Mobi-cliques for improving ergodic secrecy in fading wiretap channels under power constraints

Abstract

We consider a cooperative secret communication scenario, in which a group of mobile and power constrained nodes, acting as relays, cooperatively transmit to a destination in the presence of an eavesdropper; both destination and eavesdropper are assumed stationary. The cooperative scheme entails motion control and optimal communication, in order to achieve a prescribed level of ergodic secrecy rate. The group of motion-controlled cooperating relays is here termed as mobi-clique. Under this setting, a novel, decentralized motion control scheme is derived, which effectively drives the relays to a formation configuration, so that a prescribed expected secrecy requirement is met, while at the same time the utilization of network resources is maximized. The effectiveness of the proposed approach is verified both theoretically and through numerical simulations.

Spatially Controlled Relay Beamforming

“…In this paper, we try to overcome this major limitation, and we consider the problem of optimally and dynamically updating relay positions in one source/destination relay beamforming networks, in a dynamic channel environment. Different from [8][9][10], we model the wireless channel as a spatiotemporal stochastic field ; this approach may be seen as a versatile extension of a realistic, commonly employed "log-normal" channel model [20]. We then propose a 2-stage stochastic programming problem formulation, optimally specifying the positions of the relays at each time slot, such that the Signal-to-Interference+Noise Ratio (SINR) or QoS at the destination, at the same time slot, is maximized on average, based on causal CSI, and subject to a total power constraint at the relays.…”

“…In [9], in the framework of information theoretic physical layer security, decentralized jammer motion control has been jointly combined with noise nulling and cooperative jamming, maximizing the network secrecy rate. In [10], optimal relay positioning has been studied in systems where multiple relays deliver information to a destination, in the presence of an eavesdropper, with a goal of maximizing or achieving a target level of ergodic secrecy.…”

“…In [8][9][10], the links among the nodes of the network (or the related statistics) are assumed to be available in the form of static channel maps, during the whole motion of the jammers/relays. However, this is an oversimplifying assumption in scenarios where the channels change significantly in time and space [18][19][20].…”

Asymptotically Optimal Discrete-Time Nonlinear Filters From Stochastically Convergent State Process Approximations

“…To better motivate the reader, let us describe two problems that fit the scenario described above and can benefit from the contributions of this paper, namely, those of sequential channel state estimation and (sequential) spatiotemporal channel prediction [13] (see also [14]). The above problems arise naturally in novel signal processing applications in the emerging area of distributed, autonomous, physical layer aware mobile networks [15][16][17]. Such networks usually consist of cooperating mobile sensors, each of them being capable of observing its communication channel (under a flat fading assumption), relative to a reference point in the space.…”

Nonlinear spatiotemporal channel gain map tracking in mobile cooperative networks

“…This work is supported by the National Science Foundation (NSF) under grant CNS -1239188. evolves through space is tightly connected to the ability of a network to assess the quality of the channel at previously unexplored locations in space. Such knowledge would be highly beneficial in a number of new and important applications, such as mobile beamforming [4] and mobility enhanced physical layer security [5], [6], in which both transmission weights and node locations become decision variables.…”