??EEC 687/787 - Mobile Computing (Fall 2012)

EEC 687/787 - Mobile Computing (Fall 2012)

Instructor

Prof. Yu Chansu, e-mail: c.yu91@csuohio.edu, Office: SH 437, phone: 2584, Office hour: T 1-3pm, R 10-12am

Course Information

Download here (including syllabus, grading policy, labs, course schedule, etc.)

Final Exam

  • Lecture notes
  • Quiz #6
  • MAC and carrier sensing mechanism: IEEE 802.11 spec (1999), “DCF & PCF,” Ch. 9.2 & 9.3
  • IEEE 802.11 MAC: IEEE 802.11 spec (1999), “DCF & PCF,” Ch. 9.2 & 9.3
  • IEEE 802.11 MAC Frame Format and Procedure: IEEE 802.11 spec (1999), “MAC Frame Formats,” Ch. 7.1
  • IEEE 802.11 MAC Management: IEEE 802.11 spec (1999), “Synchronization & Power Management,” Ch. 11.1 & 11.2
  • Mobile IP: C. E. Perkins, "Mobile-IP", in Communications Magazine, May 1997 (can find a copy here)
  • MANET Routing: D. B. Johnson and D. A. Maltz, "Dynamic Source Routing in Ad Hoc Wireless Networks," Mobile Computing, edited by T. Imielinski and H. F. Korth, Kluwer Academic Publishers, 1996. (can find a copy here)
  • TCP in Mobile Networks: R. Caceres and L. Iftode, "Improving The Performance of Reliable Transport Protocols in Mobile Computing Environments," IEEE Journal of Selected Areas in Communications, pp. 850-857, June 1995. (can find a copy here)

Midterm Exam

  • MAC and carrier sensing mechanism: IEEE 802.11 spec (1999), “DCF & PCF,” Ch. 9.2 & 9.3
  • IEEE 802.11 MAC: IEEE 802.11 spec (1999), “DCF & PCF,” Ch. 9.2 & 9.3
  • IEEE 802.11 MAC Frame Format and Procedure: IEEE 802.11 spec (1999), “MAC Frame Formats,” Ch. 7.1
  • IEEE 802.11 MAC Management: IEEE 802.11 spec (1999), “Synchronization & Power Management,” Ch. 11.1 & 11.2
  • Performance anomaly in IEEE 802.11 MAC: Section II, http://academic.csuohio.edu/yuc/papers/PID999868.pdf

Class projects

  • Helpful information
    OFDM implementation examples
    FHSS implementation document
    USRP hardware explained
    Asha & Shashanka's report & slides
  • Designing OFDM Communication System Using USRP
    • Amar Srinivas Rangineedi and Durga Rao Devisetti
    • Abstract: Orthogonal frequency division multiplexing (OFDM) is a specialized frequency multiplexing (FDM) technique, which divides total available bandwidth into a number of orthogonal sub-carriers and simultaneously transfers signals on these sub-carriers with a low data rate, achieving a total data rate. Lower data rate on each sub-carrier means bandwidth, which brings several benefits to the data transmission, such as flat channel fading and simpler receiver design. In view of this principle, we are implementing the OFDM with different modulation techniques at different coding rates in order to obtain desired and optimal data rates. In addition, we are anticipating at studying PDR for different data rates that we will obtain in the above procedure.
  • Data transmission and reception with OFDM signals using GNU Radio and USRP
    • Shahana Kallarackal and Sushma Yarlagadda
    • The project aims in implementing data transmission and reception using OFDM signals over a wireless medium using GNU radio and Universal software radio peripheral (USRP). As its name reveals, OFDM is a multiplexing method, which means that different data channels share the bandwidth available. OFDM is called orthogonal because all subcarriers are orthogonal to each other. The project aims in analyzing the spectrum of OFDM signals, and to analyze the Packet Delivery ratio (PDR) in the transmission and reception of data. Also, we intend to vary the system parameters such as modulation scheme, channel of data transmission, transmission amplitude and FFT bins, to analyze the PDR in each scenario.
  • TCP versus SCTP: Comparing the Performance of Transport Protocols
    • AhlamEbjad and Heli Shah
    • Today most applications use either the Transmission Control Protocol (TCP) or the User Datagram Protocol (UDP). Applications that need a reliable in-order delivery of the bytes sent by its peer use TCP, whereas ones that can tolerate a certain degree of loss prefer UDP, primarily because UDP provides speedier delivery of packets. Most applications prefer TCP over UDP and applications using TCP include file transfer applications, electronic mail and the worldwide web. Stream Control Transfer Protocol (SCTP) is a reliable transport protocol operating on top of a potentially unreliable connectionless packet service such as IP. SCTP’s services are at the same layer as TCP and UDP. Our goal is to use ns-2 simulator to compare TCP and SCTP, we are planning to create different scenarios to figure out if SCTP demonstrate superior performance or characteristics better than TCP protocol in any scenario and when should we use TCP versus SCTP.
  • Data and power efficiency of various modulation schemes using USRP/GNU Radio
    • Dhruv Khati and Krishna Teja
    • In the past few years in mobile communications industry the main emphasis has been laid on high data throughput and reliable communication. In this project we will perform experiments by using different modulation schemes to transfer a large amount of data. The experimental setup will include two personal computers and two USRPs using GNU radio between which data will be transferred. The modulation schemes used will be that used in the industry at present namely BPSK, QPSK and 16QAM. The data rates will be studied and compared among these. Also, their behavior will be observed in a noisy channel and probability of error will be determined for different signal to noise ratios. Through this we will be able to determine the data and power efficiency of each of the modulation schemes.
  • Performance Evaluation of DSR, DSDV and ZRP
    • Vivek Trivedi and Shasvat Parikh
    • In this project we are going to discuss a detailed simulation study of Dynamic Source Routing (DSR), DSDV (Distance sequenced distance vector) and Zone Routing Protocol (ZRP) in different mobile scenarios generated for Mobile Ad hoc networks. The impact of these routing protocols is evaluated with respect to Average End-to-End Delay, Average Jitter, Average Throughput and Packet Delivery Ratio (PDR). The major goal of this study is to analyze the performance of popular MANETs routing Protocol in different scenarios.
       

Previous class projects

  • Amarnadh Oleti, Daljeet Singh, Praveen Reddy, Capture and Retransmit I/Q Data using GNU radio and USRP - Mid-report, Final report
  • Dan Levtonyuk, Emanuel Papp, Xiang Li, BER Performance of DSSS Technique under Periodic Jamming - Mid-report, Final report
  • Brian Krupp, Limei Hou, Ns2 SetDest Acceleration Enhancement - Mid-report, Final report
  • John Merchant, Kevin Kelsey, Scanner using GNU Radio - Mid-report, Final report
  • Asha Mariam Iype, Shashanka C. D., Video transmission using USRP - Mid-report, Final report
  • Rakesh Vanam, Varun Venkatraman, Implementation of Co-operative Communication using GNU-Radio - Mid-report, Final report
  • Robert Fiske, Malav Shah: Two Channel Transmitter/Receiver (report, presentation)
  • Tianning Shen, Yuanchao Lu: Research on key digital modulation techniques using GNU Radio (transmit a large amount of data with π/4-DQPSK) (report, presentation)
  • Elie Salameh: Modulate internet radio into FM radio using USRP (report, presentation)
  • Sai Gumudavally, Sachine Hirve: JPEG transfer using USRP and GNU Radio (report, presentation)
  • Derek Sean Zechman: Test different mobile communications using HP iPAQ (report, presentation)
  • Murali Marunganti: iPAQ (report, presentation)

Example class projects

 

Lecture notes and Announcements

·         Bring your 8GB thumb drive for a "LiveUSB". If you wish to install by yourself, here are the version numbers: Ubuntu 10.04, Ns-2 2.35, and GNU Radio 3.30.

  • Week 02 (Sep. 4): Lecture III: Physical layer
  • Week 02 (Sep. 6): Lab I: Mobile network simulation using ns-2
  • Week 03 (Sep. 11): Lecture IV: Medium access control
  • Week 03 (Sep. 13): Lab I: Mobile network simulation using ns-2
    • Files: ex6sta.tcl, fil.awk, fil2.awk, fil4.awk, fil6.awk
    • Lab I report (due Sep. 20) - SAMPLE REPORT (Discussions in this report are not always correct.)
      • PDR analysis: What is the overall PDR? What is the PDR for WT1-WT2, WT3-WT4 and WT5-WT6? Discuss the results.
      • Delay analysis: Send me delay.awk. What is the average packet delay? What is the packet delay for WT1-WT2, WT3-WT4, and WT5-WT6? Discuss the results.
      • Read: Mandatory reading’s
      • Read: Ns2 manual, “Mobile networking in ns,” Ch. 16 (http://www.isi.edu/nsnam/ns/doc/ns_doc.pdf)
      • Read: Ns2 manual, “Trace and Monitoring support,” Ch. 23
  • Week 04 (Sep. 18): Lecture V: 802.11
  • Week 04 (Sep. 20): Lab II: Wireless propagation using ns-2
    • Files: rate.tcl
    • Lab II Report (Due Sep. 27):
      • Run the simulations for all TX-rate. Obtain the aggregate CBR throughput for each run. And draw a chart drawing throughput versus TX-rate
      • Research: For the same data rate, change the capture threshold to see its effect on performance
      • Research: Modify mac-802_11.cc file so that the receiver gets a new packet (p) if Sp/IpktRx>CPThresh. Observe how it improves the performance. Do not forget to explain and discuss the results.
      • Read: WLAN: A. Kamerman and L. Monteban, “WaveLAN-II: A High-Performance Wireless LAN for the Unlicensed Band,” Bell Labs Technical J., pp. 118-133, Summer 1997.
  • Week 05 (Sep. 25): Lecture VI: 802.11
  • Week 05 (Sep. 27): Lab III: GNU Radio/ USRP software radio platform 
    • Lab III Report (Due Oct. 4)
      • Complete the experiment through step 17. Experiment the dial tone example and test with different values of source (frequency and magnitude). 
        a) In step 10, how much can you drop the sample rate?
        b) In step 15, explain the parameters for the high pass filter.
        c) Add “Noise” block (where to add?) to see its effect on the sound and FFT.
      • Read “The Universal Handset,” IEEE Spectrum, April 2009 (http://spectrum.ieee.org/computing/embedded-systems/the-universal-handset) and “PPR: Partial Packet Recovery for Wireless Networks” to write one-page comment on the software radio.
  • Week 06 (Oct. 2): Lecture VII: 802.11 Management
  • Week 06 (Oct. 4): Lab IV: USRP
    • Files: base.py, base_gui.py 
    • Lab IV Report (Due Oct. 11)
      • Complete the experiment through step 17.
        a) In steps 1-6, experiment the example and test with different values of source (frequency and magnitude). 
        b) In step 9, what should be the appropriate interpolation and decimation rate?
        c) Explain the results in step 12.
        d) What is the suitable decimation rate in step 13?
        e) Explain the results in step 14 and step 15.
        f) Explain the test results in step 17.
  • Week 07 (Oct. 9): Lecture VIII: 802.11 Security
  • Week 07 (Oct. 11): Lab V: Mobility and Traffic Scenarios in ns-2 & Lab procedure

·         Lab V Requirements (Due Oct. 17)

1) Mobility scenario

a) Read ~/ndep-utils/cmu-scen-gen/setdest/setdest.{cc, h}

b) Report results of step 7 (PDR and plot).

c) In step 8, explain the statistics at the end of the generated mobility file.

d) Discuss the two versions of setdest in step 9. What is the problem with the first version of setdest? (Reference: Random Waypoint Considered Harmful, IEEE Infocom, 2003)

e) Complete and report step 10.

2) Traffic scenario

a) Read ~/ndep-utils/cmu-scen-gen/cbrgen.tcl

b) A different seed in cbrgen.tcl produces different traffic pattern. Discuss what’s been changed (randomized)?

c) Complete and report step 13.

·         File: c2.tcl

·         Lab VI Report (Due Nov. 1)

a) Explain how it is implemented in ns-2

b) Present results (maximum throughput)

c) Comparison with 802.11 MAC

·         File: am_usrp710.zip

·         Lab VII Report (Due Sep. Nov. 8)

BER performance with different digital modulation schemes

a) In step 2, explain the numbers 0.707, 0.630, 0.562, 0.501, 0.446, 0.398, 0.354, 0.316, 0.282, 0.251, and 0.224 which are used to test the case for SNR=0dB, 1dB, … , 10dB, respectively.

b) Explain the symbol table for 4-ASK in step 4. What is the symbol table for QAM in step 5?

c) Complete and explain the experiment for QAM.

d) Plot the BER vs SNR curve for ASK, 4-ASK, and QAM (steps 2, 4, and 5).

e) Explain and compare your observations with the scope and constellation display in steps 3, 4 and 5.

f) Discuss the differences between BER and SER (symbol error rate).

g) Discuss the differences between SNR and Eb/No, where Eb and No denotes energy per bit and noise per Hz.

 

AM receiver (an analog modulation scheme)

h) Complete steps 6-10 to hear the AM station pre-recorded through a USRP.

i) Explain Rational Resampler in step 11.

j) What other stations (which frequency) do you hear in step 13?

k) Send me the final grc file in step 14 with your report.

·         Lab VIII Report (Due Sep. Nov. 15)

      a) In Lab#1, observe OFDM spectrum in the FFT chart and compare it to the theoretical OFDM spectrum
b) In Lab#2, explain benchmark_ofdm_tx.py options (--interp, --fft-length, --occupied-tones)
c) In Lab#2, try with different option values, take screenshots of the FFT charts and compare/discuss them
--interp: set to 256 & 32
--occupied-tones: set to 200 & 16
--fft-length: set to 512 & 64 when --occupied-tones=16
d) In Lab#2, calculate the center frequencies for the sub-carriers when fft-length is 64 and occupied-tones is 16.
e) In Lab#3, explain benchmark_ofdm_rx.py options (--decim)
f) In Lab#3, measure PDR and BER
g) In Lab#3, compare PDR & BER with two other cases (fft-length 512 and occupied-tones 128, fft-length 64 and occupied-tones 16)
h) In Lab#3, with the fft-length and occypied-tones values that provide the best PDR, change the tx-amplitude in the range of [0, 32767] to measure BER and PDR. Plot charts for BER/PDR versus tx-amplitude.

·         File: manet-test.tcl, manet-test2.tcl, manet-aodv.sh, manet-dsr.sh, mob, traffic

·         Lab requirements (Due Nov. 22)

·         1) Comparison of MANET routing algorithms

a) Answer questions in step 2: Where are initial positions of the three nodes? When do they start moving? Do they pause some time between movements? What are the three nodes’ first waypoints and their speeds? After reaching the waypoints, when do they move again? In “traffic”, which node is the source and which node is the destination? When does the traffic start? What is the packet size?

b) Describe your observations in step 3 (DSDV). Do you observe that the traffic does not go through in the middle of the simulation but does it as time progresses? What else do you observe? Also, describe your observations with AODV and DSR in steps 5 and 6. Explain the differences.

c) With DSDV, how often does node 0 send the message in step 4? Is this the same for other nodes? Are the message sizes the same? If not, why?

d) Presents PDR, average packet delay, routing control overhead in terms of bytes, and normalized routing overhead of DSDV, AODV and DSR in steps 4, 5 and 6.

e) According to DSR trace file format, which additional information about DSR does the trace tell you?

 2) Performance study in a (relatively) large-scale MANET

a) Plot performance metrics for both DSR and AODV versus pause-times, for each CBR Load. State if any peculiar behavior is observed in steps 8 and 9.

 3) Confidence in your simulation results

a) What is CI with 100% CL?

b) What is the equation for CI corresponding to 99% CL?

c) In step 10, what are the 10 seed numbers (including 1) you used and the 10 delay values? What is the mean and standard deviation of delay measurements? What is your CI with 95% CL? Discuss if this CI is satisfactory.

·         File: infra.tcl, fil_tcp.awk

·         Lab requirements (Due Nov. 29)

·         1) With infra.tcl & fil_tcp.awk, explain when and why throughput changes. Also, describe what you observed with nam animation.

2) Repeat the same scenario of infra.tcl but with UDP traffic. Plot a chart that compares throughput of TCP, UDP (0.05), UDP (0.01) and UDP (0.005) and discuss the differences.

3) With the TCP traffic in infra.tcl, trace the variable using the periodic probing method and variable tracing method. Plot the chart that shows both of them and explain the differences between the two methods.

§         Presentation must be based on powerpoint slides (~20 slides).

§         Presentation should be no more than 20 minutes excluding Q&A.

§         Try to motivate the audience by giving them enough background rather than explaining the issues in great detail. Use figures and charts as many as possible. Remember also that this project is experiment-based. Give enough information about the experiment setup, (expected) results, and discussions on the results.

§         Order of presentation: Amar & Durga, Shahana & Sushma, Ahlam & Heli, Dhruv & Teja, Vivek & Shasvat

§         Report is due Friday, Dec. 14. It must be 3-5 single-spaced, single-column pages with 11-point font. The report consists of title, author names, abstract, introduction, background, experiment setup and results, discussion and conclusions, and references (authors, title, and source).

 

 

Links and Notices

Maintained by Chansu Yu (c.yu91@csuohio.edu)