https://groups.io/g/rsgb-lf-group/topic/103975865 Strange wideband signal around 70 kHz Markus (DF6NM) 22:35 #18412 Using a 192 kHz soundcard with SpecLab and 1pps synchronisation, I could receive the three MARS-75 stations here at home in Nürnberg today. Using phase evaluation, the fundamental frequency was measured more accurately as df = 1215.154404 Hz (+- a few µHz). The transmitted frequencies are coherent harmonics at 53 to 75 times df. In the attached image, you can see the staircase signal from the master station (near Latvia) before the 4.5 s UTC frame border, preceded by two extra dots. Then come the two slaves S1 (Wyborg) and S2 (Kaliningrad) with 1.44 s spacing. Colour indicates phase, and different phase slopes correspond to different propagation delays (modulo 1/fd ~ 823 µs). To improve SNR, exponential averaging with a 10 frame time constant was employed, leading to similar-looking noise patterns in consecutive frames. If you feel like playing around yourself, I have attached my current SpecLab configuration files (albeit this setup is pretty experimental, particularly around the delayline filters). The MARS-75 chain is not active every day, so I'd recommend to check aurally on one of the Finnish Kiwi-SDRs before you start (e.g. Karelia DX, OH5AE, OH5LIZ or SM1OII). Best 73, Markus Markus (DF6NM) 22:00 #18285 Here's a somewhat nicer map from the same data, generated by pasting the attached script into PA3FWM's hflines plotter: https://pa3fwm.nl/hflines/ 73, Markus Markus (DF6NM) 20:44 #18283 https://groups.io/g/rsgb-lf-group/message/18283 The timing of the 64 - 91 kHz staircase signals are precise and reproducible, letting me suspect that we are observing an LF navigation system. Detailed analysis of wav recordings shows that the tones are sent on 23 equi-spaced frequencies from 64402.52 to 91135.81 Hz, with 1215.154 Hz spacing. Each tone is 40 ms long, followed by a 20 ms pause to the next tone, and the whole sequence is repeated accurately every 4.5 seconds. Peter's spectrogram from 2021-04-22 showed a chain of three transmitters (I, II, III), with time delays 1.62, 1.44 and 1.44 s between them. Scandinavian online receivers seem to pick up even more transmitters with different time offsets. I presume that this variant of hyperbolic navigation would be done in three steps: First, coarse time-differences can be obtained using the edges of the 40 ms blip envelopes. Then the phase information from the 23 frequencies can be combined by a Fourier transform, producing a sharper "synthetic pulse" with 36 µs duration. It has a 823 µs ambiguity which can be easily resolved by the previous coarse measurement. Like with Loran-C, skywave interference can be suppressed by evaluating only the rising pulse edge. The pulse time can then finally be used to disambiguate a precise carrier-phase measurement. Compared to pulsed Loran-C, this stepped-frequency scheme requires much less peak power, and only a relatively small narrowband TX-antenna which can be sequentially tuned. Having become curious, I then attempted to locate the transmitters, using arrival time differences from three Kiwi Web-SDRs. Unfortunately the blips are too weak for the Kiwi's built-in TDOA function, so I decided to try to do the processing myself. On Feb 25, three simultaneous one-minute I-Q recordings were made from receivers in Joensuu, Lohja and Lublin. The 12 kHz wide passband was centered on 72 kHz, so that time-signals from DCF77, RBU and BPM can be used to align the individual receiver's samplerate, LO-frequency and time offsets. Some screenshots are here http://df6nm.bplaced.net/LF/MARS-75/ . Using the "Poslines" hyperbolic plotter program from G3PLX, I could then obtain good fixes for transmitters II and III, based only on coarse blip envelope time which seem to be good to around 100 µs. However station I (the strongest here) suffered from an unfavourable geometry, apparently because it is near the extended baseline from the two Finnish receivers. Thus I made another recording today, replacing Lohja by DK0FFO in Frankfurt (Oder). This time I centered the recordings on 83 kHz, including Loran Slonim instead of RBU for somewhat improved time alignment. The attached little map shows the three transmitters at the hyperbola intersections. Estimated coordinates are all in Russia: I (red) 54.9 N 21.5 E Kaliningrad area, II (green) 57.0 N 28.3 E near Krasnogorodsk (Latvian border), III (blue) 60.7 N 29.3 E near Wyborg (Finnish border). Interestingly, this constellation might well serve the eastern Baltic area, which has recently been affected by widespread GNSS jamming. With this information in hand, a Google search indeed brought up a couple of relevant hits, mentioning a little-known Russian radionavigation system called "MARS-75". https://www.udxf.nl/Radio-Navigation-Signals.pdf "MARS-75 is the Russian answer to DECCA. This Soviet made system is also a hyperbolic LF system, but more advanced. MARS-75 was first used in the early seventies and is still in use. The former Baltic chain was closed down after the fall of the USSR. The transmitters in Wismar (E. Germany) and Tallinn were dismantled. There are still chains along the western and northern coasts of Russian like the Gulf of Finland and the Barentsz Sea. MARS-75 operates in the 64-92 kHz range. The transmitters are arranged in groups of three or four. Spreadspectrum signals are used and there are nine clusters. At the moment some 47 transmitters are still running." https://podlodka.info/education/35-technical-aids-to-navigation/710-mars75.html has a short technical description (in Russian), which seems to corroberate most of the conclusions mentioned above. 73, Markus pws Feb 25 #18053 Markus, Confirmed from Kiel. See attachment. (Firework of whistlers included...) 21:05 utc, E-Field antenna with BF982, Xonar U7, 16ys old thin client. On 25.02.24 02:16, Markus (DF6NM) via groups.io wrote: Tonight I noted that DCF77 is down, so I wanted to check on a couple of Kiwi-SDR's. When going to DL7NN's Kiwi in Oberlungwitz (Saxony), I heard the very same little blips between about 66 and 90 kHz (perhaps also higher but hidden under Loran / Chayka). Screenshot attached. Very strong in Finland http://kareliamwdx.ddns.net:8073/ Frequency 70 kHz, USB. Peter, df3lp df3lp_2024-02-24_2105_EF.jpg Markus (DF6NM) Feb 25 #18052 Tonight I noted that DCF77 is down, so I wanted to check on a couple of Kiwi-SDR's. When going to DL7NN's Kiwi in Oberlungwitz (Saxony), I heard the very same little blips between about 66 and 90 kHz (perhaps also higher but hidden under Loran / Chayka). Screenshot attached. 73, Markus kiwi_dl7nn_240225_0058_dcf77off_blips.jpg pws Jan 28 #17922 Johan, The magnitude pattern looks threefold: strong/medium/weak marked with I/II/III inside the attachment. A full sequence (I-III) indeed actually takes approx. 4.4 seconds. There are 2 or 3 "back steps" in between each single "staircase", visible between I and II. A quick look around 1400 utc today shows only "staircase" no. I. No idea what kind of signal we are observing. Maybe it's not identical but from comparable sources? 73 de df3lp, Peter df3lp_2021-04-22_13:34b.jpg Johan Bodin Jan 28 #17921 Hi Peter, thank you! The signal that you observed in 2021 is indeed similar but the timing looks different. The signal is active as I type this. The sweep interval is about 4.5 seconds. A short audio recording is attached, wav, 11025Hz, 16-bit, mono. Parts of the two sweeps encircled below can be heard in the file: Last time I heard it there were sometimes two alternating sweep patterns that sounded different. Possibly also with different signal strengths but I'm not sure. 73 Johan SM6LKM Den 2024-01-28 kl. 10:39, skrev pws: toggle quoted messageShow quoted text audio_69400Hz_12-37-58_28-01-2024.wavaudio_69400Hz_12-37-58_28-01-2024.wav pws Jan 28 #17917 Hi Johan, TNX for observation report! I have been observing this/such a signal for many years. Since my monitoring site is right next to the airfield at Kiel I always thought it was a local signal (DME, ILS etc.). Attached a screenshot from 2021. 73 from Kiel, Peter, df3lp df3lp_2021-04-22_13:34:07.jpg Johan Bodin Jan 26 #17904 Hello, A few minutes ago I spotted a strange signal with short pulses that sweeped upwards in frequency. It was visible from approximately 60 to 85 kHz where it drowned in the Loran rattle. Screenshot from my RX: It was not local interference. I heard the same signal on some online receivers around Europe including SM1OII's KiwiSDR in JO97CE: http://158.174.17.55:8073/ : The signal is gone now. Does anyone know the origin of this signal? 73 Johan SM6LKM