1

2 Enterprise WLAN setting 2 Vivek Shrivastava Wireless controller Access Point Clients Internet NSDI 2011

3 Enterprise WLAN setting 3 Vivek Shrivastava Wireless controller Access Point Clients Internet Functionalities implemented at controller Intrusion detection system Interference management (channel assignment, power control) NSDI 2011

4 Enterprise WLAN setting 4 Vivek Shrivastava Wireless controller Access Point Clients Internet Functionalities implemented at controller Intrusion detection system Interference management (channel assignment, power control) NSDI 2011

55 Vivek Shrivastava Problems with wireless* Support “Flaky how ?” “We will take a look.” “The wireless is being flaky.” “Well my connection dropped earlier and now it seems to be slow” “Wait, now it seems fine.” User NSDI 2011 * Slide borrowed from Cheng et. al (Jigsaw, Sigcomm ’06)

66 Vivek Shrivastava Problems with wireless NSDI Hidden terminals

77 Vivek Shrivastava Mismatch in data rates, slows down fast links Problems with wireless Rate anomaly 6 Mbps 54 Mbps NSDI 2011

88 Vivek Shrivastava Problems with wireless NSDI Increasing client density and mobility ….

99 Vivek Shrivastava Problems with wireless NSDI Increasing client density and mobility….

10 Vivek Shrivastava Problems with wireless NSDI …. changing interference patterns

11 Vivek Shrivastava Interference management in WLANs NSDI

12 Vivek Shrivastava Interference management in WLANs Estimate Interference dynamically Manage Interference (data scheduling, transmit power control, channel assignment) NSDI

13 Vivek Shrivastava Interference management in WLANs Manage Interference (data scheduling, transmit power control, channel assignment) NSDI Estimate Interference dynamically

14 Vivek Shrivastava How to estimate interference ? Use bandwidth tests NSDI 2011

15 Vivek Shrivastava How to estimate interference ? Use bandwidth tests Interferer AP-Client pair NSDI 2011

16 Vivek Shrivastava How to estimate interference ? NSDI )Measure AP-Client delivery in isolation

17 Vivek Shrivastava How to estimate interference ? Isolation delivery= 0.95 NSDI )Measure AP-Client delivery in isolation

18 Vivek Shrivastava How to estimate interference ? NSDI )Measure AP-Client delivery in isolation 2)Activate interferer and measure delivery

19 Vivek Shrivastava How to estimate interference ? Interference delivery= 0.66 NSDI )Measure AP-Client delivery in isolation 2)Activate interferer and measure delivery

20 Vivek Shrivastava How to estimate interference ? 1)Measure AP-Client delivery in isolation 2)Activate interferer and measure delivery NSDI 2011 Link Interference Ratio (LIR) = del Interference / del isolation

21 Vivek Shrivastava How to estimate interference ? NSDI )Measure AP-Client delivery in isolation 2)Activate interferer and measure delivery Link Interference Ratio (LIR) = 0.66 / 0.95 = 0.69

22 Vivek Shrivastava How to estimate interference ? NSDI )Measure AP-Client delivery in isolation 2)Activate interferer and measure delivery LIR 01 StrongWeak

23 But are bandwidth tests practical ? Can we use bandwidth tests in live settings Good accuracy – Network downtime required - X Not scalable (~ 1 hr for 20 AP-Client pair network) - X Not based on realistic rates and packet sizes – X Inefficient in dynamic scenario (client mobility) – X 23 Vivek Shrivastava NSDI 2011

24 But are bandwidth tests practical ? Can we use bandwidth tests in live settings Good accuracy – Network downtime required - X Not scalable (~ 1 hr for 20 AP-Client pair network) - X Not based on realistic rates and packet sizes – X Inefficient in dynamic scenario (client mobility) – X 24 Vivek Shrivastava NSDI 2011 Can we estimate interference in a passive, real-time way ?

25 PIE Outline Motivation Conventional bandwidth tests not sufficient Passive Interference Estimation (PIE) Polling period of PIE Accuracy of PIE Realistic trace replay with PIE Applications of PIE Summary 25 Vivek Shrivastava NSDI 2011

26 Vivek Shrivastava Estimating interference passively Sniffer NSDI 2011

27 Vivek Shrivastava Estimating interference passively Sniffer NSDI 2011 Sniffer could be a dedicated wireless radio Clocks synchronized using wired backplane

28 Vivek Shrivastava Estimating interference passively Sniffer reports NSDI 2011

29 Vivek Shrivastava Estimating interference passively Sniffer reports Timestamp, duration, rate, success. NSDI 2011

30 Vivek Shrivastava Estimating interference passively Hidden terminals NSDI 2011

31 Vivek Shrivastava Hidden terminals Estimating interference passively NSDI 2011

32 Vivek Shrivastava Hidden terminals Estimating interference passively NSDI 2011

33 Vivek Shrivastava 1.Carrier sense Hidden terminals Estimating interference passively NSDI 2011

34 Vivek Shrivastava 2. Channel free, transmit Hidden terminals Estimating interference passively 1)Note timestamp, rate, duration NSDI 2011

35 Vivek Shrivastava 3. Collision ! Estimating interference passively 1)Note timestamp, rate, duration 2)Note if transmission is a success (ack received ?) NSDI 2011

36 Vivek Shrivastava Estimating interference passively Timestamp: T0 Rate: 6Mbps Length: 1400 bytes Success: False Timestamp: T0 Rate: 6Mbps Length: 1400 bytes Success: False Timestamp: T0 + δ Rate: 12Mbps Length: 600 bytes Success: False Timestamp: T0 + δ Rate: 12Mbps Length: 600 bytes Success: False NSDI 2011

37 Vivek Shrivastava Estimating interference passively T0 6Mbps 1400 bytes False T0 6Mbps 1400 bytes False T0 + δ 12Mbps 600 bytes False T0 + δ 12Mbps 600 bytes False Interference estimation NSDI 2011

38 Vivek Shrivastava Estimating interference passively T0 6Mbps 1400 bytes False T0 6Mbps 1400 bytes False T0 + δ 12Mbps 600 bytes False T0 + δ 12Mbps 600 bytes False NSDI 2011

39 Vivek Shrivastava Estimating interference passively T0 6Mbps 1400 bytes False T0 6Mbps 1400 bytes False T0 + δ 12Mbps 600 bytes False T0 + δ 12Mbps 600 bytes False NSDI 2011

40 Vivek Shrivastava Estimating interference passively T0 6Mbps 1400 bytes False T0 6Mbps 1400 bytes False T0 + δ 12Mbps 600 bytes False T0 + δ 12Mbps 600 bytes False & NSDI 2011

41 Vivek Shrivastava Estimating interference passively T0 6Mbps 1400 bytes False T0 6Mbps 1400 bytes False T0 + δ 12Mbps 600 bytes False T0 + δ 12Mbps 600 bytes False & Red & Green clients interfere NSDI 2011

42 Estimating interference passively Sniffer reports Infer interference Scenarios Reception X X X X Vivek Shrivastava NSDI 2011

43 Estimating interference passively Sniffer reports Infer interference Scenarios Reception X X X X Red and Green packets overlaps => both lost Vivek Shrivastava

44 Estimating interference passively Sniffer reports Infer interference Scenarios Reception X X X X No overlap, no problem ! Vivek Shrivastava NSDI 2011

Estimating interference passively Sniffer reports Infer interference Scenarios Reception X X X X Both way hidden terminals Vivek Shrivastava 45 NSDI 2011

46 Estimating interference passively Sniffer reports Infer interference Scenarios Reception X X Vivek Shrivastava NSDI 2011

47 Estimating interference passively Sniffer reports Infer interference Scenarios Reception X X Red and Green packets overlaps => Green is lost Vivek Shrivastava

48 Estimating interference passively Sniffer reports Infer interference Scenarios Reception X X One way hidden terminals Vivek Shrivastava NSDI 2011

49 Computing interference measure in PIE Compute Isolation loss rate Fraction of non-overlapping packets lost Compute Interference loss rate Fraction of overlapping packets lost Interference measure (LIR): (1 – Interference loss) / (1 – Isolation loss ) 49 Vivek Shrivastava NSDI 2011

50 How quickly can PIE converge ? Time taken by PIE to converge depends on two key properties Periodicity with which sniffer reports are collected by the controller Traffic patterns for the links which dictate the number of interference events captured in a time interval 50 Vivek Shrivastava NSDI 2011

51 How quickly can PIE converge ? Time taken by PIE to converge depends on two key properties Periodicity with which sniffer reports are collected by the controller What is the minimum polling period ? Traffic patterns for the links which dictate the number of interference events captured in a time interval How much time does PIE take under realistic access patterns ? 51 Vivek Shrivastava NSDI 2011

52 PIE Outline Motivation Conventional bandwidth tests not sufficient Passive Interference Estimation (PIE) Polling period of PIE Accuracy of PIE Realistic trace replay with PIE Applications of PIE Summary 52 Vivek Shrivastava NSDI 2011

53 Vivek Shrivastava What is the minimum polling period ? Time I1I1 I2I2 time interval PPP NSDI 2011

54 Vivek Shrivastava Time I1I1 I2I2 time interval PPP R(I 1 ) NSDI 2011 What is the minimum polling period ?

55 Vivek Shrivastava What is the minimum polling period ? Time I1I1 I2I2 time interval PPP LIR (I 1 ) R(I 1 ) NSDI 2011

56 Vivek Shrivastava What is the minimum polling period ? Time I1I1 I2I2 time interval PPP R(I 2 ) LIR (I 1 ) NSDI 2011

57 Vivek Shrivastava What is the minimum polling period ? Time I1I1 I2I2 time interval PPP LIR (I 2 ). R(I 2 ) LIR (I 1 ) NSDI 2011

58 What is the minimum polling period ? 58 Vivek Shrivastava LIR Polling period (ms) Stability of interference measure for saturated traffic NSDI 2011

59 What is the minimum polling period ? 59 Vivek Shrivastava Measure stabilizes after ~85 ms (at least 20 overlap samples) LIR Polling period (ms) Stability of interference measure per polling period NSDI 2011

60 What is the minimum polling period ? 60 Vivek Shrivastava LIR Polling period (ms) Stability of interference measure per polling period NSDI 2011 We use a polling period of 100ms

61 PIE Outline Motivation Conventional bandwidth tests not sufficient Passive Interference Estimation (PIE) Polling period of PIE Accuracy of PIE Realistic trace replay with PIE Applications of PIE Summary 61 Vivek Shrivastava NSDI 2011

62 How accurate is PIE ? 62 Vivek Shrivastava % of link-interferer pairs Mean Error in LIR estimation

63 How accurate is PIE ? 63 Vivek Shrivastava % of link-interferer pairs Mean Error in LIR estimation

64 How accurate is PIE ? 64 Vivek Shrivastava % of link-interferer pairs Mean Error in LIR estimation 95% of link-interferer pairs, LIR computed by PIE is within +/- 0.1 of the value reported by BW test

65 PIE Outline Motivation Conventional bandwidth tests not sufficient Passive Interference Estimation (PIE) Polling period of PIE Accuracy of PIE Realistic trace replay with PIE Applications of PIE Summary 65 Vivek Shrivastava NSDI 2011

66 Vivek Shrivastava PIE with realistic access patterns NSDI 2011

67 Vivek Shrivastava NSDI 2011 PIE with realistic access patterns Evaluate PIE using realist traffic patterns on a 15 node topology (7 AP – 8 laptops) Each client laptop replays the traffic patterns of an actual client from a real wireless trace Three activity periods: heavy (> 40 % medium busy), medium (40 – 20% busy), light (< 20% busy)

68 PIE with realistic access patterns 68 Vivek Shrivastava Time to estimate (ms) NSDI 2011 Traffic period

69 PIE with realistic access patterns 69 Vivek Shrivastava Time to estimate (ms) NSDI 2011 Traffic period Convergence is faster for higher client activity Even for light activity, median time of estimate LIR is less than 650 ms

70 PIE Outline Motivation Conventional bandwidth tests not sufficient Passive Interference Estimation (PIE) Polling period of PIE Accuracy of PIE Realistic trace replay with PIE Applications of PIE Summary 70 Vivek Shrivastava NSDI 2011

71 Vivek Shrivastava What is the impact on WLAN applications? NSDI 2011 AP-Client pairs

72 Vivek Shrivastava What is the impact on WLAN applications? NSDI 2011 AP-Client pairs Evaluate usefulness of PIE for an interference mitigation mechanism (data scheduling using CENTAUR – Mobicom ‘09)

73 Vivek Shrivastava What is the impact on WLAN applications? NSDI 2011 AP-Client pairs 1.Estimate interference using PIE

74 Vivek Shrivastava What is the impact on WLAN applications? NSDI 2011 AP-Client pairs 1.Estimate interference using PIE 2.Input estimate to a centralized data scheduler

75 Vivek Shrivastava What is the impact on WLAN applications? NSDI 2011 AP-Client pairs 1.Estimate interference using PIE 2.Input estimate to a centralized data scheduler 3.Evaluate performance under dynamic scenarios

76 What is the impact on end users ? 76 Vivek Shrivastava System Throughput (Mbps) Static scenario NSDI 2011

77 What is the impact on end users ? 77 Vivek Shrivastava System Throughput (Mbps) Static scenario NSDI 2011

78 What is the impact on end users ? 78 Vivek Shrivastava System Throughput (Mbps) Static scenario Static scenarios, PIE is comparable to BW test NSDI 2011

79 What is the impact on end users ? 79 Vivek Shrivastava System Throughput (Mbps) Mobile scenario NSDI 2011

80 What is the impact on end users ? 80 Vivek Shrivastava System Throughput (Mbps) Mobile scenario NSDI 2011

81 What is the impact on end users ? 81 Vivek Shrivastava System Throughput (Mbps) Mobile scenario Mobility scenarios, PIE outperforms BW test NSDI 2011

82 What is the impact on end users ? PIE can also be used to monitor production systems (like Jigsaw) We monitored two production WLANs Use testbed nodes in proximity of production APs as sniffers Identify hidden terminals and rate anomaly problems 82 Vivek Shrivastava NSDI 2011

83 What is the impact on end users ? 83 Vivek Shrivastava NSDI 2011 WLANs WLAN1 WLAN2 Hidden terminal cases (LIR < 0.7) Rate anomaly cases (Ratio of rates < 0.2) 8% 11% 21% 22%

84 What is the impact on end users ? 84 Vivek Shrivastava NSDI 2011 WLANs WLAN1 WLAN2 Hidden terminal cases (LIR < 0.7) Rate anomaly cases (Ratio of rates < 0.2) 8% 11% 21% 22% Hidden terminals are rare, but can become pain points for clients Rate anomaly is more frequent, but do not cause drastic performance issues

85 PIE Outline Motivation Conventional bandwidth tests not sufficient Passive Interference Estimation (PIE) Polling period of PIE Accuracy of PIE Realistic trace replay with PIE Applications of PIE Summary 85 Vivek Shrivastava NSDI 2011

86 Related Work PIE leverages techniques from Jigsaw, WIT (Sigcomm 2006) and builds on their ideas Focus of Jigsaw, WIT was to understand interference, ours is to compute it in real-time CMAP also infers interference to harness exposed terminals, but requires physical layer change Active techniques like Microprobing (CoNext 2008) still require downtime and do not use realistic traffic 86 Vivek Shrivastava NSDI 2011

87 PIE Limitations Does not handle non-WiFi interferer like microwaves. Can miss external interferers if none of the enterprise APs can listen to the interferer May miss client conflicts, can use client participation in PIE to enhance the system Interference detection techniques at the physical layer may be more accurate in some scenarios where diversity is too low for PIE to function 87 Vivek Shrivastava NSDI 2011

88 PIE Summary Online interference estimation important for interference mitigation BW test incurs high overhead, requires downtime PIE is a passive mechanism, generates interference estimates in real time Leverages centralized infrastructure to collect real time reports from APs Non-intrusive with good accuracy 88 Vivek Shrivastava NSDI 2011

Vivek Shrivastava 89 Thank you ! NSDI 2011

90 Is there sufficient diversity ? 90 Vivek Shrivastava % Overlap in transmit time % of transmit pairs NSDI 2011

91 Is there sufficient diversity ? 91 Vivek Shrivastava % Overlap in transmit time % of transmit pairs Overlap in transmit times for transmitter pairs in real WLAN traces NSDI 2011

92 Is there sufficient diversity ? 92 Vivek Shrivastava % Overlap in transmit time % of transmit pairs 80% of transmit pairs have less then 10% overlap Overlap in transmit times for transmitter pairs in real WLAN traces NSDI 2011

93 How quick is PIE ? 93 Vivek Shrivastava Convergence time of PIE under varying loads Traffic load on link and interferer Convergence time NSDI 2011

94 How quick is PIE ? 94 Vivek Shrivastava Convergence time of PIE under varying loads Traffic load on link and interferer Convergence time NSDI 2011

95 How quick is PIE ? 95 Vivek Shrivastava Convergence time of PIE under varying loads Traffic load on link and interferer Convergence time Converges within 650 ms for light traffic loads as well NSDI 2011

96 Synchronization error in PIE ? 96 Vivek Shrivastava Synchronization error using TSF beacon synchronization NSDI 2011

97 Vivek Shrivastava How accurate is PIE ? Interferer AP-Client pair Non Interferer 1.Activate link, interferer and non-interferer NSDI 2011

98 Vivek Shrivastava How accurate is PIE ? Interferer AP-Client pair Non Interferer 1.Activate link, interferer and non-interferer 2.Measure LIR for interferer and non-interferer NSDI 2011

99 Vivek Shrivastava What is the impact on WLAN applications? NSDI 2011 AP-Client pairs

100 Vivek Shrivastava What is the impact on end users ? AP-Client pairs NSDI 2011

101 Vivek Shrivastava What is the impact on end users ? AP-Client pairs Evaluate PIE on a 7 AP - 8 Client topology NSDI 2011

102 Vivek Shrivastava What is the impact on end users ? AP-Client pairs Evaluate usefulness of PIE for an interference mitigation mechanism NSDI 2011

103 Vivek Shrivastava What is the impact on end users ? AP-Client pairs 1.Estimate interference using PIE NSDI 2011

104 Vivek Shrivastava What is the impact on end users ? AP-Client pairs 1.Estimate interference using PIE 2.Input estimate to a centralized data scheduler NSDI 2011

105 Vivek Shrivastava What is the impact on end users ? AP-Client pairs 1.Estimate interference using PIE 2.Input estimate to a centralized data scheduler 3.Evaluate performance under dynamic scenarios NSDI 2011

106 Vivek Shrivastava Carrier sensing (conservative) Problems with wireless Exposed terminals NSDI 2011

107 Vivek Shrivastava Interference management in WLANs Manage Interference (data scheduling, transmit power control, channel assignment) NSDI Estimate Interference

108 Vivek Shrivastava Key Challenge Q. Is there a way to dynamically identify the precise set of nodes in an enterprise WLAN that interfere with each other and the degree to which they do so ? NSDI 2011

109 PIE Outline Motivation Conventional bandwidth tests not sufficient Passive Interference Estimation (PIE) Accuracy of PIE Convergence of PIE Agility of PIE Applications of PIE (data scheduling) Summary 109 Vivek Shrivastava NSDI 2011

110 Vivek Shrivastava How agile is PIE ? Client moves from AP towards interferer Interferer AP-Client pair NSDI 2011

111 Vivek Shrivastava How agile is PIE ? Client moves from AP towards interferer Interferer AP-Client pair NSDI 2011

112 Vivek Shrivastava How agile is PIE ? Client moves from AP towards interferer Interferer AP-Client pair NSDI 2011

113 How agile is PIE ? 113 Vivek Shrivastava Interferer AP Tracking client mobility. SNR SNR (dB) NSDI 2011

114 How agile is PIE ? 114 Vivek Shrivastava Interferer AP Tracking client mobility. SNR Tput SNR (dB) Tput (Mbps) NSDI 2011

115 How agile is PIE ? 115 Vivek Shrivastava Tracking client mobility. SNR Tput LIR SNR (dB) Tput (Mbps) LIR (PIE) NSDI 2011

116 Vivek Shrivastava % Overlap in transmit timeLIR NSDI 2011 Can PIE classify interferers accurately ?

117 Can PIE classify interferers accurately ? 117 Vivek Shrivastava % Overlap in transmit timeLIR NSDI 2011

118 Can PIE classify interferers accurately ? 118 Vivek Shrivastava % Overlap in transmit timeLIR When overlap is less than 80%, PIE is able to classify accurately NSDI 2011

119 Convergence of PIE Test stability of interference measure with different polling periods Polling period of 85 ms is sufficient for stable interference measure in presence of saturated traffic Test stability of interference measure with different traffic rates Higher convergence times for low traffic rates, as number of events per polling period is low Still converges within 600ms for very lightly loaded links (when load is 2-3% of link capacity) 119 Vivek Shrivastava NSDI 2011

120 Estimating carrier sense passively Sniffer reports Infer carrier sense Vivek Shrivastava NSDI 2011

121 Estimating carrier sense passively Sniffer reports Infer carrier sense Vivek Shrivastava NSDI 2011 Determine CS by observing direction of packet overlaps for contending transmitters

122 Estimating carrier sense passively Sniffer reports Infer carrier sense Scenarios Vivek Shrivastava NSDI 2011 Observe packet overlaps in both directions

123 Estimating carrier sense passively Sniffer reports Infer carrier sense Scenarios Vivek Shrivastava NSDI 2011 Both Red and Green APs do not sense each other Observe packet overlaps in both directions

124 Estimating carrier sense passively Sniffer reports Infer carrier sense Scenarios Vivek Shrivastava NSDI 2011 Red senses Green, but not vice versa Observe packet overlaps in only one direction

125 Estimating carrier sense passively Sniffer reports Infer carrier sense Scenarios Vivek Shrivastava NSDI 2011 Green senses Red, but not vice versa Observe packet overlaps in only one direction

126 Estimating carrier sense passively Sniffer reports Infer carrier sense Scenarios Vivek Shrivastava NSDI 2011 No contending packets overlap Green and Red sense each other

127 Computing carrier sense measure in PIE Compute overlap probability for contending packets ( P overlap ) If (1 – P overlap ) > δ t, report no carrier sensing else if P overlap > δ t If P one-way > δ t, report one-way CS else report mutual carrier sensing 127 Vivek Shrivastava NSDI 2011

128 Vivek Shrivastava How accurate is PIE ? Interferer AP-Client pair NSDI 2011

129 Vivek Shrivastava How accurate is PIE ? Interferer AP-Client pair 1.Activate both link and interferer NSDI 2011

130 Vivek Shrivastava How accurate is PIE ? Interferer AP-Client pair 1. Activate both link and interferer 2. Measure LIR using PIE and BW test

131 How accurate is PIE with multiple interferers ? Scenarios Reception X X Vivek Shrivastava Red and Green packets overlap => Green is lost One way hidden terminals NSDI 2011

132 How accurate is PIE with multiple interferers ? Scenarios Reception Vivek Shrivastava Blue and Green packets overlap => None is lost No interference NSDI 2011

133 Is there sufficient diversity ? 133 Vivek Shrivastava % Overlap in transmit time % of transmit pairs NSDI 2011

134 Is there sufficient diversity ? 134 Vivek Shrivastava % Overlap in transmit time % of transmit pairs Overlap in transmit times for transmitter pairs in real WLAN traces

NSDI Is there sufficient diversity ? 135 Vivek Shrivastava % Overlap in transmit time % of transmit pairs 80% of transmit pairs have less then 5% overlap Overlap in transmit times for transmitter pairs in real WLAN traces

136 What about multiple interferers? X Vivek Shrivastava X Ambiguous of loss source (Red/Blue) NSDI 2011

137 What about multiple interferers? X Vivek Shrivastava X Overlap with Blue =>No Loss NSDI 2011

138 What about multiple interferers? X Vivek Shrivastava X Overlap with Red => Loss NSDI 2011

139 What about multiple interferers? X Vivek Shrivastava X Isolation =>No Loss NSDI 2011

140 What about multiple interferers? X Vivek Shrivastava X Isolation =>No Loss Overlap with Blue =>No Loss NSDI 2011

141 What about multiple interferers? X Vivek Shrivastava X Isolation =>No Loss Overlap with Blue =>No Loss Overlap with Red => Loss NSDI 2011

142 What about multiple interferers? X Vivek Shrivastava X PIE needs transmission diversity to identify interferers accurately NSDI 2011

143 PIE with realistic access patterns 143 Vivek Shrivastava % of link - interferer pairs NSDI 2011 Convergence time (ms)

144 PIE with realistic access patterns 144 Vivek Shrivastava % of link - interferer pairs NSDI 2011 Convergence time (ms)

145 PIE with realistic access patterns 145 Vivek Shrivastava % of link - interferer pairs NSDI 2011 Convergence time (ms) Convergence is faster for higher client activity

146 PIE with realistic access patterns 146 Vivek Shrivastava % of link - interferer pairs NSDI 2011 Convergence time (ms) Convergence is faster for higher client activity Even for light client activity, 90% of link-interferer pairs converge within ~ 10 seconds

147 Synchronization error 147 Vivek Shrivastava Synchronization error for a 19 node topology NSDI 2011 CDF Synchronization error (microsec)

148 Synchronization error 148 Vivek Shrivastava Synchronization error as a function of beacon interval NSDI 2011 Clock skew (us) Time (ms)

149