Traffic status monitoring is essential to support intelligent transportation, and demands for highly efficient and secure processing and transmission of massive data. Inspired by recent information-theoretic study, the traffic status monitoring is formulated as a privacy-preserving semantic coding problem in this work to characterize the fundamental trade-off among data compression, traffic image reconstruction, segmentation of traffic participants, and preservation of pixel regions of traffic participants. Grounded on the theoretic results, a novel generative semantic coding scheme is further developed for data-driven semantic coding design with the aid of attention mechanisms to capture correlations between sparse information. Experiments on the RCooper dataset demonstrate the effectiveness of the proposed generative semantic coding scheme and its advantages over the benchmark compression methods.

Vehicle-to-everything (V2X) communication plays an essential role in the Internet-of-Vehicles (IoV) systems. However, in practice the diverse decision objectives of different users and the highly dynamic nature of IoV cause great challenges in realizing efficient data transmission. This paper investigates a sequential power control problem in a typical V2X communication network formed by multiple data delivery links, and proposes a novel decision framework from the perspective of dynamic multi-objective optimization (DMO). The framework is developed based on the dynamic multi-objective optimization evolutionary algorithm (DMOEA). When the variation of channel fading state is detected, an innovative environment change response mechanism is implemented to predict the centroid change of the Pareto optimal set (PS) in the new environment. This allows to properly generate the initial population for the subsequent evolutionary search process, and rapidly track the time-varying Pareto optimal front (PF). The effectiveness and advantages of the proposed framework over conventional solutions are validated by simulation results.

Resource allocation in vehicle-to-everything (V2X) communication is an important yet challenging problem. Most existing works target optimizing a specific performance metric in a relatively stable condition, which may not be able to adaptively satisfy the dynamic quality of service (QoS) requirements prevalent in practical vehicular networks. In this paper, we investigate a typical V2X resource allocation problem requiring optimization of both vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communication performance. Instead of setting one single system goal, we formulate it as a multi-objective optimization problem and aim to seek the complete trade-off between the objectives, i.e., the Pareto frontier (PF). To this end, the problem is first transformed to a multi-objective Markov decision process (MOMDP) and solved by a multi-objective reinforcement learning resource allocation (MORLRA) algorithm. Simulation results demonstrate the effectiveness and advantages of the proposed method.

As an emerging data transmission paradigm, semantic communication can significantly improve transmission efficiency by focusing on extracting and preserving task-critical semantic information. However, privacy preservation of sensitive information in semantic communication has not been well studied. This letter focuses on the semantic coding and studies the fundamental trade-off among data compression, source distortion, semantic distortion, and privacy preservation from an information-theoretic perspective. Based on the theoretic results, a novel deep learning method is proposed for data-driven privacy-preserving semantic coding design. The theory and method are validated through experiments on the CelebA dataset.

This paper presents a two-stage Multiple-Model Compression (MMC) approach for sampled electrical waveforms. To limit latency, the processing is window-based, with a window length commensurate to the electrical period. For each window, the first stage compares several parametric models to get a coarse representation of the samples. The second stage then compares different residual compression techniques to minimize the norm of the reconstruction error. The allocation of the rate budget among the two stages is optimized. The proposed MMC approach provides better signal-to-noise ratios than state-of-the-art solutions on periodic and transient waveforms.

In this paper, we design privacy-preserving and cost-efficient energy management strategies for smart grid users that are equipped with renewable energy sources. The adversary is assumed to employ a factorial hidden Markov model based inference for load disaggregation, and the corresponding joint log-likelihood of the model is utilized as the privacy measure. The studied dynamic pricing model is applicable to a commodity-limited market, where the price of unit amount of energy is determined by the users’ aggregated power request. The users’ energy management strategies are designed under a non-cooperative game framework, where each user aims to optimize a weighted sum objective of both privacy measure and energy cost saving. The users’ non-cooperative game is shown to admit a unique pure strategy Nash equilibrium. As an extension, a computational-efficient distributed Nash equilibrium energy management strategy seeking method is proposed, which also avoids the privacy leakage due to the sharing of payoff functions between users. The performance of practical designs of the energy management strategies in the equilibrium is finally illustrated by numerical experiments.

Reinforcement learning (RL) has been widely employed in communications, in the areas of interference management, resource allocation, signal detection, and power control, etc. Nevertheless, RL is vulnerable under various malicious attacks, such as adversarial examples and privacy intrusions. In this paper, a falsification attack on the agent actions in a scalar linear quadratic regulator (LQR) system is studied. This adversarial problem is formulated as a novel dynamic game by introducing an adversarial belief, and subgame perfect equilibria (SPEs) are characterized under different adversarial constraints. Numerical experiments show the impact of strategic interactions and justify the theoretic results.

We design optimal privacy-enhancing and cost-efficient energy management strategies for consumers that are equipped with a rechargeable energy storage. The Kullback-Leibler divergence rate is used as privacy measure and the expected cost-saving rate is used as utility measure. The corresponding energy management strategy is designed by optimizing a weighted sum of both privacy and cost measures over a finite time horizon, which is achieved by formulating our problem into a belief-state Markov decision process problem. A computationally efficient approximated Q-learning method is proposed as a generalization to high-dimensional problems over an infinite time horizon. At last, we explicitly characterize a stationary policy that achieves the steady belief state over an infinite time horizon, which greatly simplifies the design of the privacy-preserving energy management strategy. The performance of the practical design approaches are finally illustrated in numerical experiments.

Privacy against an adversary (AD) that tries to detect the underlying privacy-sensitive data distribution is studied. The original data sequence is assumed to come from one of the two known distributions, and the privacy leakage is measured by the probability of error of the binary hypothesis test carried out by the AD. A management unit (MU) is allowed to manipulate the original data sequence in an online fashion while satisfying an average distortion constraint. The goal of the MU is to maximize the minimal type II probability of error subject to a constraint on the type I probability of error assuming an adversarial Neyman-Pearson test, or to maximize the minimal error probability assuming an adversarial Bayesian test. The asymptotic exponents of the maximum minimal type II probability of error and the maximum minimal error probability are shown to be characterized by a Kullback-Leibler divergence rate and a Chernoff information rate, respectively. Privacy performances of particular management policies, the memoryless hypothesis-aware policy and the hypothesis-unaware policy with memory, are compared. The proposed formulation can also model adversarial example generation with minimal data manipulation to fool classifiers. At last, the results are applied to a smart meter privacy problem, where the user's energy consumption is manipulated by adaptively using a renewable energy source in order to hide user's activity from the energy provider.

The trade-off of hypothesis tests on the correlated privacy hypothesis and utility hypothesis is studied. The error exponent of the Bayesian composite hypothesis test on the privacy or utility hypothesis can be characterized by the corresponding minimal Chernoff information rate. An optimal management protects the privacy by minimizing the error exponent of the privacy hypothesis test and meanwhile guarantees the utility hypothesis testing performance by satisfying a lower bound on the corresponding minimal Chernoff information rate. The asymptotic minimum error exponent of the privacy hypothesis test is shown to be characterized by the infimum of corresponding minimal Chernoff information rates subject to the utility guarantees.

Privacy-preserving energy management is studied in the presence of a renewable energy source. It is assumed that the energy demand/supply from the energy provider is tracked by a smart meter. The resulting privacy leakage is measured through the probabilities of error in a binary hypothesis test, which tries to detect the consumer behavior based on the meter readings. An optimal privacy-preserving energy management policy maximizes the minimal Type II probability of error subject to a constraint on the Type I probability of error. When the privacy-preserving energy management policy is based on all the available information of energy demands, energy supplies, and hypothesis, the asymptotic exponential decay rate of the maximum minimal Type II probability of error is characterized by a divergence rate expression. Two special privacy-preserving energy management policies, the memoryless hypothesis-aware policy and the hypothesis-unaware policy with memory, are then considered and their performances are compared. Further, it is shown that the energy supply alphabet can be constrained to the energy demand alphabet without loss of optimality for the evaluation of a single-letter-divergence privacy-preserving guarantee.

In this paper, an energy flow control strategy to reduce the smart meter privacy leakage is studied. The considered smart grid is equipped with an energy storage device. The privacy leakage is modeled as optimal Bayesian detections on the behaviors of the consumer made by an authorized adversary. To evaluate the privacy risk, a Bayesian detection-operational privacy leakage metric is proposed. The design of an optimal privacy-preserving energy control strategy can be formulated as a belief state MDP problem. Therefore, standard methods and algorithms can be utilized to obtain or to approximate the optimal control strategy. A simplified problem to design an instantaneous optimal privacy-preserving control strategy is also considered. It is shown that the problem of the instantaneous optimal control strategy design can be formulated as a set of linear programmings.

In this paper, the privacy problem of a tandem distributed detection system vulnerable to an eavesdropper is proposed and studied in the Bayesian formulation. The privacy risk is evaluated by the detection cost of the eavesdropper which is assumed to be informed and greedy. For the sensors whose operations are constrained to suppress the privacy risk, it is shown that the optimal detection strategies are likelihood-ratio tests. This fundamental insight allows for the optimization to reuse known algorithms extended to incorporate the privacy constraint. The trade-off between the detection performance and privacy risk is illustrated in an example.

In this paper, the privacy problem of a parallel distributed detection system vulnerable to an eavesdropper is proposed and studied in the Bayesian formulation. The privacy risk is evaluated by the detection cost of the eavesdropper which is assumed to be informed and greedy. It is shown that the optimal detection strategy of the sensor whose decision is eavesdropped on is a likelihood-ratio test. This fundamental insight allows for the optimization to reuse known algorithms extended to incorporate the privacy constraint. The trade-off between the detection performance and privacy risk is illustrated in a numerical example. The incorporation of physical layer privacy in the system design will lead to trustworthy sensor networks in future.