This repository is a fork of sdr4iot-ble-rx proposed in 2021 by Rtone.
The original project used a message sink block, which cannot be used that way in the new version of GNU Radio (3.10.5.1). Therefore, I will use ZMQ block instead to make the project operational again.
The operating environment of the original project:
Ubuntu 18.04.6 LTS+GNU Radio Companion 3.7.11
This tool was created to dump Bluetooth LE (BLE) packets using SDR hardware. The captured BLE packets can either be saved to a PCAP file or displayed directly in Wireshark via a named pipe (FIFO). Gnu Radio is used to receive and demodulate the incoming BLE packets. The received packet bytes are transferred to ble_dump using a common Gnu Radio Message Sink.
The following (simplified) Gnu Radio Companion (GRC) signal flow graph is used to generate the final flow-graph source code:
Usage: ble_dump.py: [opts]
Options:
-h, --help show this help message and exit
Capture settings:
-o PCAP_FILE, --pcap_file=PCAP_FILE
PCAP output file or named pipe (FIFO)
-m MIN_BUFFER_SIZE, --min_buffer_size=MIN_BUFFER_SIZE
Minimum buffer size [default=65]
-s SAMPLE_RATE, --sample-rate=SAMPLE_RATE
Sample rate [default=4000000.0]
-t SQUELCH_THRESHOLD, --squelch_threshold=SQUELCH_THRESHOLD
Squelch threshold (simple squelch) [default=-70]
Low-pass filter::
-C CUTOFF_FREQ, --cutoff_freq=CUTOFF_FREQ
Filter cutoff [default=850000.0]
-T TRANSITION_WIDTH, --transition_width=TRANSITION_WIDTH
Filter transition width [default=300000.0]
GMSK demodulation::
-S SAMPLES_PER_SYMBOL, --samples_per_symbol=SAMPLES_PER_SYMBOL
Samples per symbol [default=4]
-G GAIN_MU, --gain_mu=GAIN_MU
Gain mu [default=0.7]
-M MU, --mu=MU Mu [default=0.5]
-O OMEGA_LIMIT, --omega_limit=OMEGA_LIMIT
Omega limit [default=0.035]
Bluetooth LE::
-c CURRENT_BLE_CHANNELS, --current_ble_channels=CURRENT_BLE_CHANNELS
BLE channels to scan [default=37,38,39]
-w BLE_SCAN_WINDOW, --ble_scan_window=BLE_SCAN_WINDOW
BLE scan window [default=10.24]
-x, --disable_crc Disable CRC verification [default=False]
-y, --disable_dewhitening
Disable dewhitening [default=False]
Misc::
-i IQ_FILE, --iq-output=IQ_FILE
Filename for IQ data
Usage: iq_save.py: [opts]
Options:
-h, --help show this help message and exit
-c CSV_FILE, --csv-file=CSV_FILE
Csv file path where are recorded: [timestamp,start_trame,endtrame,frequency,sample_rate]
-d DATA_FILE, --data-file=DATA_FILE
IQ file path to be used to extract BLE IQ data
Usage: get_robot_position.py: [opts]
Options:
-h, --help show this help message and exit
-r ROBOT_NODE, --robot-node=ROBOT_NODE
Robot node corresponding to the BLE emitter
-o CSV_FILE, --csv-file=CSV_FILE
Csv file path where will be recorded: [timestamp,robot_node,x,y]
Usage: tag_iq_data.py: [opts]
Options:
-h, --help show this help message and exit
-r ROBOTCSV_FILE, --robotcsv-file=ROBOTCSV_FILE
Csv file path where are recorded robot position information: [timestamp,robot_node,x,y]
-p PACKETCSV_FILE, --packetcsv-file=PACKETCSV_FILE
Csv file path where are recorded BLE packet information: [timestamp,start_frame,end_frame,frequency,sample_rate]
-o OUTPUTCSV_FILE, --outputcsv-file=OUTPUTCSV_FILE
Csv file path which will be used to tag BLE packet information: [timestamp,start_frame,end_frame,frequency,sample_rate,robot_node,x,y]
Your SDR device should offer a minimum sample rate of 4.000.000 samples/sec. The SDR device as well should be usable as Gnu Radio osmocom/osmosdr source device. The HackRF One SDR was used for all Gnu Radio related tasks described within this document. Other SDRs should also work well.
Scan BLE advisory channels (37, 38, 39) and save received packets to PCAP file:
./ble_dump.py -o /tmp/dump1.pcap
Scan BLE advisory channels (37, 38, 39) and send received packets to FIFO:
mkfifo /tmp/fifo1
./ble_dump.py -o /tmp/fifo1
Display BLE packets from FIFO in Wireshark:
wireshark -S -k -i /tmp/fifo1
- Calculate the start/end indexes of IQ BLE data into the global IQ data
- Record timestamped robot positions into a csv file
- Record BLE packet information ['Timestamp','Start_trame','End_trame','Channel frequency','Sample_rate'] into a csv file
- Use a linear interpolation method for robot position estimation for a detected packet
- Tag the IQ BLE data by using the estimaded robot positions
- Extract IQ BLE data from the global IQ data by using the csv file and the whole IQ data file
- Save the extracted IQ BLE data and useful information (Start_Frame, Sample count, Central Frequency, Sample rate, robot positions X & Y) into a descriptive format(sigmf)
# install sigmf via pip
pip install sigmf- The captured BLE packets are stored as "Bluetooth Low Energy Link Layer" (btle) format
- BLE Data and BLE Control packets are currently not supported
- If the default hopping pattern is used, and you want to receive BLE data it should be possible to hop only a limited number of BLE data channels. Keep in mind that the initial CRC value (not 0x555555) is essential to determine the validity of incoming BLE data packets.
- Feel free to help and improve the code!
The generated Gnu Radio Companion (GRC) signal flow graph (grc/gr_ble.py) was slightly modified to avoid errors. If you re-generate the GRC flow graph please run to following command-line:(No longer needed)
sed -i -e "s/message_sink_msgq_out,/message_queue,/" -e "s/message_sink_msgq_out = virtual_sink_msgq_in/self.message_queue = message_queue/" ./grc/gr_ble.py
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent