Mesh Networking: Another Way to Bring Cell Phone Service to North Korea?

This video gives a simple explanation of the concept of mesh networking, which allows android phone users who download some additional software to connect with each other wirelessly without a base station or cell phone towers. An Australian group known as The Serval Project is trying to raise funds to test and prototype the technology, and OFK reader Josh Hansen wrote me a few weeks back to start a discussion about the potential this technology could have for bring cell phone service to North Korea, without the obvious involvement of any foreign government.

Here’s how the Serval Project’s founders explain the potential for mesh networking to penetrate closed societies:

Mesh Networks in Authoritarian Regimes, with Dr. Paul Gardner-Stephen, founder of the Serval Project by salimfadhley

Several months ago, I wrote about the potential of cheap portable base stations to cover much of North Korea with a cell signal. The obvious drawback to that concept is that this system still depends on a centralize network with base stations, which would have to be hosted on South Korean territory. South Korea probably still lacks the testicular fortitude to allow that.

Personally, I lack the technological knowledge to say whether or not this could work. I’d be interested in your thoughts below, in the comments.

10 Responses

  1. Yes, this would work. Videos, pictures, and reports could be shared even if contact between nodes were temporary and rare. If a significant network could be built up, realtime information could be available. This network would also be more resilient than the governments network which could be sabotaged.

  2. I’m sorry to be the one to pour cold water on things, but mesh networks require nodes to broadcast their presence, and to continualy exchange information about which nodes are in range of which other nodes. I can see how this would be useful during a mass uprising like in Egypt, but day to day in north korea I think the governments ability to track down nodes woul be enough to kill it. Basetations are more realistic then, because all the continulally broadcasting elements would be beyond the reach of the regime.

    (Assuming that people aren’t standing by to accept calls when not talking on the phone. It is my understanding that north koreans with phones along the northern border either prearange times to comunicate, or call out to people outside korea, so that they don’t have to risk being tracked down when not speaking on the phone.)

  3. I’m sorry, but it won’t work. It’s a good idea, but the nodes would be very quick and easy to track down. There is just no reliable and repeatable method for obfuscating the node’s location.

    Cell phones on the border are simple enough. Walk to a remote area, turn the phone on, talk a short while, turn the phone off, and return home. The problem with a mesh network is that the phone must be on at all times, and of course needs to be powered, so people will probably keep them in their homes.

    This isn’t to mention the limited range of these nodes. The mesh needs to be contiguous in order to function and I doubt that everyone involved would be able to get it more than a few miles from the border if international dialing is a key part of the plan. Then there’s the eternal problem of introducing high-value items in impoverished nations (I can either run this cell network node, or I can sell the phone and feed my family).

  4. Hmm, I’m not entirely sure I understand the question.

    From a purely technical standpoint, the concept works just fine. There are some logistics involved to work out, but it is quite plausible. The problem is protecting those who participate in the mesh from being exposed. At this time, there’s no way to do that.

    The other issue is that of population density vs. mesh phone range. Cell towers are large and high powered, so they can reach for miles and miles. I would appreciate input from a broadcast engineer more familiar with the technology, but phones that form a mesh network need to be relatively close together. So urban areas could create a great mesh, and rural areas would be very difficult. If you want to target specifically cities, then you could probably create a local mesh and have a satellite phone connect to the outside world.

  5. On a related note, are satellite phones traceable in the way that regular cell phones are? I.e., are they as dangerous to the user in terms of being discovered remotely? Been wondering this for a while.

  6. Please note that I am not a SIGINT engineer and you would want to consult one before making decisions.

    Sat phones are traceable, but not quite as easily. Cell phone signals penetrate walls and go over hills and valleys to reach towers that are attached to the ground. So the signal goes all over the place and is easily picked up.

    Sat phones on the other hand are line-of-sight and meant to reach receivers that are up in the sky. You can’t use them indoors because the frequency used has little penetration power. Tracking down a sat phone user would be plausible using properly equipped planes with its eyes on the ground. This would be a considerable expense.

    Were the user to make use of a unidirectional antenna, focus it square on the satellite, and power it down to the minimum needed for a reliable connection, it would be extremely difficult for any surveilling craft to locate the user without flying directly between the user and satellite.

  7. The need to avoid detection is definitely the weakness of mesh networking under a regime such as North Korea’s. However, I believe it’s a weakness that could be managed. What if, instead of sending voice traffic across a continuously connected mesh, users were sending text traffic across an intermittently connected mesh? Text data is vastly smaller and so can be transmitted in one quick burst from peer to peer. Mobile phones being mobile it would be easy to be on the move while such a transmission was made, thus making it even harder for officials to track users. Add to that a crowded environment such as a street or market, and the chances of remaining anonymous just might be high enough to make it sustainable.

  8. Well reasoned, but it’s too easy to jam with the same ubiquitous, off-the-shelf equipment available to the western consumer. A better chance (one that a friend and I are starting to work on) is mesh networking on bluetooth network freqs. since the range is limited detection is difficult. Moreover, jamming is hard to do in a opposition controlled area of any size on frequencies that peter out that easily at short distances.