on the fidelity of 802 11 packet traces
play

On The Fidelity of 802.11 Packet Traces Aaron Schulman, Dave Levin, - PowerPoint PPT Presentation

Jun 08, 2023 •468 likes •1.26k views

On The Fidelity of 802.11 Packet Traces Aaron Schulman, Dave Levin, Neil Spring University of Maryland, College Park PAM - April 2008 On The Fidelity of 802.11 Packet Traces 1 Uses of 802.11 packet traces MAC Layer (Mahajan et al, Jardosh

  1. On The Fidelity of 802.11 Packet Traces Aaron Schulman, Dave Levin, Neil Spring University of Maryland, College Park PAM - April 2008 On The Fidelity of 802.11 Packet Traces 1

  2. Uses of 802.11 packet traces • MAC Layer (Mahajan et al, Jardosh et al) • Performance (Rodrig et al) • Troubleshooting (Cheng et al) These studies benefit from complete packet traces PAM - April 2008 On The Fidelity of 802.11 Packet Traces 2

  3. What is an incomplete trace? Transmissions are within range of the monitor but packets are missing AP Monitor Client 1 Client 2 PAM - April 2008 On The Fidelity of 802.11 Packet Traces 3

  4. What is an incomplete trace? Transmissions are within range of the monitor but packets are missing AP Monitor Client 1 Client 2 PAM - April 2008 On The Fidelity of 802.11 Packet Traces 3

  5. What is an incomplete trace? Transmissions are within range of the monitor but packets are missing AP Monitor Client 1 Client 2 PAM - April 2008 On The Fidelity of 802.11 Packet Traces 3

  6. Capturing complete 802.11 packet traces is hard • Monitor Hardware/Software • RF Interference • Monitor Placement • Merging requires accurate timestamps (Yeo et al, Portoles-Comeras et al) PAM - April 2008 On The Fidelity of 802.11 Packet Traces 4

  7. Trace Fidelity Completeness Accuracy Did we capture all of Did we timestamp the the packets? packets correctly? PAM - April 2008 On The Fidelity of 802.11 Packet Traces 5

  8. Main finding: Both are dependent on load Trace Fidelity Completeness Accuracy Did we capture all of Did we timestamp the the packets? packets correctly? PAM - April 2008 On The Fidelity of 802.11 Packet Traces 5

  9. Main finding: Both are dependent on load Trace Fidelity Completeness Accuracy Did we capture all of the packets? Did we timestamp the packets correctly? PAM - April 2008 On The Fidelity of 802.11 Packet Traces 5

  10. Monitors can miss packets Monitor AP Client 2 Client 1 PAM - April 2008 On The Fidelity of 802.11 Packet Traces 6

  11. Monitors can miss packets Monitor AP Client 2 Client 1 PAM - April 2008 On The Fidelity of 802.11 Packet Traces 6

  12. Monitors can miss packets Both the Monitor and AP receive a packet from Client 1 Monitor Monitor AP AP Client 2 Client 1 PAM - April 2008 On The Fidelity of 802.11 Packet Traces 6

  13. Monitors can miss packets Both the Monitor and AP receive a packet from Client 1 Monitor AP Client 2 Client 1 PAM - April 2008 On The Fidelity of 802.11 Packet Traces 6

  14. Monitors can miss packets Both the Monitor and AP receive a packet from Client 1 Monitor AP Client 2 Client 1 PAM - April 2008 On The Fidelity of 802.11 Packet Traces 6

  15. Monitors can miss packets Both the Monitor and AP receive a packet from Client 1 The Monitor misses a packet from Client 2 Monitor AP AP Client 2 Client 1 PAM - April 2008 On The Fidelity of 802.11 Packet Traces 6

  16. 802.11 protocol can show completeness Retry 802.11 Header Sequence # Bit Incremented when a Set when a packet is packet is sent a retransmission (Yeo et al) PAM - April 2008 On The Fidelity of 802.11 Packet Traces 7

  17. Estimating completeness Monitor Client AP PAM - April 2008 On The Fidelity of 802.11 Packet Traces 8

  18. Estimating completeness 1 Monitor Client 1 AP PAM - April 2008 On The Fidelity of 802.11 Packet Traces 8

  19. Estimating completeness Missed 1 2 Monitor Client 1 2 AP PAM - April 2008 On The Fidelity of 802.11 Packet Traces 8

  20. Estimating completeness 2 is missing 1 2 3 Monitor Client 1 2 3 AP PAM - April 2008 On The Fidelity of 802.11 Packet Traces 8

  21. Estimating completeness Missed 1 2 3 4 Monitor Client 1 2 3 4 AP Missed PAM - April 2008 On The Fidelity of 802.11 Packet Traces 8

  22. Estimating completeness Retransmitted 1 2 3 4 4 Monitor Client 1 2 3 4 4 AP Retransmitted PAM - April 2008 On The Fidelity of 802.11 Packet Traces 8

  23. Estimating completeness 4 is missing 1 2 3 4 4 Monitor Client 1 2 3 4 4 AP PAM - April 2008 On The Fidelity of 802.11 Packet Traces 8

  24. Estimating completeness The sequence number and retransmission bit show packets 2 and 4 are missing. 1 2 3 4 4 Monitor Client 1 2 3 4 4 AP PAM - April 2008 On The Fidelity of 802.11 Packet Traces 8

  25. How accurate is the estimate? • Start with SIGCOMM ’04 trace CHI • Randomly removed packets from trace • Compute estimated # of packets missing • Relative Error of Method = Estimate - Known Total packets PAM - April 2008 On The Fidelity of 802.11 Packet Traces 9

  26. Accuracy of estimate Relative error Fraction of Non-beacon Packets Removed Error bars show 95% confidence interval PAM - April 2008 On The Fidelity of 802.11 Packet Traces 10

  27. Accuracy of estimate The relative error is < 0.02 when up to 55% of the trace is removed. Relative error Fraction of Non-beacon Packets Removed Error bars show 95% confidence interval PAM - April 2008 On The Fidelity of 802.11 Packet Traces 10

  28. Percentage for trace completeness SIGCOMM 2004 Dataset Rodrig et al PAM - April 2008 On The Fidelity of 802.11 Packet Traces 11

  29. Percentage for trace completeness 81% Using the estimate the trace has of the packets sent by the AP SIGCOMM 2004 Dataset Rodrig et al PAM - April 2008 On The Fidelity of 802.11 Packet Traces 11

  30. Percentage for trace completeness 81% Using the estimate the trace has of the packets sent by the AP 37% of the AP’s packets were beacon packets sent when the network was idle SIGCOMM 2004 Dataset Rodrig et al PAM - April 2008 On The Fidelity of 802.11 Packet Traces 11

  31. Percentage for trace completeness 81% Using the estimate the trace has of the packets sent by the AP 37% of the AP’s packets were beacon packets sent when the network was idle 70% Excluding idle beacon packets of packets sent by the AP are in the trace SIGCOMM 2004 Dataset Rodrig et al PAM - April 2008 On The Fidelity of 802.11 Packet Traces 11

  32. One number is not enough • Problem: Completeness is only interesting when the network is under load - Example: Capturing a trace from an AP overnight • Solution: Estimate completeness within small trace intervals - Beacons are sent by AP every 100ms PAM - April 2008 On The Fidelity of 802.11 Packet Traces 12

  33. Trace completeness score Packets collected i Packets expected i PAM - April 2008 On The Fidelity of 802.11 Packet Traces 13

  34. Trace completeness score For all devices in-range Packets collected i Packets expected i PAM - April 2008 On The Fidelity of 802.11 Packet Traces 13

  35. Trace completeness score For all devices in-range Packets i Packets expected i PAM - April 2008 On The Fidelity of 802.11 Packet Traces 13

  36. Trace completeness score For all devices in-range Packets i Sequence Change i + Retransmissions i PAM - April 2008 On The Fidelity of 802.11 Packet Traces 13

  37. Trace completeness score For all devices in-range Packets i Sequence Change i + Retransmissions i Quantifies the completeness of interval i PAM - April 2008 On The Fidelity of 802.11 Packet Traces 13

  38. Visualizing trace completeness Portland ug 100000 1 10000 0.8 1000 Score 0.6 100 0.4 10 0.2 0 1 0 10 20 30 40 50 Load (change in sequence number) PAM - April 2008 On The Fidelity of 802.11 Packet Traces 14

  39. Visualizing trace completeness Portland ug • Y-Axis: Score 100000 1 • Completeness of 10000 0.8 an Interval 1000 Score 0.6 100 0.4 10 0.2 0 1 0 10 20 30 40 50 Load (change in sequence number) PAM - April 2008 On The Fidelity of 802.11 Packet Traces 14

  40. Visualizing trace completeness Portland ug • Y-Axis: Score 100000 1 • Completeness of 10000 0.8 an Interval 1000 Score 0.6 • X-Axis: Load 100 0.4 • Sequence # change 10 0.2 0 1 0 10 20 30 40 50 Load (change in sequence number) PAM - April 2008 On The Fidelity of 802.11 Packet Traces 14

  41. Visualizing trace completeness Portland ug • Y-Axis: Score 100000 1 • Completeness of 10000 0.8 an Interval 1000 Score 0.6 • X-Axis: Load 100 0.4 • Sequence # change 10 0.2 0 1 • Color: Frequency 0 10 20 30 40 50 Load (change in sequence number) PAM - April 2008 On The Fidelity of 802.11 Packet Traces 14

  42. Completeness with T -Fi plot Portland ug 100000 1 10000 0.8 1000 Score 0.6 100 0.4 10 0.2 0 1 0 10 20 30 40 50 Load (change in sequence number) PAM - April 2008 On The Fidelity of 802.11 Packet Traces 15

  43. Completeness with T -Fi plot • Complete loaded Portland ug 100000 trace has dark area 1 10000 on top 0.8 1000 Score 0.6 100 0.4 10 0.2 0 1 0 10 20 30 40 50 Load (change in sequence number) PAM - April 2008 On The Fidelity of 802.11 Packet Traces 15

  44. Completeness with T -Fi plot • Complete loaded Portland ug 100000 trace has dark area 1 10000 on top 0.8 1000 • Incomplete trace has Score 0.6 100 lower dark areas 0.4 10 0.2 0 1 0 10 20 30 40 50 Load (change in sequence number) PAM - April 2008 On The Fidelity of 802.11 Packet Traces 15

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend

A Method to Compress and Anonymize Packet Traces Markus Peuhkuri 2001-11-02 Abstract Data

A Method to Compress and Anonymize Packet Traces Markus Peuhkuri 2001-11-02 Abstract Data

A Method to Compress and Anonymize Packet Traces Markus Peuhkuri 2001-11-02 Abstract Data volume and privacy issues are one of problems related to large-scale packet capture. Utilizing flow nature of

375 views • 3 slides
Eligibility Traces Chapter 12 Eligibility traces are Another way of interpolating between MC and

Eligibility Traces Chapter 12 Eligibility traces are Another way of interpolating between MC and

Eligibility Traces Chapter 12 Eligibility traces are Another way of interpolating between MC and TD methods A way of implementing compound -return targets A basic mechanistic idea a short-term, fading memory A new style of algorithm

881 views • 31 slides
Capture, conversion, and analysis of an intense NFS workload uating newer storage systems because

Capture, conversion, and analysis of an intense NFS workload uating newer storage systems because

the analysis or simulation. about 750 million operations per day. In comparison, the vironment than prior traces, we limit our comparisons Since our traces were captured in such a different en- convert, and analyze the traces. quired us to

366 views • 14 slides
Modeling interaction traces of an online panel From raw interaction traces to actionable

Modeling interaction traces of an online panel From raw interaction traces to actionable

Modeling interaction traces of an online panel From raw interaction traces to actionable indicators: lessons learned 8th European Survey Research Association conference July 2019, Zagreb Simon DELLAC Elodie PETORIN (both CDSP, Sciences Po)

435 views • 18 slides
Introduction to Packet Tracer What is Packet Tracer? Packet Tracer is a protocol simulator

Introduction to Packet Tracer What is Packet Tracer? Packet Tracer is a protocol simulator

Introduction to Packet Tracer What is Packet Tracer? Packet Tracer is a protocol simulator developed by Dennis Frezzo and his team at Cisco Systems. Packet Tracer (PT) is a powerful and dynamic tool that displays the various protocols used in

419 views • 17 slides
ACHIEVING HIGH FIDELITY ASSERTIVE COMMUNITY TREATMENT THROUGH THE IMPLEMENTATION OF FIDELITY

ACHIEVING HIGH FIDELITY ASSERTIVE COMMUNITY TREATMENT THROUGH THE IMPLEMENTATION OF FIDELITY

ACHIEVING HIGH FIDELITY ASSERTIVE COMMUNITY TREATMENT THROUGH THE IMPLEMENTATION OF FIDELITY EVALUATION AND TECHNICAL ASSISTANCE Lorna Moser, Ph.D. ACT TA Center Center for Excellence in Community Mental Health, UNC Department of Psychiatry

691 views • 46 slides
Traces Exist (Hypothetically)! Carl Pollard Structure and Evidence in Linguistics Workshop in

Traces Exist (Hypothetically)! Carl Pollard Structure and Evidence in Linguistics Workshop in

Traces Exist (Hypothetically)! Carl Pollard Structure and Evidence in Linguistics Workshop in Honor of Ivan Sag Stanford University April 28, 2013 Carl Pollard Traces Exist (Hypothetically)! Traces in Transformational Grammar (1/2) Traces

411 views • 28 slides
Tier 2 Fidelity Data: Strengthening your Tier 2 PBIS Implementation: Using Fidelity Measures to

Tier 2 Fidelity Data: Strengthening your Tier 2 PBIS Implementation: Using Fidelity Measures to

Tier 2 Fidelity Data: Strengthening your Tier 2 PBIS Implementation: Using Fidelity Measures to Promote Growth in Tier 2 Implementation March 2020 Celeste Rossetto Dickey, M.Ed. University of Oregon Introductions and Welcome School Staff

865 views • 55 slides
Network Measurement Carey Williamson Department of Computer Science University of Calgary

Network Measurement Carey Williamson Department of Computer Science University of Calgary

CPSC 641: Network Measurement Carey Williamson Department of Computer Science University of Calgary Network Traffic Measurement A focus of networking research for 30+ years Collect data or packet traces showing packet activity on the

462 views • 25 slides
Looking at DNS traces: What do we know about resolvers? lafur Gu mundsson Shinkuro

Looking at DNS traces: What do we know about resolvers? lafur Gu mundsson Shinkuro

Looking at DNS traces: What do we know about resolvers? lafur Gu mundsson Shinkuro Ogud@shinkuro.com Motivation: Open Questions What is the market share for resolver X What can we deduct from traces to a subset of authorative

623 views • 19 slides
Two 2-traces Simon Willerton University of Sheffield f Tr ( f ) := V

Two 2-traces Simon Willerton University of Sheffield f Tr ( f ) := V

Two 2-traces Simon Willerton University of Sheffield f Tr ( f ) := V Tr ( f ) := V f Traces What is a trace? Tr ( f g ) = Tr ( g f ) Tr ( f ) = Tr ( a f a 1 ) Traces in a monoidal

651 views • 13 slides
Fidelity Bank Presentation Non-Deal Roadshow May Jun 2018 www.fidelitybank.ng Profile of

Fidelity Bank Presentation Non-Deal Roadshow May Jun 2018 www.fidelitybank.ng Profile of

Fidelity Bank Presentation Non-Deal Roadshow May Jun 2018 www.fidelitybank.ng Profile of Fidelity Bank Road-Show Team Nnamdi Okonkwo Managing Director / Chief Executive Officer Appointed Managing Director of Fidelity Bank PLC in January

601 views • 38 slides
Packet Radio Lee Maddox, N4HOK What is Packet Radio? Packet radio is the connection of a computer

Packet Radio Lee Maddox, N4HOK What is Packet Radio? Packet radio is the connection of a computer

03/11/2017 Packet Radio Lee Maddox, N4HOK What is Packet Radio? Packet radio is the connection of a computer to a radio for the transmission of digital data via amateur radio. It is another mode of operation that has a multitude of uses, such as

571 views • 17 slides
How to (passively) measure? Packet Monitoring 1 What to expect? Overview / What is packet

How to (passively) measure? Packet Monitoring 1 What to expect? Overview / What is packet

How to (passively) measure? Packet Monitoring 1 What to expect? Overview / What is packet monitor? How to acquire the data Handling performance bottlenecks Case Study: Packet Capture Performance Analyzing the transport and

487 views • 37 slides
1 Validating Replica Streams

1 Validating Replica Streams

Link Utilization Detection and Analysis of Routing Loops Internet backbone link in Packet Traces IMW 2002 Routing loop causes Urs Hengartner, Sue Moon, traffic increase by 25%! Richard Mortier, Christophe Diot 1 2 Advanced

352 views • 3 slides
LN-10 Searching for traces of Norwegian economic thought Intro When trying to find the traces of

LN-10 Searching for traces of Norwegian economic thought Intro When trying to find the traces of

LN-10 Searching for traces of Norwegian economic thought Intro When trying to find the traces of Norwegian economic thought, we should be aware that most economic questions until very recently were dealt with within a framework of religious,

409 views • 39 slides
David W. Butler High School 2019-2020 Course Registration General Information Students should

David W. Butler High School 2019-2020 Course Registration General Information Students should

David W. Butler High School 2019-2020 Course Registration General Information Students should have the following: Course Selection Card Transcript of your grades Graduation Requirements Chart Career and Technical Education

461 views • 20 slides
PILOT WIRELESS NETWORK PILOT WIRELESS NETWORK FOR ACCESS TO THE INTERNET FOR ACCESS TO THE

PILOT WIRELESS NETWORK PILOT WIRELESS NETWORK FOR ACCESS TO THE INTERNET FOR ACCESS TO THE

PILOT WIRELESS NETWORK PILOT WIRELESS NETWORK FOR ACCESS TO THE INTERNET FOR ACCESS TO THE INTERNET IN IN POZNAN POZNAN Tadeusz Szkudlarz Tadeusz Szkudlarz - - The The City of City of Poznan Poznan tadeusz_szkudlarz@um.poznan.pl

311 views • 17 slides
Lectures 34: Consumer Theory Alexander Wolitzky MIT 14.121 1 Consumer Theory Consumer theory

Lectures 34: Consumer Theory Alexander Wolitzky MIT 14.121 1 Consumer Theory Consumer theory

Lectures 34: Consumer Theory Alexander Wolitzky MIT 14.121 1 Consumer Theory Consumer theory studies how rational consumer chooses what bundle of goods to consume. Special case of general theory of choice. Key new assumption: choice sets defined

1.04k views • 42 slides
MA/CSSE 473 Day 08 Randomized Primality Testing Carmichael Numbers Miller-Rabin test MA/CSSE

MA/CSSE 473 Day 08 Randomized Primality Testing Carmichael Numbers Miller-Rabin test MA/CSSE

MA/CSSE 473 Day 08 Randomized Primality Testing Carmichael Numbers Miller-Rabin test MA/CSSE 473 Day 08 Student questions Fermat's Little Theorem Implications of Fermats Little Theorem What we can show and what we cant

211 views • 7 slides
6.2 Bump Mapping &amp; Clipping Hao Li http://cs420.hao-li.com 1 Bump Mapping 2 A long time

6.2 Bump Mapping &amp; Clipping Hao Li http://cs420.hao-li.com 1 Bump Mapping 2 A long time

Fall 2018 CSCI 420: Computer Graphics 6.2 Bump Mapping &amp; Clipping Hao Li http://cs420.hao-li.com 1 Bump Mapping 2 A long time ago, in 1978 3 bump mapping was born courtesy by ZBrush 4 For Meshes vertex normal interpolation

738 views • 54 slides
Inference Suppose you are given a Bayesian network with the graph structure and the parameters

Inference Suppose you are given a Bayesian network with the graph structure and the parameters

Inference Suppose you are given a Bayesian network with the graph structure and the parameters CS 331: Artificial Intelligence all figured out Bayesian Networks (Inference) Now you would like to use it to do inference You need

417 views • 5 slides
e : Evolving Databases for New-Gen Big Data Applications R. Barber, C. Garcia-Arellano, R.

e : Evolving Databases for New-Gen Big Data Applications R. Barber, C. Garcia-Arellano, R.

Wildfire e : Evolving Databases for New-Gen Big Data Applications R. Barber, C. Garcia-Arellano, R. Grosman, R. Mueller, V. Raman , R. Sidle, M. Spilchen, A. Storm, Y. Tian, P. Tozun, D. Zilio, M. Huras, G. Lohman, C. Mohan, F. Ozcan, H.

286 views • 16 slides
Welcome! We'll be getting started promptly at the top of the hour, and until then you will

Welcome! We'll be getting started promptly at the top of the hour, and until then you will

Welcome! We'll be getting started promptly at the top of the hour, and until then you will NOT be hearing any audio (aside from a sound check). In the meantime, feel free to reach out via the Chat/Questions panel. @fdncenter

626 views • 40 slides

More recommend