Python decoder for Mode S and ADS-B signals





GitHub Stars



Last Commit

3mos ago










The Python ADS-B/Mode-S Decoder

PyModeS is a Python library designed to decode Mode-S (including ADS-B) message. It can be imported to your python project or used as a standalone tool to view and save live traffic data.

This is a project created by Junzi Sun, who works at TU Delft <>, Aerospace Engineering Faculty <>, CNS/ATM research group <>. It is supported by many contributors <> from different institutions.


pyModeS supports the decoding of following types of messages:

  • DF4 / DF20: Altitude code

  • DF5 / DF21: Identity code (squawk code)

  • DF17 / DF18: Automatic Dependent Surveillance-Broadcast (ADS-B)

    • TC=1-4 / BDS 0,8: Aircraft identification and category
    • TC=5-8 / BDS 0,6: Surface position
    • TC=9-18 / BDS 0,5: Airborne position
    • TC=19 / BDS 0,9: Airborne velocity
    • TC=28 / BDS 6,1: Airborne status [to be implemented]
    • TC=29 / BDS 6,2: Target state and status information [to be implemented]
    • TC=31 / BDS 6,5: Aircraft operational status [to be implemented]
  • DF20 / DF21: Mode-S Comm-B messages

    • BDS 1,0: Data link capability report
    • BDS 1,7: Common usage GICB capability report
    • BDS 2,0: Aircraft identification
    • BDS 3,0: ACAS active resolution advisory
    • BDS 4,0: Selected vertical intention
    • BDS 4,4: Meteorological routine air report (experimental)
    • BDS 4,5: Meteorological hazard report (experimental)
    • BDS 5,0: Track and turn report
    • BDS 6,0: Heading and speed report

If you find this project useful for your research, please considering cite this tool as::

@article{sun2019pymodes, author={J. {Sun} and H. {V\^u} and J. {Ellerbroek} and J. M. {Hoekstra}}, journal={IEEE Transactions on Intelligent Transportation Systems}, title={pyModeS: Decoding Mode-S Surveillance Data for Open Air Transportation Research}, year={2019}, doi={10.1109/TITS.2019.2914770}, ISSN={1524-9050}, }


Check out and contribute to this open-source project at:

Detailed manual on Mode-S decoding is published at:

The API documentation of pyModeS is at:

Basic installation

Installation examples::

stable version

pip install pyModeS

conda (compiled) version

conda install -c conda-forge pymodes

development version

pip install git+

Dependencies numpy, and pyzmq are installed automatically during previous installations processes.

If you need to connect pyModeS to a RTL-SDR receiver, pyrtlsdr need to be installed manually::

pip install pyrtlsdr

Advanced installation (using c modules)

If you want to make use of the (faster) c module, install pyModeS as follows::

conda (compiled) version

conda install -c conda-forge pymodes

stable version (to be compiled on your side)

pip install pyModeS[fast]

development version

git clone cd pyModeS pip install .[fast]

View live traffic (modeslive)

General usage::

$ modeslive [-h] --source SOURCE [--connect SERVER PORT DATAYPE][--latlon LAT LON] [--show-uncertainty][--dumpto DUMPTO]

arguments: -h, --help show this help message and exit --source SOURCE Choose data source, "rtlsdr" or "net" --connect SERVER PORT DATATYPE Define server, port and data type. Supported data types are: ['raw', 'beast', 'skysense'] --latlon LAT LON Receiver latitude and longitude, needed for the surface position, default none --show-uncertainty Display uncertainty values, default off --dumpto DUMPTO Folder to dump decoded output, default none

Live with RTL-SDR

If you have an RTL-SDR receiver connected to your computer, you can use the rtlsdr source switch (require pyrtlsdr package), with command::

$ modeslive --source rtlsdr

Live with network data

If you want to connect to a TCP server that broadcast raw data. use can use net source switch, for example::

$ modeslive --source net --connect localhost 30002 raw $ modeslive --source net --connect 30005 beast

Example screenshot:

.. image:: :width: 700px

Use the library

.. code:: python

import pyModeS as pms

Common functions

.. code:: python

pms.df(msg) # Downlink Format pms.icao(msg) # Infer the ICAO address from the message pms.crc(msg, encode=False) # Perform CRC or generate parity bit

pms.hex2bin(str) # Convert hexadecimal string to binary string pms.bin2int(str) # Convert binary string to integer pms.hex2int(str) # Convert hexadecimal string to integer pms.gray2int(str) # Convert grey code to integer

Core functions for ADS-B decoding

.. code:: python

pms.adsb.icao(msg) pms.adsb.typecode(msg)

Typecode 1-4


Typecode 5-8 (surface), 9-18 (airborne, barometric height), and 20-22 (airborne, GNSS height)

pms.adsb.position(msg_even, msg_odd, t_even, t_odd, lat_ref=None, lon_ref=None) pms.adsb.airborne_position(msg_even, msg_odd, t_even, t_odd) pms.adsb.surface_position(msg_even, msg_odd, t_even, t_odd, lat_ref, lon_ref) pms.adsb.surface_velocity(msg)

pms.adsb.position_with_ref(msg, lat_ref, lon_ref) pms.adsb.airborne_position_with_ref(msg, lat_ref, lon_ref) pms.adsb.surface_position_with_ref(msg, lat_ref, lon_ref)


Typecode: 19

pms.adsb.velocity(msg) # Handles both surface & airborne messages pms.adsb.speed_heading(msg) # Handles both surface & airborne messages pms.adsb.airborne_velocity(msg)

Note: When you have a fix position of the aircraft, it is convenient to use position_with_ref() method to decode with only one position message (either odd or even). This works with both airborne and surface position messages. But the reference position shall be within 180NM (airborne) or 45NM (surface) of the true position.

Decode altitude replies in DF4 / DF20

.. code:: python

pms.common.altcode(msg) # Downlink format must be 4 or 20

Decode identity replies in DF5 / DF21

.. code:: python

pms.common.idcode(msg) # Downlink format must be 5 or 21

Common Mode-S functions

.. code:: python

pms.icao(msg) # Infer the ICAO address from the message pms.bds.infer(msg) # Infer the Modes-S BDS register

Check if BDS is 5,0 or 6,0, give reference speed, track, altitude (from ADS-B)

pms.bds.is50or60(msg, spd_ref, trk_ref, alt_ref)

Check each BDS explicitly

pms.bds.bds10.is10(msg) pms.bds.bds17.is17(msg) pms.bds.bds20.is20(msg) pms.bds.bds30.is30(msg) pms.bds.bds40.is40(msg) pms.bds.bds44.is44(msg) pms.bds.bds50.is50(msg) pms.bds.bds60.is60(msg)

Mode-S Elementary Surveillance (ELS)

.. code:: python

pms.commb.ovc10(msg) # Overlay capability, BDS 1,0 pms.commb.cap17(msg) # GICB capability, BDS 1,7 pms.commb.cs20(msg) # Callsign, BDS 2,0

Mode-S Enhanced Surveillance (EHS)

.. code:: python

BDS 4,0

pms.commb.selalt40mcp(msg) # MCP/FCU selected altitude (ft) pms.commb.selalt40fms(msg) # FMS selected altitude (ft) pms.commb.p40baro(msg) # Barometric pressure (mb)

BDS 5,0

pms.commb.roll50(msg) # Roll angle (deg) pms.commb.trk50(msg) # True track angle (deg) pms.commb.gs50(msg) # Ground speed (kt) pms.commb.rtrk50(msg) # Track angle rate (deg/sec) pms.commb.tas50(msg) # True airspeed (kt)

BDS 6,0

pms.commb.hdg60(msg) # Magnetic heading (deg) pms.commb.ias60(msg) # Indicated airspeed (kt) pms.commb.mach60(msg) # Mach number (-) pms.commb.vr60baro(msg) # Barometric altitude rate (ft/min) pms.commb.vr60ins(msg) # Inertial vertical speed (ft/min)

Meteorological reports [Experimental]

To identify BDS 4,4 and 4,5 codes, you must set mrar argument to True in the infer() function:

.. code:: python

pms.bds.infer(msg. mrar=True)

Once the correct MRAR and MHR messages are identified, decode them as follows:

Meteorological routine air report (MRAR) +++++++++++++++++++++++++++++++++++++++++

.. code:: python

BDS 4,4

pms.commb.wind44(msg) # Wind speed (kt) and direction (true) (deg) pms.commb.temp44(msg) # Static air temperature (C) pms.commb.p44(msg) # Average static pressure (hPa) pms.commb.hum44(msg) # Humidity (%)

Meteorological hazard air report (MHR) +++++++++++++++++++++++++++++++++++++++++

.. code:: python

BDS 4,5

pms.commb.turb45(msg) # Turbulence level (0-3) pms.commb.ws45(msg) # Wind shear level (0-3) pms.commb.mb45(msg) # Microburst level (0-3) pms.commb.ic45(msg) # Icing level (0-3) pms.commb.wv45(msg) # Wake vortex level (0-3) pms.commb.temp45(msg) # Static air temperature (C) pms.commb.p45(msg) # Average static pressure (hPa) pms.commb.rh45(msg) # Radio height (ft)

Customize the streaming module

The TCP client module from pyModeS can be re-used to stream and process Mode-S data as you like. You need to re-implement the handle_messages() function from the TcpClient class to write your own logic to handle the messages.

Here is an example:

.. code:: python

import pyModeS as pms from pyModeS.extra.tcpclient import TcpClient

define your custom class by extending the TcpClient

- implement your handle_messages() methods

class ADSBClient(TcpClient): def init(self, host, port, rawtype): super(ADSBClient, self).init(host, port, rawtype)

  def handle_messages(self, messages):
      for msg, ts in messages:
          if len(msg) != 28:  # wrong data length

          df = pms.df(msg)

          if df != 17:  # not ADSB

          if pms.crc(msg) !=0:  # CRC fail

          icao = pms.adsb.icao(msg)
          tc = pms.adsb.typecode(msg)

          # TODO: write you magic code here
          print(ts, icao, tc, msg)

run new client, change the host, port, and rawtype if needed

client = ADSBClient(host='', port=30005, rawtype='beast')

Unit test

To perform unit tests, pytest must be install first.

Build Cython extensions ::

$ make ext

Run unit tests ::

$ make test

Clean build files ::

$ make clean

Rate & Review

Great Documentation0
Easy to Use0
Highly Customizable0
Bleeding Edge0
Responsive Maintainers0
Poor Documentation0
Hard to Use0
Unwelcoming Community0