<p>Distributed real-time system consists of many tasks on different processors communicating with</p><p>each other. The communication between these tasks must occur in bounded time. In our thesis we</p><p>are considering a holistic approach to estimate the end-to-end delays between two processors</p><p>communicating with each other. For both tasks scheduled by fixed priority and dynamic priority,</p><p>the system model considered is time token passing ring. The method is described to calculate</p><p>worst-case communication delay of the message sent by certain tasks over the network. Once the</p><p>worst-case communication delay of the message is obtained, and all required parameters are</p><p>provided i.e. deadline of the message, transmission time of the packet etc, one can figure out how</p><p>many messages will successfully meet their deadline. Thus this will enable us to select</p><p>appropriate algorithm according to system requirement.</p>

Distributed real-time system consists of many tasks on different processors communicating with each other. The communication between these tasks must occur in bounded time. In our thesis we are considering a holistic approach to estimate the end-to-end delays between two processors communicating with each other. For both tasks scheduled by fixed priority and dynamic priority, the system model considered is time token passing ring. The method is described to calculate worst-case communication delay of the message sent by certain tasks over the network. Once the worst-case communication delay of the message is obtained, and all required parameters are provided i.e. deadline of the message, transmission time of the packet etc, one can figure out how many messages will successfully meet their deadline. Thus this will enable us to select appropriate algorithm according to system requirement.

Communication delay in distributed teams is salient problem, especially in operational setting where communication is critical to team safety and success. The present study investigated the impact of communication delays affect distributed team performance and processes, and if being able to see one's team member would lessen the effects of delays. In addition, team gender composition was investigated, to see how delays affected the interactions of same and mixed-gender teams, as well as teams with familiar and unfamiliar members. Lastly, a supplemental was performed on a subset of the experimental data to determine if teams with familiar members' communicated more efficiently than unfamiliar teams when coordination complexity was high. Thirty distributed dyads, were assigned the role of intelligence analysts, and performed a collaborative problem solving task, using audioconferencing and videoconferencing technologies. During the task, participants verbally shared and discussed information in order to solve a fictitious terror plot. Communication between team members was delayed by 0 ms, 800, or 1600 ms. Linear mixed models showed that participants took longer to solve the task at the 800 ms delay. Task accuracy was not affected by delays. At the 1600 ms delay, participants shared less information with each other, and rated their frustration higher compared to the 0 ms delay. Audiovisual technology affected overall workload scores, with lower scores at the 0 ms delay compared to the 800 ms delay. Although delays did not have the anticipated effect on familiar and same-gender teams, there were some interactive effects of interest. Specifically, in gender-diverse teams task accuracy was higher with audiovisual technology than audio-alone, but this effect was independent of delays. Also, familiar teams exhibited higher levels of cognitive trust across all levels of delay and technology. Results of the supplemental analysis showed no differences in communication efficiency between familiar and unfamiliar teams when coordination complexity was high. Based on the results of this work, recommendations were proposed for strategies to lessen the effects of communication delays and future research directions were outlined. / Doctor of Philosophy / Communication delay in distributed teams is salient problem, especially in operational settings where communication is critical to team safety and success. In previous work, communication delays have been shown to disrupt turn-taking in conversations and create instances of overlaps or interruptions. The present study was conducted to further investigate the effects of communication delays on various aspects of distributed team performance and to determine if being able to see one’s team member via video technology may potentially lessen the effects of delays. In addition, team gender composition was investigated, to see how delays affected the interactions of same and mixed-gender teams, as well as teams with familiar and unfamiliar members. Lastly, a supplemental analysis was performed using a subset of the experimental data to determine if teams with familiar members’ communicated more efficiently than unfamiliar teams when coordination complexity was high. Thirty distributed teams of two members or dyads, performed a collaborative problem solving task, using audio conferencing and videoconferencing technologies. During the task, participants verbally shared and discussed information in order to identify the solution to a fictitious terror plot. Communication between team members was delayed by 0 ms, 800, or 1600 ms. Overall, results showed that participants took longer to solve the task at the 800 ms delay, with no effects on task accuracy. At the 1600 ms delay, participants shared less information with each other, and rated their frustration higher compared to the 0 ms delay. When teams used audiovisual technology, workload scores were lower at 0 ms compared to the 800 ms delay. Although delays did not have the anticipated effects on familiar and same-gender teams, there were some other interesting effects that emerged. Namely, gender-diverse teams scored higher accuracy with audiovisual technology than audio-alone, but this effect was independent of delays. Also, teams with familiar members exhibited higher levels of cognitive trust across all levels of delay and technology. Results of the supplemental analysis showed that unfamiliar teams communicated more efficiently with audiovisual technology, but only when coordination complexity was low.

The decentralized consensus control of a formation of rigid-body spacecraft is studied in the framework of geometric mechanics while accounting for a constant communication time delay between spacecraft. The relative position and attitude (relative pose) are represented on the Lie group SE(3) and the communication topology is modeled as a digraph. The consensus problem is converted into a local stabilization problem of the error dynamics associated with the Lie algebra se(3) in the form of linear time-invariant delay differential equations with a single discrete delay in the case of a circular orbit, whereas it is in the form of linear time-periodic delay differential equations in the case of an elliptic orbit, in which the stability may be assessed using infinite-dimensional Floquet theory. The proposed technique is applied to the consensus control of four spacecraft in the vicinity of a Molniya orbit.

Vehicle platoons are efficient, closely-spaced groups of robotically controlled vehicles which travel at high speeds down the road, similar to carts in a train. Within this thesis, a promising control algorithm for vehicle platooning is explored. The control algorithm was previously demonstrated in a sterile setting which significantly reduced the challenges facing full-scale implementation of platoons, most notably loss of shared data and imprecision within the data. As found within this work, transmission loss and imprecise position, velocity, and acceleration data significantly degraded the control algorithm's performance. Vehicles in the platoon became more closely spaced, changed speeds more frequently, and expended far more energy than necessary. Introducing a measure of each following vehicle's position with respect to the lead vehicle into the control algorithm noticeably reduced platoon contraction. Adjusting the control algorithm's responsiveness based on what data was successfully received reduced the speed-variations by vehicles. Finally, using past behavior to predict the next acceleration reduced the energy used by each vehicle. Combining these modifications with a model of the proposed communication scheme shows platoons of up to 25 vehicles are feasible.

In the recent decade, peer-to-peer (P2P) technology has greatly enhanced the scalability of multimedia streaming on the Internet by enabling efficient cooperation among end-users. However, existing streaming applications are plagued by the problems of long playback latency and long churn-induced delays. First of all, many streaming applications, such as IPTV and video conferencing, have rigorous constraints on end-to-end delays. Moreover, churn-induced delays, including delays from channel switching and streaming recovery, in current P2P streaming applications are typically in the scale of 10-60 seconds, which is far below the favorable user experience as in cable TV systems. These two issues in terms of playback latency and churn-induced delays have hindered the extensive commercial deployment of P2P systems. Motivated by this, in this dissertation, we focus on reducing delays in P2P live streaming systems. Specifically, we propose solutions for reducing delays in P2P live streaming systems in four problem spaces: (1) minimizing the maximum end-to-end delay in P2P streaming; (2) minimizing the average end-to-end delay in P2P streaming; (3) minimizing the average delay in multi-channel P2P streaming; and (4) reducing churn-induced delays. We devise a streaming scheme to minimize the maximum end-to-end streaming delay under a mesh-based overlay network paradigm. We call this problem, the MDPS problem. We formulate the MDPS problem and prove its NP-completeness. We then present a polynomial-time approximation algorithm, called Fastream-I, for this problem, and show that the performance of Fastream-I is bounded by a ratio of O(SQRT(log n)), where n is the number of peers in the system. We also develop a distributed version of Fastream-I that can adapt to network dynamics. Our simulation study reveals the effectiveness of Fastream-I, and shows a reasonable message overhead. While Fastream-I yields the minimum maximum end-to-end streaming delay (within a factor of O(SQRT(log n)), in many P2P settings, users may desire the minimum average end-to-end P2P streaming delay. Towards this, we devise a streaming scheme which optimizes the bandwidth allocation to achieve the minimum average end-to-end P2P streaming delay. We call this problem, the MADPS problem. We first develop a generic analytical framework for the MADPS problem. We then present Fastream-II as a solution to the MADPS problem. The core part of Fastream-II is a fast approximation algorithm, called APX-Fastream-II, based on primal-dual schema. We prove that the performance of APX-Fastream-II is bounded by a ratio of 1+w, where w is an adjustable input parameter. Furthermore, we show that the flexibility of w provides a trade-off between the approximation factor and the running time of Fastream-II. The third problem space of the dissertation is minimizing the average delay in multi-channel P2P streaming systems. Toward this, we present an algorithm, called Fastream-III. To reduce the influence from frequent channel-switching behavior, we build Fastream-III for the view-upload decoupling (VUD) model, where the uploaded content from a serving node is independent of the channel it views. We devise an approximation algorithm based on primal-dual schema for the critical component of Fastream-III, called APX-Fastream-III. In contrast to APX-Fastream-II, APX-Fastream-III addresses the extra complexity in the multichannel scenario and maintains the approximation bound by a ratio of 1+w. Besides playback lag, delays occurring in P2P streaming may arise from two other factors: node churn and channel switching. Since both stem from the re-connecting request in churn, we call them churn-induced delays. Optimizing churn-induced delays is the dissertation's fourth problem space. Toward this, we propose NAP, a novel agent-based P2P scheme, that provides preventive connections to all channels. Each channel in NAP selects powerful peers as agents to represent the peers in the channel to minimize control and message overheads. Agents distill the bootstrapping peers with superior bandwidth and lifetime expectation to quickly serve the viewer in the initial period of streaming. We build a queueing theory model to analyze NAP. Based on this model, we numerically compare NAP's performance with past efforts. The results of the numerical analysis reveal the effectiveness of NAP. / Ph. D.

Esta dissertação de mestrado apresenta um estudo sobre problemas de escalonamento de tarefas com atrasos de comunicação. Mais precisamente, são abordados problemas de escalonar um conjunto de tarefas em um conjunto de máquinas paralelas de número limitado ou não, e tarefas de tempo de processamento unitário, sujeitas a relações de precedência, e com atrasos de comunicação estabelecidos para cada par de tarefas precedentes, assumindo valores extremos, ou seja, podendo ser desprezíveis ou infinitamente grandes, isto com o objetivo de minimizaro o tempo em que a última tarefa escalonada termina seu processamento - minimização do makespan. Sendo assim, dois problemas são demostrados serem da classe NP-difícil. Para o primeiro, a quantidade de processadores é indicada a cada instância, sendo este resultado válido ainda que as relações de precedência formem um conjunto de cadeias (P|chains; cij ∈ {0, ∞}; pj = 1|Cmax). O segundo problema admite relações de precedência arbitrárias e é válido para qualquer quantidade fixa de processadores diferente de um (P2|prec;cij ∈ {0, ∞}; pj = 1|Cmax). Por outro lado, neste trabalho, dois outros problemas são demonstrados serem solúveis em tempo polinomial, ou seja, estarem na classe P, ambos quando uma quantidade ilimitada de processadores está disponível. É visto que, se a ordem de precedência das tarefas é limitada a uma árvore descendente, o problema é polinomial (P∞|tree; cij ∈ {0, ∞}; pj = 1|Cmax). O outro caso polinomial demonstrado é válido quando é permitido processar a mesma tarefa em mais de um processador (P∞|prec; cij ∈ {0, ∞}; pj = 1|Cmax). Para ambos os casos são apresentados os algoritmos polinomiais. Finalmente, são apresentados resultados para o problema de escalonar tarefas particionadas em conjuntos para os quais todas as tarefas devem ser processadas no mesmo processador. O problema é NP-difícil quando a quantidade de processadores é determinada a cada instância. Esse resultado é válido ainda que a precedência seja restrita a duas cadeias. O problema se torna polinomial quando o conjunto de partições é limitado por constante e as cadeias são restritas em uma das duas formas: pela quantidade delas ou pela quantidade de tarefas em cada uma delas. Como trabalho futuro, este estudo deixa em aberto a NP-Completude do problema de escalonar sob tais atrasos de comunicação de valores extremos, para uma quantidade fixa de processadores, quando a ordem de precedência é de alguma forma restrita, por exemplo, uma árvore descendente (Pm|out-tree;cij ∈ {0, ∞}; pj = 1|Cmax). / This Master’s Thesis presents a study on scheduling problems subject to communication delays. More precisely, this work involves job scheduling problems with a number of parallel machines, limited or not, and where the tasks (or jobs) have unit execution time, and are subject to some precedence relation. Communication delays are imposed at each pair of preceding tasks, taking extreme values, which may be negligible or infinitely large. The objective is minimize the completion time of the latest job to be processed, that is, to get the minimum makespan. Thus, NP-hard results are demonstrated for two cases. For the first, when the number of processors is indicated in the instance of the problem, and this result holds even when the precedence relation is restricted to a set of chains (P|chains; cij ∈ {0, ∞}; pj = 1|Cmax). The second results is valid when arbitrary precedence relations are allowed, and any fixed number of processors (greater than one) is available (P2|prec;cij ∈ {0, ∞}; pj = 1|Cmax). Two other problems are demonstrated to have polynomial time solutions, both when an unlimited number of processors are available. The first result imposes the precedence relation to be an out-tree (P∞|tree; cij ∈ {0, ∞}; pj = 1|Cmax). The second result is valid when the execution of the same job on multiples processors are allowed (P∞|prec; cij ∈ {0, ∞}; pj = 1|Cmax). For both cases, polynomial algorithms are presented. Finally, results are presented for the problem of job scheduling that are partitioned in sets which must be executed on the same processors. The problem is demonstrated to be NP-hard even if the precedence relation consists of two chains. Also, it is shown that the problem becomes solvable in polynomial time if the number of partitions is limited by a constant and the chains are restricted by a constant on either their number, or the number of tasks that each chain may have. As future work, this study leaves open whether is NP-hard the case to schedule tasks subject to such communication delays with extreme values, when a fixed number of processors is available, and the precedence relations are some how restricted, for example, by an out-tree (Pm|out-tree;cij ∈ {0, ∞}; pj = 1|Cmax).

Esta dissertação de mestrado apresenta um estudo sobre problemas de escalonamento de tarefas com atrasos de comunicação. Mais precisamente, são abordados problemas de escalonar um conjunto de tarefas em um conjunto de máquinas paralelas de número limitado ou não, e tarefas de tempo de processamento unitário, sujeitas a relações de precedência, e com atrasos de comunicação estabelecidos para cada par de tarefas precedentes, assumindo valores extremos, ou seja, podendo ser desprezíveis ou infinitamente grandes, isto com o objetivo de minimizaro o tempo em que a última tarefa escalonada termina seu processamento - minimização do makespan. Sendo assim, dois problemas são demostrados serem da classe NP-difícil. Para o primeiro, a quantidade de processadores é indicada a cada instância, sendo este resultado válido ainda que as relações de precedência formem um conjunto de cadeias (P|chains; cij ∈ {0, ∞}; pj = 1|Cmax). O segundo problema admite relações de precedência arbitrárias e é válido para qualquer quantidade fixa de processadores diferente de um (P2|prec;cij ∈ {0, ∞}; pj = 1|Cmax). Por outro lado, neste trabalho, dois outros problemas são demonstrados serem solúveis em tempo polinomial, ou seja, estarem na classe P, ambos quando uma quantidade ilimitada de processadores está disponível. É visto que, se a ordem de precedência das tarefas é limitada a uma árvore descendente, o problema é polinomial (P∞|tree; cij ∈ {0, ∞}; pj = 1|Cmax). O outro caso polinomial demonstrado é válido quando é permitido processar a mesma tarefa em mais de um processador (P∞|prec; cij ∈ {0, ∞}; pj = 1|Cmax). Para ambos os casos são apresentados os algoritmos polinomiais. Finalmente, são apresentados resultados para o problema de escalonar tarefas particionadas em conjuntos para os quais todas as tarefas devem ser processadas no mesmo processador. O problema é NP-difícil quando a quantidade de processadores é determinada a cada instância. Esse resultado é válido ainda que a precedência seja restrita a duas cadeias. O problema se torna polinomial quando o conjunto de partições é limitado por constante e as cadeias são restritas em uma das duas formas: pela quantidade delas ou pela quantidade de tarefas em cada uma delas. Como trabalho futuro, este estudo deixa em aberto a NP-Completude do problema de escalonar sob tais atrasos de comunicação de valores extremos, para uma quantidade fixa de processadores, quando a ordem de precedência é de alguma forma restrita, por exemplo, uma árvore descendente (Pm|out-tree;cij ∈ {0, ∞}; pj = 1|Cmax). / This Master’s Thesis presents a study on scheduling problems subject to communication delays. More precisely, this work involves job scheduling problems with a number of parallel machines, limited or not, and where the tasks (or jobs) have unit execution time, and are subject to some precedence relation. Communication delays are imposed at each pair of preceding tasks, taking extreme values, which may be negligible or infinitely large. The objective is minimize the completion time of the latest job to be processed, that is, to get the minimum makespan. Thus, NP-hard results are demonstrated for two cases. For the first, when the number of processors is indicated in the instance of the problem, and this result holds even when the precedence relation is restricted to a set of chains (P|chains; cij ∈ {0, ∞}; pj = 1|Cmax). The second results is valid when arbitrary precedence relations are allowed, and any fixed number of processors (greater than one) is available (P2|prec;cij ∈ {0, ∞}; pj = 1|Cmax). Two other problems are demonstrated to have polynomial time solutions, both when an unlimited number of processors are available. The first result imposes the precedence relation to be an out-tree (P∞|tree; cij ∈ {0, ∞}; pj = 1|Cmax). The second result is valid when the execution of the same job on multiples processors are allowed (P∞|prec; cij ∈ {0, ∞}; pj = 1|Cmax). For both cases, polynomial algorithms are presented. Finally, results are presented for the problem of job scheduling that are partitioned in sets which must be executed on the same processors. The problem is demonstrated to be NP-hard even if the precedence relation consists of two chains. Also, it is shown that the problem becomes solvable in polynomial time if the number of partitions is limited by a constant and the chains are restricted by a constant on either their number, or the number of tasks that each chain may have. As future work, this study leaves open whether is NP-hard the case to schedule tasks subject to such communication delays with extreme values, when a fixed number of processors is available, and the precedence relations are some how restricted, for example, by an out-tree (Pm|out-tree;cij ∈ {0, ∞}; pj = 1|Cmax).

Esta dissertação de mestrado apresenta um estudo sobre problemas de escalonamento de tarefas com atrasos de comunicação. Mais precisamente, são abordados problemas de escalonar um conjunto de tarefas em um conjunto de máquinas paralelas de número limitado ou não, e tarefas de tempo de processamento unitário, sujeitas a relações de precedência, e com atrasos de comunicação estabelecidos para cada par de tarefas precedentes, assumindo valores extremos, ou seja, podendo ser desprezíveis ou infinitamente grandes, isto com o objetivo de minimizaro o tempo em que a última tarefa escalonada termina seu processamento - minimização do makespan. Sendo assim, dois problemas são demostrados serem da classe NP-difícil. Para o primeiro, a quantidade de processadores é indicada a cada instância, sendo este resultado válido ainda que as relações de precedência formem um conjunto de cadeias (P|chains; cij ∈ {0, ∞}; pj = 1|Cmax). O segundo problema admite relações de precedência arbitrárias e é válido para qualquer quantidade fixa de processadores diferente de um (P2|prec;cij ∈ {0, ∞}; pj = 1|Cmax). Por outro lado, neste trabalho, dois outros problemas são demonstrados serem solúveis em tempo polinomial, ou seja, estarem na classe P, ambos quando uma quantidade ilimitada de processadores está disponível. É visto que, se a ordem de precedência das tarefas é limitada a uma árvore descendente, o problema é polinomial (P∞|tree; cij ∈ {0, ∞}; pj = 1|Cmax). O outro caso polinomial demonstrado é válido quando é permitido processar a mesma tarefa em mais de um processador (P∞|prec; cij ∈ {0, ∞}; pj = 1|Cmax). Para ambos os casos são apresentados os algoritmos polinomiais. Finalmente, são apresentados resultados para o problema de escalonar tarefas particionadas em conjuntos para os quais todas as tarefas devem ser processadas no mesmo processador. O problema é NP-difícil quando a quantidade de processadores é determinada a cada instância. Esse resultado é válido ainda que a precedência seja restrita a duas cadeias. O problema se torna polinomial quando o conjunto de partições é limitado por constante e as cadeias são restritas em uma das duas formas: pela quantidade delas ou pela quantidade de tarefas em cada uma delas. Como trabalho futuro, este estudo deixa em aberto a NP-Completude do problema de escalonar sob tais atrasos de comunicação de valores extremos, para uma quantidade fixa de processadores, quando a ordem de precedência é de alguma forma restrita, por exemplo, uma árvore descendente (Pm|out-tree;cij ∈ {0, ∞}; pj = 1|Cmax). / This Master’s Thesis presents a study on scheduling problems subject to communication delays. More precisely, this work involves job scheduling problems with a number of parallel machines, limited or not, and where the tasks (or jobs) have unit execution time, and are subject to some precedence relation. Communication delays are imposed at each pair of preceding tasks, taking extreme values, which may be negligible or infinitely large. The objective is minimize the completion time of the latest job to be processed, that is, to get the minimum makespan. Thus, NP-hard results are demonstrated for two cases. For the first, when the number of processors is indicated in the instance of the problem, and this result holds even when the precedence relation is restricted to a set of chains (P|chains; cij ∈ {0, ∞}; pj = 1|Cmax). The second results is valid when arbitrary precedence relations are allowed, and any fixed number of processors (greater than one) is available (P2|prec;cij ∈ {0, ∞}; pj = 1|Cmax). Two other problems are demonstrated to have polynomial time solutions, both when an unlimited number of processors are available. The first result imposes the precedence relation to be an out-tree (P∞|tree; cij ∈ {0, ∞}; pj = 1|Cmax). The second result is valid when the execution of the same job on multiples processors are allowed (P∞|prec; cij ∈ {0, ∞}; pj = 1|Cmax). For both cases, polynomial algorithms are presented. Finally, results are presented for the problem of job scheduling that are partitioned in sets which must be executed on the same processors. The problem is demonstrated to be NP-hard even if the precedence relation consists of two chains. Also, it is shown that the problem becomes solvable in polynomial time if the number of partitions is limited by a constant and the chains are restricted by a constant on either their number, or the number of tasks that each chain may have. As future work, this study leaves open whether is NP-hard the case to schedule tasks subject to such communication delays with extreme values, when a fixed number of processors is available, and the precedence relations are some how restricted, for example, by an out-tree (Pm|out-tree;cij ∈ {0, ∞}; pj = 1|Cmax).

L'objectif principal de cette thèse est de proposer les méthodes permettant d'obtenir le délai de transmission de bout en bout pire cas exact d'un réseau AFDX. Actuellement, seules des bornes supérieures pessimistes peuvent être calculées en utilisant les approches de type Calcul Réseau ou par Trajectoires. Pour cet objectif, différentes approches et outils existent et ont été analysées dans le contexte de cette thèse. Cette analyse a mis en évidence le besoin de nouvelles approches. Dans un premier temps, la vérification de modèle a été explorée. Les automates temporisés et les outils de verification ayant fait leur preuve dans le domaine temps réel ont été utilisés. Ensuite, une technique de simulation exhaustive a été utilisée pour obtenir les délais de communication pire cas exacts. Pour ce faire, des méthodes de réduction de séquences ont été définies et un outil a été développé. Ces méthodes ont été appliquées à une configuration réelle du réseau AFDX, nous permettant ainsi de valider notre travail sur une configuration de taille industrielle du réseau AFDX telle que celle embarquée à bord des avions Airbus A380. The main objective of this thesis is to provide methodologies for finding exact worst case end to end communication delays of AFDX network. Presently, only pessimistic upper bounds of these delays can be calculated by using Network Calculus and Trajectory approach. To achieve this goal, different existing tools and approaches have been analyzed in the context of this thesis. Based on this analysis, it is deemed necessary to develop new approaches and algorithms. First, Model checking with existing well established real time model checking tools are explored, using timed automata. Then, exhaustive simulation technique is used with newly developed algorithms and their software implementation in order to find exact worst case communication delays of AFDX network. All this research work has been applied on real life implementation of AFDX network, allowing us to validate our research work on industrial scale configuration of AFDX network such as used on Airbus A380 aircraft. / The main objective of this thesis is to provide methodologies for finding exact worst case end to end communication delays of AFDX network. Presently, only pessimistic upper bounds of these delays can be calculated by using Network Calculus and Trajectory approach. To achieve this goal, different existing tools and approaches have been analyzed in the context of this thesis. Based on this analysis, it is deemed necessary to develop new approaches and algorithms. First, Model checking with existing well established real time model checking tools are explored, using timed automata. Then, exhaustive simulation technique is used with newly developed algorithms and their software implementation in order to find exact worst case communication delays of AFDX network. All this research work has been applied on real life implementation of AFDX network, allowing us to validate our research work on industrial scale configuration of AFDX network such as used on Airbus A380 aircraft.