Federated learning has recently emerged as a paradigm promising the benefits of harnessing rich data from diverse sources to train high quality models, with salient features that training datasets never leave local devices. Only model updates are locally computed and shared for aggregation produce global model. While federated greatly alleviates privacy concerns opposed centralized data, sharing still poses risks. In this paper, we present system design which offers efficient protection...

In this paper we present a solution to one of the location-based query problems. This problem is defined as follows: (i) user wants database location data, known Points Of Interest (POIs), and does not want reveal his/her server due privacy concerns; (ii) owner that is, server, simply distribute its data all users. The desires have some control over since asset. We propose major enhancement upon previous solutions by introducing two stage approach, where first step based on Oblivious...

The vast development of the Internet Things (IoT) and cloud-enabled data processing solutions provide opportunity to build novel fascinating smart, connected healthcare systems. Smart systems analyze IoT-generated patient both enhance quality care reduce costs. A major challenge for these is how Cloud can handle generated from billions IoT devices. Edge computing infrastructure offers a promising solution by operating as middle layer between devices cloud computing. (EoT) offer small-scale...

In mobile communication, spatial queries pose a serious threat to user location privacy because the of query may reveal sensitive information about user. this paper, we study approximate k nearest neighbor (kNN) where location-based service (LBS) provider points interest (POIs) on basis his current location. We propose basic solution and generic for preserve in kNN queries. The proposed solutions are mainly built Paillier public-key cryptosystem can provide both privacy. To privacy, our...

In recent years, wireless sensor networks have been widely used in healthcare applications, such as hospital and home patient monitoring. Wireless medical are more vulnerable to eavesdropping, modification, impersonation replaying attacks than the wired networks. A lot of work has done secure The existing solutions can protect data during transmission, but cannot stop inside attack where administrator database reveals sensitive data. this paper, we propose a practical approach prevent by...

The main goal of a personalized recommender system is to provide useful recommendations on various items the users. In order generate recommendations, service needs access types user data such as previous product purchasing history, demographic and biographical information. However, users are sensitive disclosure personal information it can be easily misused by malicious third parties. Consequently, there unavoidable security concerns which will become known through attempted unauthorized...

This paper proposes a blockchain-based Federated Learning (FL) framework with Intel Software Guard Extension (SGX)-based Trusted Execution Environment (TEE) to securely aggregate local models in Industrial Internet-of-Things (IIoTs). In FL, can be tampered by attackers. Hence, global model generated from the erroneous. Therefore, proposed leverages blockchain network for secure aggregation. Each node hosts an SGX-enabled processor that performs FL-based aggregation tasks generate model....

Over the past decade, Internet of Things (IoT) is widely adopted in various domains, including education, commerce, government, and healthcare. There are also many IoT-based applications drawn significant attentions recent years. With increasing numbers connected devices IoT system, one challenging tasks to ensure devices' authenticity, which allows users have a high confidence decision. In addition, due heterogeneity system resource-constrained devices, how efficiently manage such guarantee...

Internet of Vehicles (IoV), a special form Things (IoT), is an important enabler intelligent transportation system, which one the most strategic applications in smart city initiatives. In order to achieve its intended functionalities, IoV requires anytime anywhere connectivity, cannot be satisfied by traditional networking technologies. Space–air–ground-integrated network (SAGIN) widely believed ideal infrastructure for connecting IoV. this article, we present approach understanding security...

Due to the rising awareness of privacy and security in machine learning applications, federated (FL) has received widespread attention applied several areas, e.g., intelligence healthcare systems, IoT-based industries, smart cities. FL enables clients train a global model collaboratively without accessing their local training data. However, current schemes are vulnerable adversarial attacks. Its architecture makes detecting defending against malicious updates difficult. In addition, most...

Background: Circular RNAs (circRNAs) represent a subset of non-coding implicated in the regulation diverse biological processes, including tumorigenesis. However, expression and functional implications circ0060467 hepatocellular carcinoma (HCC) remain elusive. In this study, we aimed to elucidate role modulating progression HCC. Methods: Differentially expressed circRNAs HCC tissues were identified through circRNA microarray assays. Quantitative reverse transcription polymerase chain...

We come up with an ID-based signature scheme from the Weil pairing. Our is secure if Diffie-Hellman problem hard. Furthermore, our size only about a quarter of Guillou-Quisquater (1988) size.

In cryptographic applications, hash functions are used within digital signature schemes to provide data integrity (e.g., detect modification of an original message). this brief, we propose a new 2/spl lscr/ -bit iterated function based on chaotic tent maps. This can be expected have at least the same computational security against target attack, free-start collision semi-free-start and attack as DM scheme.

Private Information Retrieval (PIR) allows a user to retrieve the ith bit of an n-bit database without revealing server value i. In this paper, we present PIR protocol with communication complexity O(γ logn) bits, where -y is ciphertext size. Furthermore, extend private block retrieval (PBR) protocol, natural and more practical extension in which retrieves instead retrieving single bit. Our protocols are built on state-of-the-art fully homomorphic encryption (FHE) techniques provide privacy...

It is attractive for an organization to outsource its data analytics a service provider who has powerful platforms and advanced skills. However, the (data owner) may have concerns about privacy of data. In this paper, we present method that allows owner encrypt with homomorphic encryption scheme perform k-means clustering directly over encrypted since ciphertexts resulting from do not preserve order distances between objects cluster centers, propose approach enables compare trapdoor...

This brief proposes a circuit structure that performs stateful logic operation on memristor memory based nanocrossbar. Through analysis and comparison of multiple schemes, achievable condition is demonstrated, the feasibility duplication proved. The proposed provides with function in situ thus can potentially reduce amount accessing actions provide possible solution to wall problem.

In mobile communication, spatial queries pose a serious threat to user location privacy because the of query may reveal sensitive information about user. this paper, we study k nearest neighbor (kNN) where location-based service (LBS) provider points interest (POIs) on basis his current location. We propose solution for preserve in kNN queries. The proposed is built Paillier public-key cryptosystem and can provide both data privacy. particular, our allows retrieve one type POIs, example, car...