Secrecy Capacity in Large Cooperative Networks in Presence of Eavesdroppers with Unknown Locations
This addresses security challenges in realistic wireless networks for applications like IoT and military communications, though it is incremental by building on prior work with more practical assumptions.
The paper tackles the problem of achieving secure communication in large wireless networks with unknown eavesdropper locations, proposing a stochastic virtual beamforming scheme that achieves an unbounded secure rate with any desired outage level as legitimate user density increases.
In this paper, an extended large wireless network under the secrecy constraint is considered. In contrast to works which use idealized assumptions, a more realistic network situation with unknown eavesdroppers locations is investigated: the legitimate users only know their own Channel State Information (CSI), not the eavesdroppers CSI. Also, the network is analyzed by taking in to account the effects of both fading and path loss. Under these assumptions, a power efficient cooperative scheme, named \emph{stochastic virtual beamforming}, is proposed. Applying this scheme, an unbounded secure rate with any desired outage level is achieved, provided that the density of the legitimate users tends to infinity. In addition, by tending the legitimate users density to the infinity, the tolerable density of eavesdroppers will become unbounded too.