03-17-2025, 04:15 PM
More Drones. I can't help myself :-).
It appears that Russian troops still prefer the DJI Mavic 3 to the "similar" drones made in Russia from Chinese parts. That is a bit counter-intuitive because the Russian built drones operate in different frequency bands and therefore should be less prone to jamming. There are other possible interpretations of the Russian troops asking the public to supply DJI drones. First and most compelling one is that DJI drones are available to the general public. The bespoke ones provided by Russian manufactures are generally not. Therefore asking the public for them is somewhat pointless. Even if these drones were available DJI's production undoubtedly dwarfs anything being done by Russia.
Second, the "official" drone units of the Russian Army are battalion level assets, although often they are treated as regimental assets. Any given infantry company is therefore at the mercy of higher formations as to whether they have drone support at all, and if they do, how much. The higher formation's priorities may not be the infantry company's priorities. Many (most?) of the troops soliciting drones from the general public are probably "unofficial" drone units created by company-sized organizations trying to survive. They aren't eligible for drones through the regular supply systems and have no training on operating them if the could get them. Thus DJI products are the only viable solution for them. The existence of such units may be tenuous if one accepts this ISW analysis.
The article even more importantly confirms that Russia has the ability to flash (program the onboard non-volatile memory) in DJI drones with new firmware. This supports my conjecture that Russia has a copy of the source code. The article also references Ukraine using a USB stick to upload new (or patch the existing) firmware to disable the DJI "house" ID messages. The ability to do this confirms Ukraine either has the source code or has substantially reverse engineered it. It is a little disconcerting that different Ukrainian units seem to have software of different origins. That flexibility can speed up solving problems, but if the solutions are not propagated quickly to other units the benefits could be wasted. It may also lead to duplication of effort. The early stages of this effort are described here.
Other organizations had reverse engineered the DJI "house" ID messages prior to the start of the Ukraine war. The existence of these messages was also confirmed by hackers who reverse engineered their content. The message includes both the GPS location of the drone and of the drone operator(!). This message type was transmitted for at least three months by all DJI drones operated by Ukraine. DJI sells their AeroScope system to track DJI drones. It uses the DJI "house" messages to do so. The DJI "house" message is one of TWO different drone ID messages routinely broadcast by DJI drones. The second message type is mandated by the FCC/EU. Putting the drone in "FCC mode" to increase power will enable the FCC message type. Presumably Ukraine and Russia have also modified the code to prevent that message from being transmitted.
The section of the prefer the DJI Mavic 3 article titled "Avoiding The Jam" states
Quote:What the firmware hack cannot change is the drone’s operating frequencies. This makes the DJI drones easier to jam than locally made FPVs which are made with an increasing variety of different frequencies, as well as using frequency-hopping and other techniques to evade jamming.If read literally this statement is false. The DJI Mavic 3 series is equipped to operate in the 2.4-2.5 GHz ISM band and also the 5.725-5.875 GHz ISM band. These frequency BANDS are fixed by hardware installed. However the exact FREQUENCIES used in these bands (and the pattern of their use) can and has been changed by Ukrainian extensions of the DJI the firmware. There are indeed hardware limitations on the resolution of frequencies that are available to the firmware. However these limitations are much less restrictive than the set actually used in a "stock" DJI product.
The ability to alter the exact frequencies and the hopping pattern can defeat some kinds of jammers. A barrage noise jammer won't care. A multi-tone jammer may be defeated. Jammers that take advantage of "known" DJI signal characteristics will very probably be defeated. As a trivial example a DJI drone with modified firmware can move the control signal between the 2.4 GHz ISM band and the 5.8 GHz ISM band at will to avoid jamming. This is a "macro" example. There are many more subtle approaches. An example of this extracted from a relatively recent article about Ukrainian drones on the battlefield.
Quote:The Kremlin is shipping new electronic-warfare radio jammers in a desperate effort to protect assault groups, the blogger explained. But they don’t work very well. “The enemy adapts extremely quickly, changing frequencies,” according to the blogger, “while our E.W. systems at best cover less than half the frequencies they use.”
The locally made FPVs in general attempt to avoid jamming by operating in BANDS different from the ones used by the Mavic drones. Most jammers are targeted to a specific frequency range. This may be an entire band or a subset thereof. For instance, a jammer may cover the 2.4-2.5 GHz ISM band. If the control link is moved to the 5.8 GHz ISM band, that jammer becomes useless. If an additional 5.8 GHz ISM band jammer is brought online, the DJI drones may have no answer, depending on the type of jammer used. The drone has only two bands, and with the "stock" DJI drone hardware one cannot chose what bands they are. Of course, the opponent needs twice as many jammers to jam both bands simultaneously. Locally made drones can chose a different band for the "second" frequency range. In principle, it can be anywhere in the 0.3 GHz to 18 GHz frequency range. Going lower in frequency creates problems with antenna size. Going higher creates problems with increasing path loss. In a previous post we see Russian using the 1.4 GHz frequency range. Jammers are a finite resource. It is prohibitively expensive (even for the US) to have even 20 watt jammers contiguously covering 0.3 GHz to 18 GHz located everywhere Ukrainian drones may be encountered.
Once again, it must be pointed out that "stock" DJI drones utilize frequency hopping. So did (do) GSM cell phones. Noting all that special about it. Hopping can defeat some types of jammers but not others. In general, fast hoping works "better" than slow hopping. DJI drones can't hop fast enough to matter much. Anything slower than 1000 hops per second was considered "slow" in the early 1990s.
The following are "leftovers" from my previous efforts that are IMO interesting but that did not fit well into the narrative.
At 26:19 into the video [Ref 1] shows a simple, compact "drone detector". It has two antennas. The longer one appears to be an appropriate length (3.15 inches) for a 1/4 wavelength antenna in the 850-930 MHz frequency range used by the Orlan-10. The shorter one appears to be a an appropriate length (2.41 inches) for a 1/2 wavelength antenna in the 2400-2500 MHz frequency range used by DJI compatible drones (as well as most other commercial drones). Hard to judge exactly from the pictures. The electronics inside probably consists of two band pass filters and two successive detection log video amplifiers (SDLVA), most probably an AD8313 (readily available on eBay or commercially for more $). A small micro-controller (ESP32-<something>) and a few "piece" parts completes the device. Parts cost about $75, less if you hunt around a bit. There are alternative approaches but I am guessing a "broadband" approach is desirable/necessary for the frequency hopping control signals. The 70 dB dynamic range of the SDLVA allows for high sensitivity and "good" dynamic range.