Open-source modular toolkit for simulating high-fidelity pulsar X-ray events.
OpenXNAV is designed to aid development and testing of Pulsar-based Autonomous Navigation (XNAV) Positioning, Navigation, and Timing (PNT) solutions.
-You can read our IEEE paper to learn more: IEEE.org.
+This tool is a flexible, cost-effective testbed to enable the next generation of XNAV solutions. OpenXNAV includes three modular components that allow engineers simulate high-fidelity pulsar X-ray events along a flight trajectory over a user-defined mission timeline. Hardware-in-the-loop functionality can be configured using software-defined radios (SDRs), and users can query for pulsar candidates using the query tool (powered by the ATNF Pulsar Catalogue).
+For more information, you can read our IEEE paper: IEEE.org.
Open-source modular toolkit for simulating high-fidelity pulsar X-ray events.
"},{"location":"#introduction-to-openxnav","title":"Introduction to OpenXNAV","text":"OpenXNAV is designed to aid development and testing of Pulsar-based Autonomous Navigation (XNAV) Positioning, Navigation, and Timing (PNT) solutions.
You can read our IEEE paper to learn more: IEEE.org.
"},{"location":"#components","title":"Components","text":"Each component of OpenXNAV can be used individually, or strung together as an end-to-end pipeline. To get started with OpenXNAV, select the module you are interested in leveraging and access the documentation.
"},{"location":"components/#pulsar-querying","title":"Pulsar Querying","text":"Query for all known pulsars catalogued in the ATNF database by providing a coordinate in space, and a search radius.
"},{"location":"components/#mission-planning","title":"Mission Planning","text":"Model the desired flight path, taking into account spacecraft orientation, viewing angle and line of sight.
"},{"location":"components/#timing-event-generation","title":"Timing & Event Generation","text":"Generate representative photon arrival events at the location of interest along the user-defined trajectory.
"},{"location":"components/1__pulsar_querying/pq_overview/","title":"Pulsar Querying","text":""},{"location":"components/1__pulsar_querying/pq_overview/#installation","title":"Installation","text":"Before using the OpenXNAV Pulsar Querying Tool, you will need to install the necessary python packages. You can do this using your preferred python package manager, but instructions are provided below for using Anaconda
Inside the open-xnav/1__pulsar_querying/ directory, there is an environment.yml file. This file can be passed to your preferred package manager to create a new environment that can run the tool.
The default environment name in environment.yml is openXNAV-gui. If you wish to change the environment name, simply edit environment.yml before creating the environment.
Open your Anaconda Prompt (or activate Anaconda in a terminal window). Navigate to open-xnav/`1__pulsar_querying/ and run the following command:
conda env create -f environment.yaml\nTo activate your environment, run the following command:
conda activate openXNAV-gui\nThe tool also depends on KivyMD, which is a library built off of the popular Kivy library. This package is not available through conda, so you must install it using pip.
First, make sure you have activated the openXNAV-gui environment in your terminal. Then, run the following command:
pip install kivymd\nIf you are getting a ReadTimeoutError response during installation, you might be using a VPN or have a slower internet connection. If this happens, try running the following instead:
pip install --default-timeout=1000 kivymd\nIf that still does not work, you can troubleshoot your issues further using the pip documentation here.
Once you have successfully installed all the required packages, you are ready to run the tool!
First, ensure you have navigated to open-xnav/1__pulsar_querying/ in your terminal and have your openXNAV-gui environment activated.
Then, simply run the following in your terminal:
python run.py\nUpon startup, the tool will need to pull the current ATNF Pulsar Catalogue from the web. This process only need to be done the first time you open the application.
However, if you wish to update the database, simply delete the .../pulsar_database/ directory from your file system. Then, relaunch the application. Once the database has been generated, you can continue begin querying.
You should see the OpenXNAV launch window with the blue Launch Application button active. In the future, when you open the application, the tool will locate the previously-generated database and you can quickly continue to the querying functionality.
After you click Launch Application, the query tool will display.
There are three fields that you need to provide to execute a query:
These values will create a query, searching for all known pulsars that are within a given search radius of the provided coordinates in space.
The OpenXNAV tool leverages the open-source psrqpy library. More information about this package, and how queries are executed, can be found here.
The query results will display to the GUI interface, and will also be stored in a sub-directory within .../query_results/.
For information about how these sub-directories are tagged, see below.
"},{"location":"components/1__pulsar_querying/pq_overview/#tagging-results","title":"Tagging Results","text":"There is an optional Tag field that OpenXNAV also provides. This allows you to specify the tag for this query you are about to execute. If a tag is provided, that will be used to name the sub-directory within .../query_results/ where all pulsars returned by the query are stored.
Sample Queries
Below are couple sample queries that you can run to get familiar with the OpenXNAV tool. There are additional sample query results contained in the .../query_results/ directory.
Right now, the OpenXNAV Pulsar Querying Tool only outputs pulsar data into the .st format. This is the format that is required by STK to perform subsequent mission planning.
To display the format of this file, here is J00012_5431.st as an example:
stk.v.12.0\nWrittenBy OpenXNAV\n\nBEGIN Star\n\n Name J0012_5431\n\n BEGIN PathDescription\n\n Epoch 58912.0\n RefFrame J2000\n RightAscension 3.097083333333333\n Declination +54:31:47\n ProperMotionRAPerYr 0\n ProperMotionDecPerYr 0\n Parallax 0\n RadialVelocity 0.0000000000000000e+00\n\n END PathDescription\n\n BEGIN PhysicalData\n\n Magnitude None\n\n END PhysicalData\n\n BEGIN IdentityData\n\n Id 0\n\n END IdentityData\n\n\n BEGIN Extensions\n\n BEGIN ExternData\n END ExternData\n\n BEGIN ADFFileData\n END ADFFileData\n\n BEGIN AccessConstraints\n LineOfSight IncludeIntervals\n\n UsePreferredMaxStep No\n PreferredMaxStep 360\n END AccessConstraints\n\n BEGIN Desc\n END Desc\n\n BEGIN Crdn\n END Crdn\n\n BEGIN Graphics\n\n BEGIN Attributes\n\n MarkerColor #00ff00\n LabelColor #00ff00\n MarkerStyle 2\n FontStyle 0\n\n END Attributes\n\n BEGIN Graphics\n\n Show On\n Inherit On\n ShowLabel On\n ShowMarker On\n\n END Graphics\n END Graphics\n\n BEGIN VO\n END VO\n\n END Extensions\n\nEND Star\nYou will need to provide these .st file(s) to STK to successfully execute OpenXNAV's mission planning functionality.
However, if you would like pulsar information to be stored/saved in a different format, you can modify the source code.
Navigate to queryPulsar.py and locate the following function:
def saveToFile(self, root_directory) ...\n.../query_results/."},{"location":"components/2__mission_planning/mp_overview/","title":"Mission Planning","text":"In this Python notebook, the mission_planning module from the OpenXNAV library is demonstrated.
We start by importing the necessary libraries:
from astropy import units as u\nfrom astropy.coordinates import SkyCoord,get_body\nfrom astropy.time import Time,TimeDelta\nfrom astropy.table import Table,QTable\n\nimport numpy as np\nimport matplotlib.pyplot as plt\nimport pandas as pd\nNext we define the necessary variables that we will use to instantiate the orbit. We will define an elliptical orbit directly by its orbital elements in this example. Other options, which are currently commented out, are a circular orbit (special case of elliptical orbit where eccentricity = 0) and an elliptical orbit defined by its burnout trajectory.
# User definitions for elliptical orbital elements\n\ne = 0.3 # eccentricity of the orbit\na = 750000*u.m # semi-major axis \ninc = 20*u.deg # inclination of orbit\nw = 90*u.deg # argument of periapsis\nv_0 = 0*u.deg # initial eccentric anomaly\nOmega = 90*u.deg # longitude of ascending node\n\nimport warnings\nwarnings.filterwarnings('ignore',message='leap-second file is expired')\nt = Time.now() + TimeDelta(120*u.day) + TimeDelta(np.linspace(0,800,51)*u.s)\n\n# Alternative: User definitions for elliptical orbital insertion \n# (note that this is just an example and would not generate the \n# same trajectory as the one above defined directly by orbital elements)\n\n#e = 0.1 # eccentricity of the orbit, also user defined\n#a = 750000*u.m # m semi-major axis \n#B = 70*u.deg # azimuth heading measured in degrees clockwise from north\n#dec = 20*u.deg # geocentric latitude in degrees (declination) of burnout\n#v_0 = 0*u.deg # initial eccentric anomaly\nWe can now do the same thing with a low-fidelity elliptical orbit, starting with the EllipticalOrbit class:
from mission_planning import EllipticalOrbit\n\neo = EllipticalOrbit(t,a,e,v_0=v_0,\n inc=inc,w=w,\n Omega=Omega)\n\nplt.figure()\nplt.plot(eo.x.to(u.km),eo.y.to(u.km),'b.')\n#plt.xlim([-a/u.km,a/u.km])\n#plt.ylim([-a/u.km,a/u.km])\nplt.show()\nWhile the above plot is represented in two dimensions as the projection of the orbit onto the XY-plane, this orbit was actually instantiated in 3D. To show the trajectory in the 3D Cartesian reference frame with the center of Earth at the origin, we just have to plot it in the 3D figure:
fig = plt.figure()\n\nax = fig.add_subplot(projection='3d')\nax.scatter(eo.x.to(u.km),eo.y.to(u.km),eo.z.to(u.km))\n\nplt.show()\nIn order to better demonstrate the pulsar access calculation functions in the OpenXNAV mission_planning module, we will create a SkyCoord object representing the moon, but with a distorted size and orbit to ensure some obfuscation of a chosen pulsar (whose coordinates are also exaggerated for demonstration purposes).
from mission_planning import plot_accesses\n\nmoon = SkyCoord(\n x = -284405*u.m,\n y = 249415*u.m,\n z = 0*u.m,\n frame='gcrs',\n representation_type = 'cartesian'\n)\n\n#moon_rad = 1740*u.km # actual radius of the moon \nmoon_rad = 150*u.km # fake news \n\n\n# assuming we have the location of the pulsars, using one as an example\npulsar = SkyCoord(\n x = -853215*u.m,\n y = 748245*u.m,\n z = 0*u.km,\n frame='gcrs',\n representation_type = 'cartesian'\n)\n\naccess = eo.pulsar_access(pulsar,moon,moon_rad)\n\n# create figure to plot pulsar access\nfig = plt.figure()\n\nax = fig.add_subplot(projection='3d')\nax.scatter(pulsar.x.to(u.km),pulsar.y.to(u.km),pulsar.z.to(u.km),label='pulsar')\nax.scatter(eo.x.to(u.km),eo.y.to(u.km),eo.z.to(u.km),label='spacecraft')\nax.scatter(moon.x.to(u.km),moon.y.to(u.km),moon.z.to(u.km),label='moon',s=500)\nplt.legend()\n\nplt.show()\n\nfig = plt.figure()\nax = fig.add_subplot()\nax = plot_accesses(ax,t,{'pulsar':access})\nWe can also instantiate an EllipticalOrbit object with high fidelity. This version of the class calculates the eccentric anomaly at each time in the time array in order to calculate a more exact position.
There is currently a bug in this version that causes it to only output positive x and positive z values, so we'll need to fix that. (Negative values shown below are there because x and z are positive before coordinate rotation to account for longitude of ascending node, orbital inclination, and argument of periapsis)
eo_h = EllipticalOrbit(t,a,e,v_0=v_0,\n inc=inc,w=w,\n Omega=Omega,\n hifi=True)\n\nplt.figure()\nplt.plot(eo_h.x,eo_h.y,'b.')\nplt.xlim([-a/u.m,a/u.m])\nplt.ylim([-a/u.m,a/u.m])\nplt.show()\n\nfig = plt.figure()\n\nax = fig.add_subplot(projection='3d')\nax.scatter(eo_h.x,eo_h.y,eo_h.z)\n\nplt.show()\nNext we will take advantage of the get_body and pulsar_access_export functionalities of Astropy and OpenXNAV Mission Planning, respectively.
Astropy allows us to create a SkyCoord object using the actual location of the moon at any time or array of times. We will make our moon much bigger than in real life to ensure that we obfuscate some pulsars for demonstration purposes. We will define a new orbit circular orbit using the CircularOrbit class, which is a special case of the EllipticalOrbit class in which eccentricity and orbital inclination are zero.
from mission_planning import CircularOrbit\n\n# User definitions for circular orbit\nr = 360000*u.km\nt0 = Time('2023-08-24 12:12:15.932083')\nt = t0 + TimeDelta(np.linspace(0,25,100)*u.day)\n\nmoon = get_body('moon',t)\nmoon.representation_type='cartesian'\nMOON_RAD = 1740*u.km\n\nearth = get_body('earth',t)\nearth.representation_type = 'cartesian'\nEARTH_RAD = 6378.14*u.km\n\nco = CircularOrbit(t,r)\nNext we will create an astropy.table.Table object with some pulsars that we want to load in to our pulsar access export function. The table is shown and created here for demonstration purposes; however, tables with this format can be created in the OpenXNAV pulsar query module.
name = [ 'J0002+6216' , 'J0006+1834' , 'J0007+7303' , 'J0011+08' , 'J0012+5431']\nraJ_deg = [ 0.74238 , 1.52 , 1.7571 , 2.9 , 3.0971 ] * u.deg\ndecJ_deg = [ 62.26928 , 18.5831 , 1.7571 , 8.17 , 54.5297 ] * u.deg\ndist_kpc = [ 6.357 , 0.86 , 1.4 , 5.399 , 5.425 ] * u.kpc\n\nt = Table([name,raJ_deg,decJ_deg,dist_kpc],\n names = ['NAME','RAJD','DECJD','DIST'])\n\nt.write('some_pulsars.fits',overwrite=True)\nprint(t)\n NAME RAJD DECJD DIST\n deg deg kpc \n---------- ------- -------- -----\nJ0002+6216 0.74238 62.26928 6.357\nJ0006+1834 1.52 18.5831 0.86\nJ0007+7303 1.7571 1.7571 1.4\n J0011+08 2.9 8.17 5.399\nJ0012+5431 3.0971 54.5297 5.425\npulsar_access_export method","text":"Now that we have our spacecraft, moon, and pulsar table, we can use the pulsar_access_export method to create a table and plot of pulsar accesses. Both this method and the pulsar_access function allow you to account for obfuscation from multiple celestial bodies; just make sure to define them as shown below, entered before the keyword arguments and alternating between the SkyCoord object representing the celestial body and the radius of the body represented as an astropy.units.Quantity object.
pulsars_qtbl = QTable.read('some_pulsars.fits')\n\n# 3D plot of satellite orbit and moon\nfig1 = plt.figure()\nax1 = fig1.add_subplot(projection='3d')\nax1.scatter(moon.x.to(u.km),moon.y.to(u.km),moon.z.to(u.km),label='moon',s=100)\nax1.scatter(co.x.to(u.km),co.y.to(u.km),co.z.to(u.km),label='satellite in cirular orbit')\nax1.scatter(earth.x.to(u.km),earth.y.to(u.km),earth.z.to(u.km),label='earth',s=500)\nplt.legend(loc='upper left')\n\n# Plot of pulsar accesses\n_,(fig2,ax2) = co.pulsar_access_export(pulsars_qtbl,\n moon,MOON_RAD,\n earth,EARTH_RAD,\n make_csv=False,\n make_fig=True,save_fig=False)\nNext we will define a new elliptical orbit from the EllipticalOrbit class in which eccentricity and orbital inclination are non-zero.
# User definitions for elliptical orbital elements\n\ne = 0.3 # eccentricity of the orbit\na = 30000*u.km # semi-major axis \ninc = 20*u.deg # inclination of orbit\nw = 90*u.deg # argument of periapsis\nv_0 = 0*u.deg # initial eccentric anomaly\nOmega = 90*u.deg # longitude of ascending node\n\nt0 = Time('2023-08-24 12:12:15.932')\nt = t0 + TimeDelta(np.linspace(0,25,1000)*u.day)\n\nmoon = get_body('moon',t)\nmoon.representation_type='cartesian'\nMOON_RAD = 1740*u.km\n\nearth = SkyCoord(x=0*u.m,y=0*u.m,z=0*u.m,frame='gcrs',representation_type='cartesian')\nEARTH_RAD = 6378.14*u.km\n\neo2 = EllipticalOrbit(t,a,e,v_0=v_0,\n inc=inc,w=w,\n Omega=Omega)\n\nfig = plt.figure()\n\nax = fig.add_subplot(projection='3d')\nax.scatter(eo2.x.to(u.km),eo2.y.to(u.km),eo2.z.to(u.km))\n\nplt.show()\nNext we will create a CSV with some pulsars that we want to load in to our pulsar access export function. The CSV is shown and created here for demonstration purposes:
name = [ 'J0002+6216' , 'J0006+1834' , 'J0007+7303' , 'J0011+08' , 'J0012+5431']\nraJ_deg = [ 0.74238 , 1.52 , 1.7571 , 2.9 , 3.0971 ] * u.deg\ndecJ_deg = [ 62.26928 , 18.5831 , 1.7571 , 8.17 , 54.5297 ] * u.deg\ndist_kpc = [ 6.357 , 0.86 , 1.4 , 5.399 , 5.425 ] * u.kpc\n\ntbl = Table([name,raJ_deg,decJ_deg,dist_kpc],\n names = ['NAME','RAJD','DECJD','DIST'])\n\ntbl.write('some_pulsars.fits',overwrite=True)\nprint(tbl)\n NAME RAJD DECJD DIST\n deg deg kpc \n---------- ------- -------- -----\nJ0002+6216 0.74238 62.26928 6.357\nJ0006+1834 1.52 18.5831 0.86\nJ0007+7303 1.7571 1.7571 1.4\n J0011+08 2.9 8.17 5.399\nJ0012+5431 3.0971 54.5297 5.425\npulsar_access_export method","text":"Now that we have our spacecraft, moon, and pulsar CSV, we can use the pulsar_access_export method to create a table and plot of pulsar accesses. Both this method and the pulsar_access function allow you to account for obfuscation from multiple celestial bodies; just make sure to define them as shown below, entered before the keyword arguments and alternating between the SkyCoord object representing the celestial body and the radius of the body represented as an astropy.units.Quantity object.
As part of the pulsar_access_export method call, we can write all pulsar access data to a CSV file. This CSV can then be used in the pulsar photon time-of-arrival simulation module to model the X-ray pulse sequence that the spacecraft would observe at those coordinates.
pulsars_qtbl = QTable.read('some_pulsars.fits')\n\n# 3D plot of satellite orbit\nfig1 = plt.figure()\nax1 = fig1.add_subplot(projection='3d')\nax1.scatter(eo2.x.to(u.km),eo2.y.to(u.km),eo2.z.to(u.km),label='satellite in elliptical orbit')\nax1.scatter(earth.x.to(u.km),earth.y.to(u.km),earth.z.to(u.km),label='earth',s=200)\nplt.legend(loc='upper left')\n\n# 3D plot of satellite orbit and moon\nfig2 = plt.figure()\nax2 = fig2.add_subplot(projection='3d')\nax2.scatter(moon.x.to(u.km),moon.y.to(u.km),moon.z.to(u.km),label='moon',s=100)\nax2.scatter(eo2.x.to(u.km),eo2.y.to(u.km),eo2.z.to(u.km),label='satellite in elliptical orbit')\nax2.scatter(earth.x.to(u.km),earth.y.to(u.km),earth.z.to(u.km),label='earth',s=200)\nplt.legend(loc='upper left')\n\n# Plot of pulsar accesses, and save pulsar access table to CSV\naccesses,(fig3,ax3) = eo2.pulsar_access_export(pulsars_qtbl,\n moon,MOON_RAD,\n earth,EARTH_RAD,\n make_csv=True,save_csv=True,\n make_fig=True,save_fig=False)\nThe breaks in the plot above might be hard to see since they are so small compared to the scale of the timeline displayed. We have therefore represented the breaks in tabular form below.
# Create DataFrame containing all times at which access to at least one pulsar is interrupted\naccess_breaks = accesses[accesses.all(axis=1) == False]\naccess_breaks\nAs an alternative to the mission_planning module built into OpenXNAV, users can also plan their missions and generate trajectory and pulsar access inputs using the mission planning software of their choice, such as ANSYS STK. STK is very visual, and a great option if you\u2019re starting your mission plan from scratch and want a lot of flexibility in designing your spacecraft trajectory.
STK users who choose this route should follow the following steps:
In this Python notebook, the event_generation module from the PLANTS library is demonstrated.
We start by importing the necessary libraries:
%load_ext autoreload\n%autoreload 2\nfrom pint_backend import *\nimport pulsar_definitions\npint.logging.setup(level=\"INFO\")\nLoads the ephemerides kernel
pint.solar_system_ephemerides.load_kernel(\"de440\")\n\u001b[1mINFO \u001b[0m (pint.solar_system_ephemerides ): \u001b[1mSet solar system ephemeris to de440 through astropy\u001b[0m\nCreates the time vector that the user hopes to observe at and adds additional error if desired. Specifies what pulsar model to use
time_interval = astrotime.Time( ['2022-07-11T16:00:01.000', '2022-07-11T16:21:01.000'], format='isot', scale='utc')\nobs_fs = 100e6\nn_toa = 10\nerror = 5 * u.us\napply_error = False\nmodel_name = \"j0218\" \nLoads in the navigation data provided by STK and generates a spacecraft object off of that navigation data. Note that spacecraft is a custom class created in the PINT backend
time, pos, vel = load_nav_data(\"XNAVSAT_TEMEofDate_Position_Velocity_JD.txt\")\nmy_spacecraft = SpaceCraft(\"my_spacecraft\", time, pos, vel, overwrite = True)\nGenerates the times of arrival at the barycenter compared to the spacecraft. Note that printing the time of arrival generated is slightly different at the two locations
if(model_name == \"j0218\"): \n pulsar_of_interest = PulsarObj(pulsar_definitions.J0218)\nelif(model_name == \"b1821\"):\n pulsar_of_interest = PulsarObj(pulsar_definitions.B1821)\ntoas_barycenter = pint.simulation.make_fake_toas_uniform(\n time_interval[0].mjd, time_interval[1].mjd, n_toa, model=pulsar_of_interest.model, freq = 1e15*u.Hz, obs='barycenter', error=error)\ntoas_spacecraft = pint.simulation.make_fake_toas_uniform(\n time_interval[0].mjd, time_interval[1].mjd, n_toa, model=pulsar_of_interest.model, freq = 1e15*u.Hz, obs = \"my_spacecraft\", error=error)\n\n\n\nprint(\"Barycenter first TOA: \" + str(toas_barycenter.get_mjds(\"True\")[0].utc.iso))\nprint(\"Spacecraft first TOA: \" + str(toas_spacecraft.get_mjds(\"True\")[0].utc.iso))\n\u001b[33m\u001b[1mWARNING \u001b[0m (pint.logging ): \u001b[33m\u001b[1m/opt/anaconda/lib/python3.10/site-packages/pint/models/timing_model.py:373 UserWarning: PINT only supports 'T2CMETHOD IAU2000B'\u001b[0m\n\u001b[33m\u001b[1mWARNING \u001b[0m (pint.logging ): \u001b[33m\u001b[1m/opt/anaconda/lib/python3.10/site-packages/pint/models/model_builder.py:139 UserWarning: Unrecognized parfile line 'EPHVER 2'\u001b[0m\n\u001b[1mINFO \u001b[0m (pint.simulation ): \u001b[1mUsing CLOCK = TT(TAI), so setting include_bipm = False\u001b[0m\n\u001b[1mINFO \u001b[0m (pint.models.absolute_phase ): \u001b[1mThe TZRSITE is set at the solar system barycenter.\u001b[0m\n\u001b[1mINFO \u001b[0m (pint.models.absolute_phase ): \u001b[1mTZRFRQ was 0.0 or None. Setting to infinite frequency.\u001b[0m\n\u001b[1mINFO \u001b[0m (pint.simulation ): \u001b[1mUsing CLOCK = TT(TAI), so setting include_bipm = False\u001b[0m\n\n\nBarycenter first TOA: 2022-07-11 15:58:51.816031354\nSpacecraft first TOA: 2022-07-11 15:58:51.815489874\n%matplotlib inline\nresolution_factor = 1\npulsar_of_interest.lightcurve.plot_lightcurve()\ntime_vec, photon_vec, probability_arr = pulsar_of_interest.lightcurve.create_toa_vec(0, 10, k=4,resolution_factor= resolution_factor)\n%matplotlib inline\nplt.plot(time_vec[0:10000], photon_vec[0:10000])\nplt.xlabel(\"Time (s)\")\nplt.ylabel(\"Number of photons received\")\nplt.title(\"Photon vector over 25 ms\")\nText(0.5, 1.0, 'Photon vector over 25 ms')\nsnip = time_vec\nindices = np.argwhere(photon_vec != 0).flatten()\nplt.hist(snip[indices], bins = 500)\nplt.title(\"Histogram of photons received with bin resolution = 60 ms\")\nplt.xlabel(\"Time (s)\")\nplt.ylabel(\"Number of photons received\")\nText(0, 0.5, 'Number of photons received')\n%matplotlib inline\nbarycenter_toas = create_obs_time_vec(toas_barycenter.first_MJD, time_vec)\npulse_fold_barycenter = LightProfile.pulse_folding(barycenter_toas, photon_vec, pulsar_of_interest.lightcurve, mjd = True, start_offset = toas_barycenter.first_MJD, resolution_factor = resolution_factor)\n\nplt.plot(pulsar_of_interest.lightcurve.phase_list, pulse_fold_barycenter)\nplt.xlabel(\"Phase\")\nplt.ylabel(\"Count\")\nspacecraft_toas = create_obs_time_vec(toas_spacecraft.first_MJD, time_vec)\npulse_fold_spacecraft = LightProfile.pulse_folding(spacecraft_toas, photon_vec, pulsar_of_interest.lightcurve, mjd = True, start_offset = toas_barycenter.first_MJD, resolution_factor = resolution_factor)\nplt.plot(pulsar_of_interest.lightcurve.phase_list, pulse_fold_spacecraft)\nplt.legend(['Barycenter', 'Spacecraft'])\n# plt.plot(pulse_fold_spacecraft.phase_list, pulse_fold_barycenter)\n# plt.show()\n<matplotlib.legend.Legend at 0x7f9a6dc79120>\nAn example of what happens if you fold the pulse with the wrong pulsar of interest
pulse_fold_barycenter = LightProfile.pulse_folding(barycenter_toas, photon_vec, pulsar_of_interest.lightcurve, mjd = True, start_offset = toas_barycenter.first_MJD, resolution_factor = resolution_factor)\nplt.plot(pulsar_of_interest.lightcurve.phase_list, pulse_fold_barycenter)\nplt.xlabel(\"Phase\")\nplt.ylabel(\"Count\")\nText(0, 0.5, 'Count')\nnp.save(\"pulsar_vec.npy\", np.asarray(photon_vec))\nIn this notebook, we will walk through setup and operation of SDRs to transmit pulse train data for pulsar signal simulation.
"},{"location":"components/3__custom_event_generation/ceg_overview/#checking-sdr-connectivity","title":"Checking SDR Connectivity","text":"To start, it is necessary to make sure you have successfully mounted the SDRs to your PC via USB. You can check this by executing the following commands in a command prompt:
# ping 192.168.2.1\n# ping 192.168.2.2\nFor each of these commands, you should see four data packets successfully sent to the SDR and received back by the PC.
Next, it will be necessary to check that both SDRs can transmit and receive data to themselves. You can do so by executing the following code. Before running this code, make sure both SDRs are mounted to your PC and have been pinged successfully, and that each one has a BNC cable connecting its Rx port to its Tx port.
import numpy as np\nimport adi\nimport matplotlib.pyplot as plt\nimport sdr_io\n\nsample_rate = 1e6 # Hz\ncenter_freq = 915e6 # Hz\n\ntx_data = np.load('test_vec.npy')\ntx_length = 5000\n\nplot_rx_pulse_train = True\nplot_rx_samples = True\nplot_fft = True\ncheck_fid = True\n\nsdr1 = \"ip:192.168.2.1\"\nsdr2 = \"ip:192.168.2.2\"\n\nsdr_io.sdr_tx_rx(sample_rate,center_freq,\n sdr1,sdr1,\n tx_data,tx_length,\n False,False, False, False)\n\nsdr_io.sdr_tx_rx(sample_rate,center_freq,\n sdr2,sdr2,\n tx_data,tx_length,\n False,False, False, False)\nNow that you have verified both SDRs can transmit and receive data to and from your PC, connect the Tx port of the \"X-RAY EMITTER\" SDR to the Rx port of the \"X_RAY DETECTOR\" SDR and execute the following code. This will transmit the same test pulse as above, but between the two SDRs instead of within each one.
sdr_io.sdr_tx_rx(sample_rate,center_freq,\n sdr1,sdr2,\n tx_data,tx_length,\n plot_rx_pulse_train,plot_rx_samples,plot_fft,check_fid)\nOut of phase by 1141\nSuccess! Perfect Transmission!\nOnce you have verified that the SDRs can communicate with one another, you are ready to proceed with data transmission.
"},{"location":"components/3__custom_event_generation/ceg_overview/#transmitting-pulsar-simulation-data","title":"Transmitting Pulsar Simulation Data","text":"After generating a pulse train from the OpenXNAV software, transmit it from the \"Emitter\" SDR to the \"Detector\" SDR by loading it through the code below.
tx_data_name = \"pulsar_vec.npy\" # insert filename for pulse train here\ntx_data = np.load(tx_data_name)\ntx_data = tx_data[:5000]\ntx_length = len(tx_data)\n\nplot_rx_samples = True # once connectivity and fidelic transmission is verified,\n # this is redundant\nplot_fft = True # no longer concerned with freq domain - only concerned \n # with transmitted pulses\ncheck_fid = True # we have already verified connectivity - no need to check again\nprint(\"Transmitted data\")\nplt.plot(tx_data)\nplt.show()\nprint(\"Received data\")\nsdr_io.sdr_tx_rx(sample_rate,center_freq,\n sdr1,sdr2,\n tx_data,tx_length,\n plot_rx_pulse_train,plot_rx_samples,plot_fft,check_fid)\nTransmitted data\nReceived data\nOut of phase by 2890\nSuccess! Perfect Transmission!\nOpen-source modular toolkit for simulating high-fidelity pulsar X-ray events.
"},{"location":"#introduction-to-openxnav","title":"Introduction to OpenXNAV","text":"OpenXNAV is designed to aid development and testing of Pulsar-based Autonomous Navigation (XNAV) Positioning, Navigation, and Timing (PNT) solutions.
This tool is a flexible, cost-effective testbed to enable the next generation of XNAV solutions. OpenXNAV includes three modular components that allow engineers simulate high-fidelity pulsar X-ray events along a flight trajectory over a user-defined mission timeline. Hardware-in-the-loop functionality can be configured using software-defined radios (SDRs), and users can query for pulsar candidates using the query tool (powered by the ATNF Pulsar Catalogue).
For more information, you can read our IEEE paper: IEEE.org.
"},{"location":"#components","title":"Components","text":"Each component of OpenXNAV can be used individually, or strung together as an end-to-end pipeline. To get started with OpenXNAV, select the module you are interested in leveraging and access the documentation.
"},{"location":"components/#pulsar-querying","title":"Pulsar Querying","text":"Query for all known pulsars catalogued in the ATNF database by providing a coordinate in space, and a search radius.
"},{"location":"components/#mission-planning","title":"Mission Planning","text":"Model the desired flight path, taking into account spacecraft orientation, viewing angle and line of sight.
"},{"location":"components/#timing-event-generation","title":"Timing & Event Generation","text":"Generate representative photon arrival events at the location of interest along the user-defined trajectory.
"},{"location":"components/1__pulsar_querying/pq_overview/","title":"Pulsar Querying","text":""},{"location":"components/1__pulsar_querying/pq_overview/#installation","title":"Installation","text":"Before using the OpenXNAV Pulsar Querying Tool, you will need to install the necessary python packages. You can do this using your preferred python package manager, but instructions are provided below for using Anaconda
Inside the open-xnav/1__pulsar_querying/ directory, there is an environment.yml file. This file can be passed to your preferred package manager to create a new environment that can run the tool.
The default environment name in environment.yml is openXNAV-gui. If you wish to change the environment name, simply edit environment.yml before creating the environment.
Open your Anaconda Prompt (or activate Anaconda in a terminal window). Navigate to open-xnav/`1__pulsar_querying/ and run the following command:
conda env create -f environment.yaml\nTo activate your environment, run the following command:
conda activate openXNAV-gui\nThe tool also depends on KivyMD, which is a library built off of the popular Kivy library. This package is not available through conda, so you must install it using pip.
First, make sure you have activated the openXNAV-gui environment in your terminal. Then, run the following command:
pip install kivymd\nIf you are getting a ReadTimeoutError response during installation, you might be using a VPN or have a slower internet connection. If this happens, try running the following instead:
pip install --default-timeout=1000 kivymd\nIf that still does not work, you can troubleshoot your issues further using the pip documentation here.
Once you have successfully installed all the required packages, you are ready to run the tool!
First, ensure you have navigated to open-xnav/1__pulsar_querying/ in your terminal and have your openXNAV-gui environment activated.
Then, simply run the following in your terminal:
python run.py\nUpon startup, the tool will need to pull the current ATNF Pulsar Catalogue from the web. This process only need to be done the first time you open the application.
However, if you wish to update the database, simply delete the .../pulsar_database/ directory from your file system. Then, relaunch the application. Once the database has been generated, you can continue begin querying.
You should see the OpenXNAV launch window with the blue Launch Application button active. In the future, when you open the application, the tool will locate the previously-generated database and you can quickly continue to the querying functionality.
After you click Launch Application, the query tool will display.
There are three fields that you need to provide to execute a query:
These values will create a query, searching for all known pulsars that are within a given search radius of the provided coordinates in space.
The OpenXNAV tool leverages the open-source psrqpy library. More information about this package, and how queries are executed, can be found here.
The query results will display to the GUI interface, and will also be stored in a sub-directory within .../query_results/.
For information about how these sub-directories are tagged, see below.
"},{"location":"components/1__pulsar_querying/pq_overview/#tagging-results","title":"Tagging Results","text":"There is an optional Tag field that OpenXNAV also provides. This allows you to specify the tag for this query you are about to execute. If a tag is provided, that will be used to name the sub-directory within .../query_results/ where all pulsars returned by the query are stored.
Sample Queries
Below are couple sample queries that you can run to get familiar with the OpenXNAV tool. There are additional sample query results contained in the .../query_results/ directory.
Right now, the OpenXNAV Pulsar Querying Tool only outputs pulsar data into the .st format. This is the format that is required by STK to perform subsequent mission planning.
To display the format of this file, here is J00012_5431.st as an example:
stk.v.12.0\nWrittenBy OpenXNAV\n\nBEGIN Star\n\n Name J0012_5431\n\n BEGIN PathDescription\n\n Epoch 58912.0\n RefFrame J2000\n RightAscension 3.097083333333333\n Declination +54:31:47\n ProperMotionRAPerYr 0\n ProperMotionDecPerYr 0\n Parallax 0\n RadialVelocity 0.0000000000000000e+00\n\n END PathDescription\n\n BEGIN PhysicalData\n\n Magnitude None\n\n END PhysicalData\n\n BEGIN IdentityData\n\n Id 0\n\n END IdentityData\n\n\n BEGIN Extensions\n\n BEGIN ExternData\n END ExternData\n\n BEGIN ADFFileData\n END ADFFileData\n\n BEGIN AccessConstraints\n LineOfSight IncludeIntervals\n\n UsePreferredMaxStep No\n PreferredMaxStep 360\n END AccessConstraints\n\n BEGIN Desc\n END Desc\n\n BEGIN Crdn\n END Crdn\n\n BEGIN Graphics\n\n BEGIN Attributes\n\n MarkerColor #00ff00\n LabelColor #00ff00\n MarkerStyle 2\n FontStyle 0\n\n END Attributes\n\n BEGIN Graphics\n\n Show On\n Inherit On\n ShowLabel On\n ShowMarker On\n\n END Graphics\n END Graphics\n\n BEGIN VO\n END VO\n\n END Extensions\n\nEND Star\nYou will need to provide these .st file(s) to STK to successfully execute OpenXNAV's mission planning functionality.
However, if you would like pulsar information to be stored/saved in a different format, you can modify the source code.
Navigate to queryPulsar.py and locate the following function:
def saveToFile(self, root_directory) ...\n.../query_results/."},{"location":"components/2__mission_planning/mp_overview/","title":"Mission Planning","text":"In this Python notebook, the mission_planning module from the OpenXNAV library is demonstrated.
We start by importing the necessary libraries:
from astropy import units as u\nfrom astropy.coordinates import SkyCoord,get_body\nfrom astropy.time import Time,TimeDelta\nfrom astropy.table import Table,QTable\n\nimport numpy as np\nimport matplotlib.pyplot as plt\nimport pandas as pd\nNext we define the necessary variables that we will use to instantiate the orbit. We will define an elliptical orbit directly by its orbital elements in this example. Other options, which are currently commented out, are a circular orbit (special case of elliptical orbit where eccentricity = 0) and an elliptical orbit defined by its burnout trajectory.
# User definitions for elliptical orbital elements\n\ne = 0.3 # eccentricity of the orbit\na = 750000*u.m # semi-major axis \ninc = 20*u.deg # inclination of orbit\nw = 90*u.deg # argument of periapsis\nv_0 = 0*u.deg # initial eccentric anomaly\nOmega = 90*u.deg # longitude of ascending node\n\nimport warnings\nwarnings.filterwarnings('ignore',message='leap-second file is expired')\nt = Time.now() + TimeDelta(120*u.day) + TimeDelta(np.linspace(0,800,51)*u.s)\n\n# Alternative: User definitions for elliptical orbital insertion \n# (note that this is just an example and would not generate the \n# same trajectory as the one above defined directly by orbital elements)\n\n#e = 0.1 # eccentricity of the orbit, also user defined\n#a = 750000*u.m # m semi-major axis \n#B = 70*u.deg # azimuth heading measured in degrees clockwise from north\n#dec = 20*u.deg # geocentric latitude in degrees (declination) of burnout\n#v_0 = 0*u.deg # initial eccentric anomaly\nWe can now do the same thing with a low-fidelity elliptical orbit, starting with the EllipticalOrbit class:
from mission_planning import EllipticalOrbit\n\neo = EllipticalOrbit(t,a,e,v_0=v_0,\n inc=inc,w=w,\n Omega=Omega)\n\nplt.figure()\nplt.plot(eo.x.to(u.km),eo.y.to(u.km),'b.')\n#plt.xlim([-a/u.km,a/u.km])\n#plt.ylim([-a/u.km,a/u.km])\nplt.show()\nWhile the above plot is represented in two dimensions as the projection of the orbit onto the XY-plane, this orbit was actually instantiated in 3D. To show the trajectory in the 3D Cartesian reference frame with the center of Earth at the origin, we just have to plot it in the 3D figure:
fig = plt.figure()\n\nax = fig.add_subplot(projection='3d')\nax.scatter(eo.x.to(u.km),eo.y.to(u.km),eo.z.to(u.km))\n\nplt.show()\nIn order to better demonstrate the pulsar access calculation functions in the OpenXNAV mission_planning module, we will create a SkyCoord object representing the moon, but with a distorted size and orbit to ensure some obfuscation of a chosen pulsar (whose coordinates are also exaggerated for demonstration purposes).
from mission_planning import plot_accesses\n\nmoon = SkyCoord(\n x = -284405*u.m,\n y = 249415*u.m,\n z = 0*u.m,\n frame='gcrs',\n representation_type = 'cartesian'\n)\n\n#moon_rad = 1740*u.km # actual radius of the moon \nmoon_rad = 150*u.km # fake news \n\n\n# assuming we have the location of the pulsars, using one as an example\npulsar = SkyCoord(\n x = -853215*u.m,\n y = 748245*u.m,\n z = 0*u.km,\n frame='gcrs',\n representation_type = 'cartesian'\n)\n\naccess = eo.pulsar_access(pulsar,moon,moon_rad)\n\n# create figure to plot pulsar access\nfig = plt.figure()\n\nax = fig.add_subplot(projection='3d')\nax.scatter(pulsar.x.to(u.km),pulsar.y.to(u.km),pulsar.z.to(u.km),label='pulsar')\nax.scatter(eo.x.to(u.km),eo.y.to(u.km),eo.z.to(u.km),label='spacecraft')\nax.scatter(moon.x.to(u.km),moon.y.to(u.km),moon.z.to(u.km),label='moon',s=500)\nplt.legend()\n\nplt.show()\n\nfig = plt.figure()\nax = fig.add_subplot()\nax = plot_accesses(ax,t,{'pulsar':access})\nWe can also instantiate an EllipticalOrbit object with high fidelity. This version of the class calculates the eccentric anomaly at each time in the time array in order to calculate a more exact position.
There is currently a bug in this version that causes it to only output positive x and positive z values, so we'll need to fix that. (Negative values shown below are there because x and z are positive before coordinate rotation to account for longitude of ascending node, orbital inclination, and argument of periapsis)
eo_h = EllipticalOrbit(t,a,e,v_0=v_0,\n inc=inc,w=w,\n Omega=Omega,\n hifi=True)\n\nplt.figure()\nplt.plot(eo_h.x,eo_h.y,'b.')\nplt.xlim([-a/u.m,a/u.m])\nplt.ylim([-a/u.m,a/u.m])\nplt.show()\n\nfig = plt.figure()\n\nax = fig.add_subplot(projection='3d')\nax.scatter(eo_h.x,eo_h.y,eo_h.z)\n\nplt.show()\nNext we will take advantage of the get_body and pulsar_access_export functionalities of Astropy and OpenXNAV Mission Planning, respectively.
Astropy allows us to create a SkyCoord object using the actual location of the moon at any time or array of times. We will make our moon much bigger than in real life to ensure that we obfuscate some pulsars for demonstration purposes. We will define a new orbit circular orbit using the CircularOrbit class, which is a special case of the EllipticalOrbit class in which eccentricity and orbital inclination are zero.
from mission_planning import CircularOrbit\n\n# User definitions for circular orbit\nr = 360000*u.km\nt0 = Time('2023-08-24 12:12:15.932083')\nt = t0 + TimeDelta(np.linspace(0,25,100)*u.day)\n\nmoon = get_body('moon',t)\nmoon.representation_type='cartesian'\nMOON_RAD = 1740*u.km\n\nearth = get_body('earth',t)\nearth.representation_type = 'cartesian'\nEARTH_RAD = 6378.14*u.km\n\nco = CircularOrbit(t,r)\nNext we will create an astropy.table.Table object with some pulsars that we want to load in to our pulsar access export function. The table is shown and created here for demonstration purposes; however, tables with this format can be created in the OpenXNAV pulsar query module.
name = [ 'J0002+6216' , 'J0006+1834' , 'J0007+7303' , 'J0011+08' , 'J0012+5431']\nraJ_deg = [ 0.74238 , 1.52 , 1.7571 , 2.9 , 3.0971 ] * u.deg\ndecJ_deg = [ 62.26928 , 18.5831 , 1.7571 , 8.17 , 54.5297 ] * u.deg\ndist_kpc = [ 6.357 , 0.86 , 1.4 , 5.399 , 5.425 ] * u.kpc\n\nt = Table([name,raJ_deg,decJ_deg,dist_kpc],\n names = ['NAME','RAJD','DECJD','DIST'])\n\nt.write('some_pulsars.fits',overwrite=True)\nprint(t)\n NAME RAJD DECJD DIST\n deg deg kpc \n---------- ------- -------- -----\nJ0002+6216 0.74238 62.26928 6.357\nJ0006+1834 1.52 18.5831 0.86\nJ0007+7303 1.7571 1.7571 1.4\n J0011+08 2.9 8.17 5.399\nJ0012+5431 3.0971 54.5297 5.425\npulsar_access_export method","text":"Now that we have our spacecraft, moon, and pulsar table, we can use the pulsar_access_export method to create a table and plot of pulsar accesses. Both this method and the pulsar_access function allow you to account for obfuscation from multiple celestial bodies; just make sure to define them as shown below, entered before the keyword arguments and alternating between the SkyCoord object representing the celestial body and the radius of the body represented as an astropy.units.Quantity object.
pulsars_qtbl = QTable.read('some_pulsars.fits')\n\n# 3D plot of satellite orbit and moon\nfig1 = plt.figure()\nax1 = fig1.add_subplot(projection='3d')\nax1.scatter(moon.x.to(u.km),moon.y.to(u.km),moon.z.to(u.km),label='moon',s=100)\nax1.scatter(co.x.to(u.km),co.y.to(u.km),co.z.to(u.km),label='satellite in cirular orbit')\nax1.scatter(earth.x.to(u.km),earth.y.to(u.km),earth.z.to(u.km),label='earth',s=500)\nplt.legend(loc='upper left')\n\n# Plot of pulsar accesses\n_,(fig2,ax2) = co.pulsar_access_export(pulsars_qtbl,\n moon,MOON_RAD,\n earth,EARTH_RAD,\n make_csv=False,\n make_fig=True,save_fig=False)\nNext we will define a new elliptical orbit from the EllipticalOrbit class in which eccentricity and orbital inclination are non-zero.
# User definitions for elliptical orbital elements\n\ne = 0.3 # eccentricity of the orbit\na = 30000*u.km # semi-major axis \ninc = 20*u.deg # inclination of orbit\nw = 90*u.deg # argument of periapsis\nv_0 = 0*u.deg # initial eccentric anomaly\nOmega = 90*u.deg # longitude of ascending node\n\nt0 = Time('2023-08-24 12:12:15.932')\nt = t0 + TimeDelta(np.linspace(0,25,1000)*u.day)\n\nmoon = get_body('moon',t)\nmoon.representation_type='cartesian'\nMOON_RAD = 1740*u.km\n\nearth = SkyCoord(x=0*u.m,y=0*u.m,z=0*u.m,frame='gcrs',representation_type='cartesian')\nEARTH_RAD = 6378.14*u.km\n\neo2 = EllipticalOrbit(t,a,e,v_0=v_0,\n inc=inc,w=w,\n Omega=Omega)\n\nfig = plt.figure()\n\nax = fig.add_subplot(projection='3d')\nax.scatter(eo2.x.to(u.km),eo2.y.to(u.km),eo2.z.to(u.km))\n\nplt.show()\nNext we will create a CSV with some pulsars that we want to load in to our pulsar access export function. The CSV is shown and created here for demonstration purposes:
name = [ 'J0002+6216' , 'J0006+1834' , 'J0007+7303' , 'J0011+08' , 'J0012+5431']\nraJ_deg = [ 0.74238 , 1.52 , 1.7571 , 2.9 , 3.0971 ] * u.deg\ndecJ_deg = [ 62.26928 , 18.5831 , 1.7571 , 8.17 , 54.5297 ] * u.deg\ndist_kpc = [ 6.357 , 0.86 , 1.4 , 5.399 , 5.425 ] * u.kpc\n\ntbl = Table([name,raJ_deg,decJ_deg,dist_kpc],\n names = ['NAME','RAJD','DECJD','DIST'])\n\ntbl.write('some_pulsars.fits',overwrite=True)\nprint(tbl)\n NAME RAJD DECJD DIST\n deg deg kpc \n---------- ------- -------- -----\nJ0002+6216 0.74238 62.26928 6.357\nJ0006+1834 1.52 18.5831 0.86\nJ0007+7303 1.7571 1.7571 1.4\n J0011+08 2.9 8.17 5.399\nJ0012+5431 3.0971 54.5297 5.425\npulsar_access_export method","text":"Now that we have our spacecraft, moon, and pulsar CSV, we can use the pulsar_access_export method to create a table and plot of pulsar accesses. Both this method and the pulsar_access function allow you to account for obfuscation from multiple celestial bodies; just make sure to define them as shown below, entered before the keyword arguments and alternating between the SkyCoord object representing the celestial body and the radius of the body represented as an astropy.units.Quantity object.
As part of the pulsar_access_export method call, we can write all pulsar access data to a CSV file. This CSV can then be used in the pulsar photon time-of-arrival simulation module to model the X-ray pulse sequence that the spacecraft would observe at those coordinates.
pulsars_qtbl = QTable.read('some_pulsars.fits')\n\n# 3D plot of satellite orbit\nfig1 = plt.figure()\nax1 = fig1.add_subplot(projection='3d')\nax1.scatter(eo2.x.to(u.km),eo2.y.to(u.km),eo2.z.to(u.km),label='satellite in elliptical orbit')\nax1.scatter(earth.x.to(u.km),earth.y.to(u.km),earth.z.to(u.km),label='earth',s=200)\nplt.legend(loc='upper left')\n\n# 3D plot of satellite orbit and moon\nfig2 = plt.figure()\nax2 = fig2.add_subplot(projection='3d')\nax2.scatter(moon.x.to(u.km),moon.y.to(u.km),moon.z.to(u.km),label='moon',s=100)\nax2.scatter(eo2.x.to(u.km),eo2.y.to(u.km),eo2.z.to(u.km),label='satellite in elliptical orbit')\nax2.scatter(earth.x.to(u.km),earth.y.to(u.km),earth.z.to(u.km),label='earth',s=200)\nplt.legend(loc='upper left')\n\n# Plot of pulsar accesses, and save pulsar access table to CSV\naccesses,(fig3,ax3) = eo2.pulsar_access_export(pulsars_qtbl,\n moon,MOON_RAD,\n earth,EARTH_RAD,\n make_csv=True,save_csv=True,\n make_fig=True,save_fig=False)\nThe breaks in the plot above might be hard to see since they are so small compared to the scale of the timeline displayed. We have therefore represented the breaks in tabular form below.
# Create DataFrame containing all times at which access to at least one pulsar is interrupted\naccess_breaks = accesses[accesses.all(axis=1) == False]\naccess_breaks\nAs an alternative to the mission_planning module built into OpenXNAV, users can also plan their missions and generate trajectory and pulsar access inputs using the mission planning software of their choice, such as ANSYS STK. STK is very visual, and a great option if you\u2019re starting your mission plan from scratch and want a lot of flexibility in designing your spacecraft trajectory.
STK users who choose this route should follow the following steps:
In this Python notebook, the event_generation module from the PLANTS library is demonstrated.
We start by importing the necessary libraries:
%load_ext autoreload\n%autoreload 2\nfrom pint_backend import *\nimport pulsar_definitions\npint.logging.setup(level=\"INFO\")\nLoads the ephemerides kernel
pint.solar_system_ephemerides.load_kernel(\"de440\")\n\u001b[1mINFO \u001b[0m (pint.solar_system_ephemerides ): \u001b[1mSet solar system ephemeris to de440 through astropy\u001b[0m\nCreates the time vector that the user hopes to observe at and adds additional error if desired. Specifies what pulsar model to use
time_interval = astrotime.Time( ['2022-07-11T16:00:01.000', '2022-07-11T16:21:01.000'], format='isot', scale='utc')\nobs_fs = 100e6\nn_toa = 10\nerror = 5 * u.us\napply_error = False\nmodel_name = \"j0218\" \nLoads in the navigation data provided by STK and generates a spacecraft object off of that navigation data. Note that spacecraft is a custom class created in the PINT backend
time, pos, vel = load_nav_data(\"XNAVSAT_TEMEofDate_Position_Velocity_JD.txt\")\nmy_spacecraft = SpaceCraft(\"my_spacecraft\", time, pos, vel, overwrite = True)\nGenerates the times of arrival at the barycenter compared to the spacecraft. Note that printing the time of arrival generated is slightly different at the two locations
if(model_name == \"j0218\"): \n pulsar_of_interest = PulsarObj(pulsar_definitions.J0218)\nelif(model_name == \"b1821\"):\n pulsar_of_interest = PulsarObj(pulsar_definitions.B1821)\ntoas_barycenter = pint.simulation.make_fake_toas_uniform(\n time_interval[0].mjd, time_interval[1].mjd, n_toa, model=pulsar_of_interest.model, freq = 1e15*u.Hz, obs='barycenter', error=error)\ntoas_spacecraft = pint.simulation.make_fake_toas_uniform(\n time_interval[0].mjd, time_interval[1].mjd, n_toa, model=pulsar_of_interest.model, freq = 1e15*u.Hz, obs = \"my_spacecraft\", error=error)\n\n\n\nprint(\"Barycenter first TOA: \" + str(toas_barycenter.get_mjds(\"True\")[0].utc.iso))\nprint(\"Spacecraft first TOA: \" + str(toas_spacecraft.get_mjds(\"True\")[0].utc.iso))\n\u001b[33m\u001b[1mWARNING \u001b[0m (pint.logging ): \u001b[33m\u001b[1m/opt/anaconda/lib/python3.10/site-packages/pint/models/timing_model.py:373 UserWarning: PINT only supports 'T2CMETHOD IAU2000B'\u001b[0m\n\u001b[33m\u001b[1mWARNING \u001b[0m (pint.logging ): \u001b[33m\u001b[1m/opt/anaconda/lib/python3.10/site-packages/pint/models/model_builder.py:139 UserWarning: Unrecognized parfile line 'EPHVER 2'\u001b[0m\n\u001b[1mINFO \u001b[0m (pint.simulation ): \u001b[1mUsing CLOCK = TT(TAI), so setting include_bipm = False\u001b[0m\n\u001b[1mINFO \u001b[0m (pint.models.absolute_phase ): \u001b[1mThe TZRSITE is set at the solar system barycenter.\u001b[0m\n\u001b[1mINFO \u001b[0m (pint.models.absolute_phase ): \u001b[1mTZRFRQ was 0.0 or None. Setting to infinite frequency.\u001b[0m\n\u001b[1mINFO \u001b[0m (pint.simulation ): \u001b[1mUsing CLOCK = TT(TAI), so setting include_bipm = False\u001b[0m\n\n\nBarycenter first TOA: 2022-07-11 15:58:51.816031354\nSpacecraft first TOA: 2022-07-11 15:58:51.815489874\n%matplotlib inline\nresolution_factor = 1\npulsar_of_interest.lightcurve.plot_lightcurve()\ntime_vec, photon_vec, probability_arr = pulsar_of_interest.lightcurve.create_toa_vec(0, 10, k=4,resolution_factor= resolution_factor)\n%matplotlib inline\nplt.plot(time_vec[0:10000], photon_vec[0:10000])\nplt.xlabel(\"Time (s)\")\nplt.ylabel(\"Number of photons received\")\nplt.title(\"Photon vector over 25 ms\")\nText(0.5, 1.0, 'Photon vector over 25 ms')\nsnip = time_vec\nindices = np.argwhere(photon_vec != 0).flatten()\nplt.hist(snip[indices], bins = 500)\nplt.title(\"Histogram of photons received with bin resolution = 60 ms\")\nplt.xlabel(\"Time (s)\")\nplt.ylabel(\"Number of photons received\")\nText(0, 0.5, 'Number of photons received')\n%matplotlib inline\nbarycenter_toas = create_obs_time_vec(toas_barycenter.first_MJD, time_vec)\npulse_fold_barycenter = LightProfile.pulse_folding(barycenter_toas, photon_vec, pulsar_of_interest.lightcurve, mjd = True, start_offset = toas_barycenter.first_MJD, resolution_factor = resolution_factor)\n\nplt.plot(pulsar_of_interest.lightcurve.phase_list, pulse_fold_barycenter)\nplt.xlabel(\"Phase\")\nplt.ylabel(\"Count\")\nspacecraft_toas = create_obs_time_vec(toas_spacecraft.first_MJD, time_vec)\npulse_fold_spacecraft = LightProfile.pulse_folding(spacecraft_toas, photon_vec, pulsar_of_interest.lightcurve, mjd = True, start_offset = toas_barycenter.first_MJD, resolution_factor = resolution_factor)\nplt.plot(pulsar_of_interest.lightcurve.phase_list, pulse_fold_spacecraft)\nplt.legend(['Barycenter', 'Spacecraft'])\n# plt.plot(pulse_fold_spacecraft.phase_list, pulse_fold_barycenter)\n# plt.show()\n<matplotlib.legend.Legend at 0x7f9a6dc79120>\nAn example of what happens if you fold the pulse with the wrong pulsar of interest
pulse_fold_barycenter = LightProfile.pulse_folding(barycenter_toas, photon_vec, pulsar_of_interest.lightcurve, mjd = True, start_offset = toas_barycenter.first_MJD, resolution_factor = resolution_factor)\nplt.plot(pulsar_of_interest.lightcurve.phase_list, pulse_fold_barycenter)\nplt.xlabel(\"Phase\")\nplt.ylabel(\"Count\")\nText(0, 0.5, 'Count')\nnp.save(\"pulsar_vec.npy\", np.asarray(photon_vec))\nIn this notebook, we will walk through setup and operation of SDRs to transmit pulse train data for pulsar signal simulation.
"},{"location":"components/3__custom_event_generation/ceg_overview/#checking-sdr-connectivity","title":"Checking SDR Connectivity","text":"To start, it is necessary to make sure you have successfully mounted the SDRs to your PC via USB. You can check this by executing the following commands in a command prompt:
# ping 192.168.2.1\n# ping 192.168.2.2\nFor each of these commands, you should see four data packets successfully sent to the SDR and received back by the PC.
Next, it will be necessary to check that both SDRs can transmit and receive data to themselves. You can do so by executing the following code. Before running this code, make sure both SDRs are mounted to your PC and have been pinged successfully, and that each one has a BNC cable connecting its Rx port to its Tx port.
import numpy as np\nimport adi\nimport matplotlib.pyplot as plt\nimport sdr_io\n\nsample_rate = 1e6 # Hz\ncenter_freq = 915e6 # Hz\n\ntx_data = np.load('test_vec.npy')\ntx_length = 5000\n\nplot_rx_pulse_train = True\nplot_rx_samples = True\nplot_fft = True\ncheck_fid = True\n\nsdr1 = \"ip:192.168.2.1\"\nsdr2 = \"ip:192.168.2.2\"\n\nsdr_io.sdr_tx_rx(sample_rate,center_freq,\n sdr1,sdr1,\n tx_data,tx_length,\n False,False, False, False)\n\nsdr_io.sdr_tx_rx(sample_rate,center_freq,\n sdr2,sdr2,\n tx_data,tx_length,\n False,False, False, False)\nNow that you have verified both SDRs can transmit and receive data to and from your PC, connect the Tx port of the \"X-RAY EMITTER\" SDR to the Rx port of the \"X_RAY DETECTOR\" SDR and execute the following code. This will transmit the same test pulse as above, but between the two SDRs instead of within each one.
sdr_io.sdr_tx_rx(sample_rate,center_freq,\n sdr1,sdr2,\n tx_data,tx_length,\n plot_rx_pulse_train,plot_rx_samples,plot_fft,check_fid)\nOut of phase by 1141\nSuccess! Perfect Transmission!\nOnce you have verified that the SDRs can communicate with one another, you are ready to proceed with data transmission.
"},{"location":"components/3__custom_event_generation/ceg_overview/#transmitting-pulsar-simulation-data","title":"Transmitting Pulsar Simulation Data","text":"After generating a pulse train from the OpenXNAV software, transmit it from the \"Emitter\" SDR to the \"Detector\" SDR by loading it through the code below.
tx_data_name = \"pulsar_vec.npy\" # insert filename for pulse train here\ntx_data = np.load(tx_data_name)\ntx_data = tx_data[:5000]\ntx_length = len(tx_data)\n\nplot_rx_samples = True # once connectivity and fidelic transmission is verified,\n # this is redundant\nplot_fft = True # no longer concerned with freq domain - only concerned \n # with transmitted pulses\ncheck_fid = True # we have already verified connectivity - no need to check again\nprint(\"Transmitted data\")\nplt.plot(tx_data)\nplt.show()\nprint(\"Received data\")\nsdr_io.sdr_tx_rx(sample_rate,center_freq,\n sdr1,sdr2,\n tx_data,tx_length,\n plot_rx_pulse_train,plot_rx_samples,plot_fft,check_fid)\nTransmitted data\nReceived data\nOut of phase by 2890\nSuccess! Perfect Transmission!\n