Important Notes
Please note that this year's SDRA is spanning over two days.
Please consider that the times are in the German timezone MESZ (UTC+1+1)! For GMT please deduct 2 hours.
The YouTube stream is available at: https://youtube.sdra.io
SDRA Programme Day 1: 27.06.2020
12:30 - 13:00 Prof. Dr. Michael Hartje DK5HH, Markus Heller DL8RDS: Introduction
13:00 - 14:00 Prof. Dr. Ulrich Rohde N1UL: Keynote Interview
14:00 - 15:00 Laurence Barker G8NJJ: Using Xilinx Vivado for SDR Development
This paper presents my incomplete work to develop an SDR transceiver compatible with the HPSDR projects. The work has been inspired by that of Pavel Demin but it is not a copy.
Xilinx “Vivado” FPGA firmware tools include the IP Integrator. This tool allows a design to be implemented as a visual block diagram, with functional blocks joined together to create the overall flow graph. Functional blocks can be picked from a library, or can be written using VHDL or Verilog. A wide range of DSP functional blocks are provided. Important enablers for the block design approach are the AXI Bus and AXI Stream interconnect. The paper describes the AXI Stream, and how it enables DSP modules to be interconnected easily.
I present my flow graph for an SDR receiver and transmitter module implemented using IP integrator, and demonstrate using IP integrator. I demonstrate using the simulator to test parts of the block design, followed by the code running on Xilinx hardware. The simulator allows gate-level simulation, but the block diagram sometimes needs to be redrawn to get access to the signals required.
15:00 - 15:30 Edwin Richter DC9OE, Crt Valentincic S56GYK: Usage of higher order Nyquist Zones with Direct Sampling Devices
This talk is about the Principle of using higher order Nyquist Zones with Direct Sampling Devices, and the Benefits of doing so. It deals with the Practical Problems with examples (practical results from measurements wit STEMlab 16) and discusses Filtering required to make use of these Nyquist Zones. The talk also gives results and practical examples to show where it would make sense to go this path.
15:30 - 16:00 Prof. Dr. Michael Hartje DK5HH: Signalprocessing in the man made noise measurement system ENAMS
The German Amateur Radio Club, DARC, developed a man made noise measurement system which will measure the MMN from 50 kHz to 30 MHz for the next 5 years with more than 50 stations spread all over the country to track the developement of MMN levels in different locations ordered by categories rural, residential and industrial.
The measurement system is based on a Red Pitaya. The software is written in python. Measurements are made according the ITU rules for spectrum mesurement.
The talk will give an overview and a short look into a selected part of the measurement system algorithms.
16:00 - 16:30 Bart Somers PE1RIK: Long term spectrum monitoring using GNUradio and Python
"With specmon I monitored the usage of the 2M amateur radio band for several months. After aggregating the frequency-usage and plotting the data, the usage over days, weeks and months becomes visual. Software can be found here: https://github.com/SDRWaveRunner/specmon"
16:30 - 17:00 Florian Brauchle: Implementation and Applications of the Spectral Correlation Density
The Spectral Correlation Density (SCD) is a function that shows certain characteristics of modulated signals and can be used to detect and analyze unknown signals. Over the last year we have implemented a Tool that efficiently calculates and visualizes the SCD and used it to analyze signals.
This talk is a summary of our experiences using the SCD for signal identification. After a short introduction to the spectral correlation we will look at multiple signals to show how the spectral correlation changes from modulation to modulation and finish with an actual problem we have solved using the SCD.
17:00 - 18:00 Adrian Musceac YO8RZZ: FT8 with GNU radio and the PlutoSDR on the 2 meters band
This talk is about using the ADALM-Pluto SDR transceiver to make long distance FT8 contacts on the 2 meter amateur radio band using WSJT-X and the GNU radio software stack to drive the PlutoSDR and some external hardware to maximize the performance. Despite the minimal cost of the hardware and the modest technical specifications, and with the help of good tropospheric propagation conditions, distances of 800-1000 km were within reach. Also discussed in this talk is the impact of the processing chain delay on the latency of this communication mode.
18:00 - 18:30 Dr. Jean-Michael Friedt: Noise RADAR implementation using software defined radio hardware
We demonstrate an UHF noise radar using readily available hardware limited to two television Yagi-Uda antennas, Analog Devices PlutoSDR and Ettus Research B210 platforms and a coupler. Noise RADAR is based on searching for delayed copies of a pseudo-random pattern transmitted by the PlutoSDR and recorded on one channel of the B210 as reference while the receiving antenna is connected to the second channel. Cross-correlation is used for searching for the reference pattern in the noisy received signal. Thanks to the excellent noise rejection of the matched filter, a building located 50 meters from the setup is readily detected when emitting only -30 dBm in a 2 MHz bandwidth. Due to the poor range resolution of such a narrowband signal, the local oscillator is swept on more than 150 MHz bandwidth to improve resolution below one meter.
SDRA Programme Day 2: 28.06.2020
13:30 - 14:00 Dr. Jean-Michael Friedt: Bitstream clock synchronization in an ACARS receiver: porting gr-acars to GNU Radio 3.8
ACARS -- Aircraft Communication Addressing and Reporting System -- is a protocol used by pilots to communicate telemetry and free text messages from planes. AM modulated around 131 MHz with bits encoded as AFSK at 1200 and 2400 Hz, it is perfectly suited as an introductory GNU Radio processing block development. As part of porting gr-acars to GNU Radio 3.8, various improvements have been brought including bitrate clock synchronization which improves long message handling and increases the successful decoding rate by about 30% and running multiple processing blocks in parallel to decode multiple ACARS primary and secondary streams. The latest release of gr-acars is available at https://sourceforge.net/projects/gr-acars/ and the description of the results is summarized at http://jmfriedt.org/acars_clock.pdf
14:00 - 15:00 Dr. Bastian Bloessl DF1BBL: GNU Radio on Android
With the ever-increasing performance of smartphones and tablets, they become viable platforms for applications that were, in the past, only possible on desktops or laptops. In this talk, we will look into using GNU Radio on Android, enabling highly mobile SDR applications. While Android applications are mostly implemented in Java, all data processing happens in C++ domain, driven by GNU Radio's parallelized runtime environment. The Android port supports popular USB-based SDRs (RTL-SDR, HackRF, and USRP B200 series) and can fully benefit from available accelerators (SIMD and the GPU), allowing non-trivial SDR implementations to run in real-time on a phone. To show how GNU Radio can be used on Android, we will look into examples that implement an FM and a WLAN receiver.
15:00 - 15:30 Prof. Dr. Alberto Dassatti, Oscar Rodriguez Zalona: Deep learning inference in GNU Radio with ONNX
A GNU Radio Out Of Tree (OOT) block to run deep learning inference inside GNU Radio using ONNX Runtime. Deep Learning (DL) is a branch of machine learning that has been used successfully to solve complex problems in different domains like image processing, natural language processing and speech recognition. In recent years, deep learning techniques have been also applied to wireless communication and spectrum sensing showing that it can outperform classical approaches in tasks like automatic modulation classification. Integrating such technology in a Software Defined Radio (SDR) context is very promising and could lead to opening new possibilities in the field, for instance in cognitive radio and signal classification. We have developed an OOT module that integrates a deep learning inference engine that uses the Open Neural Network Exchange (ONNX) format and different graph optimizations and accelerators to improve inference performance. ONNX has support for most of the major deep learning frameworks and provides framework interoperability. ONNX Runtime supports different optimization techniques, hardware accelerators and backends. We demonstrate its functionality in a GNU Radio flowgraph running classification using a simple deep learning model, raw I/Q values acquired from a PlutoSDR.
15:30 - 16:00 Manuel Milla Peinado: An OFDM channel sounder with GNURadio
When deploying a radio system it is necessary study the properties of the propagation channel. There are two main elements be characterized: the channel coherence bandwidth and the channel coherence time. They are respectively related with the multipath components and their Doppler spectrum.
In this context, a channel sounder is an instrument designed to study the propagation channel by the emission of a test signal. This test signal is known by the receiver in advance. Thus, the channel properties can be obtained by analyzing how the channel has affected this signal.
We propose the use of Orthogonal Frequency-Division Multiplexing Technique (OFDM) signals and the capabilities of GNURadio to implement a real time channel sounder based on this transmission technique.
16:00 - 16:30 Lukas Ostendorf: HNAP for Pluto-SDR - Implementing the HAMNET Access Protocol on the Adalm Pluto SDR
The HAMNET Access protocol (HNAP) provides spectrally efficient access with low latencies to the HAMNET using a 200kHz duplex channel in the 70cm band.
The complete HNAP stack has been implemented on an Adalm Pluto SDR, allowing standalone operation without any additional processing PC.
The implementation was done in plain C using the liquid-dsp library and the Pluto's ARM CPU.
This talk will introduce to HNAP and then show how the implementation on the Adalm Pluto was achieved.
16:30 - 17:00 Clement Campo: Positioner steering and synchronization with antenna weighting using GNU Radio
Software-Defined Radio (SDR) is of major interest for antenna array weighting applications. In the context of phase coherent applications, measurement in an anechoic environment with precise control of the angle between the measured antenna array and the reference source is a mandatory step. However, such measurement while steering the array using SDR can prove difficult, as available equipment is often incompatible with the additional hardware.
The following work therefore presents a motor positioner and array weighting GNU Radio blocks developed in order to allow for precise measurements in phase coherent applications using commercial SDR. For GNU Radio to handle the automated measurement, a client architecture is implemented in a block in the form of a state machine. The client commands are sent to the implemented server steering the motor using a TCP/IP protocol, for precise control of the angle between the antenna array and reference source. Beam steering and null steering blocks are also implemented in GNU Radio, and measurements are compared to numerical simulations to validate the proposed experimental setup. Measured radiation patterns are found to be in accordance with expectations, hence validating the array weighting blocks. Previous observable 3 to 5° angular shifts of the measured radiation patterns are reduced to less than 1°, showing the implemented motor positioner efficiently takes advantage of the precise rotation offered by the proprietary equipment.
Authors: Clément Campo, Manuel Milla Peinado, Loïc Bernard, Hervé Boeglen, Sébastien Hengy, Jean-Marie Paillot
17:00 - 17:30 Burkhard Kainka DK7JD: The Arduino SDR-Shield 0.1 ... 30 MHz
The Arduino SDR-Shield 0.1 ... 30 MHz - Simple design, Hi Z antenna input, two additional PLL outputs - Hardware IQ-Decoder and filter for stand alone usage - QRP-transceiver 2 W CW and WSPR - 200 mW mini WSPR TRX - SWR meter using Arduino and the SDR shield
17:30 - 18:00 Kjell Karlsen LA2NI: ANDROMEDA HPSDR Radio Demonstration
We will present and demonstrate the prototype of the new "Andromeda" radio. This has been developed from the existing HPSDR hardware, with an added front panel and integrated ATU.
We will describe the software changes that have made this possible. The software uses the "Thetis" program from the HPSDR project, with a new screen layout matched to a 7" 1024x600 touchscreen display. The front panel controller is similar to the Odin controller we presented at SDRA2018, using an Arduino controller. The "Aries" ATU uses another Arduino, and is integrated into the Thetis program.
We will provide a video demonstrating the completed radio, with its front panel control and touchscreen display.
18:00 - 18:30 Prof. Dr. Harald Gerlach DL2SAX: Operational SDR Benefits in Contesting
This talk is for operators who want to make use of the additional benefits of an SDR transceiver in a contest. While SDR transceivers are mainly seen from a technical perspective and thus from the standpoint of technological progress, these advances can and should also be transferred to operating technique and station automation. The SDR radio should not just be reduced on the doubtlessly useful waterfall diagram: The systematic use of an SDR changes the overall station concept and the mode of operating. While a personal computer has ever been a useful extension of a radio station, it has become a central and essential component of a station scenario.
This talk is on how to use an SDR transceiver as the centerpoint of a modern contest station with the operating goal "SO2R mode" (i.e. single operator, two radios), however only using a single SDR transceiver. The main focus will be on station layout and operative procedures and we will shed some light on the latest strategy of 2BSIQ (two band synchronous interleaved QSO).