1 / 29

SoNIC:Classifying Interference in 802.15.4 Sensor Networks

SoNIC:Classifying Interference in 802.15.4 Sensor Networks. IPSN’13. Introduction. Devices access to 2.4 GHz band Knowing the interference source helps mitigation e.g. BuzzBuzz protocol against Wi-Fi interference [ SenSys’10]. Introduction. SoNIC. Incomming packet.

chandler
Download Presentation

SoNIC:Classifying Interference in 802.15.4 Sensor Networks

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. SoNIC:Classifying Interference in 802.15.4 Sensor Networks IPSN’13

  2. Introduction • Devices access to 2.4 GHz band • Knowing the interference source helps mitigation • e.g. BuzzBuzz protocol against Wi-Fi interference [SenSys’10]

  3. Introduction SoNIC Incomming packet Mitigation Strategy Application Interfernce State Classifier • Existing approaches for interference detection • Active sampling • Channel hopping • Additional hardware • SoNIC: • use a supervised learning approach to create a classifier that classifies corrupted packets • light weight,energy efficiency

  4. Characterizing Interference • Experiment Setup: • Sources of Interference • Wifi, Microwave ovens, Bluetooth,Non-interfered weak links

  5. LQI threshold 14dBm 11dBm Wifi(54M) Bluetooth 10dBm 2dBm Weak Link Microwave

  6. Error bursts Wifi(54M) Bluetooth Weak Link Microwave

  7. Error bursts • Definition: • a sequence of corrupted symbols that may contain subsequences of at most four consecutive correct symbols. • Temporal behavior of different interferers

  8. RSSI-based features Wifi(54M) Bluetooth Weak Link Microwave

  9. Features of Corrupted Packets

  10. Overview of SoNIC Interference state Voter Classifier Application Feature calculation Mitigation strategies On match Storage Matching crc fail crc ok Radio Driver Incoming packet

  11. Overview of SoNIC Interference state Voter Classifier Application Feature calculation Mitigation strategies On match Storage Matching crc fail crc ok Radio Driver Incoming packet

  12. Classifying Corrupted Packets Interference state Voter Classifier Application Feature calculation Mitigation strategies On match Storage Matching crc fail crc ok Radio Driver Incoming packet

  13. Classifying Corrupted Packets • Decision Tree Classifier • Low computational complexity • Accuracy of decision tree

  14. Voting Classifier rx a corrupted packet 30s time wifi bluetooth mircowave weaklink

  15. Voting Classifier rx a corrupted packet 30s time wifi bluetooth mircowave weaklink the dominant interferer causes a significant amount of packet corruption

  16. Evaluation • SoNIC implementation • Based on Contiki • Environment • An office corridor of 32m long • With some uncontrolled interferer

  17. Classifying Results Weak Link Microwave Bluetooth Wifi

  18. Classifying Results Weak Link Microwave Bluetooth Wifi

  19. Classifying Results Weak Link Microwave Bluetooth Wifi

  20. Classifying Results Weak Link Microwave Bluetooth Wifi

  21. Classifying Results Weak Link Microwave Bluetooth Wifi

  22. Detection Results We conclude that when experiencing packet error ratios of 20% and more, SoNIC correctly detects the interference state 87.5% of the time on average

  23. Conclusion • Key Contribution • SoNIC Attribute corrupted packets to an interference type • SoNIC correctly detects the predominant interferer • Future work • Deal with packets form different interferers are transmitted simultaneously

  24. Using and Extending SoNIC Voter Classifier Application Feature calculation Mitigation strategies On match Storage Matching crc fail crc ok Radio Driver Incoming packet

  25. AUGMENTING A MOBILE SINK WITHSONIC

  26. AUGMENTING A MOBILE SINK WITHSONIC

  27. Features of Corrupted Packets Automatic Gain Control

  28. Evaluation mean feature calculation time of 26.5 ms

More Related