SepEvaluationofSoftwareDefinedRadioforParasiticTracking

Chair for Computer Aided Medical Procedures & Augmented Reality
Lehrstuhl für Informatikanwendungen in der Medizin & Augmented Reality

THIS WEBPAGE IS DEPRECATED - please visit our new website

Evaluation of Software Defined Radio for Parasitic Tracking

Thesis by: Reissner Ulrich
Advisor: Manuel Huber
Director: Gudrun Klinker
Due date: 26.09.2007

Abstract

This Sep focus on meassuring IEEE 802.11(Wireless Lan) signal strength with the Universal Software Radio Peripheral(USRP) and GNU-Radio.
The 802.11 signal is converted from analog to digital with the USRP, an later demodulated with GNU-Radio. This demodulated signal is used to extract the needed data, like signal strength, type and MAC-Address.
On the other side, data received with an Intel Corporation PRO/Wireless 3945ABG controler integrated into a Laptop, is used for comparsion.
At the end, a visualisation of both signals is done with MagicMap.

Parasitic Tracking

  • tracking infrastructure is not everywhere
  • use all posibilities for tracking
  • standard situation:
    • extra hardware
    • fixed application programming interface

  • in our case:
    • use only one hardware for all kind of tracking
    • use detail information to get better results

Hardware - Universal Software Radio Peripheral(USRP)

  • Altera FPGA
  • USB 2.0
  • up to 4 daughterboars with range form 1MHz to 3GHz

USRP

Software - GNU-Radio

  • software defined radio
  • turns radio hardware problems into software problems

Signal Measurement

  • IEEE 802.11
    • Receive Signal Strength Indicator(RSSI) specified as optinal
    • value between 0 and RSSI_MAX
    • RSSI_MAX diffrent for each vendor

  • USRP
    • calculated over a definable number of bits received
    • RSSI_MAX can be self defined

Evaluation

  • Comparsion USRP - INTEL 3945ABG
    • empty room, only tables and chairs
    • 4 access points with fixed positions
    • 7 independent measurements, each 300 s long
    • USRP
      • 62699 values
      • standard deviation : 4,67
      • values per access point per second : 7,46
    • INTEL
      • 9905 values
      • standard deviation : 4,11
      • values per access point per second : 1,18

  • Advantages of USRP and GNU-Radio
    • use all packets, not only beacons
    • focus only on needed channels
    • calculate the RSSI over the complete frame
    • higher packet rate
    • combination with rfid, dect, bluetooth

  • Disadvanteges of USRP and GNU-Radio
    • need of extra hardware
    • more expensive than off the shelf hardware
    • lower range
    • higher power usage
    • high cpu workload
    • a bit higher standard deviation
    • only works with 1 Mbit/s at the moment

Further Work

  • implement channel hopping for USRP
  • improve RSSI calculation
    • use new gnuradio version
    • implement new functions
  • use new USRP version, to work with more than 1Mbit/s
  • expand for dect, rfid, bluetooth tracking
  • implement positioning algorithm
  • implement api for Ubitrack

Students.ProjectForm
Title: Evaluation of Software Defined Radio for Parasitic Tracking
Abstract: This Sep focus on meassuring IEEE 802.11(Wireless Lan) signal strength with the Universal Software Radio Peripheral(USRP) and GNU-Radio. The 802.11 signal is converted from analog to digital with the USRP, an later demodulated with GNU-Radio. This demodulated signal is used to extract the needed data, like signal strength, type and MAC-Address. On the other side, data received with an Intel Corporation PRO/Wireless 3945ABG controler integrated into a Laptop, is used for comparsion. At the end, a visualisation of both signals is done with MagicMap.
Student: Reissner Ulrich
Director: Gudrun Klinker
Supervisor: Manuel Huber
Type: SEP
Area: Industrial Augmented Reality
Status: finished
Start: 2006/11/01
Finish: 2007/09/31


Edit | Attach | Refresh | Diffs | More | Revision r1.4 - 26 May 2009 - 14:19 - MarcusToennis