Hunting a red demon or satellite navigation direction finder

Do not think that I am some criminal element, paranoid or unfaithful husband, but once I decided to buy a GPS jammer. Firstly, the process itself was interesting. After all, the case seems to be illegal, the seller seems to be taking risks. I wanted to take part in it, and not just listen or read. Secondly, it was technically and economically interesting to see how it was done and how much it costs.



What it led to read on.





Why GPS jammers and GPS spoofers are needed, there is no need to tell. Today, everyone knows that even the "new aristocracy" uses them for greater security. Especially know the residents of Moscow. What can we say about ordinary people? Well, when we had someone cares about the minor inconveniences of a neighbor? "If" They "do not care, then I am all the more. If I need - cut in and drove off. ”

')

But the world is seriously stuck in technology. Today, the navigator is a continuation of the brain of a considerable number of drivers, especially the "fair-haired" of their representatives. You can smile and complain, but this process is developing. And what will happen if they all chop off the navigation?



Now imagine the future robo-driver ...



Not that I personally urged not to use jammers. On the contrary, I personally would be less interested in doing the techniques described below without them. Who is war, and who is ...



Summary - the enemy must be known in person. I know that jamming is a small thing and that there is the work of Professor Ebinuma on more “fine tuning” of satellite navigation, but the simplest method of the enemy is also interesting to me.



The cheapest way to cut down the navigation cost me 2500 rubles.





The thing is simple, but in a package, the kind of which gives brain diseases. I, at least, immediately became infected. Here, I think, is a good sarcophagus for your favorite mouse ( ambulance! ). And it probably costs as much as the device itself inside.



And how rich is his spectrum (video)!





Further, this infection in me progressed into destructive actions and I disassembled it:





Everything is very simple and tasteful, almost rustic. And now you all understand why the demon is red. (Although I also like The Story of the Red Demon.)



But what is it in action!





A smashing fire of the Inquisition! Now, every kid, having saved 30 times on a beer, can “burn” all the navigation in the area. Here I want to note that I ran this thing outside the city. No motorist was hurt.



And in obedience to fate, I was kindled by the desire to reclaim this fiend. I began to do the direction finder using my favorite augmented reality and the wonderful NT1065 chip.



NT1065 contains four radio paths that can be configured to use a single local oscillator. This is evident in the structural scheme:





Then all four channels will be in phase with the accuracy of non-identities of the radio paths. The figure below shows the spectrum of phase differences.





It can be seen that the phase differences are stable, so it is possible to create a four-channel phase finder based on the NT1065.



We take the fee from the previous article , add to it four antennas with low-noise amplifiers (LNA). It was possible without the LNA, but I wanted to slightly raise the signal in the bands where the antenna is not well matched.





The high-pass filters recommended for installation at the NT1065 input are omitted for simplicity and for the flexibility of the direction finder. The antennas are tuned only to the upper navigation range (L1), but, with eyes closed for sensitivity, you can try using the direction finder on the lower ranges (L2, L3, L5, etc.). We'll see later what happens.



Here such turned out a piece of iron:





I had to tinker. to put everything on one side, because on the other hand only the antenna. Without ignoring the recommendations for wiring circuits has not done. I will lick then.



And now - software (video).







The source is here .



Briefly, the software works like this: we take some continuous samples, make them FFT (with a window, of course), average the FFT frequency counts (bins) in the user-selected band and calculate the bearing. All these operations are standard, which are boring to describe to the educated public, except for calculating the bearing.



Here everything is done in a simple way: the algorithm can be called “beamforming”, in English there will probably be a beamformer - we find the direction of the main lobe of the pattern that most corresponds to the signal received by the antenna array. The formula will be as follows:





For specific code, see GitHub for anyone interested.



Further, we postpone 1 percent from the maximum of the radiation pattern and make a cut. We obtain an angular region, a certain “cloud” of angles at which the signal source is most likely located. This area of ​​the corners is shaded in real video from the camera of the tablet. In general, the piece of iron looks like this (as usual mounted on the back of the tablet):







And it works like this: (video) .



You can also select the position and width of the processed frequency band and set the threshold for the signal level. There is also an indicator of the total power in the band, which allows you to first roughly find a sector.



If you compare this phase direction finder with my previous amplitude for WiFi , then this one is fundamentally more sensitive. Well, it is no longer manual, but automatic. That is, in principle, it is not necessary for them to rotate. But it is still advisable to do this, since accuracy is best of all in focus. And on the reverse side, it does not work either. There are less than any internal antenna distortion.



But, as the video shows, nobody canceled the interference. If there is no direct beam, the distortion can be significant. And there is also some polarization dependence.



This is what I want to direct my efforts further.



PS: On the lower range, in the navigation wording L2, L3, L5, etc., while not very working. Need to smoke calibration.