In this paper, secure communication over a broadcast channel with multiple legitimate receivers and an external eavesdropper is investigated. Two different secrecy measures are considered. The first criterion is a conservative one known as joint secrecy, where the mutual leakage of all confidential messages must be small. The second criterion is a less conservative constraint known as individual secrecy, where the individual leakage of each confidential message must be small. At first, we consider the degraded multi-receiver wiretap broadcast channel and manage to establish the individual secrecy capacity region. Our encoding scheme applies a careful combination of the standard techniques of wiretap random coding and Shannon's one time pad encoding, where the confidential messages of the weak receivers are used as secret keys for the stronger ones. The validity of this technique is due to the properties of the degraded broadcast channel and the secrecy requirements of the individual secrecy criterion. Our result indicates that the individual secrecy capacity region is in fact larger than the joint one established in earlier literature. The established capacity region is then used to derive the individual secrecy capacity regions of the Gaussian single-input single-output and degraded Gaussian multiple-input multiple-output multi-receiver wiretap broadcast channels. Furthermore, we present an achievable rate region for the general two-receiver wiretap broadcast channel under both the joint and the individual secrecy criterion. Comparing these two rate regions suggests that even for the general case, the individual secrecy criterion might be able to provide a larger rate region compared with the joint one.
In this paper, secure communication over a broadcast channel with multiple legitimate receivers and an external eavesdropper is investigated. Two different secrecy measures are considered. The first criterion is a conservative one known as joint secrecy, where the mutual leakage of all confidential messages must be small. The second criterion is a less conservative constraint known as individual secrecy, where the individual leakage of each confidential message must be small. At first, we consider the degraded multi-receiver wiretap broadcast channel and manage to establish the individual secrecy capacity region. Our encoding scheme applies a careful combination of the standard techniques of wiretap random coding and Shannon's one time pad encoding, where the confidential messages of the weak receivers are used as secret keys for the stronger ones. The validity of this technique is due to the properties of the degraded broadcast channel and the secrecy requirements of the individual secrecy criterion. Our result indicates that the individual secrecy capacity region is in fact larger than the joint one established in earlier literature. The established capacity region is then used to derive the individual secrecy capacity regions of the Gaussian single-input single-output and degraded Gaussian multiple-input multiple-output multi-receiver wiretap broadcast channels. Furthermore, we present an achievable rate region for the general two-receiver wiretap broadcast channel under both the joint and the individual secrecy criterion. Comparing these two rate regions suggests that even for the general case, the individual secrecy criterion might be able to provide a larger rate region compared with the joint one.
We study secure communication over a degraded wiretap broadcast channel with multiple receivers and an eavesdropper. We consider two different secrecy measures: the traditional joint secrecy where the mutual information leakage of all messages must be small; and individual secrecy where the sum of the information leakage of each individual message must be small. At first, we investigate the joint secrecy criterion, where we present the capacity region established before and provide a simpler converse proof. We then consider the individual secrecy criterion and establish its capacity region, by combining the techniques of wiretap random coding and Shannon's one time pad principle. Our results indicate that the individual secrecy capacity region is bigger than the joint one. Further, we show that the established capacity regions are valid for any degraded wiretap broadcast channel, regardless of the degradedness order of the eavesdropper. Finally, we extend our results to Gaussian channels.
We study secure communication in which a transmitter wants to send confidential messages to an arbitrary number of receivers in the presence of an external eavesdropper. We consider two classes of Gaussian channels: the Gaussian single-input single-output (SISO) multi-receiver wiretap channel and the degraded Gaussian multiple-input multiple-output (MIMO) multi-receiver wiretap channel. These two channels belong to the class of degraded multi-receiver wiretap channels. In previous literature the secrecy capacity regions of these two channels were established under the conservative joint secrecy constraint. Thus, we focus our work on the less conservative individual secrecy constraint, which is characterized by a higher throughput. We establish the individual secrecy capacity region for these two cases. The achievability follows from the individual secrecy capacity established for the discrete memoryless degraded wiretap channels where coding is performed using Gaussian signals. On the other hand, the converse is established by adapting the techniques used for formulating the joint secrecy capacity region to the individual secrecy constraint.
This paper studies the problem of secure communication over broadcast channels under the lens of individual secrecy constraints (i.e., information leakage from each message to an eavesdropper is made vanishing). It is known that, for the communication over the degraded broadcast channels, the stronger receiver is able to decode the message of the weaker receiver. In the individual secrecy setting, the message for the weaker receiver can be further utilized to secure the partial message that is intended to the stronger receiver. With such a coding spirit, it is shown that more secret bits can be conveyed to the stronger receiver. In particular, for the corresponding Gaussian model, a constant gap (i.e., 0.5 bits within the individual secrecy capacity region) result is obtained. Overall, when compared with the joint secrecy constraint, the results allow for trading-off secrecy level and throughput in the system.
We study secure communication in which two confidential messages are transmitted over a broadcast channel to two legitimate receivers, while keeping an eavesdropper ignorant. Each legitimate receiver is interested in decoding one confidential message, while having the other one as side information. In order to measure the secrecy of the communication, we investigate two different secrecy criteria: joint secrecy and individual secrecy. For both criteria, we provide an achievable rate region and a matching multi-letter outer bound presenting a multi-letter description for the capacity region. We further investigate the class of more capable channels and provide a single-letter converse establishing the secrecy capacity region, not only for more capable channels but less noisy and degraded channels as well. Our results indicate that the secrecy capacity for individual secrecy is higher than the one for joint secrecy, as one message can be used as a secret key for the other one.
We study the multi-receiver wiretap channel with public and confidential messages. In this channel, there is a transmitter that wishes to communicate with two legitimate users in the presence of an external eavesdropper. The transmitter sends a pair of public and confidential messages to each legitimate user. While there are no secrecy constraints on the public messages, confidential messages need to be transmitted in perfect secrecy. We study the discrete memoryless multi-receiver wiretap channel as well as its Gaussian multi-input multi-output (MIMO) instance. First, we consider the degraded discrete memoryless channel, and obtain an inner bound for the capacity region by using an achievable scheme that uses superposition coding and binning. Next, we obtain an outer bound, and show that this outer bound partially matches the inner bound, providing a partial characterization for the capacity region of the degraded channel model. Second, we obtain an inner bound for the general, not necessarily degraded, discrete memoryless channel by using Marton's inner bound, superposition coding, rate-splitting and binning. Third, we consider the degraded Gaussian MIMO channel, and show that, to evaluate both the inner and outer bounds, considering only jointly Gaussian auxiliary random variables and channel input is sufficient. Since the inner and outer bounds partially match, these sufficiency results provide a partial characterization of the capacity region of the degraded Gaussian MIMO channel. Finally, we provide an inner bound for the capacity region of the general, not necessarily degraded, Gaussian MIMO multi-receiver wiretap channel.
In this paper, we consider the Gaussian multiple-input multiple-output (MIMO) multi-receiver wiretap channel in which a transmitter wants to have confidential communication with an arbitrary number of users in the presence of an external eavesdropper. We derive the secrecy capacity region of this channel for the most general case. We first show that even for the single-input single-output (SISO) case, existing converse techniques for the Gaussian scalar broadcast channel cannot be extended to this secrecy context, to emphasize the need for a new proof technique. Our new proof technique makes use of the relationships between the minimum-mean-square-error and the mutual information, and equivalently, the relationships between the Fisher information and the differential entropy. Using the intuition gained from the converse proof of the SISO channel, we first prove the secrecy capacity region of the degraded MIMO channel, in which all receivers have the same number of antennas, and the noise covariance matrices can be arranged according to a positive semi-definite order. We then generalize this result to the aligned case, in which all receivers have the same number of antennas; however, there is no order among the noise covariance matrices. We accomplish this task by using the channel enhancement technique. Finally, we find the secrecy capacity region of the general MIMO channel by using some limiting arguments on the secrecy capacity region of the aligned MIMO channel. We show that the capacity achieving coding scheme is a variant of dirty-paper coding with Gaussian signals.