DSPIRA

Correlation and the Two-Element Interferometer

Sat, 15 Aug 2026 00:00:00 +0000

Every measurement a radio telescope makes is a correlation of some kind. A single horn correlates a signal with itself; two horns correlate one against the other. This lesson explains convolution, autocorrelation, and cross-correlation. It then uses these operations to explain what a two-element interferometer measures.

It was written by Pranav Sanghavi at this lab in 2018. Everything here can be run yourself: the notebook is at code/interferometry/ and needs nothing but numpy, scipy and matplotlib.

Convolution
Cross-correlation and autocorrelation
Why interferometry
The two-element interferometer
The effect of bandwidth
Where to go next
Credits

Convolution

“In mathematics (and, in particular, functional analysis) convolution is a mathematical operation on two functions (f and g) to produce a third function, that is typically viewed as a modified version of one of the original functions, giving the integral of the pointwise multiplication of the two functions as a function of the amount that one of the original functions is translated”

— Wikipedia

\[(f * g)(x) \,=\, \int_{-\infty}^{+\infty} f(x-t)\,g(t)\,dt\]

for multiple dimensions

\[\begin{align} (f * g)(x_1,\ldots,x_n ) \,&=\, (f * g)({\bf x})\\ \,&=\, \int_{-\infty}^{+\infty} \ldots \int_{-\infty}^{+\infty} f(x_1-t_1, \ldots , x_n-t_n)\,g(t_1, \ldots, t_n) \,d^nt\\ \,&=\, \int_{-\infty}^{+\infty} f({\bf x}-{\bf t})\,g({\bf t}) \,d^nt\end{align}\]

Properties of convolution

\[\begin{align} f * g \,&=\, g * f&\qquad (\text{commutativity})\\ (f * g)* h \,&=\, f * (g * h)&\qquad (\text{associativity})\\ f * (g + h) \,&=\, (f * g) + (f * h) &\qquad (\text{distributivity})\\ (a\, g) * h \,&=\, a \, (g * h)&\qquad (\text{associativity with scalar multiplication})\\ \end{align}\]

Convolution smooths. Convolving a rectangle with itself produces a triangle. Convolving that result with the rectangle again approaches a Gaussian. That is the central limit theorem happening in front of you — repeated convolution drives almost anything towards a bell curve.

The reason this matters for a telescope is the next picture. Point sources on the sky are spikes. Your telescope has a beam — a response pattern roughly 15° across for a DSPIRA horn. What you record is the sky convolved with the beam.

Two sources closer together than the beam merge into one bump. No amount of integration time separates them; the information is gone at the moment of measurement. That single fact is the whole motivation for interferometry.

↑ Go to the Top of the Page

Cross-correlation and autocorrelation

Cross-correlation asks how much two signals resemble each other when one is slid past the other:

\[\begin{split} (f\star g)(x) \,&=\, ({f_-}^*\circ g)(x)\\ &=\, \int_{-\infty}^{+\infty} f^*(t-x)\,g(t)\,dt\\ &\underset{t^\prime = t-x}{=}\, \int_{-\infty}^{+\infty} f^*(t^\prime)\,g(t^\prime+x)\,dt^\prime\\ \end{split}\qquad \text{,}\] \[\begin{align} (f\star g)(x_1,\ldots,x_n ) \,&=\, (f\star g)({\bf x})\\ &=\, ({f_-}^*\circ g)(x)\\ &=\, \int_{-\infty}^{+\infty} \ldots \int_{-\infty}^{+\infty} f^*(t_1-x_1, \ldots , t_n-x_n)\,g(t_1, \ldots, t_n) \,d^nt\\ \end{align}\]

Autocorrelation is the same operation with one signal played against itself:

\[\begin{split} R\{f\}(x) \,&=\, (f\star f)(x)\\ &=\, (f_-^* * f)(x)\\ &=\, \int_{-\infty}^{+\infty} f^*(t-x)\,f(t)\,dt\\ &\underset{t^\prime = t-x}{=}\, \int_{-\infty}^{+\infty} f^*(t^\prime)\,f(t^\prime+x)\,dt^\prime\\ \end{split}\qquad \text{.}\]

Unlike convolution, cross-correlation does not flip a function. Thus, $ f \star g \neq g \star f $, while $ f * g = g * f $.

Autocorrelation is how a radio telescope finds signal underneath noise. Below is a sine wave three times smaller than the noise it sits in — invisible in the raw trace. Its autocorrelation still reveals the period. The noise is uncorrelated at nonzero lags, while the sine wave remains correlated.

This is not an analogy for what the spectrometer does — it is what the spectrometer does. spectrometer_w_cal.grc displays a power spectrum. This is the Fourier transform of the horn voltage’s autocorrelation. The Wiener–Khinchin theorem establishes this equivalence. It explains how the hydrogen line emerges from a much larger noise floor.

↑ Go to the Top of the Page

Why interferometry

The resolution of a single dish is set by diffraction:

\[\theta \,\approx\, 1.22 \frac{\lambda}{D}\]

At 21 cm that is a brutal constraint. Rearranged for the diameter you would need:

You want	At 21 cm you need a dish
1 degree	15 m
1 arcminute	881 m
1 arcsecond	53 km
15 milliarcseconds	3,523 km

The last row is the resolution the VLA reaches routinely. A single dish that did the same would be a quarter of the diameter of the Earth. A dish of that size is impractical, regardless of budget.

The way out is to stop trying to fill the aperture and instead sample it at a few points. Two horns separated by $ b $ resolve detail on the scale $ \lambda / b $. They match that large dish’s resolution, but not its collecting area.

↑ Go to the Top of the Page

The two-element interferometer

Take two antennas and combine their signals. There are two ways to do it:

by addition, to form a sum interferometer, or $ \sum $-interferometer
by multiplication, to form a product interferometer, or $ \prod $-interferometer

The sum interferometer

\[P ={2 V_0^2(1+ \cos\Delta \Phi)}\]

The product interferometer

\[R_c= \langle V_1 V_2 \rangle_t\] \[R_c =\frac{V_0^2}{2}\cos{\omega \tau}=\frac{V_0^2}{2}\cos \Delta \Phi\]

where $ \tau $ is the time delay in the signal path. A source at angle $ \theta $ from the zenith reaches one antenna first. The geometric delay is $ \tau = b\cos\theta / c $. As the Earth turns, that delay sweeps, and the output oscillates — the fringe.

This is a cosine correlator. Add a phase delay of $ \pi/2 $ to one signal path to make a sine correlator:

\[V_1=V_{01} \cos (\omega (t + \tau)) \quad V_2 = V_{02} \cos (\omega t + \frac{\pi}{2} )\] \[\begin{align} R&=\langle V_{01} V_{02} \cos{(\omega t + \frac{\pi}{2})} \cos{[\omega (t + \tau) ]} \rangle_t\\ R&= V_0^2 \frac{\langle \cos(2 \omega t + \tau + \frac{\pi}{2})+\cos (\omega \tau - \frac{\pi}{2})\rangle_t}{2} \end{align}\] \[R_s =\frac{V_0^2}{2}\sin{\omega \tau}\]

Run both at once and you have a complex correlator, whose output is the visibility:

\[R = \int_{\Omega} I_\nu(\mathbf{s}) e^{-\imath 2\pi \frac{\mathbf{b}\cdot\mathbf{s}}{\lambda}} d\Omega = V\]

That integral is a Fourier transform of the sky brightness. An interferometer measures the Fourier transform of the sky, one spatial frequency per baseline. Everything in aperture synthesis follows from it.

Here is the fringe pattern of a two-element interferometer with a baseline three wavelengths long — the left panel. The lobes are the directions where the two paths arrive in phase.

↑ Go to the Top of the Page

The effect of bandwidth

The right panel above is the same interferometer observing across a wide band rather than at one frequency. Across the band, the baseline length in wavelengths changes. Each frequency places its lobes differently, so averaging washes out the pattern. Near $ \theta = 90^\circ $, the geometric delay is zero and all frequencies agree.

This is bandwidth decorrelation. A broadband interferometer sees clearly only near the direction where its delays match. Arrays compensate with a delay that tracks the source. The zero-delay direction follows the target instead of remaining at the zenith.

DSPIRA’s two-horn setup produces its sharpest fringes for sources transiting near the phase center. Narrowing the band preserves fringes across a wider region, at the cost of sensitivity.

↑ Go to the Top of the Page

Where to go next

Two elements give you one spatial frequency at a time. Earth’s rotation sweeps the baseline through the Fourier plane. More antennas fill that plane faster. This process is aperture synthesis.

See Fundamentals of Radio Interferometry for the standard treatment. Rhodes University and the National Astrophysics and Space Science Programme produced this course book. It is free to read and uses runnable notebooks. Chapter 4, Visibility Space, continues from this page. Start there.

Closer to home, and worth doing in this order:

Additive Interferometry Using Two DSPIRA Radio Horn Telescopes — LightWork Memo 31. This is a measured version of the fringe pattern above. A solar transit on a 5.0 m baseline produced 0.043 rad spacing, compared with 0.042 rad predicted. This is what the theory on this page looks like when it comes out of a real horn.
Setting up a 2 Horn Interferometer — building one and running it
Interferometry — the rest of the interferometry material on this site
Fourier Analysis and Expert Mode — the transforms this page leans on

↑ Go to the Top of the Page

Credits

Pranav Sanghavi wrote this page’s text, equations, and code at WVU’s Radio Astronomy Instrumentation Laboratory on May 21, 2018. They are published under this site’s MIT license. The notebook was ported from Python 2 to Python 3 in 2026. The autocorrelation figure was added then; the remaining material is his.

The definition of convolution is quoted from Wikipedia under CC BY-SA 4.0.

Fundamentals of Radio Interferometry is a separate work licensed under GPL v2. Its authors are the Rhodes University Centre for Radio Astronomy Techniques & Technologies and the NASSP community. It is linked above rather than reproduced.

Interferometry

Thu, 03 Aug 2023 00:00:00 +0000

Two horns a few metres apart, pointed at the same patch of sky, make an interferometer. As a source drifts across the beams, the path length to each horn changes. The two signals move in and out of step. Their combined power rises and falls in a fringe pattern. The distance between the horns sets the fringe spacing. That relationship makes this arrangement useful.

Correlation and the Two-Element Interferometer - Where the fringes come from. Covers convolution, autocorrelation, and cross-correlation. Explains resolution limits for a single dish, adding and multiplying arrangements, and the effects of finite bandwidth. Comes with a notebook you can run.
Additive Interferometry Using Two DSPIRA Radio Horn Telescopes - LightWork Memo 31, by John Makous. Derives additive interferometry from the path length difference. Describes the two-horn system tested at Green Bank and its solar transit on a 5.0 m baseline. The measured fringe spacing was 0.043 rad, compared with the predicted 0.042 rad. The ending mentions signal multiplication as work the group had completed but had not yet written up.
Setting Up a 2 Horn Interferometer is the practical companion. It covers equipment, wiring, east-west alignment, and operation of the adding or multiplying spectrometer.

Installing gr-radio_astro on Ubuntu 20.04

Tue, 27 Jun 2023 00:00:00 +0000

This historical guide uses the gr38 release line. Its application paths refer to that preserved version. For current software, use the DSPIRA software guide.

This installs spectrometer_w_cal.grc and other GNU Radio astronomy programs on Ubuntu 20.04.

Complete the following steps:

Open a terminal window in Ubuntu.
Install GNU Radio external python dependencies and SDR drivers by typing the following and hit enter:
```
sudo apt install gnuradio gr-osmosdr airspy python3-h5py python3-ephem git cmake liborc-0.4-dev -y
```
To clone the repository: in the terminal, type and Enter: git clone https://github.com/WVURAIL/gr-radio_astro.git
Switch to the gr-radio_astro directory: cd gr-radio_astro
Switch to the gr-radio_astro 3.8 branch by typing: git checkout gr38
Make a build directory: mkdir build, and then move to it: cd build
Then run the following in the build directory:
```
cmake ..
sudo make
sudo make install
```
Additional Steps for setting the proper Python environment:
Open a terminal window.
Make sure you are at the home directory (type and Enter cd ). Then type gedit .bashrc to open the .bashrc file in an editor.
Scroll to the file’s bottom and add a blank line. Paste this code: export PYTHONPATH=/usr/local/lib/python3/dist-packages:/usr/local/lib/python3.8/dist-packages:$PYTHONPATH
Save and close (x in upper right corner).
Additionally you can create appropriate symbolic links
1. Open gnuradio-companion in a terminal, then click Help --> About. Note the Python version in the dialog.
2. For Python 3.8, type cd /usr/local/lib/python3.8/dist-packages. For Python 3.9, use cd /usr/local/lib/python3.9/dist-packages.
3. Type ln -s /usr/local/lib/python3/dist-packages/radio_astro

Check that the Installation was Successful

Restart your terminal window. Run the program in GNU Radio:
- In a terminal window type gnuradio-companion
- Open the spectrometer_w_cal.grc program as follows:
  
  File --> Open --> gr-radio_astros --> examples --> *spectrometer_w_cal.grc*
- Plug an Airspy radio, with the LNA attached, into the USB port. Run the program by hitting the start triangle (“execute the flowgraph”) on the menu bar at top. If no errors occur, you are all set!

How to Update files from the gr-radio_astro gr38 Repository

Open the terminal window.
From your home directory (cd ), go to the gr-radio_astro folder: cd gr-radio_astro
Type git status. Check the “On branch …” statement at the top. You want to be in the gr38 branch. To get there, type git checkout gr38.
If a warning message shows up about local changes made that could overwrite files, type git stash.
Type git status to check that you are “On branch gr38”.
Type git pull.
- If a warning message shows up about local changes made that could overwrite files, type git stash.
- Then type git pull again.
Change to the build directory: cd build
Type rm -rf *. NOTE: Make sure you are in the build directory before typing rm -rf *!
Then run the following:
```
cmake ..
sudo make
sudo make install
```
The update is complete.

To Run the __spectrometer_w_cal.grc__ program in GNU Radio After Updating:

In a terminal window type gnuradio-companion.
Close any previous version of spectrometer_w_cal.grc that might be open in GNU Radio.
Open the new version of spectrometer_w_cal.grc from the folder /gr-radio_astro/examples/
- under File select Open;
- Navigate to the gr-radio_astro folder.
- Under examples open DSPIRA; then select spectrometer_w_cal.grc.
Connect an Airspy SDR to USB. Start the program (click the black triangle at the top center of the ribbon bar.)

Installing Ubuntu

Tue, 27 Jun 2023 00:00:00 +0000

Our software works under Ubuntu versions 22.04 LTS and older.

Installing Ubuntu on a Windows computer requires three steps. Partition the hard drive, create a bootable Ubuntu flash drive, and install Ubuntu on the partition:

A. Partition a Hard Drive - Before installing Ubuntu on the hard disk, you need to partition the hard disk in Windows.

NOTE: A hard drive with at least 250 GB is recommended.

Before proceeding, check that Device Encryption is OFF. (This is also known as Bitlocker.) For most Windows systems, this can be done as follows:

Boot up the computer in Windows.
Open the Start menu.
- In the search box type “encryption”.
- In the menu that pops up, click on Manage Bitlocker.
- Verify that it is off and turn off if needed.
- If you disable encryption, wait for Windows to finish decrypting the drive before proceeding. This could take some time.

Now you are ready to partition the hard drive.

Open a Command Prompt window with admin rights, as follows:
- In the search field, type Command Prompt, or just CMD.
- Click on Run as Administrator from the selections on the right side of the window that pops up.
- Click Yes on the pop-up to allow the app to make changes to your device.
In the Command Prompt window that pops up, enter the command diskmgmt.msc to open the Disk Management utility.
In Disk Management, right-click OS (C:). Select Shrink Volume to reduce the partition size.
Wait for the partition-size query to finish. Enter the amount of space to reclaim, then click Shrink. The amount chosen to shrink will be the amount that will be allotted to Ubuntu. This should be an absolute minimum of 50 GB; we recommend at least 100 GB.

After the shrink process completes, a new unallocated space will be present in your drive. We’ll use this free space to install Ubuntu alongside Windows, as described below.

B. Download Ubuntu 22.04 onto a Bootable Flashdrive

You need to download Ubuntu Desktop 22.04 ISO onto the hard drive in Windows. You want to choose the Desktop version. It will take some time to download this file.
Download BalenaEtcher for Windows and install it if you don’t already have it.
Install a 32 GB or larger flash drive in the usb port.
Run BalenaEtcher. Choose the Ubuntu 22.04.iso file as the image and the flash drive as the target. It will take some time for this to run.

C. Install Ubuntu 22.04 Alongside Windows

Turn off the computer. Place the bootable flash drive in the USB port. Turn on the computer, holding down the bootable key (F12) in order to boot from the Ubuntu USB bootable image.
On the first installation screen, select Install Ubuntu and hit Enter to start the installation process. Complete the prompts as needed.
After the installation is complete, you will restart the computer. Choose the Ubuntu option from the menu that appears.
Check the date and time displayed on your computer. If it is not correct, it must be corrected in order for the installation to be completed successfully. This can be changed in the Date & Time menu under Settings.
Open a Terminal window. This can be found in the Show Applications waffle in the lower left corner of the screen.

NOTE: Most of the commands we use are entered in a Terminal window. We recommend adding the Terminal to your Favorites to be accessed easily.
With the time being correct, in a Terminal window type and enter sudo apt update. Then type and enter sudo apt upgrade. This will bring the Ubuntu installation up to date.

The Ubuntu OS is now ready to use.

Setting Up a 2 Horn Interferometer

Mon, 11 Jul 2022 00:00:00 +0000

Procedure for Setting Up a 2 Horn Interferometer

Set up and run a two-horn interferometer with a Lime SDR and GNU Radio. The GNU Radio spectrometer programs write time-stamped data files once every integration time. The adding spectrometer is Interferometer_SimpleSpectrometer_Adding.grc. Its data includes spectra for horns A and B, plus power spectra from their summed signals. The multiplying spectrometer is Interferometer_SimpleSpectrometer_Multiplying.grc. Its data includes spectra for horns A and B, plus interference magnitude and phase.

List of Equipment
- 2 Horn telescopes, each consisting of:
  - horn and can antennae
  - cradle
  - stand
  - bolts for cradle axle
  - clamps - recommended to secure the cradle position after the horn is aligned
- Electronics
  - computer with the GNU Radio spectrometer program on it:
    - adding spectrometer program: Interferometer_SimpleSpectrometer_Adding.grc
    - multiplying spectrometer program: Interferometer_SimpleSpectrometer_Multiplying.grc
  - Lime SDR with USB cable to connect to a computer – This SDR can digitize 2 channels.
  - 2 bias T’s for powering the LNA’s; the Lime does not provide power to the LNA’s.
  - 5 V transformer to power the bias T’s
  - 2 SMA cables to connect the Lime Rx outputs to the “RF” inputs of the bias T’s
  - 2 long SMA cables to connect the “RF+dc” outputs of the bias T’s to the LNA’s
  - 2 LNA’s
- Accessories for setting up
  - duct tape - recommended for taping the devices and cables in place
  - inclinometer - recommended for setting the telescope angle
  - E-W (N-S) line - critical!
    - best method: determined, prior to a run, from the shadow of a vertical pole or string at sun’s transit
    - or determine using an accurate compass
  - Large protractor to help orient the telescopes
  - pliers for tightening the horn rotation axis
  - string - can be useful for aiding the alignment of the horns relative to each other
  - tape measure
  - tent in case of rain
Schematic Diagram and photo of the electrical connections
Setting up the horns
- Assemble 2 horns as would be done for a single horn observation.
- For an east-west baseline, align the horns in parallel toward the same sky position (e.g. the transiting Sun).
- Here are some tips:
  - Sometime before doing interferometry, establish the N-S direction marking the shadow of a vertical pole when the sun transits.
  - For an east-west baseline, use a large compass or square. Mark the east-west direction perpendicular to north-south. A tape measure or string can be useful for doing this.
Setting up the electrical connections - REFER TO THE PHOTO ABOVE
- The cable connected to RX1 on the Lime SDR collects “Horn A” data. RX2 collects Horn B data.
- Plug in the 5 V dc transformer that powers the bias-T’s. Check that the power strip is on.
- The “RF” terminal of each bias-T is connected to the respective RX1 and RX2 on the Lime.
- The “RF + dc” terminal of each bias-T is connected to the respective LNA.
NOTE ABOUT COMPUTER POWER: Be sure the computer power settings are such that it does not go to sleep. Also, for longer runs it is recommended to have the computer plugged in during the run.
For Adding Interferometry:
- Run the spectrometer program Interferometer_SimpleSpectrometer_Adding.grc
- Plug the Lime SDR into a USB port.
- On the computer set up the file directories for data collection:
  - Create a folder appropriately titled for the data collection, e.g. “Sun_scan_July21”. Be sure any folders and filenames have no spaces in them.
  - Run GNU Radio: $ gnuradio-companion
  - Open the Variable boxes described below in Interferometer_SimpleSpectrometer_Adding.grc. They are at the upper left of the GNU Radio canvas:
    - Open the “prefix_hornA” Variable block (double-click it). For “Value” type the pathname to the subfolder that will be used to collect the spectrum of Horn A. This must be in quotes. In this example, horn A data goes into “magnitude_july21”. That folder is inside “sun_july21”, which is inside “interferometer_data”. This will be written in the prefix-hornA Variable box as: “/home/john/dspira_2021/interferometer_data/sun_july21/hornA_july21/”
    - Set up similar file pathnames in the Variable blocks “prefix_hornB” and “prefix_adding”.
    - Be sure these sub-folders on the computer have been created.
- If all is good, you are ready to go. Delete the test files in each folder (not required but recommended).
- Start the program. Select “Long Integration” and “Writing to File”. If all is good, the graphs should show signals being collected.
For Multiplying Interferometry:
- Run the spectrometer program Interferometer_SimpleSpectrometer_Multiplying.grc
- Plug the Lime SDR into a USB port.
- On the computer set up the file directories for data collection:
  - Create a folder appropriately titled for the data collection, e.g. “Sun_scan_July21”. Be sure any folders and filenames have no spaces in them.
  - Run GNU Radio: $ gnuradio-companion
  - Open the Variable boxes described below in Interferometer_SimpleSpectrometer_Multiplying.grc. They are at the upper left of the GNU Radio canvas:
    - Open the “prefix_mag” Variable block (double-click it). For “Value” type the pathname to the subfolder that will be used to collect the spectrum of Horn A. This must be in quotes. In this example, interference magnitude data goes into “magnitude_july21”. That folder is inside “sun_july21”, which is inside “interferometer_data”. This will be written in the prefix-mag Variable box as: “/home/john/dspira_2021/interferometer_data/sun_july21/magnitude_july21/”
    - Set up similar file pathnames in the Variable blocks “prefix_phase”, “prefix_hornA”, and “prefix_hornB”.
    - Be sure these sub-folders on the computer have been created.
- If all is good, you are ready to go. Delete the test files in each folder (not required but recommended).
- Start the program. Select “Long Integration” and “Writing to File”. If all is good, the graphs should show signals being collected.

PlutoSDR software Installation

Wed, 04 Aug 2021 00:00:00 +0000

The steps below outline the installation of software that is needed to run the Pluto SDR in GNU Radio. Find GNU Radio and IIO device information at: Analog Devices documentation

Step 1: Install Dependencies

Open a terminal window. Then type (or copy and paste) and enter the following:

sudo apt install libxml2 libxml2-dev bison flex cmake git libaio-dev libboost-all-dev
sudo apt install libusb-1.0-0-dev
sudo apt install libavahi-common-dev libavahi-client-dev
git clone https://github.com/analogdevicesinc/libiio.git
cd libiio
cmake .
make
sudo make install
cd ..

Download and build libad9361-iio

git clone https://github.com/analogdevicesinc/libad9361-iio.git
cd libad9361-iio
cmake .
make
sudo make install
cd ..

Additional Installations

sudo apt install bison flex cmake git libgmp-dev
sudo apt install liborc-dev

Build and install gr-iio from source:

git clone -b upgrade-3.8 https://github.com/analogdevicesinc/gr-iio.git
cd gr-iio
cmake .
make
sudo make install
cd ..
sudo ldconfig

Your system should now be ready to install the PlutoSDR Source block in GNU Radio. If there are any problems, refer to the information at: https://wiki.analog.com/resources/tools-software/linux-software/gnuradio

Other Horn Designs

Thu, 29 Jul 2021 00:00:00 +0000

Baby DSPIRA Horn is a smaller version of the design above. Find instructions in this PDF and these videos. This includes a cradle and stand as part of its construction. This design uses the same components and software as the standard DSPIRA horn described here. It provides easy use and is good for demonstrations.
Radio Telescope Base Construction Guide - Instructions for building a radio horn telescope similar in design to the DSPIRA horn.
Cone Horn Construction - Instructions for building a cylindrical radio horn with a cylindrical can.

Build a Simple Spectrometer

Sun, 04 Jul 2021 00:00:00 +0000

Introductory Lessons on Using GNU Radio and Some Basic DSP
- Lesson 1 - Introduction to GNU Radio basics.
- Lesson 2 - Learning more GNU Radio tools building a multiple waveform source.
- Lesson 3 - Demonstration of Fourier series.
- Lesson 4 - Demonstration of how an FFT block works.
- Lesson 5 - Filter basics.
Build a Simple Spectrometer - Directions on building a simple spectrometer for a horn telescope, with detailed explanations of the blocks.

Raspberry Pi

Thu, 01 Jul 2021 00:00:00 +0000

Raspberry Pis are inexpensive. Newer models can handle the GNU Radio signal processing needed for radio astronomy.

Supported Raspberry Pi Devices
Installing Ubuntu image with radio astronomy preinstalled on a Raspberry Pi
Setting OS and installing software from scratch

Supported Raspberry Pi Devices

All devices must have RAM greater than 4GB

Raspberry Pi 4 Model B
Raspberry Pi 400

Installing Ubuntu image with radio astronomy preinstalled on a Raspberry Pi

This image requires a minimum of 16GB of space on the SD card.

Download the image here
Unzip the .zip file.
Use Raspberry Pi Imager to install this image.
1. Insert SD card into your card reader on your computer
2. Open Raspberry Pi imager.
3. Click Choose OS, and choose Use Custom.
4. Select the correct image file downloaded in step 1 and 2 from your system.
5. Click Choose storage and select your inserted SD card.
6. Click write.
7. More info here and a video
Insert SD card to Raspberry Pi and power it up.
The default user name is pi, with password raspberry. Change the password after first boot.

Setting OS and installing software from scratch.

Use the sections below to choose an operating system, install Ubuntu, and set up the radio astronomy software.

Operating system requirement

Ubuntu for Raspberry Pi Click here for more Info

This historical setup guide describes older GNU Radio versions. For the maintained library and classroom applications, use the software guide.

Installing Ubuntu on Raspberry Pi

Prerequisites

Support Raspberry Pi
A microSD card (9GB minimum, 16GB recommended)
A computer with a microSD card drive
A micro USB-C power cable
A monitor with an HDMI interface (Optionally for installing Ubuntu Server)
A micro HDMI cable (Optionally for installing Ubuntu Server)
A USB keyboard and mouse (Optionally for installing Ubuntu Server)

Instructions from Ubuntu to install the OS on the Raspberry Pi 4

NOTE: Firstly, for Ubuntu desktop a monitor and mouse and keyboard is required. For command-line access without Ubuntu’s desktop interface, install Ubuntu Server. See instruction for Ubuntu server installation here. Secondly, Raspberry Pi is built on an ARM based architecture any version of Ubuntu will not work. The OS has to specific to the Pi.

Post OS Installation actions:

Interacting with the Raspberry Pi

Using a monitor and mouse and keyboard like any other desktop.
Using SSH:
1. This method can be used on any machine with ssh server enabled.
2. Install X server on your computer. VcXsrv on windows and XQuartz for macOS.
3. Power up the Raspberry Pi and connect your computer via ethernet cable
4. First we need to determine the IP address of the raspberry PI: How to determine Raspberry PI IP address
5. If the determined IP address is <IP address> then in the terminal type ssh -Y pi@<IP address>
6. More info on SSH here
Using PuTTY on Windows:
1. Download PuTTY.
2. Add IP address in the hostname field.
Using VNC: More info here

Installing gr-radio_astro

Notes: run volk_profile after installing everything to make sure GNU Radio is optimized to work on your device.

Installing gr-radio_astro

Thu, 01 Jul 2021 00:00:00 +0000

The shared gr-radio_astro library supplies GNU Radio blocks. Classroom applications now come from dspira-software. Install the library first, then download the applications separately.

These instructions target GNU Radio 3.10 on Ubuntu. For an older installation, see the Ubuntu 20.04 and GNU Radio 3.8 guide.

Install the shared library

Follow the current library build instructions to install GNU Radio, receiver drivers, and build dependencies. Clone and build the shared library:

git clone https://github.com/WVURAIL/gr-radio_astro.git
cd gr-radio_astro
cmake -S . -B build
cmake --build build
sudo cmake --install build

Check the repository’s Python environment instructions if GNU Radio cannot find the installed blocks.

Get the DSPIRA applications

Use the software download page or clone the application repository from your home directory:

cd
git clone https://github.com/WVURAIL/dspira-software.git

Open dspira-software/flowgraphs/spectrometer_w_cal.grc in GNU Radio Companion. The file no longer lives in gr-radio_astro/examples/DSPIRA/. Review the known compatibility findings before attempting to run it. Configure your receiver, output folders, and calibration settings.

Update the applications

Keep a separate copy of any local receiver settings or customized flowgraphs before updating. From the application checkout, run:

git status
git pull --ff-only

If Git reports local changes, review and preserve them before continuing. Application updates do not require rebuilding the shared library.

Update the shared library

From the gr-radio_astro checkout, review local changes and pull the update:

git status
git pull --ff-only
cmake -S . -B build
cmake --build build
sudo cmake --install build

Close and reopen GNU Radio Companion after installing updated blocks. Test the application with your equipment before relying on it in class.