基于饱和TCP流量的Wi-Fi级联吞吐量模型

doi:10.11959/j.issn.2096-3750.2023.00356

摘要/Abstract

摘要：

目前，无线保真（Wi-Fi, wireless fidelity）技术的重要性与商业价值已经获得了广泛认可，同时人们也追求更好的服务质量（QoS, quality of service）。传输控制协议（TCP, transmission control protocol）作为一种面向连接的、提供端到端可靠性服务的通信协议，其重要性也日益上升，是互联网中占主导地位的传输协议。此外，由于单个接入点（AP, access point）的覆盖范围有限，在许多场景下往往需要通过级联来扩展覆盖范围。然而，现有的关于级联场景的研究均基于用户数据报协议（UDP, user datagram protocol），即认为场景中仅存在下行流量，而许多场景中上下行流量往往同时存在，即运行的是TCP流量，在级联场景下用户获得TCP吞吐量应如何计算一直是一个尚待解决的问题。针对该问题，首先在级联场景下建立了饱和TCP流量条件下的Wi-Fi吞吐量模型，该模型通过两个马尔可夫链实现对用户吞吐量的计算；其次，搭建了级联系统进行实测，通过对比理论仿真结果与实际测试结果，对模型的有效性进行了验证。

关键词: Wi-Fi, QoS, TCP, 吞吐量

Abstract:

At present, the importance and commercial value of wireless fidelity (Wi-Fi) technology have been widely recognized.And people are also pursuing better quality of service (QoS).Transmission control protocol (TCP) as a connection-oriented communication protocol that provides end-to-end reliability services, its importance is also increasing, and it is the dominant transmission protocol in the Internet.Besides, since the coverage of a single access point (AP) is limited, it is often necessary to extend the coverage through cascading in many scenarios.However, the current research on cascading scenarios is based on the user datagram protocol (UDP), which means that only downlink traffic exists in the scenario.Actually, uplink and downlink traffic often exist at the same time in many scenarios, which means TCP traffic is running.How to calculate the TCP throughput obtained by users in the cascading scenario has always been an unresolved problem.To solve this problem, firstly, a Wi-Fi throughput model under the condition of saturated TCP traffic was established in the cascading scenario, and the calculation of user throughput was realized through two Markov chains.Secondly, a cascading system was built and actual tests were conducted.The validity of the model is verified by comparing the theoretical simulation results with the actual test results.

Key words: Wi-Fi, QoS, TCP, throughput

中图分类号:

TN92

张俊雄, 刘剑钊, 葛晓虎, 吴伟民. 基于饱和TCP流量的Wi-Fi级联吞吐量模型[J]. 物联网学报, 2023, 7(3): 53-61.

Junxiong ZHANG, Jianzhao LIU, Xiaohu GE, Weimin WU. Throughput model of Wi-Fi cascading based on saturated TCP traffic[J]. Chinese Journal on Internet of Things, 2023, 7(3): 53-61.

图/表 8

图1

图2

图3

图4

图5

表1

图6

图7

参考文献 25

[1]	华为. 释放Wi-Fi的潜能,2019～2023企业级Wi-Fi 6产业发展与展望白皮书[EB]. 2019.
	HUAWEI. Unleash the potential of Wi-Fi,2019～2023 enterprise Wi-Fi 6 industry development and prospect white paper[EB]. 2019.
[2]	林玉梅, 章喜字 . 高校校园 WLAN 安全方案设计与实施[J]. 软件导刊, 2016,15(6): 200-201.
	LIN Y M , ZHANG X Z . Design and implementation of WLAN security scheme in university campus[J]. Software Guide, 2016,15(6): 200-201.
[3]	俞晓辉, 何伟 . 无线局域网(WLAN)模块化建设思想在南京联通工程建设中的实践和探索[J]. 中国新通信, 2019,21(17): 55-56.
	YU X H , HE W . Practice and exploration of modular construction of WLAN in Nanjing unicom project construction[J]. China New Telecommunications, 2019,21(17): 55-56.
[4]	BIANCHI G . Performance analysis of the IEEE 802.11 distributed coordination function[J]. IEEE Journal on Selected Areas in Communications, 2000,18(3): 535-547.
[5]	TAY Y C , CHUA K C . A capacity analysis for the IEEE 802.11 MAC protocol[J]. Wireless Networks, 2001,7(2): 159-171.
[6]	WU H T , PENG Y , LONG K P ,et al. Performance of reliable transport protocol over IEEE 802.11 wireless LAN:analysis and enhancement[C]// Proceedings of Twenty-First Annual Joint Conference of the IEEE Computer and Communications Societies. Piscataway:IEEE Press, 2002: 599-607.
[7]	XIAO Y . Backoff-based priority schemes for IEEE 802.11[C]// Proceedings of IEEE International Conference on Communications,2003.ICC＇03. Piscataway:IEEE Press, 2003: 1568-1572.
[8]	HE J H , ZHENG L , YANG Z K ,et al. Performance analysis and service differentiation in IEEE 802.11 WLAN[C]// Proceedings of 28th Annual IEEE International Conference on Local Computer Networks,2003.LCN ＇03.Proceedings. Piscataway:IEEE Press, 2003: 691-697.
[9]	HUI J , DEVETSIKIOTIS M . Designing improved MAC packet schedulers for 802.11e WLAN[C]// Proceedings of GLOBECOM ＇03.IEEE Global Telecommunications Conference (IEEE Cat.No.03CH37489). Piscataway:IEEE Press, 2004: 184-189.
[10]	ZHU H , CHLAMTAC I . Performance analysis for IEEE 802.11e EDCF service differentiation[J]. IEEE Transactions on Wireless Communications, 2005,4(4): 1779-1788.
[11]	LEE J Y , LEE H S . A performance analysis model for IEEE 802.11e EDCA under saturation condition[J]. IEEE Transactions on Communications, 2009,57(1): 56-63.
[12]	ZHENG J , WU Q . Performance modeling and analysis of the IEEE 802.11p EDCA mechanism for VANET[J]. IEEE Transactions on Vehicular Technology, 2016,65(4): 2673-2687.
[13]	ZHANG Y Y , LI S P , SHANG Z H ,et al. Performance analysis of IEEE 802.11 DCF under different channel conditions[C]// Proceedings of 2019 IEEE 8th Joint International Information Technology and Artificial Intelligence Conference (ITAIC). Piscataway:IEEE Press, 2019: 1904-1907.
[14]	张虹 . UDP与TCP结合实现多进程通信[J]. 中山大学学报:自然科学版, 2002(z1): 4.
	ZHANG H . Combining UDP and TCP to realize multi-process communication[J]. Journal of SUN YAT-SEN University (Social Science Edition), 2002(z1): 4.
[15]	YU J , CHOI S , QIAO D J . TCP dynamics over IEEE 802.11E WLANs:modeling and throughput enhancement[C]// Proceedings of 2007 Fourth International Conference on Broadband Communications,Networks and Systems (BROADNETS ＇07). Piscataway:IEEE Press, 2008: 66-75.
[16]	KANEMATSU T , NGUYEN K , SEKIYA H . Throughput analysis for IEEE 802.11 multi-hop networks considering transmission rate[C]// Proceedings of 2019 IEEE 89th Vehicular Technology Conference (VTC2019-Spring). Piscataway:IEEE Press, 2019: 1-5.
[17]	GUO Y H , LI Z X , YANG Z X . Performance analysis of A-MSDU and UDP in IEEE 802.11 wireless linear multi-hop network[C]// Proceedings of 2020 IEEE 10th International Conference on Electronics Information and Emergency Communication (ICEIEC). Piscataway:IEEE Press, 2020: 22-25.
[18]	DEMIRTA? O , ILYAS M . Cloud assisted approach for determining Wi-Fi problems in field deployed mesh APs - A case study for quality problems[C]// Proceedings of 2020 28th Signal Processing and Communications Applications Conference (SIU). Piscataway:IEEE Press, 2021: 1-4.
[19]	MACABALE N A , VILLASOTO A N , RIVERA J D ,et al. CRADLE:cross-layer design for load-aware routing in IEEE 802.11-based wireless mesh and sensor networks[C]// Proceedings of 2020 10th Annual Computing and Communication Workshop and Conference (CCWC). Piscataway:IEEE Press, 2020: 970-974.
[20]	MAESAKO K , TAKAKI Y , KAMADA T ,et al. Asymmetric hidden node problem aware routing metric for wireless mesh networks[C]// Proceedings of 2019 16th IEEE Annual Consumer Communications ＆Networking Conference (CCNC). Piscataway:IEEE Press, 2019: 1-7.
[21]	RETHFELDT M , BROCKMANN T , ECKHARDT R ,et al. Extending the flexible network tester (flent) for IEEE 802.11s WLAN mesh networks[C]// Proceedings of 2022 IEEE International Symposium on Measurements ＆ Networking (M＆N). Piscataway:IEEE Press, 2022: 1-6.
[22]	BACKHAUS M , ROSSBERG M , SCHAEFER G . Towards a realistic maximum flow model in hybrid multi-channel wireless mesh networks[C]// Proceedings of 2021 Wireless Days (WD). Piscataway:IEEE Press, 2021: 1-8.
[23]	GOKALGANDHI B , TAVARES M , SAMARDZIJA D ,et al. Reliable low-latency Wi-Fi mesh networks[J]. IEEE Internet of Things Journal, 2022,9(6): 4533-4553.
[24]	KIRAN M P R S , RAJALAKSHMI P . Saturated throughput analysis of IEEE 802.11ad EDCA for high data rate 5G-IoT applications[J]. IEEE Transactions on Vehicular Technology, 2019,68(5): 4774-4785.
[25]	VALLEJOS C R A , MARTíNEZ V J M . A fast transformation of Markov chains and their respective steady-state probability distributions[J]. The Computer Journal, 2014,57(1): 1-11.

参数	说明	设定值
MSDU	MAC服务数据单元大小	1 548 byte
MSDU_N	MAC协议数据单元内的MSDU聚合数量	3
MPDU_N	物理层协议数据单元内MPDU聚合数量	64
ACK	STA回复确认报文大小	40 byte
R	各设备空口速率	2 144 Mbit/s
σ	时隙	9 μs
SIFS	最小帧间间隔	16 μs
AIFSN	各设备仲裁帧间间隔数	3
w	STA接收窗口	3