R/LPIparam.default.gdf.R

In ilkkavir/LPI.gdf: Lag Profile Inversion with gdf files

## file:LPIparam.default.gdf.R
## (c) 2010- University of Oulu, Finland
## Written by Ilkka Virtanen <ilkka.i.virtanen@oulu.fi>
## Licensed under FreeBSD license.

##
## Default LPI analysis parameters.
##
## Arguments:
##  None
##
## Returns:
##  LPIparam A named list of parameters
## 

LPIparam.default.gdf <- function()
  {
    return(
      list(
        # Cluster definition
        clusterNodes       = list(tesla1=8,tesla2=8,tesla3=8,tesla4=8,tesla5=8,localControl=TRUE),
        # Receiver filter length in us
        filterLength.us    = 10,
        # Voltage level decoding filter.
        decodingFilter     = c("none","matched","inverse"),
        # Time-lag limits in us
        lagLimits.us       = seq(50)*10,
        # Range-gate limits in km
        rangeLimits.km     = c(seq(50,149.9,by=.9),seq(154.4,298.4,by=4.5),seq(343.4,793.5,by=45)),
        # Maximum range for each time-lag, will be
        # padded with the last value as necessary
        maxRanges.km       = Inf,
        # Time resolution (integration time) in s
        timeRes.s          = 10,
        # Maximum clutter suppression range, use negative
        # value to guarantee that suppression is not used
        maxClutterRange.km = 60,
        # Ground clutter decoding integration time,
        # fraction of the full integration period
        clutterFraction    = 1,
        # Background noise autocovariance function estimation
        backgroundEstimate = TRUE,
        # Analysis start time
        beginTime          = c(2000,1,1,0,0,0),
        # Analysis end time
        endTime            = c(2050,1,1,0,0,0),
        # Maximum time to wait for new data 
        # before stopping the analysis [s]
        maxWait.s          = -1,
        # Frequency offset(s) from baseband
        freqOffset.Hz      = 0,
        # Adjustments to TX / RX indices
        # (written in this way to remind that shifts for
        # TX and RX are most probably different)
        indexShifts.us     = list( TX=c(0,0) , RX=c(0,0) ), 
        # Invserse problem solver
        # "rlips", "fishs", "dummy", "deco", or "ffts"
        solver             = "fishs",
        # Number of theory rows to produce
        # before calling the solver
        nBuf               = 10000,
        # Calculate full posterior covariance matrix
        # (set FALSE to calculate only the variances)
        fullCovar          = FALSE,
        # Options to rlips
        rlips.options      = list( type="c" , nbuf=1000 , workgroup.size=128),
        # Is the receiver data from a remote site?
        remoteRX           = FALSE,
        # Should the TX samples be normalised to common
        # amplitude after decimation? Setting this may
        # improve the first lag in matched-filter decoding
        normTX             = FALSE,
        # Number of pulses in a full modulation cycle. Used
        # only for lag-profile pre-averaging, set to NA or
        # negative value to disable the pre-averaging
        nCode              = NA,
        # Should linear interpolation be applied when
        # calculating the range-ambiguity functions
        # (setting to FALSE will work only if the receiver
        # filter is short compared to baud length,
        # or if strong codes are used)
        ambInterp          = FALSE,
        # lower limit for number of samples to average in noise
        # estimation. The global average is used for data points
        # with less than minNpower samples
        minNpower          = 100,

        # threshold for noise power detection. Points with amplitude
        # squared largern than noiseSpikeThreshold are rejected
        noiseSpikeThreshold = 10,
        
        
        # Functions for data input. Tthese functions
        # will need to be replaced in order to apply
        # new input formats. The replacement functions
        # must be included in a package listed in
        # requiredPackages, and exported in the
        # packages NAMESPACE file.
             
        # Additional package(s) needed for data input,
        # LPI and parallel are always loaded
        inputPackages      = c( "LPI.gdf" ),
        # Additional functions to call immediately after
        # the first parameter collection step with the LPI
        # parameter list as argument. These must
        # return lists, which will be concatenated to the
        # LPI parameter list.
        extraFunctions     = c( "LPIexpandInput.gdf" , "getFileLengths.gdf" ) ,
        # Function for reading full integration
        # periods of raw data
        dataInputFunction  = "readLPIdata.gdf",
        # Function for reading sampling start times
        startTimeFunction  = "getSamplingStartTimes.gdf",
        # Function for reading sampling end times
        dataEndTimeFunction    = "getDataEndTimes.gdf",
        # function for readind data sample rates
        sampleRateFunction = "getSamplingFrequencies.gdf",

        # Additional input that the gdf input routines require
        # Raw data directory(ies)
        dataDir            = '.',
        # File name prefix(es)
        fileNamePrefix     = "data-",
        # File name extension(s)
        fileNameExtension  = ".gdf",

        # Results saving function
        resultSaveFunction = "LPIsaveACF.gdf",

        # Site location information
        llhT = c( 69.58, 19.23, 86.00 ),
        llhR = c( 69.58, 19.23, 86.00 ),
        azelT = c(0,90),
        azelR = c(0,90),
        radarFreq = 224e6,

        # should the final acf be complex-conjugated (i.e. the spectrum mirrored)
        # this is needed e.g. in EISCAT VHF which mirrors the received signal
        mirrorSpectra = FALSE,
 
        # Output directory, set NA to prevent creation of a directory
        resultDir          = paste(format(Sys.time(),"%Y-%m-%d_%H:%M"),'LP',sep='_')
        
        )
      )
  }