Loading Files

The raw Schottky signals in the Radio Frequency (RF) range, after being mixed with a reference sinusoid down to the Intermediate Frequency (IF), are sampled and digitized by the spectrum analyzer into the so-called IQ (In-phase & Quadrature) data. Those data are then continuously streamed to the disk to form individual data files of various sizes according to the preset acquisition parameters.

Currently two file formats, namely the wv and the tiq, are supported by the script preprocessing.py, in which a dedicated class Preprocessing handles the IO interface with those kinds of files. Apart from reading IQ pairs from the data block, the class is also capable of extracting the acquisition-relevant meta-data from the header section, and parsing them in a suitable manner. The digital Schottky signal can thus be re-constructed from the combined information of IQ pairs and meta-data.

Table of Contents

• Class Structure
  • extract_wv
  • extract_tiq
  • display
  • load
  • draw
  • diagnosis

Class Structure

Preprocessing(file_str, verbose=True)

Instantiate the class.

• arguments:
  • file_str: str
    Name of the data file. The file format can only be recognized from the file extension, which is case insensitive. *.tiq implies a tiq file; *.wvh or *.wvd implies a wv file.
  • verbose: bool, default True
    If True, print the data acquisition parameters onto the screen.

To divert the IO load when the file is extremely large so as to improve the executive efficiency, the class sets an upper limit on the IQ pairs to be read at once, as is specified with a global attribute n_buffer. Its default value is 0.1 million.

extract_wv

extract_wv()

Extract the meta-data from a wv file and parse them.

• arguments:
  • None
• returns:
  • None
    The parsed meta-data are stored as class attributes. Some frequently referred ones can be ref_level [dBm], center_frequency [Hz], span [Hz], and sampling_rate [Hz].

extract_tiq

extract_tiq()

Extract the meta-data from a tiq file and parse them.

• arguments:
  • None
• returns:
  • None
    The parsed meta-data are stored as class attributes. Some frequently referred ones can be ref_level [dBm], center_frequency [Hz], span [Hz], and sampling_rate [Hz].

display

display()

Print all the data acquisition parameters onto the screen. The displayed message is file-format-aware.

• arguments:
  • None
• returns:
  • None

load

load(size, offset, decimating_factor=1, draw=False)

Load a number of IQ pairs from the file starting at the beginning with an offset. The actual loaded data size may be smaller than the given one, due to the insufficient samples in the file. In such a case, the remaining samples from the offset will all be loaded. The IQ pairs may also be loaded in a stride mode, where only one pair is adopted among every a few other contiguous pairs.

• arguments:
  • size: int
    Number of IQ pairs to be loaded.
  • offset: int
    Offset from the beginning of the file where the data start to load.
  • decimating_factor: int, default 1
    The stride distance between two neighboring IQ pairs to be loaded. 1 means contiguous loading. If decimating_factor is larger than 1, the size still counts the total number of loaded IQ pairs.
  • draw: bool, default False
    If True, also plot the loaded IQ pairs.
• returns:
  • times: numpy array
    A real array of time stamps in units of second.
  • data: numpy array
    A complex array of data sequences in units of volt.

draw

draw(t, signal)

Plot the signal in the in-phase and quadrature parts as a function of time.

• arguments:
  • t: numpy array
    A real array specifying the time.
  • signal: numpy array
    A complex array specifying the signal.
• returns:
  • None

diagnosis

diagnosis(n_point=None, draw=True)

Inspect the acquired data in a file by plotting them in the in-phase and quadrature parts.

• arguments:
  • n_point: int, default None
    If None, n_point will be set to n_buffer. If negative, plot all the data contained in the file, after divided into a few segments in compliance with the n_buffer constraint. Otherwise, plot the decimated data, if necessary, with size no more than n_point from the beginning of the file through the end.
  • draw: bool, default True
    If True, also plot the data in the file.
• returns:
  • times: numpy array, optional
    If n_point is None or positive, return a real array of time stamps in units of second.
  • data: numpy array, optional
    If n_point is None or positive, return a complex array of data sequences in units of volt.