20 September 2023 in IEEE Transactions on Vehicular Technology
This paper considers a multi-jammer-assisted covert communication scenario, where a transmitter communicates with a receiver without being detected by a warden under the assistance of multiple friendly jammers. The transmission rate maximization problem subjecting to the covertness constraint by jointly optimizing the jammer selection strategy and power control of the transmitter is studied under the following three channel models: 1) Additive white Gaussian noise (AWGN) channel; 2) Block fading channel without channel state information (CSI) acquisition cost; 3) Block fading channel with CSI acquisition cost. Under the AWGN channel, considering achievable schemes of the transmitter, relative entropy is adopted to lower bound the detection error probability at the warden. We decouple the joint optimization problem into two sub-problems and prove that selecting one jammer with the maximum ratio of the jamming power at the warden to that at the receiver is optimal with proper transmit power control in an interference-limited environment. Motivated by this key analytical conclusion, we then study the second channel model's optimal single jammer selection strategy. Considering the worst case that the warden employs the optimal detector, we analytically show that the optimal selection criterion in this channel model is choosing the jammer with the maximum ratio of the average jamming power at the warden to the instantaneous jamming power at the receiver. Given this selection criterion, we then study the problem of how many jammers should be explored when the acquisition cost of jammers' CSI cannot be ignored. We formulate the optimization problem as an optimal stopping problem and propose two optimal strategies for two cases that whether the previously explored jammers are allowed to select. The numerical results evaluate the theoretical results and show the superiority of the proposed jammer selection strategies. Furthermore, we also present the impact of some key parameters on the covert rate, such as the covertness constraint, the number of jammers, and the CSI acquisition time of one jammer, which provide more valuable insights.
19 September 2022 in IEEE Communications Letters
This letter studies a wireless covert communication system, where a jammer generating intermittent artificial noise (AN) is employed to confuse the warden. The transmission probabilities of information and AN are jointly optimized to maximize the communication covertness under a throughput requirement. We first analyse the detection performance of the warden and the throughput of the system, based on which we formulate the optimization problem. Then, we simplify this two-dimensional nonconvex problem to a one-dimensional problem and obtain the closed-form optimal solutions in special cases. The results illustrate that the optimal transmission probabilities of information and AN are 0.5 and 1, respectively, when the throughput requirement is less than a threshold. However, 0.5 and 1 are not optimal when the throughput requirement exceeds the aforementioned threshold, which has not been reported yet.
13 November 2024 in IEEE Transactions on Wireless Communications
The environmental shelters such as public links can enable covert communication by covering covert transmission. To further exploit shelters, this paper focuses on a two-hop system where multiple pairs of public users and one pair of covert users communicate through a multi-antenna relay. We aim to improve covertness performance while satisfying the covertness constraints of two hops and quality of service (QoS) requirements of public users. The covert throughput maximization problem is formulated via jointly optimizing transmit power and beamformer, which is challenging to solve. We introduce successive convex approximation (SCA) and semidefinite relaxation (SDR) techniques to convert the problem into convex, where the joint optimization algorithm is developed. Considering the computational complexity, we further design a block diagonalization (BD) beamformer at the relay, which translates the beamformer optimization into a power allocation problem and derives the optimal solution in a closed form. We analytically show that the covert throughput first increases and then decreases as the number of public pairs increases in BD-based design, which has been verified numerically and can be generalized in other designs. Numerical results also evaluate the superiority of the joint optimization algorithm and the effectiveness of the BD-based efficient design. In particular, the BD-based design is very close to the joint optimization under small maximum transmit power of users or large maximum transmit power of relay.
22 January 2024 in IEEE Transactions on Communications
In this paper, we consider a novel covert communication scenario where the warden does not have prior knowledge of the transmitter’s power. In this scenario, the current widely adopted likelihood ratio test (LRT)-based detector is not optimal anymore. To address this, we formulate a detection framework based on the generalized likelihood ratio test (GLRT), which replaces the unknown parameter with maximum likelihood estimation (MLE) and is proven optimal in the scenario. Based on the GLRT-based framework, two different detection models are proposed, where one only utilizes observations of the current time and the other can exploit observations of past time slots. We analyze the estimation and fisher information of the transmit power, and even the detection and covert performance under two different detection models, respectively. Based on the analytical results, we further derive the maximum number of tolerable slots under which the transmitter can remain the same transmit power while the detection error probability of the warden is still larger than the regulated threshold. Moreover, the transmit power and the number of tolerable slots are jointly optimized to maximize the transmission throughput subjecting to covertness constraint. The numerical results demonstrate the correctness of the theoretical analysis. We show how the covert rate is influenced by the number of transmitting slots and the transmit power, which can guide the design of the covert transmission strategy.
23 March 2023 in IEEE Wireless Communications Letters
In this letter, we consider a multi-server mobile edge computing (MEC) network where a user with limited computation ability aims to complete an intensive computation task in a short time under the help of multiple edge servers. We study the delay-minimization offloading schemes with and without user’s energy constraint, which involve the joint optimization of task allocation ratio for local computation, the wireless offloading manner (i.e., non-orthogonal multiple access (NOMA) offloading or single-server offloading mode), and the server pair selection. Specifically, the joint optimal solutions are derived in semi-closed forms and the critical conditions of switching offloading manners are derived in closed form. Numerical results corroborate the superiority of our proposed offloading scheme in terms of overall latency under various settings.
1 January 2022 in Wireless Communications and Mobile Computing
This paper studies the user selection problem for a cooperative nonorthogonal multiple access (NOMA) system consisting of a base station, a far user, and N near users. The selected near user receives its own message and assists the far user by relaying the far user’s message. Firstly, we propose a user selection strategy to maximize the selected near user’s data rate while satisfying the quality‐of‐service (QoS) requirement of the far user. Considering that the channel state information (CSI) of users in actual communication is usually imperfect, we then analyze the outage probability of the NOMA system based on the user selection strategy under imperfect CSI and obtain a closed‐form expression. The theoretical analysis shows that the diversity order of the NOMA system under imperfect CSI is 0, which means the multiuser diversity order disappears. In order to improve the impact of imperfect CSI on system performance, we use the deep learning method to identify and classify channels of imperfect CSI and improve the accuracy of CSI. The simulation results show that the theoretical analysis of outage performance is consistent with the numerical results. Compared with the strategy without the deep learning method, the proposed deep learning‐based user selection scheme significantly improves the system performance. Furthermore, we verify that our scheme recovers the diversity gain.