doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 1 End-to-End Performance and Fairness in Multihop Wireless Backhaul Networks Notice: This document has been prepared to assist IEEE It is offered as a basis for discussion and is not binding on the contributing individual(s) or organization(s). The material in this document is subject to change in form and content after further study. The contributor(s) reserve(s) the right to add, amend or withdraw material contained herein. Release: The contributor grants a free, irrevocable license to the IEEE to incorporate material contained in this contribution, and any modifications thereof, in the creation of an IEEE Standards publication; to copyright in the IEEE’s name any IEEE Standards publication even though it may include portions of this contribution; and at the IEEE’s sole discretion to permit others to reproduce in whole or in part the resulting IEEE Standards publication. The contributor also acknowledges and accepts that this contribution may be made public by IEEE Patent Policy and Procedures: The contributor is familiar with the IEEE 802 Patent Policy and Procedures, including the statement "IEEE standards may include the known use of patent(s), including patent applications, provided the IEEE receives assurance from the patent holder or applicant with respect to patents essential for compliance with both mandatory and optional portions of the standard." Early disclosure to the Working Group of patent information that might be relevant to the standard is essential to reduce the possibility for delays in the development process and increase the likelihood that the draft publication will be approved for publication. Please notify the Chair as early as possible, in written or electronic form, if patented technology (or technology under patent application) might be incorporated into a draft standard being developed within the IEEE Working Group. If you have questions, contact the IEEE Patent Committee Administrator at. Date: Authors:

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 2 Abstract Wireless IEEE networks in residences, small businesses, and public “hot spots” typically encounter the wireline access link (DSL, cable modem, T1, etc.) as the slowest and most expensive part of the end-to-end path. Consequently, network architectures have been proposed that employ multiple wireless hops in route to and from the wired Internet. Unfortunately, use of current media access and transport protocols for such systems can result in severe unfairness and even starvation for flows that are an increasing number of hops away from a wired Internet entry point. Our objective is to study fairness and end-to-end performance in multihop wireless backhaul networks via the following methodology. First, we develop a formal reference model that characterizes objectives such as removing spatial bias (i.e., providing performance that is independent of the number of wireless hops to a wire) and maximizing spatial reuse. Second, we perform an extensive set of simulation experiments to quantify the impact of the key performance factors towards achieving these goals. For example, we study the roles of the MAC protocol, end-to-end congestion control, antenna technology, and traffic types. Next, we develop and study a distributed layer 2 fairness algorithm which targets to achieve the fairness of the reference model without modification to TCP. Finally, we study the critical relationship between fairness and aggregate throughput and in particular study the fairness-constrained system capacity of multihop wireless backhaul networks.

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 3 TAPs: Multihop Wireless Infrastructure Transit Access Points (TAPs) are APs with –MIMO antennas –multiple air interfaces –enhanced MAC/scheduling/routing protocols Form wireless backbone with limited wired gateways

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 4 Research Challenges 1.Physical layer –Achieve 400 Mb/s among TAPs 2.Media access –Target multi-hop and exploit PHY capabilities 3.Fairness and traffic control –With TCP/WiFi, nodes farther away from wires get much less bandwidth and can starve 4.Prototypes, Testbeds, and Measurement Studies –Platforms for experimentation and proof-of-concept 3. Fairness and traffic control

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 5 Technology For All Deployment Technology For All – Houston, Texas (non-profit) Empower low income communities through technology –$10,326 per-capita income Applications –Education and work-at-home (“Learn-and-Earn” and Job-Tech)

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 6 Technology For All Deployment Multi-hop IEEE wireless network covering 40,000 residents –Single wireline Internet backhaul –Long-haul directional links

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 7 Rice TAP Platform 400 Mb/sec via 4x4 MIMO custom design –Single 20 MHz WiFi channel at 2.4 GHz and 20 bits/sec/Hz efficiency –Feedback-based algorithms for beam-forming MIMO Custom MAC design and FPGA implementation

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 8 Prototype and Testbed Deployment FPGA implementation of enhanced MAC (opportunistic, MIMO, multi-channel, QoS) Build prototypes and deploy on Rice campus and nearby neighborhoods Measurement study from channel conditions to traffic patterns

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 9 Wireless Backhaul Networks TAP Networks Residential user or small business Wireless Backhaul Network Internet

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 10 Fundamental Scenario One branch of the access tree Internet TAP1 TAP2 TAP3 TAP4 Traffic matrix –Traffic to and from Internet

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 11 Parking Lot Scenario Similar to parking lot with one exit Internet TAP1 TAP2 TAP3 TAP4

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 12 Fairness Problem

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 13 Fairness Problem

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 14 Fairness Problem Goal Ensure equal shares independent of spatial location We need multihop fairness

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 15 Contributions Fairness reference model –Objective: Remove spatial bias and max spatial reuse –Granularity: Ingress-Aggregated flows vs. source-destination pair –Resource: space and time vs. bandwidth and buffers –Precise definition in [GSK04] Performance study –TCP –Inter-TAP fairness algorithm Capacity and fairness Wireless Backhaul Network

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 16 Contributions Fairness reference model –Objective: Remove spatial bias and max spatial reuse –Granularity: Ingress-Aggregated flows vs. source-destination pair –Resource: space and time vs. bandwidth and buffers –Precise definition in [GSK04] Performance study –TCP –Inter-TAP fairness algorithm Capacity and fairness Wireless Backhaul Network

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 17 Problem Statement Fairness reference model defined Distributed algorithm –Targeted at achieving shares defined by reference model Solution space – Local solution – insufficient Example: Parking lot – Multihop solution Flow e2e – TCP Multihop wireless network e2e – Inter-TAP Fairness Algorithm (IFA)

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 18 Performance Factors (1/2) Factors investigated Fairness algorithms –Uncontrolled UDP, TCP, IFA Media access control – with two-way and four-way handshake Antenna technologies –Omni directional, sector Carrier sense range, multiple topologies and flow scenarios… Other simulation specs Channel rate constant 2 Mb/sec, 1000 byte packets Goal Study end-to-end performance and fairness

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 19 Performance Factors (2/2) Well understood topologies Increased no. of hops from destination Reduced throughput Increased no. of source-dest. pairs Reduced throughput Topology

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 20 Performance Factors (2/2) Parking lot MU-TAP and TAP-TAP transmissions on orthogonal channels Internet TAP1 TAP2 TAP3 TAP4 TA(1) TA(2) TA(3) Topology

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 21 Fairness with TCP MAC, Hidden Terminals and Information Asymmetry Idealized objective –Assumes perfect collision-free MAC ACK Traffic MUs generate long lived TCP-Sack flows Carrier sense range = transmission range TAP1 TAP2 TAP3 TAP4

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 22 ACK Traffic TAP 1 and TAP 2 traffic starved –Both are hidden terminals –Timeouts – significant throughput penalty TCP generates bursts of packets MUs generate long lived TCP-Sack flows Carrier sense range = transmission range TAP1 TAP2 TAP3 TAP4 Fairness with TCP MAC, Hidden Terminals and Information Asymmetry

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 23 RTS/CTS exchange introduces information asymmetry [KSSK02] –TAP1 has no information of TAP3-TAP4 trans. ACK Traffic Capacity and fairness need to be considered jointly –Total is up to 125% of objective while two flows are starved Fairness with TCP MAC, Hidden Terminals and Information Asymmetry MUs generate long lived TCP-Sack flows Carrier sense range = transmission range TAP1 TAP2 TAP3 TAP4

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 24 TCP and Sector Antennas TAP1 TAP2 TAP3 TAP4 ACK Traffic Impact of hidden terminals and information asymmetry mitigated Severe spatial bias –TAP 1 traffic obtains 26% of objective Total goodput increased Total goodput is 67% of the objective MUs generate long lived TCP-Sack flows TAPs use sector antennas

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 25 Inter-TAP Fairness Algorithm (IFA) Idealized version of algorithm –Omniscient calculation of fair rates Practical algorithm needs messaging and incurs delays Limit traffic rate at ingress Outcome –Approximates objectives despite MAC anomalies –Independent of transport protocols (TCP/UDP/…)

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 26 TCP and IFA MUs generate long lived TCP-Sack flows Carrier sense range = transmission range TAP1 TAP2 TAP3 TAP4 End-to-end performance considerably improved –TAP-aggregated throughput is 59% to 75% of the objective Hidden terminal problem mitigated – Contention considerably decreased –TCP cannot inject bursts of packets ACK Traffic Spatial bias – IFA alone cannot eliminate it Rates lower than the objective

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 27 Inter-TAP Performance Isolation Provide inter-TAP performance isolation independent of traffic types ACK Traffic TCP achieves 64% of idealized objective, while UDP obtains 75% Even with balanced contention TCP reduces its rate –Having more MUs per TAP TCP performance degraded Each TAP has one MU TAP1: MU transmits TCP traffic TAP2 and TAP3: MU transmits UDP traffic TAP1 TAP2 TAP3 TAP4

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 28 Summary of Findings (1/2) Starvation of upstream flows (UDP, TCP, with or w/o RTS/CTS) –“Parking Lot” scenario results in hidden terminals and information asymmetry Sector antennas and carrier sense range mitigate the hidden terminal problem –Severe spatial bias SA: Throughput as low as 26% of targeted values CSR: Throughput as low as 34% of targeted values TCP able to exploit spatial reuse

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 29 Summary of Findings (2/2) IFA approximates reference model performance The impact of hidden terminal problem and information asymmetry mitigated –Without any modifications to CSMA/CA TCP over IFA achieves 59% to 75% of idealized objective –Without any modifications to TCP Inter-TAP performance isolation

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 30 Conclusions Fairness Fairness reference model formally defined Designed for multihop wireless access networks Performance study Starvation of upstream flows Sector antennas, larger carrier sense range, IFA mitigate the problem IFA approximates performance of reference model Capacity and fairness Consider joint effects

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 31 References [GSK04] V. Gambiroza, B. Sadeghi, and E. Knightly, "End- to-End Performance and Fairness in Multihop Wireless Backhaul Networks,'' in Proceedings of ACM MobiCom 2004, Philadelphia, PA, September TAPs project: TFA: Rice Networks Group:

doc.: IEEE /0168r0 Submission March 2005 Violeta Gambiroza, Rice UniversitySlide 32 End-to-End Performance and Fairness in Multihop Wireless Backhaul Networks V. Gambiroza and E. Knightly, Rice University B. Sadeghi, Intel Corporation