Wireless Sensor Networks are fundamental backbones in industry 4.0 and their time synchronization is still a crucial issue. To achieve such a goal, several protocols are being implemented. Among the others, those based on consensus architectures are increasing their diffusion since their promising characteristics in terms of accuracy and reliability. These peculiarities make these kind of algorithms well suited for industrial applications where electromagnetic interferences are generally present in the operating environment. Together with the well-known Average Time Synchronization (ATS) algorithm in the last years other kind of consensus strategies are wide-spreading. In detail, algorithms based on the maximum consensus (CMTS) have been proposed in literature especially for industry applications. However, the performance of these algorithms are evaluated in scenarios that do not consider typical disturbances that characterize the industrial contexts. As an example, factors of influence as high and low frequency local clock noises, fixed and variable communication latencies. In addition, when real nodes are considered, the employment of commercial radio devices, the unavailability (in many applications) of timestamps calculated at Medium Access Control level and further non idealities as the local time resolution should be considered as well. In this work, a methodological analysis to evaluate the performance of maximum consensus-based algorithm when subjected to typical disturbances due to the industrial scenarios is carried out. The analysis output could then be used to build up a previsional model of achievable performance when real sensor nodes are employed and to find countermeasures able to neutralize or reduce unwanted non-idealities effects.
Time Synchronization Based on CMTS: A Performance Analysis in Industry Scenarios
Capriglione D.;Paciello V.
2018-01-01
Abstract
Wireless Sensor Networks are fundamental backbones in industry 4.0 and their time synchronization is still a crucial issue. To achieve such a goal, several protocols are being implemented. Among the others, those based on consensus architectures are increasing their diffusion since their promising characteristics in terms of accuracy and reliability. These peculiarities make these kind of algorithms well suited for industrial applications where electromagnetic interferences are generally present in the operating environment. Together with the well-known Average Time Synchronization (ATS) algorithm in the last years other kind of consensus strategies are wide-spreading. In detail, algorithms based on the maximum consensus (CMTS) have been proposed in literature especially for industry applications. However, the performance of these algorithms are evaluated in scenarios that do not consider typical disturbances that characterize the industrial contexts. As an example, factors of influence as high and low frequency local clock noises, fixed and variable communication latencies. In addition, when real nodes are considered, the employment of commercial radio devices, the unavailability (in many applications) of timestamps calculated at Medium Access Control level and further non idealities as the local time resolution should be considered as well. In this work, a methodological analysis to evaluate the performance of maximum consensus-based algorithm when subjected to typical disturbances due to the industrial scenarios is carried out. The analysis output could then be used to build up a previsional model of achievable performance when real sensor nodes are employed and to find countermeasures able to neutralize or reduce unwanted non-idealities effects.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.