-
[show abstract]
[hide abstract]
ABSTRACT: The aim of this paper is to allow for hybrid task sets in the context of stochastic real-time analysis. The paper goes beyond previous work by allowing for the presence of aperiodic tasks in the system. Instead of representing a task with a fixed activation period and a worst-case execution time (WCET), here a task is characterized by an arrival profile (AP) and an execution time profile (ETP), both given by random variables with known distributions. Any number of aperiodic tasks, with arbitrary arrival and execution time profiles, can be dealt with. To cope with the unbounded interference introduced by aperiodic tasks in the system, sporadic and aperiodic tasks are encapsulated within servers. The paper presents the calculus for obtaining the exact ETP of servers, which allows us to derive exact response time distributions of periodic tasks. Also, an example is used to show the potential and validity of the proposed approach.
Emerging Technologies and Factory Automation, 2007. ETFA. IEEE Conference on; 10/2007
-
[show abstract]
[hide abstract]
ABSTRACT: This paper addresses on-going work on providing a model of a robust real-time wireless link intended to be used to connect a mobile platform to a ground station in a land monitoring system. This work is part of a project where the mobile platform is an unmanned aerial vehicle (UAV), equipped with embedded devices which acquire and process sensor data to be sent to the ground station. Firstly, a temporal model of the communications link relying on radio modems is developed. Secondly, measurements have been made in order to determine packet loss probability. These two results are currently used to provide a robust real-time model of the wireless communications link.
Emerging Technologies and Factory Automation, 2006. ETFA '06. IEEE Conference on; 10/2006
-
[show abstract]
[hide abstract]
ABSTRACT: The growing need for providing real-time system designers with less pessimistic results than the ones given by traditional worst-case analysis motivates several recent works on stochastic analysis methods. This paper deals with the calculation of stochastic response time profiles of tasks that are hierarchically scheduled using server-based techniques in a stochastic analysis framework. Depending on how tasks are scheduled within the server, differences in temporal performance are expected. In the paper, initial results on calculating the response time profiles for these server-scheduled tasks are outlined.
Emerging Technologies and Factory Automation, 2006. ETFA '06. IEEE Conference on; 10/2006
