Although air-gap computers have no wireless connectivity, we show that attackers can use the SATA cable as a wireless antenna to transfer radio signals at the 6 GHz frequency band.
Table IV presents the signal-to-noise ratio (SNR) received with the three transmitting computers. The signal transmitted from PC-1 has a strength of 20 dB at 30 cm to 9 dB at 120 cm apart. The signal generated from PC-1 and PC-2 were significantly weaker, with 15 dB at 60 cm (PC-2) and 7 dB at 30 cm (PC-3).
另外大概是 PoC 的關係,只有簡單測一下是可行的 (對於真的有利用 air gap 的環境當作一種保護機制的威脅就夠大了),看起來沒有測極限可以跑多快:
We transmitted the data with a bit rate of 1 bit/sec, which is shown to be the minimal time to generate a signal which is strong enough for modulation.
關於反制的部份,這類的技術 (透過電磁訊號) 之前在其他的裝置上都有發生過,目前的 air gap 標準應該都有電磁訊號洩漏的防範了,這篇主要還是在展示 SATA 也可以這樣搞 XD
They had built a network on their own, which provided students Internet connection from 2002 till 2013 (when the administration effectively legalized this network).
In the beginning, there were only 10 Mbps hubs available. They were extremely slow because the packet arriving at one port was mirrored to all other ports, flooding the network with unnecessary traffic. Later they were replaced by L2 switches, that are smart enough to choose the destination port for the packet.
但宿舍網路是有競爭對手的,兩邊常常破壞對方 XDDD
Having two competing networks was a cherry on the top of it. They were cutting each other’s cables, executing DDoS attacks, stealing equipment, etc.
To connect this router to the other parts of the network, we laid fiber cables through secret passages in the walls, ceilings, floors, and ventilation shafts.
Most of the time, we managed to reduce the number of cables coming out of the window down to ~15. We could do this because of a simple trick: the regular ethernet cable (Cat 5e) has 8 wires. And to transmit 100 Mbps, you need just 4 of them. All 8 wires are required only for a 1 Gbps connection. So using a single cable, you can connect 2 clients who live close to each other. We always tried to do that if possible because the hole can’t fit so many cables.
The new pi has been released and it has a USB Type-C connector for power however people are finding some chargers are not working with it (notably macbook chargers). Some have speculated that this is due to a manufacturer limitation on the power supplies however it is actually due to the incorrect detection circuitry on the Pi end of the USB connection.
這樣說有點偏頗,但是 Macbook 的充電器一向是 Type C 裡的指標,如果這顆充電器跟其他裝置配合上有問題,通常都是代表其他裝置的實作有問題... (噗)
這次發現的電阻問題看起來有點苦 (看起來需要改版子),目前文章作者建議的 workaround 主要就是「不要用那麼好的設備」,比較簡單的包括了 Type C 的線不要那麼好 (像是找充手機用的線就好,不要找拿可以跑 5A 的線),或是透過 Type A 轉 Type C 的線也應該可以避開這個問題,最差的情況應該是找其他的充電器:
Now onto some solutions. Assuming the issue you are having is caused by the problem discussed above, using a non e-marked cable (most USB-C phone charger cables are likely this type) rather than an e-marked cable (many laptop charger/thunderbolt cables and any 5A capable cable will be in this category) will allow for the pi to be powered. In addition using older chargers with A-C cables or micro B to C adaptors will also work if they provide enough power as these don’t require CC detection to provide power. Ultimately though the best solution in the long run will be for there to be a board revision for the pi 4 which adds the 2nd CC resistor and fixes the problem.