Pub Date : 2005-05-20DOI: 10.1109/WISES.2005.1438722
Daniel Albeseder
Partially synchronous computational models fall in between synchronous models, which are demanding in terms of requirements but admit solutions to most of the important fault-tolerant distributed computing problems, and the purely asynchronous model, where this is not the case. The /spl otimes/-Model is a recently presented partially synchronous model close to pure asynchrony. It just assumes a bound on the ratio of maximum vs. minimum end-to-end delays of messages simultaneously in transit. This paper provides experimental evidence for the assumed correlation of end-to-end delays in some common type of distributed systems: using a comprehensive custom evaluation framework, we measured the end-to-end delays in a simple clock synchronization algorithm running on a fast-Ethernet network connecting Linux workstations. Our results reveal a significant correlation between maximum and minimum delay under several load conditions, and hence confirm that the bounded ratio assumption of the /spl otimes/-Model is well-founded.
{"title":"Evaluation of message delay correlation in distributed systems","authors":"Daniel Albeseder","doi":"10.1109/WISES.2005.1438722","DOIUrl":"https://doi.org/10.1109/WISES.2005.1438722","url":null,"abstract":"Partially synchronous computational models fall in between synchronous models, which are demanding in terms of requirements but admit solutions to most of the important fault-tolerant distributed computing problems, and the purely asynchronous model, where this is not the case. The /spl otimes/-Model is a recently presented partially synchronous model close to pure asynchrony. It just assumes a bound on the ratio of maximum vs. minimum end-to-end delays of messages simultaneously in transit. This paper provides experimental evidence for the assumed correlation of end-to-end delays in some common type of distributed systems: using a comprehensive custom evaluation framework, we measured the end-to-end delays in a simple clock synchronization algorithm running on a fast-Ethernet network connecting Linux workstations. Our results reveal a significant correlation between maximum and minimum delay under several load conditions, and hence confirm that the bounded ratio assumption of the /spl otimes/-Model is well-founded.","PeriodicalId":266947,"journal":{"name":"Third International Workshop on Intelligent Solutions in Embedded Systems, 2005.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2005-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128075227","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2005-05-20DOI: 10.1109/WISES.2005.1438707
B. Huber, P. Peti, R. Obermaisser, C. E. Salloum
The steady increase in electronics in automotive systems in order to meet the customers expectation of a cars functionality has led to the development of integrated architectures, as already partly deployed in avionics. Integrated architectures overcome the "1 Function - I Electronic Control Unit (ECU)" design philosophy by providing an infrastructure that allows the sharing of ECUs between multiple applications. As a consequence, integrated systems promise massive cost savings through the reduction of resource duplication. In addition, integrated systems permit an optimal interplay of application subsystems, reliability improvements with respect to wiring and connectors, and overcome limitations for spare components and redundancy management. In this paper we present an overview of the DECOS integrated architecture and describe a prototype setup. In particular, we describe how inner component partitioning is provided using the open source RTAI/LXRT Linux operating system. We exploit the functionality of this operating system to be suitable for the DECOS architecture and devise a static dispatching technique that forms the basis for the multiplexing of available resources between applications.
{"title":"Using RTAI/LXRT for partitioning in a prototype implementation of the DECOS architecture","authors":"B. Huber, P. Peti, R. Obermaisser, C. E. Salloum","doi":"10.1109/WISES.2005.1438707","DOIUrl":"https://doi.org/10.1109/WISES.2005.1438707","url":null,"abstract":"The steady increase in electronics in automotive systems in order to meet the customers expectation of a cars functionality has led to the development of integrated architectures, as already partly deployed in avionics. Integrated architectures overcome the \"1 Function - I Electronic Control Unit (ECU)\" design philosophy by providing an infrastructure that allows the sharing of ECUs between multiple applications. As a consequence, integrated systems promise massive cost savings through the reduction of resource duplication. In addition, integrated systems permit an optimal interplay of application subsystems, reliability improvements with respect to wiring and connectors, and overcome limitations for spare components and redundancy management. In this paper we present an overview of the DECOS integrated architecture and describe a prototype setup. In particular, we describe how inner component partitioning is provided using the open source RTAI/LXRT Linux operating system. We exploit the functionality of this operating system to be suitable for the DECOS architecture and devise a static dispatching technique that forms the basis for the multiplexing of available resources between applications.","PeriodicalId":266947,"journal":{"name":"Third International Workshop on Intelligent Solutions in Embedded Systems, 2005.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2005-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129378080","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2005-05-20DOI: 10.1109/WISES.2005.1438725
P. Peti, R. Obermaisser, H. Paulitsch
The automotive industry is at the verve to deploy computer systems not only for safety-related and comfort functionality, but for safety-critical by-wire systems. In the scope of the PEGASUS project a car will be equipped with time-triggered technology in order to provide not only superior car dynamics but also investigate system design and integration on the basis of a series car. As part of this project a diagnostic solution is being developed in order to tackle prevalent diagnostic problems, such as the trouble-not-identified phenomenon in electronic systems, by exploiting the error-containment properties of the underlying architecture. In this paper we present the diagnostic architecture of the PEGASUS car that operates on the distributed state of the system in order to trace experienced failures back to the origin and decide on the type of fault (e.g., transient vs. permanent, internal vs. external) that is affecting the system. A necessary prerequisite of such an integrated diagnostic infrastructure is the continuous monitoring and subsequent dissemination of diagnostic information in order to allow a meaningful analysis.
{"title":"The diagnostic architecture of the PEGASUS project car","authors":"P. Peti, R. Obermaisser, H. Paulitsch","doi":"10.1109/WISES.2005.1438725","DOIUrl":"https://doi.org/10.1109/WISES.2005.1438725","url":null,"abstract":"The automotive industry is at the verve to deploy computer systems not only for safety-related and comfort functionality, but for safety-critical by-wire systems. In the scope of the PEGASUS project a car will be equipped with time-triggered technology in order to provide not only superior car dynamics but also investigate system design and integration on the basis of a series car. As part of this project a diagnostic solution is being developed in order to tackle prevalent diagnostic problems, such as the trouble-not-identified phenomenon in electronic systems, by exploiting the error-containment properties of the underlying architecture. In this paper we present the diagnostic architecture of the PEGASUS car that operates on the distributed state of the system in order to trace experienced failures back to the origin and decide on the type of fault (e.g., transient vs. permanent, internal vs. external) that is affecting the system. A necessary prerequisite of such an integrated diagnostic infrastructure is the continuous monitoring and subsequent dissemination of diagnostic information in order to allow a meaningful analysis.","PeriodicalId":266947,"journal":{"name":"Third International Workshop on Intelligent Solutions in Embedded Systems, 2005.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2005-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115002285","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2005-05-20DOI: 10.1109/WISES.2005.1438709
N. Narendra
The rapid growth of handheld computing devices such as mobile phones, PDAs or palmtops is paving the way for the emergence of pervasive computing systems. Just as in the case of traditional computing systems, pervasive computing systems need to be tested in the large before they can be deployed in the field. As opposed to traditional computing systems, however, large-scale testing of pervasive computing systems requires the presence of dozens (perhaps hundreds) of physical devices, arranged together in a network, executing a variety of complex scenarios. In order to reduce the cost of such testing, it would be better to simulate the operation of a pervasive computing system using well-known techniques from multi-agent simulation, by representing each (hardware or software) component of the system as a software agent. In this paper we describe our ongoing work, where we extend our earlier work on multi-agent simulation, for pervasive computing systems. Since adaptation in pervasive computing systems is expected to be common, we also show that our simulation technique can model adaptation.
{"title":"Large scale testing of pervasive computing systems using multi-agent simulation","authors":"N. Narendra","doi":"10.1109/WISES.2005.1438709","DOIUrl":"https://doi.org/10.1109/WISES.2005.1438709","url":null,"abstract":"The rapid growth of handheld computing devices such as mobile phones, PDAs or palmtops is paving the way for the emergence of pervasive computing systems. Just as in the case of traditional computing systems, pervasive computing systems need to be tested in the large before they can be deployed in the field. As opposed to traditional computing systems, however, large-scale testing of pervasive computing systems requires the presence of dozens (perhaps hundreds) of physical devices, arranged together in a network, executing a variety of complex scenarios. In order to reduce the cost of such testing, it would be better to simulate the operation of a pervasive computing system using well-known techniques from multi-agent simulation, by representing each (hardware or software) component of the system as a software agent. In this paper we describe our ongoing work, where we extend our earlier work on multi-agent simulation, for pervasive computing systems. Since adaptation in pervasive computing systems is expected to be common, we also show that our simulation technique can model adaptation.","PeriodicalId":266947,"journal":{"name":"Third International Workshop on Intelligent Solutions in Embedded Systems, 2005.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2005-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133665882","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2005-05-20DOI: 10.1109/WISES.2005.1438727
José Alberto Vigo Segura, J. Sánchez, N. M. Madrid, R. Seepold
Individual user profiles can have strong influence on basic setup and maintenance of intelligent environments. Independently from architecture and technology, end-users' preferences need to be preserved, retrieved upon access and adapted over the life cycle of the profiles. The presented model proposes an integration of smart cards into a control network structure that can be used to setup, store and reconfigure a well-defined cluster of the network that depends on the environment in which the smart card is used. As a result, users can share and transfer profiles in different environments without any need to access a centralized repository. The architecture developed integrates a smart card as an additional device into the network and thus it allows access to efficiently transfer data between the card and the network.
{"title":"Integration of smart cards into automation networks","authors":"José Alberto Vigo Segura, J. Sánchez, N. M. Madrid, R. Seepold","doi":"10.1109/WISES.2005.1438727","DOIUrl":"https://doi.org/10.1109/WISES.2005.1438727","url":null,"abstract":"Individual user profiles can have strong influence on basic setup and maintenance of intelligent environments. Independently from architecture and technology, end-users' preferences need to be preserved, retrieved upon access and adapted over the life cycle of the profiles. The presented model proposes an integration of smart cards into a control network structure that can be used to setup, store and reconfigure a well-defined cluster of the network that depends on the environment in which the smart card is used. As a result, users can share and transfer profiles in different environments without any need to access a centralized repository. The architecture developed integrates a smart card as an additional device into the network and thus it allows access to efficiently transfer data between the card and the network.","PeriodicalId":266947,"journal":{"name":"Third International Workshop on Intelligent Solutions in Embedded Systems, 2005.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2005-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117175758","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2005-05-20DOI: 10.1109/WISES.2005.1438710
Marcus Thoss
This paper presents the design and realisation of a hybrid, high-accuracy, automated measurement framework. A hybrid measurement system is introduced that has been successfully used to carry out high-accuracy measurements in distributed automation environments. The application of the hybrid measurement system is not limited to such environments, though. It is characterised by monitoring distributed events with low interference and by the usage of global timestamps based on the high-resolution clock of a logic analyser. Furthermore, it exhibits an extensively automated experiment control and data analysis framework. Measurement results gained with the approach are presented as well.
{"title":"Automated high-accuracy hybrid measurement for distributed embedded systems","authors":"Marcus Thoss","doi":"10.1109/WISES.2005.1438710","DOIUrl":"https://doi.org/10.1109/WISES.2005.1438710","url":null,"abstract":"This paper presents the design and realisation of a hybrid, high-accuracy, automated measurement framework. A hybrid measurement system is introduced that has been successfully used to carry out high-accuracy measurements in distributed automation environments. The application of the hybrid measurement system is not limited to such environments, though. It is characterised by monitoring distributed events with low interference and by the usage of global timestamps based on the high-resolution clock of a logic analyser. Furthermore, it exhibits an extensively automated experiment control and data analysis framework. Measurement results gained with the approach are presented as well.","PeriodicalId":266947,"journal":{"name":"Third International Workshop on Intelligent Solutions in Embedded Systems, 2005.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2005-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126250745","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2005-05-20DOI: 10.1109/WISES.2005.1438728
A. Doblander, A. Maier, B. Rinner, H. Schwabach
In this paper, we present an approach for improving fault-tolerance and service availability in intelligent video surveillance (IVS) systems. A typical IVS system consists of various intelligent video sensors that combine image sensing with video analysis and network streaming. System monitoring and fault diagnosis followed by appropriate dynamic system reconfiguration mitigate effects of faults and therefore enhance the system's fault-tolerance. The applied monitoring and diagnosis unit (MDU) allows the detection of both node- and system-level faults. Lacking redundant hardware such reconfigurations are established by graceful degradation of the overall application. An optimizer module that performs multi-criterion optimization is used to compute a new degraded system configuration by trading off quality of service (QoS), energy consumption, and service availability. We demonstrate the functionality of our approach by an illustrative example.
{"title":"Improving fault-tolerance in intelligent video surveillance by monitoring, diagnosis and dynamic reconfiguration","authors":"A. Doblander, A. Maier, B. Rinner, H. Schwabach","doi":"10.1109/WISES.2005.1438728","DOIUrl":"https://doi.org/10.1109/WISES.2005.1438728","url":null,"abstract":"In this paper, we present an approach for improving fault-tolerance and service availability in intelligent video surveillance (IVS) systems. A typical IVS system consists of various intelligent video sensors that combine image sensing with video analysis and network streaming. System monitoring and fault diagnosis followed by appropriate dynamic system reconfiguration mitigate effects of faults and therefore enhance the system's fault-tolerance. The applied monitoring and diagnosis unit (MDU) allows the detection of both node- and system-level faults. Lacking redundant hardware such reconfigurations are established by graceful degradation of the overall application. An optimizer module that performs multi-criterion optimization is used to compute a new degraded system configuration by trading off quality of service (QoS), energy consumption, and service availability. We demonstrate the functionality of our approach by an illustrative example.","PeriodicalId":266947,"journal":{"name":"Third International Workshop on Intelligent Solutions in Embedded Systems, 2005.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2005-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133045097","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2005-05-20DOI: 10.1109/WISES.2005.1438716
B. Thallner, H. Moser
Energy efficiency and fault-tolerance are the most important issues in the development of next-generation wireless ad hoc networks and sensor networks. Topology control as a low level service (typically below the traditional layer structure) governs communication among all nodes and is hence the primary target for increasing connectivity and saving energy. In this paper, we present an improvement of our topology control algorithm for very dynamic networks and low power devices (e.g. sensor nodes). The algorithm constructs a fault-tolerant topology for energy-efficient and fault-tolerant multi-hop communication in a two-tier network consisting of a large number of wireless nodes and a few gateway nodes (e.g. base stations responsible for exchanging data with other networks). Using only local information, like distance/channel attenuation to neighbors, our fully distributed algorithm efficiently constructs and continuously maintains a k-regular overlay graph that guarantees low total transmission power, is k-node-connected and ensures failure locality. It automatically adapts to a dynamically changing environment, is guaranteed to converge, builds a hierarchy of clusters that reflects the node density and exhibits good performance as well.
{"title":"Topology control for fault-tolerant communication in highly dynamic wireless networks","authors":"B. Thallner, H. Moser","doi":"10.1109/WISES.2005.1438716","DOIUrl":"https://doi.org/10.1109/WISES.2005.1438716","url":null,"abstract":"Energy efficiency and fault-tolerance are the most important issues in the development of next-generation wireless ad hoc networks and sensor networks. Topology control as a low level service (typically below the traditional layer structure) governs communication among all nodes and is hence the primary target for increasing connectivity and saving energy. In this paper, we present an improvement of our topology control algorithm for very dynamic networks and low power devices (e.g. sensor nodes). The algorithm constructs a fault-tolerant topology for energy-efficient and fault-tolerant multi-hop communication in a two-tier network consisting of a large number of wireless nodes and a few gateway nodes (e.g. base stations responsible for exchanging data with other networks). Using only local information, like distance/channel attenuation to neighbors, our fully distributed algorithm efficiently constructs and continuously maintains a k-regular overlay graph that guarantees low total transmission power, is k-node-connected and ensures failure locality. It automatically adapts to a dynamically changing environment, is guaranteed to converge, builds a hierarchy of clusters that reflects the node density and exhibits good performance as well.","PeriodicalId":266947,"journal":{"name":"Third International Workshop on Intelligent Solutions in Embedded Systems, 2005.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2005-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124106782","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2005-05-20DOI: 10.1109/WISES.2005.1438715
Roberto Casas, Héctor J. Gracia, Á. Marco, Jorge L. Falcó
Wireless sensor networks (WSNs) can take advantage of versatility, completeness, and low prices of standard wireless protocols; Bluetooth as we show later is a candidate suitable for WSNs. The fusion of data collected over a WSN is just an evident application of time synchronization. Bringing together these two issues, we find that synchronization using standard protocols poses an important drawback. In this paper, we present a simple method that allows clock synchronization in Bluetooth WSNs, down to few microseconds.
{"title":"Synchronization in wireless sensor networks using Bluetooth","authors":"Roberto Casas, Héctor J. Gracia, Á. Marco, Jorge L. Falcó","doi":"10.1109/WISES.2005.1438715","DOIUrl":"https://doi.org/10.1109/WISES.2005.1438715","url":null,"abstract":"Wireless sensor networks (WSNs) can take advantage of versatility, completeness, and low prices of standard wireless protocols; Bluetooth as we show later is a candidate suitable for WSNs. The fusion of data collected over a WSN is just an evident application of time synchronization. Bringing together these two issues, we find that synchronization using standard protocols poses an important drawback. In this paper, we present a simple method that allows clock synchronization in Bluetooth WSNs, down to few microseconds.","PeriodicalId":266947,"journal":{"name":"Third International Workshop on Intelligent Solutions in Embedded Systems, 2005.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2005-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120968349","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}