Yohan Baga, Morgane Richaud, Ghaffari Fakhreddine, E. Zante, D. Declercq, Michael Nahmiyace
{"title":"用于终端系统积压评估的AFDX帧接收概率模型","authors":"Yohan Baga, Morgane Richaud, Ghaffari Fakhreddine, E. Zante, D. Declercq, Michael Nahmiyace","doi":"10.1109/SIES.2017.7993400","DOIUrl":null,"url":null,"abstract":"The Avionics Full-Duplex Switched Ethernet (AFDX) network has established itself as a reference in aeronautical embedded communications to comply with the growth of bandwidth needs and the reliability requirements. Its determinism property imposes that AFDX frames have to transit through the network in a limited amount of time. Due to potential delays in the processing frames at a reception End-System, the frames have to be stored in a reception buffer. Generally, the buffer is dimensioned enough large to avoid prohibitive frame losses, but it results a waste of memory resources. The buffer dimensioning issue requires an analysis of the reception flow to determine the worst frame backlog. As the frame backlog is maximal when a sequence of back-to-back frames is received by the ES, the key point is to estimate the occurrences of such SBFs on a representative sample of received frames. In this paper, we address this issue using a probabilistic model based on local Gaussian distributions, and we propose results for a range of typical configuration tables of the reception End-System.","PeriodicalId":262681,"journal":{"name":"2017 12th IEEE International Symposium on Industrial Embedded Systems (SIES)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"5","resultStr":"{\"title\":\"Probabilistic model of AFDX frames reception for end system backlog assessment\",\"authors\":\"Yohan Baga, Morgane Richaud, Ghaffari Fakhreddine, E. Zante, D. Declercq, Michael Nahmiyace\",\"doi\":\"10.1109/SIES.2017.7993400\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The Avionics Full-Duplex Switched Ethernet (AFDX) network has established itself as a reference in aeronautical embedded communications to comply with the growth of bandwidth needs and the reliability requirements. Its determinism property imposes that AFDX frames have to transit through the network in a limited amount of time. Due to potential delays in the processing frames at a reception End-System, the frames have to be stored in a reception buffer. Generally, the buffer is dimensioned enough large to avoid prohibitive frame losses, but it results a waste of memory resources. The buffer dimensioning issue requires an analysis of the reception flow to determine the worst frame backlog. As the frame backlog is maximal when a sequence of back-to-back frames is received by the ES, the key point is to estimate the occurrences of such SBFs on a representative sample of received frames. In this paper, we address this issue using a probabilistic model based on local Gaussian distributions, and we propose results for a range of typical configuration tables of the reception End-System.\",\"PeriodicalId\":262681,\"journal\":{\"name\":\"2017 12th IEEE International Symposium on Industrial Embedded Systems (SIES)\",\"volume\":\"24 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2017-06-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"5\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2017 12th IEEE International Symposium on Industrial Embedded Systems (SIES)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/SIES.2017.7993400\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2017 12th IEEE International Symposium on Industrial Embedded Systems (SIES)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/SIES.2017.7993400","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Probabilistic model of AFDX frames reception for end system backlog assessment
The Avionics Full-Duplex Switched Ethernet (AFDX) network has established itself as a reference in aeronautical embedded communications to comply with the growth of bandwidth needs and the reliability requirements. Its determinism property imposes that AFDX frames have to transit through the network in a limited amount of time. Due to potential delays in the processing frames at a reception End-System, the frames have to be stored in a reception buffer. Generally, the buffer is dimensioned enough large to avoid prohibitive frame losses, but it results a waste of memory resources. The buffer dimensioning issue requires an analysis of the reception flow to determine the worst frame backlog. As the frame backlog is maximal when a sequence of back-to-back frames is received by the ES, the key point is to estimate the occurrences of such SBFs on a representative sample of received frames. In this paper, we address this issue using a probabilistic model based on local Gaussian distributions, and we propose results for a range of typical configuration tables of the reception End-System.