Skip to Main content Skip to Navigation

Scheduling Analysis of Tasks Constrained by Time-Division Multiplexing: Application to Software Radio Protocols

Abstract : The work, presented in this thesis, aims at performing automatically scheduling analysis of Time Division Multiple Access (TDMA) based communication systems. Products called software radio protocols, developed at Thales Communications & Security, are examples of such systems.TDMA is a channel access method based on the division of time into several time slots. TDMA-based software radio protocols are real-time embedded systems. They are implemented by tasks that are statically allocated on multiple processors. A task may have an execution time, a deadline, and a release time that depend on TDMA. The tasks also have dependencies through precedence and shared resources.TDMA-based software radio protocols have architecture characteristics that are not handled by scheduling analysis methods of the literature. A consequence is that existing methods give either optimistic or pessimistic analysis results. Furthermore, existing architecture models at Thales do not contain enough information to be used for scheduling analysis. The information is only available in specification documents. These issues impact the possibility to perform scheduling analysis, but also the possibility to perform it automatically.The propositions of this thesis solve these problems. An experimental architecture model is proposed in the UML MARTE modeling language. The architecture model is transformed to a task model called Dependent General Multiframe (DGMF). The DGMF task model describes, in particular, the different jobs of a task, and task dependencies. To analyze DGMF tasks, they are transformed to another model called tree-shaped transaction. Transactions are precedence dependent tasks. Transactions that result from the transformation have non-immediate tasks. These tasks are not necessarily released immediately by their predecessor task. To consider the effects of non-immediateness, this thesis proposes the WCDOPS+NIM schedulability test for tree-shaped transactions. The general analysis method is implemented as a toolchain that can be used by engineers at Thales.Experimental results show that the propositions give less pessimistic schedulability results, compared to fundamental methods. The results are less pessimistic for both randomly generated systems and real case-studies from Thales. Furthermore, experiments show that scheduling analysis can be applied automatically to a TDMA-based software radio protocol.Less pessimistic results are important for engineering work, in order to limit the over-dimensioning of resources. The automatic analysis is a gain in productivity. These are advantages for engineers in the more and more competitive market of software radios.Experimental results show that the DGMF gives less pessimistic analysis results for TDMA-based software radio protocols. The results are less pessimistic for both simulated systems and real case-studies developed at THALES. Furthermore, the approach is scalable to real systems and the analysis can be performed automatically on such systems.
Complete list of metadatas

Cited literature [145 references]  Display  Hide  Download
Contributor : Shuai Li <>
Submitted on : Wednesday, December 10, 2014 - 5:34:48 PM
Last modification on : Wednesday, September 16, 2020 - 9:56:31 AM
Long-term archiving on: : Wednesday, March 11, 2015 - 11:46:01 AM


Distributed under a Creative Commons Attribution 4.0 International License


  • HAL Id : tel-01093536, version 1


Shuai Li. Scheduling Analysis of Tasks Constrained by Time-Division Multiplexing: Application to Software Radio Protocols. Computer Science [cs]. Université de Bretagne Occidentale, 2014. English. ⟨tel-01093536⟩



Record views


Files downloads