We address the problem of in-network distributed estimation for sparse vectors. In order to exploit the underlying sparsity of the vector of interest, we incorporate the ℓ₁- and ℓ₀-norm constraints into the cost function of the standard diffusion least-mean squares (LMS). This technique is equivalent to adding a zero-attracting term in the iteration of the LMS-based algorithm, which accelerates the convergence rates of the zero or near-zero components. The rules for selecting the intensity of the zero-attracting term are derived and verified. Simulation results show that the performances of the proposed schemes depend on the degree of sparsity. Provided that suitable intensities of the zero-attracting term are selected, they can outperform the standard diffusion LMS when the considered vector is sparse. In addition, a practical application of the proposed sparse algorithms in spectrum estimation for a narrow-band source is presented.

Anomaly detection has attracted much attention in recent years since it plays a crucial role in many domains. Various anomaly detection approaches have been proposed, among which one-class support vector machine (OCSVM) is a popular one. In practice, data used for anomaly detection can be distributively collected via wireless sensor networks. Besides, as the data usually arrive at the nodes sequentially, online detection method that can process streaming data is preferred. In this paper, we formulate a distributed online OCSVM for anomaly detection over networks and get a decentralized cost function. To get the decentralized implementation without transmitting the original data, we use a random approximate function to replace the kernel function. Furthermore, to find an appropriate approximate dimension, we add a sparse constraint into the decentralized cost function to get another one. Then we minimize these two cost functions by stochastic gradient descent and derive two distributed algorithms. Some theoretical analysis and experiments are performed to show the effectiveness of the proposed algorithms. Experimental results on both synthetic and real datasets reveal that both of the proposed algorithms achieve low misdetection rates and high true positive rates. Compared with other state-of-the-art anomaly detection methods, the proposed distributed algorithms not only show good anomaly detection performance, but also require relatively short running time and low CPU memory consumption.

Federated learning (FL), which enables multiple distributed devices (clients) to collaboratively train a global model without transmitting their private data, has attracted much attention in the Internet of Things (IoT) domain. Compared with centralized learning, FL has obvious privacy advantages because it can protect the clients’ raw data from direct access by adversaries. Furthermore, to prevent the adversaries from inferring private information from the transmitted parameters, several FL algorithms based on differential privacy (DP) have been proposed, where the clients add artificial noise to their local parameters for privacy protection. Nevertheless, the added noise would disrupt the learning process and degrade the performance of the trained model. Considering this, in this paper, we develop a performance-enhanced differential privacy-based federated learning (PEDPFL) algorithm, where a classifier-perturbation regularization method is proposed to improve the robustness of the trained model against DP-injected noise. We derive the theoretical privacy and convergence analysis of the proposed algorithm, and also demonstrate the influence of some hyperparameters on the convergence performance. Simulation results on real-world datasets show that the proposed algorithm has better classification performance than the existing DP-based FL algorithms at the same level of privacy protection, and thus it is more applicable to IoT applications.

The problem of distributed estimation over wireless sensor networks in an adversarial environment with the presence of attacks on sensed and communicated information is considered. To tackle with this problem, a secure diffusion least-mean squares (S-dLMS) algorithm is proposed. The proposed S-dLMS can be considered as a hybrid system, which consists of a noncooperative LMS (nc-LMS) subsystem and a diffusion LMS (dLMS) subsystem. The nc-LMS subsystem is used to provide a reliable reference estimate, which is further used for constructing the threshold test to detect the trust neighbors of each node. Then, based on the detected secure network topology, the dLMS subsystem is performed by combining the received information from the trust neighbors. The performance of the proposed S-dLMS algorithm in the mean and mean-square senses is analyzed, and then an adaptive rule is suggested to select the threshold for detection. Finally, some simulations are performed to show the effectiveness of the proposed S-dLMS algorithm under fixed and time-varying attacks, respectively.

Federated learning (FL) has attracted widespread attention in the Internet of Things domain recently. With FL, multiple distributed devices can cooperatively train a global model by transmitting model updates without disclosing the original data. However, the distributed nature of FL makes it vulnerable to data poisoning attacks. In practice, malicious clients can launch the label-flipping attack (LFA) by simply tampering with the labels of local data, thus causing the global model to misclassify the samples of a selected class as the target class. Although some defense mechanisms have been proposed, they rely on specific assumptions about data distribution, and their performance degrades significantly when the data on clients are non-IID. Besides, most existing methods require clients to upload model updates in plaintext so that the server can identify and remove the malicious updates. But, direct transmission of model updates may still reveal private information. Considering these issues, we develop a label-flipping-robust and privacy-preserving FL (LFR-PPFL) algorithm, which is applicable to both independent and identically distributed (IID) and non-IID data. We first propose a detection method based on temporal analysis on cosine similarity to distinguish malicious clients from benign clients. Then, we propose a privacy-preserving computation protocol based on homomorphic encryption to implement this detection method and perform federated aggregation while protecting the privacy of clients. Besides, a detailed theoretical analysis is given to demonstrate the privacy guarantee of the proposed protocol. Experimental results on real-world data sets show that the proposed algorithm can effectively defend against LFAs under various data distributions.