We investigate the problem of secure communication in a simple network with three communicating parties, two distributed sources who communicate over orthogonal channels to one destination node. The cooperation between the sources is restricted to a rate limited common random source they both observe. The communication channels are erasure channels with strictly causal channel state information of the destination available publicly. A passive adversary is present in the system eavesdropping on any one of the channels. We design a linear scheme that ensures secrecy against the eavesdropper. By deriving an outer bound for the problem we prove that the scheme is optimal in certain special cases.
We investigate the problem of secure communication in a simple network with three communicating parties, two distributed sources who communicate over orthogonal channels to one destination node. The cooperation between the sources is restricted to a rate limited common random source they both observe. The communication channels are erasure channels with strictly causal channel state information of the destination available publicly. A passive adversary is present in the system eavesdropping on any one of the channels. We design a linear scheme that ensures secrecy against the eavesdropper. By deriving an outer bound for the problem we prove that the scheme is optimal in certain special cases.
We characterize the secret message capacity of the triangle network, that consists of a source, a relay and a destination connected through orthogonal erasure channels. A passive eavesdropper, Eve, wiretaps any one of the three channels. The source and the relay can each generate unlimited private randomness; the relay and the destination can publicly provide strictly causal channel state information. Our achievable scheme is expressed through a linear program (LP) with 11 inequalities that captures a minimal set of secret key generation methods and the use of them for message encryption. Our outer bound is expressed also through a linear program, in this case with 41 constraints, constructed from general information inequalities. We prove that the optimal value of the outer bound LP is no larger than that of the scheme LP, which implies that the solution of the achievable scheme LP is the capacity. We find that equipping the relay with private randomness increases the secrecy rate by more than 40% in some cases and that cut-set bounds, directly applied in the network, are not always tight. Because the derivation of the inner and outer bound are both lengthy, we describe in this paper the achievability scheme, outline the outer bound, and provide the full derivations online [1]. We also make available Matlab functions that take as input the erasure probabilities and evaluate the inner and outer bounds.
We present achievability schemes for secure message transmission in the presence of a passive eavesdropper, over arbitrary networks, where each edge corresponds to an independent erasure channel and we have available public channel state feedback. Our schemes are efficiently described through linear programming formulations, that generalize the information flow linear programs to networks with security constraints; different sources of secrecy show up as parameters in our linear programs, and solving the optimization problem enables to select the optimal (in terms of rate) scheme that integrates them. As far as we know, these are the first linear programming formulations for secret message transmission over arbitrary networks with erasures and channel state feedback.
We investigate the problem of information theoretically secure communication in a line network with erasure channels and state feedback. We consider a spectrum of cases for the private randomness that intermediate nodes can generate, ranging from having intermediate nodes generate unlimited private randomness, to having intermediate nodes generate no private randomness, and all cases in between. We characterize the secret message capacity when either only one of the channels is eavesdropped or all of the channels are eavesdropped, and we develop polynomial time algorithms that achieve these capacities. We also give an outer bound for the case where an arbitrary number of channels is eavesdropped. Our work is the first to characterize the secrecy capacity of a network of arbitrary size, with imperfect channels and feedback. As a side result, we derive the secret key and secret message capacity of an one-hop network, when the source has limited randomness.
The growth of wireless networks has been considerable over the last decades. Due to the broadcast nature of these networks, security issues have taken a critical role in today’s communications. A promising direction towards achieving secure communications is information theoretic secrecy, which is an approach exploiting the randomness of the channels to ensure secrecy. Based on this approach, there has been a recent surge of interest in a potential cooperation between users to enhance the secrecy of communications. In this thesis we investigate the interaction between cooperation and secrecy. In particular the contributions of the thesis can be divided into two parts. In the first part, we study cooperative strategies for secrecy for wireless channels. Our goal is to evaluate the effect of fading and limited CSI on the eavesdropper’s channels. In that purpose we consider a scenario where a helper aims at increasing the secrecy of the communication between a source and destination in the presence of an eavesdropper. Several strategies are discussed for the helper, namely decode-and-forward, amplify-and-forward, and cooperative jamming. We introduce the secrecy outage probability, the conditional secrecy outage probability and the secure throughput as secrecy measures. For each measure, we investigate and compare the secrecy performance of cooperation. We furthermore elaborate a system optimization in terms of strategy selection, node positioning, power allocation and rate design. In the second part, we consider cooperation in the 4-node scenario against a more sophisticated adversary: an active eavesdropper, which can either passively eavesdrop, or jam the transmission. A game-theoretic perspective is a natural way to analyze the competitive interaction between the helper and the eavesdropper. Therefore we define several secrecy games, for which we find the Nash and Stackelberg equilibria as well as the corresponding secrecy rate outcomes. Another important consideration in this scenario is the interaction between the source and the helper, which we model and solve as a Stackelberg game, and we illustrate its impact on the achievable secrecy rates.
With the advent of mobile telephony, complemented by the availability of economic infrastructure devices has provided users the freedom to communicate and perform everyday tasks wirelessly. The access to information anywhere anytime provide users more flexibility to communicate, nonetheless it also makes communication more vulnerable to eavesdropping. Security is now considered as an important issue in the design of modern wireless communication system. In contrast to the conventional systems where security issues are dealt in the upper\nlayers of the protocol stack, in this thesis we focus our attention to construct security schemes by exploiting the attributes of the physical layer. \nIn the first part of the thesis, we consider a two-user multiacces channel where the two encoders, Encoder 1 and Encoder 2, are allowed to communicate with each other over a unidirectional noiseless bit-pipe of finite capacity in the presence of an adversary. In this model, the role of Encoder 2 is restricted to only help Encoder 1 in securing information from adversary. For this model, we\nestablish fundamental performance limits on secure communication and characterizes the secrecy capacity in some extreme cases of cooperation between the encoders. The second part of the thesis considers an elementary\nnetwork, which consists of four nodes – a source, a relay, a legitimate receiver and a passive eavesdropper. In this network, the link between two nodes consists of multiple parallel channels as subchannel. Towards this end, we establish bounds on secure communication. We show that for the case in which the relay does not listens to the source, under certain conditions secrecy capacity is established. In the last part of the thesis, we studied a two-user multi-input multioutput X-channel with local feedback and delayed channel state information\navailable at the transmitter in high signal-to-noise ratio regimes. In this setting each transmitter transmits independent messages to both receivers that need to be kept secret from each other. We characterize the optimal sum secure degrees of freedom region of this model.
We investigate the problem of information theoretically secure communication in a line network with erasure channels and state feedback. We consider a spectrum of cases for the private randomness that intermediate nodes can generate, ranging from having intermediate nodes generate unlimited private randomness, to having intermediate nodes generate no private randomness, and all cases in between. We characterize the secret message capacity when either only one of the channels is eavesdropped or all of the channels are eavesdropped, and we develop polynomial time algorithms that achieve these capacities. We also give an outer bound for the case where an arbitrary number of channels is eavesdropped. Our work is the first to characterize the secrecy capacity of a network of arbitrary size, with imperfect channels and feedback.