WiSlow: A WiFi Network Performance Troubleshooting Tool for End
Contents
1. No interference e Baby monitor Microwave oven 3 Packet loss 0 10 20 30 40 50 Available bit rates Mbit s a Different interference sources Baby monitor ex1 Baby monitor ex2 Packet loss 0 10 20 30 40 50 Available bit rates MBit s b The same device a baby monitor in different environ ments Fig 4 The distribution of the correlation of bit rates and the estimated packet loss phones that send audio only thus causing more interference Channel contention shows less packet loss because of the 802 11 collision avoidance functions such as random back off and RTS CTS that force each client to occupy the medium in separate time slots In this case the degradation of throughput is caused by the shared medium rather than noise from other sources Furthermore we found that the correlation between bit rate and the estimated number of packet losses shows clearer differences among various problem sources In Figure 4a the majority of the samples from a clean environment are distributed in a healthy zone higher bit rate and lower packet loss while the samples of baby monitors and microwave ovens are widely dispersed WiSlow uses the correlation of these two variables to distinguish the level of interference As described above the problem sources each have their own distribution patterns on the scatter plot However an end user cannot infer a root cause by simply matching the me2. WiSlow examines which representative CDF is the most similar one to the CDF measured on the user s machine To compare the CDFs WiSlow uses the two sample Kolmogorov Smirnov test K S test a widely used statistical method that tests whether two empirical CDFs obtained from separate experiments have the same distribution 11 If the p value of this test is close to 1 the two CDFs are likely to come from the same distribution however if the p value is close to 0 they are likely to come from different distributions Since the K S test not only considers the average and variance of the samples but also takes into account the shape of the CDFs it best fits the purpose of WiSlow where it is used to pick the most similar distribution from multiple data sets Our evaluation proves that IEEE INFOCOM 2014 IEEE Conference on Computer Communications Group1 e Group2 Group3 CDF 5 10 15 Euclidean distance d bit rate packet loss Fig 6 Three groups categorized by the packet loss analysis 1 a no interference environment 2 contention and FHSS cordless phones and 3 microwave ovens and baby monitors The number of 802 11 ACKs 1 45 1 455 1 46 1 465 Time milliseconds x 10 a Time domain Magnitude pos 100 200 300 700 500 Frequency Hz b Frequency domain Fig 7 The number of 802 11 ACKs with interference of a microwave oven the approach explained above successfully distinguishes these g
3. rate type I error of WiSlow for each problem source The diagnostic accuracy of a problem source P is the ratio of the number of correct diagnostics to the total number of experiments in which P is injected as a problem source The false positive rate of P is the ratio of the number of cases that the cause is misidentified as P to the total number of experiments in which P is not actually the cause 868 Injected Problem Distance Accuracy False from the AP Positive o J ooa aa v No interference Channel contention Non Wi Fi 0 0 m interference 0 5 m baby monitor bes 3 9 cordless phone aan 2 0 m and microwave oven TABLE I The accuracy of WiSlow for distinguishing between a clean environment channel contention and non Wi Fi interference Table I shows that WiSlow successfully distinguishes them with high accuracy over 90 for no interference and channel contention In the non Wi Fi interference case the accuracy was also over 90 when the interfering device was close to the AP however it notably decreased when the distance between the AP and the device increased We found that this inaccuracy was mostly caused by the FHSS cordless phones In the following sections we explain the reason for this inaccuracy and the method WiSlow employed to reduce it A Identifying the root cause Table II shows the detailed diagnostic results of identifying each type of non Wi Fi device First WiSlow could cl
4. as UDP and 802 11 packets Because the mechanisms of these protocols are not significantly differ ent for many Wi Fi devices we believe that WiSlow s user level approach can help a wider range of end users C Lack of monitoring data Another restriction in the end user environment is the lack of a monitoring history If we assume that we have been monitoring the machine up to the moment when a performance problem happens the diagnosis will be easier because we can obtain several important clues such as the average quality of the link the time when the problem started and whether it has happened in the recent past However although the overhead of network monitoring is not heavy on modern machines it is difficult to expect that end users will continuously run such a tool The more common scenario is that a user launches a troubleshooting tool like WiSlow and requests a diagnostic only after he she has noticed a severe performance problem Therefore we need to design the tool assuming little or no previous monitoring data In the next section we explain how WiSlow estimates the problem source without knowing the baseline quality of the network IV WISLOW In this section we elaborate on the details of probing methods First to investigate the behavior of Wi Fi networks in each problem scenario we artificially inject problems while transmitting UDP packets between a client laptop and an AP We capture every packet on the client
5. device hops This frequency could be 50 Hz in other countries e g Europe and most of Asia where 50 Hz AC power is used IEEE INFOCOM 2014 IEEE Conference on Computer Communications Magnitude Magnitude 300 400 500 s 16 a2 0 100 200 Frequency Hz Time ms a A baby monitor frequency b A baby monitor time domain domain top 10 frequencies gnitude Magnitude 0 8 16 24 32 Time ms 0 100 200 300 400 500 Frequency Hz c An FHSS cordless phone fre d An FHSS cordless phone quency domain time domain top 10 frequencies Fig 8 The number of 802 11 ACKs per100 KB of UDP packets with a baby monitor and a cordless phone far from the current Wi Fi channel and it is relatively small when it hops to a nearby frequency If the device hops into the exact range of the Wi Fi channel the number of 802 11 ACKs drops almost to zero In other words there are multiple levels of interference which depend on how closely in frequency the device hops to the frequency used by the Wi Fi channel These multiple levels of interference create several pulses that have different magnitudes and frequencies Finally because the hopping interval of the device is fixed the frequencies of the created pulses are synchronized such that the periods of the cycles are multiples of a specific value The FHSS cordless phone which also uses the frequency hopping technique showed a similar result multiple peaks wi
6. is feasible for finding the relative location of an interfering device Although WiSlow shows errors of several meters in pinpointing a lo cation we believe that this level of error is not critical for a home network environment VII RELATED WORK Airshark 5 uses a commodity Wi Fi network adapter to identify the source of interference It leverages a spectral scan to obtain signal information from multiple frequency ranges It identifies the interference sources very accurately over 95 by analyzing the spectrum data using various methods However we suppose that it would be difficult to IEEE INFOCOM 2014 IEEE Conference on Computer Communications apply this approach for typical end users because collecting high resolution signal samples across the spectrum is impos sible if the network card does not support this functionality WiFiNet 6 identifies the impact of non Wi Fi interference and finds its location using observations from multiple APs that are running Airshark However this approach seems difficult to be used in a common home network environment that has a single AP In contrast WiSlow focuses on identifying the location of the interference source by cooperating end users Kanuparthy et al 16 propose an approach similar to WiSlow in terms of using user level information They distin guish congestion channel contention from hidden terminals and low SNR by measuring the one way delay of different packet sizes Th
7. machine and then trace the transport layer UDP the 802 11 MAC layer and some user accessible 802 11 PHY layer information to ascertain each problematic scenario s interference levels and characteristics To capture 802 11 packets WiSlow uses monitor mode of wireless adapters It provides the Radiotap 8 header which is a standard for 802 11 frame information The headers are used to extract the lower layer information such as FCS errors and bit rates Sniffing the wireless packets is supported by most Linux and all Mac OS X machines without additional drivers or kernel modification Therefore if we can successfully characterize each performance degrading source by probing the transmitted packets the same probes will enable WiSlow to identify the problem sources on most platforms However IEEE INFOCOM 2014 IEEE Conference on Computer Communications No interference e Contention FHSS phone Baby monitor Microwave oven No interference Contention u FHSS phone CDF Baby monitor Microwave oven Ra id O 10 20 30 40 50 0 The number of retries per 100 KB a The CDF of number of 802 11 retries 0 20 30 Available bit rates Mbit s b The CDF of available bit rates No interference Contention FHSS phone Baby monitor Microwave oven CDF eo 40 50 0 10 20 30 40 The number of FCS errors p
8. ovens is 50 and the dwell time is 16 6 ms 60 Hz 12 This implies that it stays in the ON mode producing microwaves for the first 8 3ms and the OFF mode for the next 8 3ms This feature can be observed by various means such as using a spectrum analyzer 2 or by signal measurement 5 Our hypothesis was that a user level probe could also detect this on off pattern if the network packets were monitored on a millisecond timescale because the packets would be lost only when the interferer was active on mode To validate this assumption we implemented the above method and plotted the number of successfully received 802 11 ACKs per millisecond As a result a clearly perceptible waveform with a 50 duty cycle is observed Figure 7a the number of ACKs is over five for the first 8ms and zero during the next 8ms This pattern repeats while the microwave oven is running This result becomes clearer when it is converted to the frequency domain Figure 7b using a fast Fourier transform FFT The highest peak is at 60 Hz which means the cycle is 16 6 ms This number is exactly the same as the known duty cycle of microwave ovens Consequently if a perceptible cycle is detected from this probing method and the period matches a well known value WiSlow determines that the current interference is due to a particular type of device e g 60 Hz for microwave ovens 2 Frequency hopping baby monitors and cordless phones The duty cycle of ty
9. scenario of locating interference e020 Interference Location Laptop A Q 192 168 1 135 ay Laptop B 4 al Laptop C 192 168 1 103 Eo21081 181 CS 1 Sag Ww b A real time result of WiSlow Fig 10 Locating the interference source time the location was changed Figure 10b shows an actual real time screenshot of WiSlow detecting the location of the baby monitor For the first location laptops A and B reported no interference but laptop C detected the baby monitor suc cessfully For the second location the three laptops all detected the baby monitor and reported similar interference strengths because the interference source was close to the AP and thus the entire wireless network was affected by the baby monitor In this particular case WiSlow could infer that the problem source was likely to be a device placed near the AP For the third location only laptop B detected the baby monitor and thus WiSlow placed the baby monitor icon close to laptop B For the fourth location the three laptops all detected the baby monitor but the measured interference strengths were distinct Therefore based on Equation 2 WiSlow pointed the location of the baby monitor as being relatively close to laptop B For the last spot since none of the laptops detected any interference only a green check icon was displayed which indicates that the state of the network is good This experiment proves that our approach
10. 5 GHz if customized to an 802 11n environment B Ad Hoc mode and mobile devices We also tested WiSlow on an ad hoc network using two laptops which enables WiSlow to run independently without communicating with an AP Since ad hoc networks also use the same 802 11 protocol we did not see any differences from the experiments with an AP We expect that using WiSlow with ad hoc networks will be especially helpful in independently discovering nearby interference sources when used with multiple mobile devices such as smartphones DISCUSSION AND FUTURE WORK IX CONCLUSION We designed WiSlow a Wi Fi performance trouble shoot ing application specialized to detect non Wi Fi interference WiSlow distinguishes 802 11 channel contention from non Wi Fi interference and identifies the type of interfering devices present WiSlow was designed to exploit user level probing only which enables a software only approach For this pur pose we developed two novel methods that use user accessible packet information such as UDP and 802 11 ACKs 870 The accuracy of WiSlow exceeds 90 when the sources are close to a Wi Fi device WiSlow becomes less accurate when the devices are located farther However this inaccuracy can be removed if we take into account the known characteristics of each device Also we proved that the collaborative approach is feasible for determining the relative location of an interfering device X ACKNOWLEDGMENT This m
11. 978 14799 3360 0 14 31 00 2014 IEEE IEEE INFOCOM 2014 IEEE Conference on Computer Communications WiSlow A Wi Fi Network Performance Troubleshooting Tool for End Users Kyung Hwa Kim Hyunwoo Nam and Henning Schulzrinne Department of Computer Science Columbia University New York NY Department of Electrical Engineering Columbia University New York NY Abstract Slow Internet connectivity is often caused by poor Wi Fi performance The main reasons of such performance degra dation include channel contention and non Wi Fi interference Although these problem sources can be easily removed in many cases once they are discovered it is difficult for end users to identify the sources of such interference We investigated the characteristics of different sources that can degrade Wi Fi performance and developed WiSlow a software tool that diagnoses the root causes of poor Wi Fi performance using user level network probes and leveraging peer collaboration to identify the physical location of these causes WiSlow uses two main methods packet loss analysis and 802 11 ACK number analysis The accuracy of WiSlow exceeds 90 when the sources are close to Wi Fi devices Also our experiment proves that the collaborative approach is feasible for determining the relative location of an interfering device I INTRODUCTION Today it is common for households to put together home networks with a private wireless router access point that
12. UDP throughput to almost zero 5 In this study we focus on Wi Fi channel contention and common non Wi1 Fi interference sources II CHALLENGES In this section we describe the reasons why analyzing wireless networks is difficult for end users A Inaccurate RSSI and SINR measurements Received signal strength indication RSSI and Signal to interference plus noise ratio SINR are generally considered to be the key factors that indicate the quality of a wire less link However according to Vlavianos et al 7 RSSI inaccurately captures the link quality and it is difficult to accurately compute SINR with commodity wireless cards We also observed a similar result when monitoring RSSI and SINR values during our experiments We placed various types of interference sources close to the AP and measured the values on a general client machine In Figure la RSSI values with a baby monitor were higher than those obtained from a no interference environment which should be reversed when the measured UDP throughput is considered In Figure 1b the SINR values with a cordless phone were also higher than those obtained from a no interference case Furthermore these results varied for each experiment Based on this observation we conclude that RSSI and SINR values captured by a general end user s wireless card do not correctly represent the level of interference B No specific network adapter or driver We do not make any assumptions abo
13. alent to the mass in the formula of the center of mass WiSlow first obtains the coordinates of cooperative clients based on the input from end users and calculates the coordinates of the interference source using the following formula M Silke R X Miri 2 k 1 M is the strength of interference on the zth client and fi denotes the function of the measured magnitudes for each frequency kx where x is the smallest frequency caused by the interfering device The coordinate of the interference source R can be calculated based on the sum of each client s weighted M coordinates r VI EVALUATION In this section we describe the accuracy of WiSlow in identifying the root cause of a Wi Fi performance problem First we placed a laptop 8m away from an AP where Wi Fi performance is not affected by weak signal strength Then we located the interfering devices between them one at a time We repeated the experiments altering the distance between the interfering device and the AP We ran WiSlow on the laptop 15 times each at six different locations a total of 90 measurements for each interfering device and counted the number of times that WiSlow correctly diagnosed the root cause First without considering the type of the non Wi Fi device we tested the capability of WiSlow to distinguish between no interference channel contention and non Wi1 Fi interference We evaluate the diagnostic accuracy and the false positive
14. ars that a baby monitor located close to your router is interfering with your Wi Fi network We focus on building software that does not require any additional spectrum analysis hardware unlike e g WiSpy 2 AirSleuth 3 or AirMaestro 4 In addition WiSlow does 978 1 4799 3360 0 14 31 00 2014 IEEE 862 not depend on a specific network adapter such as the Atheros chipsets which were used to achieve similar goals in other studies 5 6 These features enable WiSlow to run on common end user machines First we investigate behaviors of 802 11 networks such as retries frame check sequence FCS errors packet loss and bit rate adaption which can be observed on ordinary operating systems Our experimental results show that the statistical patterns of the above variables vary depending on the problem sources For example with the interference that caused by non Wi Fi devices we observed a greater number of retried packets fewer FCS errors and larger variations in the bit rates compared to channel contention Correlating these variables we can categorize the sources of performance problems into several distinct groups In addition the non W1 Fi devices such as baby monitors cordless phones and microwave ovens show different patterns when the number of UDP packets and 802 11 ACKs are plotted over time Based on these observations we developed two methods packet loss analysis and 802 11 ACK pattern analysis These met
15. asured Statistics with the results of our experiments This is because the measurement of a wireless network is highly affected by the client s own environment such as a distance from the AP signal power or fading multi path and shadowing In other 865 1 0 8 Baby monitor ext Baby monitor ex2 e E E T A Baby monitor ex1 Baby monitor ex2 LL Q O 0 0 30 35 5 10 15 20 25 0 10 20 30 The number of packet loss per 100 KB Euclidean distance q bit rate packet loss a The estimated number of b The Euclidean distance be packet loss tween each sample and the mean Fig 5 The CDFs obtained from two experiments with the same baby monitor in different environment words even though they have the same type of problem the statistics of the measured metrics can vary depending on each end user s own situation Note that this is the reason why simple measurements such as the higher layer throughput e g TCP or UDP or number of 802 11 retries are not enough to identify the level of interference and the type of interferers We found that even if the underlying environment changes the extent of the area over which a set of samples correlated packet loss and bit rate are dispersed remains similar if the problem source is the same Figure 4b shows that even though the two groups of samples from discrete environments are dis tributed on different spots on the coordi
16. aterial is based upon work supported by the National Science Foundation under Grant No CNS 1218977 REFERENCES 1 S Gollakota F Adib D Katabi and S Seshan Clearing the RF smog making 802 11n robust to cross technology interference in Proc of ACM SIGCOMM Toronto Ontario Canada Aug 2011 2 Wi Spy http www metageek net Online accessed May 2013 3 AirSleuth http nutsaboutnets com airsleuth spectrum analyzer Online accessed May 2013 4 AirMaestro http www bandspeed com products products php On line accessed May 2013 5 S Rayanchu A Patro and S Banerjee Airshark Detecting non WiFi RF Devices Using Commodity WiFi Hardware in Proc of ACM IMC Berlin Germany Nov 2011 6 S Rayanchu A Patro and S Banerjee Catching Whales and Min nows Using WiFiNet Deconstructing non WiFi Interference Using WiFi Hardware in Proc of USENIX NSDI San Jose CA USA Apr 2012 7 A Vlavianos L K Law I Broustis S V Krishnamurthy and M Faloutsos Assessing link quality in IEEE 802 11 wireless networks Which is the right metric in Proc of PIMRC Cannes France Sep 2008 8 Radiotap http www radiotap org Online accessed May 2013 9 WLAN packet capture http wiki wireshark org CaptureSetup WLAN Online accessed May 2013 10 S Biaz and S Wu Rate adaptation algorithms for IEEE 802 11 net works A s
17. bined interference of multiple devices Note that the client downloaded 100 MB of UDP packets for each experiment to collect a statistically meaningful amount of samples but when actually probing on an end user s machine WiSlow only needs to transmit 5 MBytes of UDP packets to identify the root cause which takes a reasonable amount of time 10 30s e Retry and available bit rate Since an 802 11 retry and bit rate reduction are both initiated by a packet loss their temporal changes are closely correlated when a packet loss occurs the bit rate decreases by the 802 11 rate adaptation algorithm 10 The probability of packet loss then decreases due to the reduced bit rate which lowers the number of retries After that the bit rate gradually increases again owing to the reduced packet loss which leads to a higher probability of packet loss and retries In other words if contention or interference exists it causes packet losses and then the bit rate and the number of retried packets repeatedly fluctuate during the subsequent data transmission Because of this fluctuation the measured statistics of retries and bit rates do not represent the characteristics of interference sources correctly Figure 2a and 2b shows that the cumulative distribution functions CDFs of the values do not distinguish each device except the baby 864 monitor e Frame check sequence errors Another variable that we trace is the number of FCS errors per b
18. ch model It appears to be impractical to collect the patterns from every product However we found that different models of the same type of product likely have common characteristics For example we tested four FHSS cordless phones produced by two different manufacturers and each one showed the same ACK number frequencies multiples of 100 Hz Therefore we believe that collecting a small amount of information can cover the majority of devices if they follow the industry standards or use similar technologies In conclusion WiSlow successfully detected the root cause of Wi Fi performance degradation with a high probability over 90 in most cases although it frequently misidentified the type of certain non Wi Fi interfering devices when they were not located near the Wi Fi device However this inaccuracy can be removed if we take into account the pre obtained ACK number pattern of each device B Locating interfering devices We set up three laptops and one 802 11g AP in a building at Columbia University We placed a baby monitor between them and changed its location over time Figure 10a illustrates our experimental scenario The circled numbers indicate the movement path of the baby monitor We ran WiSlow each 3These inaccuracies can be ignored because the throughput shows there was actually no interference even though the cordless phone was active 4Motorola and Panasonic 869 6 Schapiro Bldg 7 Floor a Experiment
19. daresan Y Grunenberger N Feamster D Papagiannaki D Levin and R Teixeira WTF Locating performance problems in home networks in SCS Technical Report GT CS 13 03 Jun 2013 19 20 Myths of Wi Fi Interference http tinyurl com 9rbe2f7 Online accessed July 2013
20. early detect interference caused by a microwave oven regardless of the distance average 98 In our extra experiments WiSlow could detect the duty cycle of the microwave oven even when located relatively far from the AP and laptop 11 m and 16 m However in these cases since the microwave oven did not severely interfere with the Wi Fi network we do not elaborate further on the results in the present paper Second the diagnostic accuracy of detecting baby monitors was also very high when it was close to the AP However it dropped to under 6 7 when the distance was greater than 1m Table II In most cases it was misidentified as a FHSS cordless phone which contributed the high false positive rate of this device 24 8 This result occurred because these two devices have the same characteristic frequency hopping and WiSlow considers their level of interference to distinguish them In other words if a baby monitor is far from a Wi Fi device and causes less interference it can mislead WiSlow s identification The accuracy of detecting FHSS cordless phones was also low when it was not close to the AP 6 7 at 2 5m However this was because the cordless phone caused insignificant interference at this spot the average UDP throughput was 13 28 Mb s at 2 5m the average throughput with no interference was 14 Mb s in the same environment With this small interference WiSlow did not observe the expected hopping patterns As a result t
21. er 100 KB c The CDF of FCS errors Fig 2 802 11 statistics with various interference sources it is not always possible to capture wireless packets on some types of OS e g Microsoft Windows 9 Instead Windows provides several APIs that report 802 11 packet statistics to user applications Those APIs enable WiSlow to run on Windows because they provide all the information that WiSlow must extract from the 802 11 packets In the following sections we explain WiSlow s two main diagnostic methods packet loss analysis and 802 11 ACK pattern analysis A Method 1 packet loss analysis First we found that each problem source varies in their packet loss characteristics represented by three statistics 1 the number of 802 11 retries 2 the available bit rates and 3 the number of FCS errors In each experiment we measured these values on a client laptop while downloading UDP packets from an AP The values were recorded for each 100KB of UDP packets received We repeated this experiment for different scenarios including channel contention and non W1 Fi interference To simulate channel contention we set up several laptops sending bulk UDP packets to the AP To generate non Wi Fi interference we placed each interfering device baby monitors microwave ovens and cordless phones close to the AP about 20cm and measured the effect on the client placed at various distances from the AP In this study we did not consider the com
22. ey then investigate the delay patterns to dis tinguish hidden terminals from low SNR While their approach intentionally avoids using layer 2 information WiSlow actively exploits 802 11 information in order to obtain a more de tailed identification e g device type causing the interference Spectrum MRI 17 also isolates interference problems The authors discuss that the link occupancy and retransmission rate is different depending on the sources of interference They measure and compare those metrics to identify Bluetooth channel congestion and the slow link on same AP problem Sundaresan at el 18 present a tool that identifies whether a performance bottleneck exists inside the home network or on the access link by measuring variation of packet interar rival time It also evaluates the state of the wireless link by monitoring the bitrate and throughput on an AP While this tool focuses on identifying where a bottleneck exists WiSlow focuses on identifying the type of interference source within the wireless network VIII A 802 lln 802 11n uses both 2 4 and 5 GHz bands Although fewer non Wi Fi devices are operating at 5GHz and thus less interference presently exists at that band Cisco has anticipated that more devices will use the 5GHz band in the future and therefore a similar interference will likely occur 19 We believe that our basic approach will also be feasible for discovering non Wi Fi interference sources at
23. he majority of incorrect diagnostic results were no interference which explains its high false positive rate 14 1 shown in Table I The low accuracy of detecting baby monitors and FHSS cordless phones can be improved if we take into account their specific ACK number frequency values which were discussed in Section IV B Recall that the ACK number frequencies of the baby monitor were a multiple of 43 Hz and those of the FHSS cordless phone were a multiple of 100Hz When WiSlow is adapted to consider these specific values the detection accuracy increases dramatically Table III shows that the accuracy was 100 most of the time except when IEEE INFOCOM 2014 IEEE Conference on Computer Communications Non Wi Fi Interference Diagnostic False Accuracy Positive Distance Avg from the AP Throughput Microwave oven Baby monitor 6 76 Mb s 9 65 Mb s FHSS Cordless 10 02 Mb s phone 10 05 Mb s 12 44 Mb s 13 28 Mb s TABLE II The accuracy of WiSlow for identifying non Wi Fi devices Non Wi Fi Interference Diagnostic Accuracy Distance from the AP Baby monitor Cordless phone TABLE III The accuracy of WiSlow for identifying baby monitors and cordless phones the FHSS cordless phone was placed at locations farther than 1 5m However the disadvantage of this approach is that WiSlow needs to learn the ACK number frequency value of the particular product in advance because the pattern depends on ea
24. hods successfully distinguish channel contention from non Wi1 Fi interference and infer the product type of the interfering device We believe that this technology will be useful to end users since it can inform them of what needs to be done in order to improve the performance of their networks whether to change the Wi Fi channel or remove a device that is emitting the interference In non Wi Fi interference scenarios another goal is to identify the physical location of the source of interference Although it is difficult to pinpoint the exact physical location of the source without a spectrum analyzer or additional support of wireless access points APs we could infer the relative location of the problem source by collaborating with other end users connected to the same wireless network WiSlow collects probing results from peers and determines whether others observe the interference If all the machines observe the same interference it is highly likely that the problematic source is close to the wireless AP However if only one of the peers observes the interference the source is likely to be located close to that peer Our experimental results clearly show that this approach is feasible In summary WiSlow 1 distinguishes channel contention from non Wi Fi interference 11 infers the product type of the interfering device e g a microwave oven cordless phone or baby monitor by analyzing network packets and finally iii points
25. nate plane their extent is similar Thus we first quantify how widely the samples are dispersed by calculating the Euclidean distances between each sample and the mean M Sz My S mean Mz My sample Sz Sy Figure 5 compares the CDFs obtained from two experiments that were conducted with the same baby monitor in two discrete environments The CDFs of packet loss estimation Figure 5a show different distribu tions while the CDFs of the Euclidean distances between the samples and the mean show similar distribution Figure 5b Therefore WiSlow can use the CDFs of the Euclidean dis tances to identify the root causes of network interference We prepare these CDFs of each problem source in advance which are obtained from our experiments Then WiSlow traces the wireless packets on an end user s machine generates a CDF of the distances and compares it to the pre obtained CDFs of each problem source For the convenience of identification we group the problem sources into three groups by the shape of the CDFs no interferers group 1 light interferers group 2 and heavy interferers group 3 Each group has its rep resentative CDFs that are determined by multiple experiments Figure 6 In our data sets group 1 indicates a no interference environment group 2 includes channel contention and cordless phones that use frequency hopping spread spectrum FHSS and group 3 contains microwave ovens and baby monitors
26. ontention I Non Wi Fi Method 2 ___Interference___ Duty cycle 50 Frequency Fixed ACK number frequency 60 Hz No interference Method 2 Hopping Frequency Hopping FHSS cordless phone _ Method 1 Group 3 Baby monitors Microwave Ovens Method 2 Duty cycle 100 H No ACK cycle x Analog cordless phone Fig 9 The classification of problem sources by WiSlow s methods few channels 4 Bluetooth Bluetooth is another widely used wireless standard that operates in the 2 4GHz spectrum Hopping within the entire 2 4 GHz band it interferes with every channel of an 802 11 network However algorithms such as Adapted Frequency Hopping AFH which is used to automatically avoid busy channels mitigate this interference Consequently Bluetooth inconsiderably affects the performance of 802 11 networks In a measurement by Rayanchu et al 5 Bluetooth was shown to degrade the UDP throughput by about 10 as a worst case Since we also verified that Bluetooth does not in terfere much with 802 11g networks based on our experimental result we excluded Bluetooth in our identification algorithm C Classification WiSlow takes into account the combination of the results from the first method packet loss analysis and the second method ACK pattern analysis to identify the device type precisely For example the resul
27. out the approximate location of the source of interference by exploiting user collaboration We evaluated WiSlow with various interference sources and it showed quite high diagnostic accuracy It also proved that our approach locating the interference source is feasible IEEE INFOCOM 2014 IEEE Conference on Computer Communications The remainder of this paper is structured as follows In Section II we describe the common sources of Wi Fi perfor mance degradation In Section HI we discuss the restrictions of an end user s environment and how WiSlow attempts to overcome them Section IV explains the detailed methods of WiSlow and Section VI evaluates our approach II BACKGROUND Common sources that cause Wi Fi performance degrada tion include e Wi Fi channel contention degradation due to a channel crowded by multiple Wi Fi devices that compete to transmit data through an AP It also includes interference due to nearby APs that are using the same channel or adjacent channels e Non Wi Fi interference interference due to non Wi1 Fi devices that use the same 2 4 GHz spectrum as the 802 1 1b g networks The devices include microwave ovens cordless phones baby monitors and Bluetooth devices e Weak signal when the signal is not strong enough due to distance or obstacles packets can be lost or corrupted Although the extent varies all the above sources result in severe performance degradation some of them even drop the TCP
28. ovens neither monitor signals IEEE INFOCOM 2014 IEEE Conference on Computer Communications nor communicate with Wi Fi devices Second owing to the limited capability of the hardware end user devices cannot detect signals emitted from the devices precisely To overcome these circumstances we leverage multiple Wi Fi devices a probing client end user machine requests cooperative clients to perform a WiSlow diagnostics as described in previous sections It then receives the diagnostic result containing the type of detected device and its interference strength from each client We calculate the interference strength using the magni tude of the particular ACK number frequency that was used to detect the device as described in the previous section Method 2 For example the interference strength of a microwave oven can be determined based on a magnitude of 60Hz in the FFT of the ACK number analysis In the case of an FHSS device it can be determined by the sum of the magnitudes of the multiple frequencies caused by the frequency hopping pattern After collecting the strength values from the clients we use the same method of obtaining the center of mass to find the location of the interference If the interference strength detected by a particular client is greater than the interference strength detected by other clients it means that the interference source is closer to that client Therefore interference strength can be considered equiv
29. pical audio and video trans mitters such as baby monitors is known to be 100 It means that they send and receive data constantly implying that they continuously interfere with Wi Fi networks without any off period Therefore intuitively we do not expect to observe similar ACK patterns as those observed in the microwave oven experiment However when converting the plot from the time domain to the frequency domain we observe another notable pattern Figure 8a shows that there are multiple high peaks set apart by a specific interval i e 43 Hz occurring at 43 86 129 and 172 Hz This is in contrast to the microwave ovens that showed only one significant peak at 60 Hz Figure 7b We conjecture that these peaks are caused by frequency hopping a frequency hopper switches its frequency periodically and interference occurs when it hops to a nearby frequency of the current Wi Fi channel Since the frequency hopping chooses the next frequency using a pseudorandom sequence it creates diverse pulses with different magnitudes that are randomly positioned in the ACK number plot For clarity we plot a quantized time domain graph Figure 8b that is converted back from the frequency domain graph We used the 10 highest frequencies from Figure 8a In the time domain graph the number of ACKs y axis fluctuates periodically however note that the heights of the peaks vary The possible explanation is as follows the number of ACKs is large when the
30. roups minimizing the impact of the end user s underlying environment B Method 2 802 11 ACK pattern analysis The first method is able to determine which type of loss pattern a problem source has However because multiple prob lem sources are categorized into each group we need another method that further narrows down the root causes In this section we explain the second method designed to distinguish several detailed characteristics of non Wi Fi devices such as frequency hopping and duty cycle WiSlow sends bulk UDP packets to the AP and counts the received 802 11 ACKs to check the quality of a wireless link within a given period In order to detect patterns on the scale of milliseconds we use a very small size of UDP packets 12 bytes that reduces potential delays such as propagation and processing delays and we transmit as many UDP packets as possible to reduce the intervals between samples As a result we received 0 7 ACKs per millisecond In the following sections we describe the results of the 866 above method when performed with various non Wi Fi inter ferers and we explain how WiSlow identifies the devices based on the results 1 Duty cycle microwave ovens Microwave ovens gener ate severe interference in almost every channel of the 2 4 GHz band We identify this heavy interferer using its duty cycle which is the ratio of the active duration to the pulse period It is known that the duty cycle of microwave
31. sup ports multiple wireless devices However the increasing usage of wireless networks inevitably results in more contention and interference which causes unsatisfactory Wi Fi performance Furthermore non Wi Fi devices such as microwave ovens cordless phones and baby monitors severely interfere with Wi Fi networks because these devices operate on the same 2 4 GHz spectrum as 802 11b g 1 Although these problem sources can be easily removed in many cases e g by relocating the interfering device choosing a different channel or moving to the 5GHz band it is difficult for technically non savvy users to even notice the existence of channel contention or interference caused by non Wi Fi devices non W1 Fi interfer ence Instead properly working routers or service providers are frequently misidentified as the culprit while the actual root cause remains unidentified However isolating the root causes of poor Wi Fi performance is nontrivial even for a network expert because they show very similar symptoms at the user level and special devices are required in order to investigate the lower layers of the protocol stack We introduce WiSlow Why is my Wi Fi slow a software tool that diagnoses the root causes of poor Wi Fi performance with user level network probes and leverages peer collaboration to identify their physical locations In other words the goal of this tool is to report the problem source to users such as It appe
32. t of the first method is Group 3 and that of the second method is frequency hopping we consider the problem source to be a baby monitor In addition WiSlow looks into the source and destination addresses of the captured 802 11 packets in order to examine the channel occupancy rate If the channel is highly occupied by other clients or nearby APs but WiSlow does not detect any non Wi Fi interference it considers the root cause to be channel contention Figure 9 describes the classification algorithm that WiSlow uses to conclude the root cause Currently we aim to provide the best estimate of suspicious devices that we are aware of but we believe that more types of devices can be covered easily once they are characterized in a similar manner V LOCATING INTERFERING DEVICES In this section we describe a method to determine the physical locations of interfering devices A number of re search studies on indoor location tracking have attempted to pinpoint the location of laptops or smartphones through various methods 13 15 While these studies focus on locating client devices using signal information such as RSSI and SINR values we focus on locating interference sources using multiple collaborative end user devices Compared to locating Wi Fi devices there are several difficulties in locating non Wi Fi devices for end users First it is impossible to obtain measurement data such as RSSI and throughput from such devices e g microwave
33. t would be more reasonable to compare the combinations of these statistics together instead of investigating each variable individually There are two cases that can cause a retry First a packet was not delivered i e it was lost Second a packet was delivered but it had an FCS error We can estimate the number of packets lost the first case by subtracting the number of FCS errors from the number of retries Eq 1 N PacketLoss N Retries _ IN BOS errors 1 We found that this estimated number of packet losses represents the level of interference more reliably than the individual statistics of retries bit rates and FCS errors In other words the number of packet losses provides relatively consistent results in repeated experiments while the others varied for each experiment Figure 3 shows that the CDF of the estimated number of packet loss clearly distinguishes each device compared to the CDFs in Figure 2 It can be seen that a baby monitor causes the most severe amount of packet loss while cordless phones cause a relatively small amount of packet loss Since baby monitors send video and audio data at the same time they use more bandwidth than cordless IEEE INFOCOM 2014 IEEE Conference on Computer Communications No interference e Contention FHSS phone Baby monitor Microwave oven 10 20 30 40 50 The number of packet loss per 100 KB Fig 3 The CDFs of the estimated packet loss
34. th a fixed interval 100 Hz Figure 8 This verifies that our method is suitable to identify frequency hopping devices Consequently we can distinguish frequency hopping de vices by determining whether the number of 802 11 ACKs has multiple high peaks with a certain interval in the frequency domain We check this by linear regression of the peak frequencies if the correlation coefficient is greater than 0 99 we consider it to be a frequency hopping device 3 Fixed frequency analog cordless phones Since typi cal analog cordless phones use a fixed frequency they usually interfere only with a small number of channels The analog phones we tested only interfered with Channel 1 Because they do not change frequency severe interference occurs if the current Wi Fi channel overlaps with the frequency of the phone In addition their duty cycle is close to 100 which implies that no ACK cycle exists In our experiments the UDP throughput stayed very low and no explicit ACK cycle no hopping was observed as expected Therefore WiSlow concludes that an analog cordless phone is the interferer if there is a heavy interference pattern but no explicit ACK cycle or duty cycle is detected Then we can inform the user that switching the Wi Fi channel can improve the performance in this case because this kind of device is likely to affect only a 867 Method 1 Group 2 High channel Occupancy Method 1 Group 1 Channel C
35. urvey and comparison in Proc of IEEE ISCC Marrakech Morocco 2008 11 EF J Massey Jr The Kolmogorov Smirnov test for goodness of fit Journal of the American statistical Association vol 46 no 253 pp 68 78 1951 12 A Kamerman and N Erkocevic Microwave oven interference on wireless LANs operating in the 2 4 GHz ISM band in Proc of PIMRC Helsinki Finland Sep 1997 13 G V Zaruba M Huber F A Kamangar and I Chlamtac Indoor location tracking using RSSI readings from a single Wi Fi access point Wireless Networks vol 13 no 2 pp 221 235 Apr 2007 14 A Ali L Latiff and N Fisal GPS free indoor location tracking in mobile ad hoc network MANET using RSSI in Proc of IEEE RFM Selangor Malaysia Oct 2004 15 J Hightower R Want and G Borriello SpotON An indoor 3D location sensing technology based on RF signal strength Technical Report UW CSE 00 02 02 University of Washington Seattle WA vol 1 2000 16 P Kanuparthy C Dovrolis K Papagiannaki S Seshan and P Steenkiste Can user level probing detect and diagnose common home WLAN pathologies Computer Communication Review vol 42 no 1 pp 7 15 2012 17 A Baid S Mathur I Seskar S Paul A Das and D Raychaudhuri Spectrum MRI Towards diagnosis of multi radio interference in the unlicensed band in Proc of IEEE WCNC Quintana Roo Mexico Mar 2011 18 S Sun
36. ut the specific net work adapters or drivers that end users may have Some Atheros chipsets which are widely used in research studies support a spectral scan that provides a spectrum analysis of multiple frequency ranges Rayanchu et al developed Air shark 5 and WiFiNet 6 leveraging this feature to distinguish non Wi1 Fi interferers using a commodity network card without We used a MacBook Pro 2013 network card AirPort Extreme chipset Broadcom BCM43 series in this measurement 863 No interference e Contention Jeen FHSS phone Baby monitor Microwave oven No interference ke x Contention og 8 ve FHSS phone m Baby monitor aeta Microwave oven 40 30 235 25 30 45 50 55 a 35 35 0 RSSI dBm SINR dBm a RSSI measurement b SINR measurement Fig 1 The CDFs of RSSI and SINR values specialized hardware Although this approach achieved quite high accuracy in identifying the interfering devices to the best of our knowledge only a few chipsets e g Atheros currently provide this feature In addition we failed to discover references to this feature for any operating system OS other than Linux Since there are hundreds of products that use a different chipset and or OS it is impractical to assume that a general end user has this specific setup Therefore we focus instead on analyzing the quality of a link observing user accessible packets such
37. yte In our experiments we counted the number of FCS errors per 100KB of data Intuitively it can be predicted that non Wi Fi interference introduces more FCS errors than channel contention or a no interference environment This is because the packet corrup tions are likely to occur more frequently when a medium is noisy However in our experiment it turned out that a large number of FCS errors are not necessarily correlated with severe interference On the contrary we often observed that fewer FCS errors occur in a severe interference environment e g interference caused by a baby monitor than in a no interference environment Figure 2c This paradox can be explained by the low bit rates in the interference case which implies that a smaller number of bits are transmitted in the same bandwidth Consequently the number of FCS errors per byte alone is not sufficient to characterize interference sources 1 Packet loss estimation As we stated above although the number of retries bit rate and FCS errors are affected by the current state of the wireless network they often show very different statistics for each experiment set We conjecture several reasons the environment is not exactly the same in every experiment the occurrence of packet loss is probabilistic rather than deterministic and the individual variables fluctuate over time affecting each other and leading to different statistics for a certain period of time Therefore i
Download Pdf Manuals
Related Search