skip to main content

Title: Bounding network-induced delays for time-critical services in avionic systems using measurements and network calculus
In this work, we propose to derive realistic, accurate bounds on network-induced delays for time-critical tasks running on Avionics Full-Duplex Switched Ethernet. In the WiP poster, we present preliminary evaluation results showing that through measurement-based modeling and refining network-calculus-based analysis with measurements, tight delay bounds can be obtained for AFDX networks with realistic traffic patterns and network workloads.
Authors:
; ;
Award ID(s):
1646458 2146968
Publication Date:
NSF-PAR ID:
10120465
Journal Name:
Proceedings of the 10th ACM/IEEE International Conference on Cyber-Physical Systems (ICCPS '19)
Page Range or eLocation-ID:
338 to 339
Sponsoring Org:
National Science Foundation
More Like this
  1. Evaluation of end-to-end network performance using realistic traffic models is a challenging problem in networking. The classical theory of queueing networks is feasible only under rather restrictive assumptions on the input traffic models and network elements. An alternative approach, first proposed in the late 1980s, is to impose deterministic bounds on the input traffic that can be used as a basis for a network calculus to compute end-to-end network delay bounds. Such deterministic bounds are inherently loose as they must accommodate worst case scenarios. Since the early 1990s, efforts have shifted to development of a stochastic network calculus to providemore »probabilistic end-to-end performance bounds. In this paper, we capitalize on the approach of stochastically bounded burstiness (SBB) which was developed for a general class of bounding functions, and was demonstrated for a bound that is based on a mixture distribution. We specialize the SBB bounds to bounds based on the class of phase-type distributions, which includes mixture distributions as a particular case. We develop the phase-type bounds and demonstrate their performance.« less
  2. This paper proposes a strategy to control a group of thermostatically controlled loads (TCLs) such that the variability in their aggregate load is reduced. This strategy could be deployed in areas of a distribution network that experience voltage excursions due to net load fluctuations, such as areas with high penetrations of photovoltaic (PV) generation and/or electric vehicles (EVs), We limit variation in the power consumption of a group of TCLs using a control strategy previously developed for large aggregations of switched systems. Using this strategy, we constrain the number of TCLs that are on (i.e., actively consuming power) between uppermore »and lower bounds. In simulations, the control strategy successfully decreases the range over which TCL power consumption varies. Percent reductions in range are greatest for medium group sizes: we find a median reduction of 82% for groups of 50 TCLs, 74% for groups of 1000 TCLs, and 59% for groups of 5 TCLs. Reducing the variability of a distribution network's power injections helps to reduce voltage variability. In a simulation of a distribution line supplying 25 households, half with PV systems, the control strategy reduces the total range of voltage by 0.02 p.u. and prevents a violation of the 0.95 p.u. limit. Lastly, we propose a new control strategy for a more realistic TCL model that includes compressor lockout. The new strategy performs comparably to the original strategy and is demonstrated through simulation.« less
  3. Recently, switched Ethernet has become increasingly popular in networked cyber-physical systems (NCPS). In an Ethernet-based NCPS, network-connected devices (e.g., sensors and actuators) realize time-critical tasks by exchanging miscellaneous information, such as sensor readings and control commands. To ensure reliable control and operation, network-induced delays for time-critical NCPS applications must be carefully examined. In this work, we propose a framework combining network delay measurements and network-calculus-based delay performance analysis to obtain accurate, deterministic worst-case delay bounds for NCPS. By modeling traffic sources and networking devices (e.g., Ethernet switches) through measurements, we establish accurate traffic and device models for network-calculus-based analysis. Tomore »obtain worst-case delay bounds, different network-calculus-based analytical methods can be leveraged, allowing CPS architects to customize the proposed delay analysis framework to suit application-specific needs. Our evaluation results show that the proposed approach derives accurate delay bounds, making it a valuable tool for architects designing NCPSs supporting time-critical applications.« less
  4. Network traffic is difficult to characterize due to its random, bursty nature. Even if a traffic source could be fit to a stochastic model with reasonable accuracy, analysis of end-to-end network performance metrics for such traffic models is generally intractable. In prior work, an approach to characterize traffic burstiness using a bound based on the class of phase-type distributions was proposed. Such phase-type bounds could be applied in conjunction with stochastic network calculus to derive probabilistic end-to-end delay bounds for a traffic stream. In this paper, we focus on the problem of estimating a tight phase-type burstiness bound for amore »given traffic trace. We investigate a method based on least squares and another based on the expectation-maximization algorithm. Our numerical results compare the two approaches in the scenario of a heavy-tailed M/G/1 queue. We find that while both methods are viable approaches for deriving phase-type bounds on traffic burstiness, the least squares approach performs better, particularly when a tail limit is imposed.« less
  5. The bursty nature of network traffic makes it difficult to characterize accurately, and may give rise to heavy-tailed queue distributions within the network. Building on prior work in stochastic network calculus, we propose traffic burstiness bounds based on the class of phase-type distributions and develop an approach to estimate the parameter of such bounds using the expectation-maximization (EM) algorithm. By limiting the tail of the burstiness bound, our approach achieves a better fit of the phase-type distribution to the empirical data from heavy-tailed traffic. The proposed tail-limited phase-type burstiness bounds fall within the framework for stochastic network calculus based onmore »generalized stochastically bounded burstiness. We demonstrate the effectiveness of the proposed methodology with a numerical example involving a heavy-tailed M/G/1 queue.« less