{ "cells": [ { "attachments": {}, "cell_type": "markdown", "metadata": { "editable": true, "slideshow": { "slide_type": "slide" }, "tags": [] }, "source": [ "
\n", "\n", "# Demo\n", "\n", "In this demo we will\n", "\n", "1. setup the spectrometer, \n", "2. record a free induction decay (FID) and \n", "3. do some simple analysis.\n", "\n", "To prevent any confusion we'll look at all the code, but glance over some parts more quickly.\n" ] }, { "cell_type": "markdown", "metadata": { "editable": true, "slideshow": { "slide_type": "slide" }, "tags": [] }, "source": [ "## Imports\n", "\n", "* First, some standard `imports` you might already be familiar with\n", "* These make the imported `modules` / `classes` / `functions` available later\n", "* The `%matplotlib widget` makes the graphs interactive\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "editable": true, "slideshow": { "slide_type": "fragment" }, "tags": [] }, "outputs": [], "source": [ "# Python imports\n", "from pathlib import Path\n", "\n", "# Standard science imports\n", "import numpy as np\n", "import matplotlib as mpl\n", "import matplotlib.pyplot as plt\n", "import nmrglue as ng\n", "\n", "# Spectrometer imports\n", "from spectrometer import (\n", " FID1D,\n", " ConnectionSettings,\n", " Delay,\n", " NMRSequence,\n", " Pulse,\n", " Record,\n", " Server,\n", " Spectrometer,\n", " plot,\n", " process,\n", ")\n", "\n", "%matplotlib widget" ] }, { "cell_type": "markdown", "metadata": { "editable": true, "slideshow": { "slide_type": "slide" }, "tags": [] }, "source": [ "## Setup \n", "\n", "### Prepare the hardware\n", "\n", "The basic steps are:\n", "\n", "1. Compiling and flashing the FPGA firmware\n", "2. Setting up the Spectrometer server\n", "3. Connecting to the server\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "editable": true, "slideshow": { "slide_type": "fragment" }, "tags": [] }, "outputs": [], "source": [ "# Connect the server platform (i.e. the RedPitaya)\n", "server = Server(\"192.168.1.100\")\n", "\n", "# Flash the FPGA bitstream (or \"low level server\")\n", "server.flash_fpga()\n", "\n", "# Compile the server on the spectrometer (or \"high level server\")\n", "server.setup()\n", "\n", "# Start the server on the spectrometer\n", "server.start()" ] }, { "cell_type": "markdown", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "### Connect to the spectrometer server\n", "\n", "Now that the server is running we can connect to it.\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "editable": true, "slideshow": { "slide_type": "fragment" }, "tags": [] }, "outputs": [], "source": [ "# Setup the spectrometer connection\n", "connection_settings = ConnectionSettings(ip_address=\"192.168.1.100\")\n", "\n", "# Create the spectrometer object\n", "spectrometer = Spectrometer(\n", " tx_freq=25_091_000, # Center transmission frequency\n", " rx_freq=None, # Receive frequency\n", " sample_rate=320_000, # samples/second\n", " server_config=connection_settings,\n", ")\n", "\n", "# Connect to the spectrometer server\n", "spectrometer.connect()" ] }, { "cell_type": "markdown", "metadata": { "editable": true, "slideshow": { "slide_type": "slide" }, "tags": [] }, "source": [ "## Perform an experiment\n", "\n", "Now we want to perform a simple experiment.\n", "\n", "Here we'll send a single pulse and look at the received signal\n", "\n", " 1. Prepare the pulse sequence\n", " 2. Send the sequence and save the received signal\n", " 3. Processing\n" ] }, { "cell_type": "markdown", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "### Preparing a pulse sequence\n", "\n", "We have multiple options for defining sequences:\n", "\n", "1. Use the simple built-in ones (`simple`, `spin_echo`)\n", "2. Build our own from a list of instructions\n", "3. Define the control arrays manually \n" ] }, { "cell_type": "markdown", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "We'll define a pulse with\n", "\n", "* 9us length\n", "* 25us delay\n", "* 20ms record\n", "\n", "using all three methods.\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "editable": true, "slideshow": { "slide_type": "fragment" }, "tags": [] }, "outputs": [], "source": [ "seq_simple = NMRSequence.simple(pulse_length_us=9, delay_us=25, record_length_us=20_000)\n", "# same as\n", "seq_built = NMRSequence.build(\n", " [Pulse(duration_us=9, power=1), Delay(duration_us=25), Record(duration_us=20_000)]\n", ")\n", "# or\n", "seq_manual = NMRSequence(\n", " tx_sequence=(np.array([0, 9]), np.array([1, 0])), rx_sequence=np.array([34, 20034])\n", ")\n", "\n", "assert seq_simple == seq_built == seq_manual" ] }, { "cell_type": "markdown", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "### Sending the pulse sequence\n", "\n", "Since all the sequences are the same we can sent any of them\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "outputs": [], "source": [ "# Send the sequence\n", "data = spectrometer.send_sequence(seq_simple, debug=True)" ] }, { "cell_type": "markdown", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "`data` contains the complex QI-demodulated data as a one dimensional `numpy` array.\n", "\n", "Now we can put the data into a simple container with the relevant metadata and save it to a file for later processing.\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "editable": true, "slideshow": { "slide_type": "fragment" }, "tags": [] }, "outputs": [], "source": [ "# Create FID object\n", "fid = FID1D(\n", " data=data,\n", " spectral_width=spectrometer.sample_rate,\n", " carrier_freq=0.0,\n", " observation_freq=spectrometer.rx_freq,\n", " label=\"1H\",\n", " sample=\"Water\",\n", " pulse=\"single_90_degree_pulse,length=9us,delay=30us\",\n", " spectrometer=\"magnETHical v0.1\",\n", ")\n", "\n", "# Save to file\n", "time = fid.timestamp.strftime(\"%Y%m%d-%H%M%S\")\n", "file = Path(f\"../../data/{time}-{fid.sample}-{fid.label}-{fid.pulse}.fid\")\n", "fid.to_file(file)" ] }, { "cell_type": "markdown", "metadata": { "editable": true, "slideshow": { "slide_type": "fragment" }, "tags": [] }, "source": [ "The data is stored in standard `nmrpipe` format, thus further analysis can be conducted in any other tooling capable of reading that format.\n", "\n", "For more features and further information, see the various `record_*` and `process_*` scripts in the `scripts` folder and the library reference.\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "editable": true, "slideshow": { "slide_type": "fragment" }, "tags": [] }, "outputs": [], "source": [ "# Stop the server\n", "server.stop()" ] }, { "cell_type": "markdown", "metadata": { "editable": true, "slideshow": { "slide_type": "slide" }, "tags": [] }, "source": [ "## Analyzing the received signal\n", "\n", "Having conducted all of our experiments, we can load the data from file to analyze it. \n", "\n", "* outside of Python, e.g. in `NMRPipe`, \n", "* in Python using e.g. `nmrglue`, `numpy`, `scipy`, `matplotlib`, ... or\n", "* any tooling that can read `NMRPipe` `.fid` files\n" ] }, { "cell_type": "markdown", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "### Loading the data back into Python\n", "\n", "We'll now quickly look at the data with some small helper functions built around `nmrglue`\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "editable": true, "slideshow": { "slide_type": "fragment" }, "tags": [] }, "outputs": [], "source": [ "# Load data back from file\n", "fid = FID1D.from_file(file)" ] }, { "cell_type": "markdown", "metadata": { "editable": true, "slideshow": { "slide_type": "slide" }, "tags": [] }, "source": [ "### Plot the data\n", "\n", "Now we can plot the time domain data\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "editable": true, "slideshow": { "slide_type": "fragment" }, "tags": [] }, "outputs": [], "source": [ "# Take a quick look at the time domain data (i.e. the FID)\n", "fig = fid.plot()" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "outputs": [], "source": [ "# Try to plot simple least squares fit for a decaying sinusoid\n", "decaying_sinus = process.decaying_sinus.fit(fid.ms_scale, fid.data.real)\n", "ms_scale_fine = np.linspace(fid.ms_scale[0], fid.ms_scale[-1], len(fid.ms_scale) * 10)\n", "\n", "# Plot raw values again\n", "fig = fid.plot()\n", "\n", "# Plot fit\n", "_ = fig.axes[0].plot(\n", " ms_scale_fine,\n", " decaying_sinus(ms_scale_fine),\n", " linestyle=\"--\",\n", ")" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "outputs": [], "source": [ "# Take a look at the fit parameters\n", "print(decaying_sinus)" ] }, { "cell_type": "markdown", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "There's also a function to quickly get a simple spectrum.\n", "\n", "This function does by default\n", "* zero-fill\n", "* complex fft\n", "* simple automatic phase correction\n", " - by minimizing the difference of minima around the highest maximum\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "outputs": [], "source": [ "# Take a quick look at the spectrum\n", "# (includes an auto zero-fill, complex fft and simple auto phase correction)\n", "spectrum, p0 = fid.spectrum()\n", "print(\"Zero-order phase correction:\", p0)\n", "\n", "# Plot the real part\n", "fig = spectrum.hz.plot()" ] }, { "cell_type": "markdown", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "We could - for example - crop the data a bit \n" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "editable": true, "slideshow": { "slide_type": "fragment" }, "tags": [] }, "outputs": [], "source": [ "# FFT again cause we can\n", "# We can also manually specify the zero-order phase shift by passing `phase_shift_kwargs={\"p0\": 68}`\n", "spectrum, _ = fid.spectrum(phase_shift_kwargs={\"p0\": p0 + 0})\n", "\n", "# Crop\n", "cropped_spectrum = spectrum.hz.crop_around(spectrum.hz.max_peak, 2000)\n", "fig = cropped_spectrum.hz.plot(linestyle=\"-\", marker=\"\")" ] }, { "cell_type": "markdown", "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "source": [ "### Do a fit\n", "\n", "Now we can take the peak and try to do a Lorentz fit to get some interesting paramters:\n", "\n", "* amplitude\n", "* position\n", "* peak width (gamma)\n" ] }, { "cell_type": "code", "execution_count": null, "metadata": { "editable": true, "slideshow": { "slide_type": "subslide" }, "tags": [] }, "outputs": [], "source": [ "# Perform least squared fit of a lorentzian mixture on detected peaks above 3*noise threshold\n", "fit_spectrum, fitpeaks = cropped_spectrum.hz.fit()\n", "peaks = cropped_spectrum.hz.peaks()\n", "\n", "# Plot\n", "fig = cropped_spectrum.hz.plot(linestyle=\"-\", marker=\"\", label=\"raw\")\n", "fig.axes[0].plot(cropped_spectrum.hz.scale, fit_spectrum, linestyle=\"--\", label=\"fit\")\n", "fig.axes[0].plot(\n", " [peak.location for peak in peaks],\n", " [peak.amplitude.real for peak in peaks],\n", " linestyle=\"\",\n", " marker=\"o\",\n", " color=\"red\",\n", " label=\"peaks\",\n", ")\n", "fig.legend()" ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": [ "# Look at the first peak\n", "print(peaks[0])" ] }, { "cell_type": "markdown", "metadata": { "editable": true, "slideshow": { "slide_type": "slide" }, "tags": [] }, "source": [ "## That's it!\n" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.11.5" } }, "nbformat": 4, "nbformat_minor": 4 }