The emergence of edge computing guarantees the combination of the Internet of Things (IoT) and artificial intelligence (AI). The vertical federated learning (VFL) framework, usually deployed by split learning, can analyze and integrate information on different features collected by different terminals in the IoT. The complete model is divided into a top model and multiple bottom models in a specific middle layer. Each passive party as a terminal with certain features owns a bottom model, and an active party as an edge server with labels holds the top model. Feature inference attack aims to infer the party’s features from the model predictions during prediction in VFL. Existing attacks considered the adversary an active party under the white-box or black-box model. However, an attacker usually is a passive party in practice because terminals are more vulnerable than edge servers. Therefore, this article discusses a practical feature inference attack in VFL during prediction in IoT under this setting. We design an adversary builds an inference model to minimize the distance between the predictions from the inferred features and target features. Because the information on the top model and other bottom models is unknown, the adversary cannot directly train the inference model. Therefore, we utilize the zeroth-order gradient estimation method to calculate the parameters’ gradients to train the inference model. Experimental results demonstrate that the performance of our attack is comparable to that of the white-box attacks while retaining apparent advantages over the existing black-box attacks.

Abstract Machine learning has become an integral part of modern intelligent systems in all aspects of life. Membership inference attacks (MIAs), as the significant model attacks, also jeopardize the privacy of the intelligent systems. Previous works on defending MIAs concentrate on the model output perturbation or tampering with the training process. However, data and model reuse are common in intelligent systems, which results in the lack of scalability of previous defending works. This paper proposes a new privacy‐preserving framework for images to transform source data into synthetic data to train models against MIAs. The synthetic data makes it easy to defend MIAs during data and model reuse to improve the scheme's scalability. The framework generates synthetic data satisfying differential privacy through the variational autoencoder model's information extraction and data generation capabilities to improve model accuracy. A noise addition mechanism with metric privacy for the latent code generated from source data is proposed, where noise is the product of Γ‐distribution and unit hyper‐sphere samples. Moreover, it is proved that the synthetic data also satisfies metric privacy. The experimental evaluations demonstrate that the framework reduces MIAs' attack accuracy to about 0.5 and maintains higher utility than DP‐SGD under the same setting.

With the explosive growth of data volume in the cloud computing environment, data owners are increasingly inclined to store their data on the cloud. Although data outsourcing reduces computation and storage costs for them, it inevitably brings new security and privacy concerns, as the data owners lose direct control of sensitive data. Meanwhile, most of the existing ranked keyword search schemes mainly focus on enriching search efficiency or functionality, but lack of providing efficient access control and formal security analysis simultaneously. To address these limitations, in this article we propose an efficient and privacy-preservingMulti-keywordRankedSearch scheme withFine-grained access control (MRSF). MRSF can realize highly accurate ciphertext retrieval by combining coordinate matching with Term Frequency-Inverse Document Frequency (TF-IDF) and improving the secure$k$NN method. Besides, it can effectively refine users’ search privileges by utilizing the polynomial-based access strategy. Formal security analysis shows that MRSF is secure in terms of confidentiality of outsourced data and the privacy of index and tokens. Extensive experiments further show that, compared with existing schemes, MRSF achieves higher search accuracy and more functionalities efficiently.

With the development of machine learning, it is popular that mobile users can submit individual symptoms at any time anywhere for medical diagnosis. Edge computing is frequently adopted to reduce transmission latency for real-time diagnosis service. However, the data-driven machine learning, which requires to build a diagnosis model over vast amounts of medical data, inevitably leaks the privacy of medical data. It is necessary to provide privacy preservation. To solve above challenging issues, in this article, we design a lightweight privacy-preserving medical diagnosis mechanism on edge, called LPME. Our LPME redesigns the extreme gradient boosting (XGBoost) model based on the edge-cloud model, which adopts encrypted model parameters instead of local data to remove amounts of ciphertext computation to plaintext computation, thus realizing lightweight privacy preservation on resource-limited edge. In addition, LPME provides secure diagnosis on edge with privacy preservation for private and timely diagnosis. Our security analysis and experimental evaluation indicates the security, effectiveness, and efficiency of LPME.

With the development of Mobile Healthcare Monitoring Network (MHMN), patients’ data collected by body sensors not only allows patients to monitor their health or make online pre-diagnosis but also enables clinicians to make proper decisions by utilizing data mining technique. However, sensitive data privacy is still a major concern. In this article, we propose practical techniques for searching and making online pre-diagnosis over encrypted data. First, we propose a new Diverse Keyword Searchable Encryption (DKSE) scheme which supports multi-dimension digital vectors range query and textual multi-keyword ranked search to gain a broad range of applications in practice. In addition, a framework called PRIDO based on the DKSE is designed to protect patients’ personal data in data mining and online pre-diagnosis. According to the PRIDO framework, we achieve privacy-preserving naïve Bayesian and decision tree classifiers and discuss its potential applications in actual deployments. Security analysis proves that patients’ data privacy can be well protected without loss of data confidentiality, and performance evaluation demonstrates the efficiency and accuracy in the diverse keyword search, data mining, and disease pre-diagnosis, respectively.

In the era of machine learning, mobile users are able to submit their symptoms to doctors at any time, anywhere for personal diagnosis. It is prevalent to exploit edge computing for real-time diagnosis services in order to reduce transmission latency. Although data-driven machine learning is powerful, it inevitably compromises privacy by relying on vast amounts of medical data to build a diagnostic model. Therefore, it is necessary to protect data privacy without accessing local data. However, the blossom has also been accompanied by various problems, i.e., the limitation of training data, vulnerabilities, and privacy concern. As a solution to these above challenges, in this paper, we design a lightweight privacy-preserving medical diagnosis mechanism on edge. Our method redesigns the extreme gradient boosting (XGBoost) model based on the edge-cloud model, which adopts encrypted model parameters instead of local data to reduce amounts of ciphertext computation to plaintext computation, thus realizing lightweight privacy preservation on resource-limited edges. Additionally, the proposed scheme is able to provide a secure diagnosis on edge while maintaining privacy to ensure an accurate and timely diagnosis. The proposed system with secure computation could securely construct the XGBoost model with lightweight overhead, and efficiently provide a medical diagnosis without privacy leakage. Our security analysis and experimental evaluation indicate the security, effectiveness, and efficiency of the proposed system.

With the development of the Mobile Healthcare Monitoring Network (MHMN), patients’ data collected by body sensors not only allows patients to monitor their health or make online pre-diagnosis but also enables clinicians to make proper decisions by utilizing data mining techniques. In MHMN, patients’ personal data are collected by sensors per second and uploaded to the cloud server as multi-dimension vectors, cloud server stores the personal data as well as sends monitoring information to the hospital when the real-time data is abnormal. Hospital users (i.e., doctors, etc.) may query some samples which contain certain textual keywords or digital keywords in certain ranges for disease diagnosis or medical research. For example, a certain hospital user may query all samples with textual keywords ‘cancer; diabetes’ and digital vectors ‘age’ ∈ [30,50], ‘blood sugar’ ∈ [4,8], ‘heart rhythm’ ∈ [70,80]. Besides, the potential value of massive medical data has attracted considerable interests recently, for example, valuable results in diagnosis model can be yield with large-scale aggregation analysis of personal medical data. The cloud server can build a diagnosis model using data mining technology over massive data, so that hospital users or pre-diagnosis users upload medical data (i.e., age, blood pressure, blood sugar, etc.) to the cloud for diagnosis.

With the development of Mobile Healthcare Monitoring Network (MHMN), patients' personal data collected by body sensors not only allows patients to monitor their health or make online pre-diagnosis but also enable clinicians to make proper decisions by utilizing data mining technique. However, the sensitive data privacy is still a major concern. In this paper, we first propose an Efficient Privacy-preserving Sensor data Monitoring and online Diagnosis (EPSMD) system for outsourced computing, then furnish an improved Multidimensional Range Query Technique (MRQT) to gain a broad range of applications in practice. In addition, a privacy-preserving naive Bayesian classifier based on MRQT is designed to protect patients' data in data mining and online diagnosis efficiently. Security analysis proves that patients' data privacy can be well protected without loss of data confidentiality, and performance evaluation demonstrates the efficiency and accuracy in data monitoring and disease pre-diagnosis, respectively.

When moving a speech recognition system whose models were trained in a clean laboratory condition to real environments, one of most important issues is how to modify the models according to the changing environments. Using an HMM composition technique we present an algorithm to compensate the dynamic cepstral coefficients for HMM based speech recognition systems in noise environments. Noise compensation for acceleration parameters and for dynamic parameters which are calculated using longer linear regression are discussed. The experimental results show a clear improvement when the algorithm was applied to a speech database recorded in a car. A noise compensation system based realtime speech recognizer using the TMS320C40 was implemented and achieves a good performance in noisy environments.

We propose a novel class of learning vector quantizers (LVQs) based on multivariate data ordering principles. A special case of the novel LVQ class is the median LVQ, which uses either the marginal median or the vector median as a multivariate estimator of location. The performance of the proposed marginal median LVQ in color image quantization is demonstrated by experiments.

A noise compensation algorithm for HMM based speech recognition systems, which utilizes the parallel model combination concept, is presented. The algorithm was tested using the TIDIGITS database with artificially added car noise. Very promising results were obtained. The results show that at -10 dB SNR the recognition accuracy could be improved from 34% to 89%. The noise compensation algorithm was also tested using a database which was recorded in a car. Improved performance was obtained, but the improvement; was clearly smaller than with the artificially added noise.

Proposes two simple and efficient architectures for the realization of stack filters in hardware: one for pipelined implementation and the other for non-pipelined implementation. Both architectures are interchangeable, with very minor modifications. A programmable stack filter has been layed out based on the proposed pipelined architecture in the AMS 1 micron technology using Mentor Graphics Generator Design Tools for five words of twelve bits each and simulations have been carried out successfully using the Mentor Graphics Lsim simulator. Results are encouraging that the simulations are error-free at a frequency as high as 166 MHz.>

A novel class of Learning Vector Quantizers (LVQs) based on multivariate order statistics is proposed in order to overcome the drawback that the estimators for obtaining the reference vectors in LVQ do not have robustness either against erroneous choices for the winner vector or against the outliers that may exist in vector-valued observations. The performance of the proposed variants of LVQ is demonstrated by experiments. In the case of marginal median LVQ, its asymptotic properties are derived as well.>

Earlier research has shown that the problem of optimal weighted median filtering with structural constraints can be formulated as a nonconvex nonlinear programming problem in general. However, its high computational complexity and poor performance due to its nonconvex nature prohibit it from practical applications. In this paper, we shall show that the design problem can be formulated as a convex quadratic programming problem. The new algorithm is very efficient in the sense of computational complexity. The algorithm is also efficient in the sense of its capability to approach the global minimum. Using the algorithm optimal 1-D weighted median filters preserving pulses of length 3, 4 and 5 are tabulated.>

In this paper, we study the convergence properties of weighted order statistic filters. Weighted order statistics filters are divided into five categories making their convergence properties easily understand. It is shown that any symmetric weighted order statistics filters will make the input sequence converge to a root or oscillate in a cycle of period 2. This result is significant since a restriction posed by earlier research is eliminated making the result applicable for the whole class of symmetric weighted order statistics filters. A condition to guarantee convergence of symmetric weighted order statistics filters is derived.>

The authors study the convergence properties of weighted order statistics filters. Based on a set of parameters, weighted order statistics filters are divided into five categories making their convergence properties easily understood. They show that any symmetric weighted order statistics filters will make the input sequence converge to a root or oscillate in a cycle of period 2. This result is significant since a restriction imposed by an earlier research is eliminated making the result applicable for the whole class of symmetric weighted order statistics filters. A condition to guarantee convergence of symmetric weighted order statistics filters is also derived.>