Share on facebook
Share on twitter
Share on linkedin
Share on reddit
Share on whatsapp

Parasitic Communications


One of the challenges of operating in built-up areas is communications, urban canyons make conventional HF communications difficult and satellite communications even more difficult. It is a very broad brush statement but the lower frequency the greater the range but higher attenuation and lower bandwidth.

This is why mobile telephone networks, especially 3G, with a requirement for many users in a small geographic area and high bandwidth use higher frequencies, GSM is typically 900MHZ for example, compared to 3 to 30MHZ for HF and 30 to 300MHZ for VHF. With a lower frequency also comes a higher wavelength and it is the wavelength that dictates antenna size.

The downside to a mobile network though is the need to have a fixed infrastructure of transmitters with a relatively small geographic footprint. This makes it wholly unsuitable to the needs of military communication. Military communications, therefore, must bring its infrastructure with it and with this comes cost, complexity, weight and the need for lots of power. That power is usually provided by vehicular or portable generators that need fuel, fuel needs transport, transport needs fuel, people and security, and they need feeding and well, you get the picture.

So anything that reduces power reduces the logistic tail and is ‘a good thing’

A story this week caught my eye on the subject of hijacking domestic wireless routers to create an ad-hoc network for emergency responders.

The paper describes an experiment to measure the density of wireless access routers and those with open access in a typical small town in Germany. It then discusses having a remote switch capability built-in that allows authorised first responders to utilise those devices in an ad-hoc peer to the peer mesh network. Once the network is established any type of traffic can be passed over it, encrypted voice or data for example.

WiFi has very low power requirements in comparison with HF or VHF for example, as I said above, low power is always a good thing.

Of course, this is not any form of solution on its own and there is the small issue of the absence of any form of first responder switch in current designs, no legislative or regulatory framework and the lack of WiFi hotspots in the Upper Gereshk Valley but the principle is nonetheless interesting.

The Future Character of Conflict recognises the increasing likelihood of operations in urban areas and urban areas are likely to have an increasing number of wireless routers, even those is non-western countries, whilst still being a problematical environment for communications as today.

In the future, communication devices may be able to hijack any number of networks to create an ad-hoc mesh, WiFi, GSM, 3G, Tetra and smart meter networks in addition to using traditional HF/VHF/UHF radios.

Click here to read the paper

Share on facebook
Share on twitter
Share on linkedin
Share on pinterest

11 Responses

  1. Very interesting post TD – will probably respond once I’ve read the report. The future character of conflict document also looks interesting.

  2. Wi-fi may be the way to go for urban combat communications but I don’t see civilian routers being of much use; for one there may be power issues.

  3. wasn’t that the drive for unmanned blimps for 4G (LTE) as they would be pointing down with no obstructions, and requiring little power, they could rely on solar power, plus being 60,000 feet up above the cloud for good line of sight of the sun.

  4. Headlines of “I forgot to set security on my wi-fi & found the army on my roof”.

  5. TD,

    Zounds – a non-naval post…. Well done, that blog.

    A couple of observations: there’s a general, if not yet complete move away from 800/850/900 MHz to the 1800/1900/2100 MHz bands for mobile telephony, so you are getting roughly double the bandwidth. No one’s phone these days has an extendable antenna, a by-product of this shift.

    I’m not convinced about the hijacking of scattered divvy routers. Working for a company that does cyber security to a class-leading level (as well as other engineering shit) I can confidently state that a WEP password is crackable in less than a minute from a laptop bought from PC world for £300, but a WPA2 password is several orders of magnitude more difficult. There are ways around that: a “Government” password that could in theory bypass the WPA2, but that’s difficult to coordinate among mostly Asian manufacturers of the routers and seems likely to spawn a black market of “non-Government” equipment, plus all sorts of privacy issues.

    Then there’s the arms race of encryption versus cracking, and it is one that cracking is losing. Double-encryption is pretty easy these days, and while the most advanced cracking technologies are both not available on the open market and can still defeat all known commercial encryption (I’m not going into detail on that), they are unlikely to be available to an infantry section hurrying past a tower block. If I really wanted to secure a wireless link I’d double encrypt using a local and an international double-encryption – perhaps an American algorithm then itself encrypted with an Israeli algorithm employing a different character set.

    It’s probably easier to tap into a local exchange and set up a local mesh network from that. You need a Landrover of equipment to do that plus some ECE’s, but in the UK’s case, you then have access to the full local loop, giving you small town coverage.

    You’ve got a good point on Smart Meters: there is currently no security layer or encryption built into the nascent standards. We’re talking with some utilities companies about that. At the moment, any hacker can access the data and run a small script to show your pattern of life (ie electricity usage, so when you are not at home in order to have an undisturbed burglary. Word to the often very wise TD community: don’t sign up for anything wifi on a Smart Meter just yet, if you must go for the wired solutions).

    Love the StratSat / airship type comments above. You get megalopolis coverage from a single airship at 70,000 feet, if someone is brave enough to fund the initial development. Malaysia damned nearly did in 2002.

Comments are closed.