Changes between Version 3 and Version 4 of DSC/Hurdle2


Ignore:
Timestamp:
Mar 30, 2013, 3:35:27 AM (11 years ago)
Author:
seskar
Comment:

—

Legend:

Unmodified
Added
Removed
Modified

  • DSC/Hurdle2

    v3 v4  
    1 = DARPA Spectrum Challenge Hurdle 2 (H2) =
     1== DARPA Spectrum Challenge Hurdle 2 (H2) ==
    22
    3 == Objective ==
     3[[TOC(DSC*,depth=2)]]
     4
     5=== Objective ===
    46
    57The main objective of Hurdle 2 is to demonstrate an understanding of the basic principles of wireless communications and familiarity with USRP2/USRP N210 and GNU radio software tools by properly configuring and executing a basic transmitter/receiver communication system on a pair of radios on the ORBIT grid. The objective is to deliver a minimum of 2000 error-free packet in 12 seconds or less.  This is a "pass/fail" objective, and no benefit is obtained by achieving higher performance than other teams.  Upon successful completion, instructions for Hurdle 3 will be sent to the team leader. Teams are referred to [wiki:Documentation/GNURadio/Tutorials/Sandbox3 USRP tutorial] for general information regarding USRP2's or N210's on ORBIT.
    68
    7 == Instructions ==
     9=== Instructions ===
    810
    911WINLAB has created 16 virtual domains containing USRP2s and USRP N210s. Given the large quantity of teams participating in the Challenge hurdles, teams have been assigned specific times slots during which they may access specified virtual domains containing the USRP2 and USRP N210s. To ensure fairness, domains and time slots have been assigned such that teams rotate through different domains, and hence no team will have a significantly larger amount of time on a particular domain compared to other teams.  At the end of each 2-hour time slot, teams will be automatically removed from their domain to make that domain available for the next team. Please be advised that this means that experiments that have not completed will be terminated.
    … …  
    7173Teams should be aware that the receiver has been modified to timeout after 1 minute.  If teams have not chosen their parameters appropriately, then the receiver will terminate without ever reaching 2000 successful packets.   The 1 minute timer starts as soon as benchmark_rx2 is initiated.  Teams should initiate benchmark_tx2 promptly after initiating benchmark_rx2 to avoid having the receiver terminate prematurely.  Also, teams should note that the packet source used by the transmitter only supports a limited number of simultaneous packet streams, and thus teams are advised that they may need to retry benchmark_tx2 if there are many teams making simultaneous demands on the packet server.
    7274
    73 == Validation ==
     75=== Validation ===
    7476
    7577The radio modules benchmark_tx2 and benchmark_rx2 have been modified to log packets and parameters to the OML server with a database name that has the node name in it. Teams are evaluated on whether their first 2000 successfully delivered packets were delivered in 12 seconds or less. The evaluation time is calculated from the time the first packet is received to the time the 2000th correct packet is received.  After reception of the 2000th correct packet, benchmark_rx2 reports the total elapsed time. The receiver benchmark_rx2 has been modified to provide a final summary report describing the number of packets successfully received in one minute (at timeout) in case teams do not successfully receive 2000 packets. Teams are also advised that they may use the verbose flag --v  in the command line to debug their parameter selection. The use of the verbose flag causes some noticeable degradation in USRP performance, so teams should take this fact into account during testing.