The goal of covert communication is to establish a communication link between two parties (Alice and Bob) while assuring that a warden (Willie) will not be able to detect the very presence of communication. In this work, we address the covert communication problem over a packet insertion channel. Assuming a Poisson channel model, we first consider an all-aware warden, who wishes to minimize the sum of his miss detection and false alarm probabilities. We show that the covertness extent versus Alice's transmission rate can be computed exactly. We also derive simplified bounds which are tighter than previously known results. Such a warden, however, has to be aware of Alice's transmission rate. A more practical scenario, is when the warden does not have this information, hence wishes to operate at a certain false alarm rate. For this case, we discuss Alice's strategy to ensure the warden's miss detection probability is lower bounded. Finally, we study covertness in the case where Willie has uncertainty regarding the nominal channel rate as well, and show that in this case, Alice's covert transmission rate can be even higher.
The goal of covert communication is to establish a communication link between two parties (Alice and Bob) while assuring that a warden (Willie) will not be able to detect the very presence of communication. In this work, we address the covert communication problem over a packet insertion channel. Assuming a Poisson channel model, we first consider an all-aware warden, who wishes to minimize the sum of his miss detection and false alarm probabilities. We show that the covertness extent versus Alice's transmission rate can be computed exactly. We also derive simplified bounds which are tighter than previously known results. Such a warden, however, has to be aware of Alice's transmission rate. A more practical scenario, is when the warden does not have this information, hence wishes to operate at a certain false alarm rate. For this case, we discuss Alice's strategy to ensure the warden's miss detection probability is lower bounded. Finally, we study covertness in the case where Willie has uncertainty regarding the nominal channel rate as well, and show that in this case, Alice's covert transmission rate can be even higher.
Consider a channel where authorized transmitter Jack sends packets to authorized receiver Steve according to a Poisson process with rate λ packets per second for a time period T. Suppose that covert transmitter Alice wishes to communicate information to covert receiver Bob on the same channel without being detected by a watchful adversary Willie. We consider two scenarios. In the first scenario, we assume that warden Willie cannot look at packet contents but rather can only observe packet timings, and Alice must send information by inserting her own packets into the channel. We show that the number of packets that Alice can covertly transmit to Bob is on the order of the square root of the number of packets that Jack transmits to Steve; conversely, if Alice transmits more than that, she will be detected by Willie with high probability. In the second scenario, we assume that Willie can look at packet contents but that Alice can communicate across an M/M/1 queue to Bob by altering the timings of the packets going from Jack to Steve. First, Alice builds a codebook, with each codeword consisting of a sequence of packet timings to be employed for conveying the information associated with that codeword. However, to successfully employ this codebook, Alice must always have a packet to send at the appropriate time. Hence, leveraging our result from the first scenario, we propose a construction where Alice covertly slows down the packet stream so as to buffer packets to use during a succeeding codeword transmission phase. Using this approach, Alice can covertly and reliably transmit O(λT) covert bits to Bob in time period T over an M/M/1 queue with service rate μ > e · λ.
This dissertation focuses on covert communication in channels where the communication takes place by the transmission of packets. Consider a channel where authorized transmitter Jack sends packets to authorized receiver Steve according to a Poisson process with rate λ packets per second for a time period T. Jack's transmitted packet visit Alice, Willie, Bob and Steve, respectively. Suppose that covert transmitter Alice wishes to communicate information to covert receiver Bob without being detected by a watchful adversary Willie. We consider three sets of assumptions for this channel. For each set of assumptions, we present a technique for establishing covert communication and analyze its fundamental limits: (1) if Willie cannot verify the sources of the packets but he is aware of the channel's packet rate λ, Alice can insert packets of her own in the channel covertly and Bob can successfully extract them (packet insertion); (2) if Willie can verify the sources of the packets but Alice and Bob can share a codebook and Bob receives the packets after they pass through an M/M/1 queue, Alice can establish covert communication with Bob by embedding information in packet arrival timings (timing channel); and (3) if packets have available space and Willie knows the probability density function (pdf) of the packet sizes, Alice can insert her own bits in the packets covertly and Bob can extract them successfully (bit insertion). We discuss the extensions of the first and second techniques to packet channels whose packet timings are governed by a renewal process. Finally, we consider the use of timing channels for network fingerprinting which is a traffic analysis method applicable to breaking the anonymity of networks. Assume Alice and Bob have access to the input and the output links of a network, respectively, where the timings of the flows are governed by independent Poisson point processes. The connections between the input and output links are hidden. We show the degree to which Alice can use flow identifiers (fingerprints) and transmit them via a timing channel to Bob to reveal the input-output connections without being detected by Willie who protects the network. We model the network as parallel M/M/1 queues where each queue is shared by a flow from Alice to Bob and other flows independent of the flow from Alice to Bob.
In recent years, the information theory of covert communications, where the very presence of the communication is undetectable to a watchful adversary Willie, has been of great interest. Our recent work introduced the information-theoretic limits for communication by covert users Alice and Bob, over packet channels where the packet timings of legitimate users Jack and Steve are governed by a Poisson point process. Here we consider the extension to timing channels characterized by more general renewal processes of rate λ. We consider two scenarios. In the first scenario, the source of the packets on the channel cannot be authenticated by Willie, and therefore Alice can insert packets into the channel. We show that if the total number of transmitted packets by Jack is N, Alice can covertly insert O(√N)packets and, if she transmits more, she will be detected by Willie. In the second scenario, packets are authenticated by Willie but we assume that Alice and Bob share a secret key; hence, Alice alters the timings of the packets according to a pre-shared codebook with Bob to send information to him over a G/M/1 queue with service rate μ > λ. We show that Alice can covertly and reliably transmit O(N) bits to Bob when the total number of packets sent from Jack to Steve is N.
Covert communication is achieved when a transmitter Al-ice can successfully transmit a message to a receiver Bob without being detected by an attentive and capable adversary Willie. Early results demonstrated the difficulty of the covert communications problem: with AWGN discrete-time channels between all parties, onlyO(n)bits can be sent innchannel uses. But it was soon recognized that uncertainty about the environment at Willie, for example, uncertainty in his own noise statistics, could allow for a positive rate:O(n)bits can be sent covertly innchannel uses. However, most covert communication results, including this promising positive rate result, have been obtained for a discrete-time communications channel. Here, we demonstrate that the assumption of a discrete-time channel is problematic when trying to exploit Willie's noise uncertainty. In particular, we demonstrate that if Alice transmitsω(T)bits in a lengthTinterval to Bob on a continuous-time channel, then there exists a detector at Willie that can detect her transmission, as the probability of false alarm and missed detectionPMD+PFA→0asT→∞. In other words, the communication is not covert, unlike the case of a discrete-time channel.
Covert communications, where a transmitter Alice wishes to hide the presence of her transmitted signal from a watchful adversary Willie, has been considered extensively in recent years. Those investigations have generally considered physical-layer models, where the adversary has access to a sophisticated (often optimal) receiver to determine whether a transmission has taken place, and have addressed the question of what rate can information be communicated covertly. More recent investigations have begun to consider the change in covert rate when Willie has uncertainty about the physical layer environment. Here, we move up the protocol stack to consider the covert rate when Willie is watching the medium-access control (MAC) layer in a network employing a random access MAC such as slotted ALOHA. Based on the rate of collisions and potentially the number of users involved in those collisions, Willie attempts to determine whether unauthorized (covert) users are accessing the channel. In particular, we assume different levels of sophistication in Willie's receiver, ranging from a receiver that only can detect whether there was a collision or not, to one that can always tell exactly how many packets were on the channel in the random access system. In each case, we derive closed-form expressions for the achievable covert rates in the system. The achievable rates exhibit significantly different behavior than that observed in the study of covert systems at the physical layer.
Covert communication considers the ability of transmitter Alice to communicate reliably to receiver Bob without being detected by warden Willie. Previous work has generally considered a discrete-time model, and, in the standard Alice-Bob-Willie scenario, it has been shown that a discrete-time model captures the salient aspects of the underlying continuous-time covert communications system. However, in the presence of a jammer assisting Alice, where it has been shown in previous work that Alice can achieve a positive covert rate on a discrete-time model, we demonstrate here that a straightforward extension of the discrete-time construction to the continuous-time system does not guarantee a positive covert rate, hence indicating that the discrete-time model does not capture the salient aspects of the continuous-time system in such a scenario. This is because Willie is able to exploit excess bandwidth for co-channel interference suppression, as we demonstrate with an interference cancellation receiver. Hence, for the case when Alice is assisted by an uninformed jammer, we consider the continuous-time channel directly and study whether efficient covert communication can still be achieved against any possible receiver that Willie might employ. By presenting and characterizing an approach much different than that suggested by previous work for discrete-time systems in such a scenario, we establish thatO(WT)information bits can be transmitted covertly and reliably on a continuous-time channel of asymptotic bandwidthWinTseconds, regardless of the receiver that Willie employs. This is done for two separate scenarios: 1) when there is perfect frame synchronization between Alice and the jammer’s signals; 2) when there is no frame synchronization between Alice and the jammer’s signals. Numerical results are provided to validate the theory.