION relies on an agent-based WSN middleware, able to generate agents governing the sensor nodes and to interact with a model-driven approach. The generation is completely automatic and, whenever the agent has been obtained, it is sent to the sensor node and executed over a virtual machine. This way, the sensor node logics is automatically adapted without experiencing service interruption. The approach has been integrated in the VISION infrastructure to bridge the QoS management and the sensor nodes to guarantee best QoS in the presence of context and/or resource changes.The middleware (MW) is able to support a set of intelligent services (audio, video, sensing) for the applications and, at the same time, all the mechanisms needed to manage QoS of such services In particular, for RFD nodes, existing mobile-agents based MW for WSN, called Agilla, has been selected as a starting point and improved in several ways.The different MW components belonging to the VISION MW Framework have been exploited to develop the MW services allowing the resource&context-aware QoS management. Two classes of services have been developed: basic services related to audio streaming and classical sensing activities (i.e. temperature, light, etc…), and advanced services related to video streaming for real-time 3D and security issues (i.e. secure communications and intrusion detection system). Moreover, basic services have been enriched with voice-recognition capability, Morse code generation and smoke detection. We also developed a complete framework, named MAIA (FraMework for Adaptaptive wIreless sensor network Applications), which allows applying Software Engineering techniques to WSN domain, in order to evaluate the QoS of the applications, by performing non-functional analysis (Timing, Performance, Energy Consumption).Usually, the analysis is carried on by transforming an application model into an analysis model, with different syntax, tools and with the need for more knowledge. Our framework integrates the modelling, analysis, code generation and adaptation features for WSNs in a single environment, with no need of knowing other than the OMG fUML standard modelling notation.The QoS evaluation in VISION is based on innovative approaches that exploit the conceptual model we derived that provides a representation of the context and resource attributes in software applications for WSNs. The interpretation of QoS analysis results and the generation of feedback to software developers, is based on a logic-based framework aimed at representing and detecting performance antipatterns in software models, on the automated process of detecting and solving performance antipatterns in Palladio Component Model and on refactoring software models to remove the detected performance antipatterns we proposed so far.We jointly analyse multiple quality attributes and conduct a trade-off analysis through a model-based methodology to quantitatively estimate the system performance while introducing fault-tolerant techniques and/or security mechanisms in order to minimize performance penalties. Finally, we specified and implemented a novel Property Meta-Model that allows us to define and specify quantitative and qualitative (possibly new) properties and metrics in order to allow the applications running on VISION to provide the specification of non-functional requirements, of specific complex events to be monitored and the automatic generation of configurations for the VISION monitoring system. The feedback process may be quite complex since architects may have to assess several design options before achieving the architectural model that best fits the end-user expectations. We introduced a ranking methodology that identifies, among a set of detected antipatterns, the “guilty” ones, i.e. the antipatterns that more likely contribute to the violation of specific performance requirements. The introduction of our ranking process leads the system to converge towards the desired performance improvement by discarding a consistent part of design alternatives. We presented also a lightweight infrastructure that is able to detect performance antipatterns at runtime through monitoring. The proposed approach pre-calculates such predicates and identifies antipatterns whose static, dynamic and deployment sub-predicates are validated by the current system configuration and brings at runtime the verification of performance sub-predicates. The proposed infrastructure leverages model-driven techniques to generate probes for monitoring the performance sub-predicates and detecting antipatterns at runtime.A prototype of the innovative platform for the Intelligent Ubiquitous Sensing has been realized by integrating all the results coming from WP3 to WP5 to realize system integration, practical testing and evaluation of the integrated system and of its sub-systems. The work has been driven by the implementation of one demonstrator acting as proof-of-concept of the VISION intelligent ubiquitous sensing system. It has been evaluated the general efficacy of the approach followed in the project considering a distributed wireless sensor network simultaneously transmitting audio, video, and sensing data. In particular, it has been realized a complete demo scenario for the VISION ERC project, which has been presented in SENSORNETS 2015, in the context of the European Project Space. To show runtime application-dependent adaptation of a WSN, the demo was organized in three main scenarios of operation for the system: Monitoring, Reconfiguration and Video Streaming. It has shown the integration of the VISION architecture elements and it has demonstrated that runtime adaptation in WMSNs is possible, allowing adding new features, previously absent, whenever the need raises, without service interruption.Finally, VISION's achievements are disseminated through many channels, including: − the VISION website (http://www.vision-ercproject.eu);− 43 published/accepted conference papers and 3 journal paper− 11 book chapter acknowledging VISION− the organization of a Session dedicated to VISION at the Third International Workshop on Software Engineering for Sensor Network Applications (http://sesena.info/).− Plenary Talk: D. Cassioli, “Millimeter-waves wireless communications,” at the XI Internat. Conf. ETAI 2013, Ohrid, Macedonia. 26-28 Sept. 2013. (http://etai13.feit.ukim.edu.mk/en/index/pt)− Participation of Dajana Cassioli in the Panel Session “European Project Space on Pervasive Systems, Physiological Computing and Information Systems Security and Privacy” at SENSORNETS 2015 (http://www.sensornets.org/EPSProgram.aspx?y=2015 )− Demo of the Video-oriented Intelligent Sensing (VISION - Video-oriented UWB-based Intelligent Ubiquitous Sensing) at SENSORNETS 2015 – European Project Space (http://www.sensornets.org/EPSProgram.aspx?y=2015 )− Advanced Lecture in Doctoral School: Vittorio Cortellessa, Antinisca Di Marco, Catia Trubiani: Software Performance Antipatterns: Modeling and Analysis. Formal Methods for Model-Driven Engineering - 12th International School on Formal Methods for the Design of Computer, Communication, and Software Systems, SFM 2012, Bertinoro, Italy, June 18-23, 2012.For further information, please, visit the VISION website http://www.vision-ercproject.eu or contact info@vision-ercproject.eu.
