I'm not so sure... in cellular systems, cells must know where the handsets are located in order to send incoming calls. Your transmitter has a physical location which could presumably be tracked in the normal manner, and I would expect the overall routing information in a net to be susceptible to traffic analysis in any case. An individual who is using the system to communicate wouldn't be able to find the physical address of another user, but e.g. an intelligence agency which was looking at the entire network would. Even assuming spread-spectrum and various link encryption techniques on top of whatever end-user encryption is supplied; with enough traffic and enough time, it should be possible to do TA.
Cell systems have to know which cell you're in (visualize each cell as a circle centered on the cell site) to know how to route a call _to_ you. I believe the MTSO (== cellular CO) will route the call directly to the cell site for the cell you're in. Of course, the cell will also know what cell you're in when you originate a call. This knowledge is useful for traffic analysis, but it's also required for the system to be able to route incoming and outgoing calls. An alternative is the ham packet radio-style addressing of user@node@node..., where "user" represents the call sign of the intended receiver and each node represents the call of a digipeater between the sender and recipient. The hard part here is that you must be able to dynamically generate a route between Alice and Bob if either of them move from their last known location. Of course, there's always store-and-forward. A spread-spectrum network of small digipeaters, combined with crypto remailing and pool software, would really be something. I'm not sure that it would work well for spread-spectrum SLIP, though. -Paul -- Paul Robichaux, KD4JZG | "Change the world for a better tomorrow. But perobich@ingr.com | watch your ass today." - aaron@halcyon.com Intergraph Federal Systems | Be a cryptography user- ask me how.