This project is a data preprocessor for DAS systems. It is designed to receive packets from DAS system, concatenate them in chunks of defined size and save them as separate files and.
To run the project you need to have Python 3.9 or higher installed on your machine.
Before running the project you have to configure parameters in the config.ini
file:
NAME
- name of the DAS system (e.g. Mekorot
or Prisma
)
LOCALPATH
is absolute PATH to the LOCAL directory, DAS client will write packets to LOCALPATH/YYYYMMDD
.
NASPATH_final
is absolute PATHs to the NAS directory which will contain concatenated hdf5
files.
CHUNK_SIZE
is the size of the chunk in seconds. By default 300.
SPS
is expected time frequency after data downsamling (in Hz). By default 100.
DX
is expected spatial spacing after data downsampling (in m). By default 9.6
- Files are named according to the following convention:
- YYYY/YYYYMMDD/.h5
- YYYY - year of the recording in UTC
- YYYYMMDD - date of the recording in UTC
- - timestamp of the beginning of the chunk
- YYYY/YYYYMMDD/.h5
- Data is stored in .h5 format
- Data is located in data_down dataset
- Each point stored as float32
- Data is located in data_down dataset
- Metadata saved in attributes. Contents of the metadata can vary depending on the system and date of recording.
- Always present:
- DX_down - spatial sampling rate after downsampling
- PRR_down - temporal sampling rate after downsampling
- down_factor_space, down_factor_time - downsampling factors
- May be present:
- Gauge_m - (Prisma specific) gauge length of the system
- Index, Origin, Spacing - (Mekorot specific) packet-wise original data descriptors
- Always present: