Πλοήγηση ανά Συγγραφέας "Ninidakis, Stylianos"

Τώρα δείχνει 1 - 1 of 1
  • Μικρογραφία εικόνας
    Τεκμήριο
    Efficient gateway implementations for cyber-physical systems
    (ΕΛΜΕΠΑ, Σχολή Μηχανικών (ΣΜΗΧ), ΠΜΣ Μηχανικών Πληροφορικής, 2026-11-04) Ninidakis, Stylianos; Νινιδάκης, Στυλιανός; Grammatikakis, Miltiadis; Γραμματικάκης, Μιλτιάδης
    Smart devices with numerous sensors and actuators connected to servers via heterogeneous networks form an increasing technology trend. The transmission, storage, and manipulation or analysis of big data raise many issues which the research community should address. A common practice to lighten a server’s workload is adding an intermediate device (gateway) to (1) synchronize various networks, (2) gather, filter, and preliminarily analyze the data from the sensors, (3) control any local actuators and (4) send the extracted data to the server. Since gateway devices are often portable embedded systems with restricted processing capability to easily function close to sensors, software optimization plays a crucial role in ensuring data integrity and real-time functionality. Parallel execution and concurrent buffers could easily contribute toward that direction. In this study, we investigate the way that concurrent buffers implemented as wait-free single-producer single-consumer queues perform on two different use cases: (1) a Bluetooth-to-Wireless healthcare gateway transferring biosensor (ECG) data, and (2) a CAN-to-CAN in-vehicle gateway transmitting CAN frames. In the Bluetooth-to-Ethernet gateway, we focus on enhancing the performance of biosensor data flows across the gateway by examining lock-free concurrent queues and traditional lock-based circular FIFOs that embody single-producer single-consumer principles. Our results show that for large biosensor rates, the waiting time of the concurrent queue is smaller. Large rates, above 256 pulses/s, are not supported by the biosensor but are examined by designing a self-similar, digital twin process. In the CAN-to-CAN gateway, we examine efficient frame transfer across incoming and outgoing CAN interfaces at the gateway. In our embedded platform prototype, each CAN interface is controlled independently by a corresponding POSIX thread. Inter-thread communication is implemented using shared data structures, namely lock-free concurrent queues, and traditional lock-based circular FIFOs. Our experimentation relies on increasing communication reliability and performance by minimizing frame loss and improving the egress CAN frame rate. Our experiments show that the frame waiting time in the queue is smaller for the concurrent queue, while relaying messages is equally fast, reaching a maximum output rate of ~380 frames/s for large enqueue rates near saturation. Power consumption is higher for the faster concurrent queue implementation.