Untrusted Third-parties Research Articles

Hardware Trojan Detection using Graph Neural Networks

The globalization of the Integrated Circuit (IC) supply chain has moved most of the design, fabrication, and testing process from a single trusted entity to various untrusted third party entities around the world. The risk of using untrusted third-Party Intellectual Property (3PIP) is the possibility for adversaries to insert malicious modifications known as Hardware Trojans (HTs). These HTs can compromise the integrity, deteriorate the performance, and deny the functionality of the intended design. Various HT detection methods have been proposed in the literature; however, many fall short due to their reliance on a golden reference circuit, a limited detection scope, the need for manual code review, or the inability to scale with large modern designs. We propose a novel golden reference-free HT detection method for both Register Transfer Level (RTL) and gate-level netlists by leveraging Graph Neural Networks (GNNs) to learn the behavior of the circuit through a Data Flow Graph (DFG) representation of the hardware design. We evaluate our model on a custom dataset by expanding the Trusthub HT benchmarks trusthub1. The results demonstrate that our approach detects unknown HTs with 97% recall (true positive rate) very fast in 21.1ms for RTL and 84% recall in 13.42s for Gate-Level Netlist.

Diagnosis of Malicious Bitstreams in Cloud Computing FPGAs

Multi-tenant FPGAs are increasingly being used in cloud computing technologies. Users are able to access the FPGA fabric remotely to implement custom accelerators in the cloud. However, the sharing of FPGA resources by untrusted third-parties can lead to serious security threats. Attackers can configure a portion of the FPGA with a malicious bitstream. Such malicious use of the FPGA fabric may lead to severe voltage fluctuations and denial-of-service. In this work, we consider FPGAs that support time-based multi-tenancy i.e., a single user has access to the FPGA at a time. We propose a convolutional neural network (CNN)-based approach to detect malicious RO-like circuits that are configured on an FPGA by learning features from the data-series representation of the bitstreams of malicious circuits. We use the classification accuracy, true-positive rate, and false-positive rate metrics to quantify the effectiveness of CNN-based classification of malicious bitstreams. Our threat model includes a variety of power-wasting circuits that are used to configure FPGAs in the cloud. We propose a two-stage malicious bitstream detection framework for classification and diagnosis of the type of malicious circuit implemented by a particular bitstream. We further propose a novel window-merging technique to improve model performance in the second stage of the detection framework. Experimental results on Xilinx FPGAs demonstrate the effectiveness of the proposed method.

Towards high performance homomorphic encryption for inference tasks on CPU: An MPI approach

Fully homomorphic encryption (FHE) is a powerful tool enabling secure outsourcing to untrusted third-parties. As AI technologies mature and companies start offering AI-as-a-service (AIaaS), the privacy of their customers’ data needs to be considered but the high overhead of existing FHE schemes is a major limitation. Prior work tried to address this but suffered accuracy degradation, lack of scalability, ciphertext expansion, and impractical performance issues. In this paper, we explore the utility of High-Performance Computing systems for accelerating the secure evaluation of deep convolutional neural networks (CNN) with FHE. We examine previous encrypted neural network architectures and propose a holistic re-design for the HPC environment, which introduces a new consideration for the latency of inter-node communication. The new architecture was implemented with MPI and OpenMP and its performance benchmarked against state-of-the-art FHE-based CNN evaluation methods. Our solution achieved improvements of between 6−34× for various CIFAR10 CNN architectures and more than 1500× for AlexNet, bringing FHE-based evaluation of AlexNet down from several hundred days to within 14 h. Such findings pave the way towards practical inference on encrypted medium resolution images through the use of HPC systems.

Privacy protection for Wi-Fi location positioning systems

With the democratization of mobile devices embedding different positioning capabilities, location information is used for a variety of applications. On mobile devices, the geolocation can be obtained via GPS or by leveraging surrounding network infrastructure such as Wi-Fi access points. Despite a lower accuracy, Wi-Fi based geolocation has several advantages over GPS such as reduced energy consumption and availability in indoor and availability in indoor environments. To enable this network-based geolocation, mobile devices need to interact with a location positioning system that will resolve a list of visible Wi-Fi access points into a position. By doing so, mobile users are revealing their mobility to the location provider, potentially exposing sensitive information to an untrusted third-party.In this paper, we propose a novel solution to preserve users’ privacy when requesting users’ location from Wi-Fi while supporting high utility. The key idea behind our online approach is to combine a caching strategy (for limiting the exposure of the user's position for already visited locations) and a random sampling (for controlling the precision of revealed information). We exhaustively evaluate our solution with a real dataset of mobility traces. We show that the proposed approach drastically reduces the exposure of the user's location to positioning systems (up to 95%). Indeed, by leveraging a caching strategy, requests are only sent when users visit new areas. Consequently, the capacity of positioning systems to extract points of interest of users from received requests is highly limited (a decrease of 50% on average). In addition, our privacy protection provides a trade-off between privacy (i.e., avoid revealing its true location) and utility (i.e., still benefiting from services such as places recommendation) fully controllable by the users.

Security of continuous-variable measurement-device-independent quantum key distribution with imperfect state preparation

Abstract The state preparation operation of continuous-variable measurement-device-independent quantum key distribution (CV-MDI-QKD) protocol may become imperfect in practical applications. We address the security of the CV-MDI-QKD protocol based on imperfect preparation of the coherent state under realistic conditions of lossy and noisy quantum channel. Specifically, we assume that the imperfection of Alice's and Bob's practical state preparations equal to the amplification of ideal modulators and lasers at both Alice's and Bob's sides by untrusted third-parties Fred and Gray employing phase-insensitive amplifiers (PIAs), respectively. The equivalent excess noise introduced by the imperfect state preparation is comprehensively and quantitatively calculated by adopting the gains of PIAs. Security analysis shows that CV-MDI-QKD is quite sensitive to the imperfection of practical state preparation, which inevitably deteriorates the performance and security of CV-MDI-QKD system. Moreover, a lower bound of the secret key rate is derived under arbitrary collective attacks, and the upper threshold of this imperfection tolerated by the system is obtained in the form of the specific gains of PIAs. In addition, the methods presented will improve and perfect the practical security of CV-MDI-QKD protocol.

A Game-Theoretic Framework to Preserve Location Information Privacy in Location-based Service Applications.

Recently, the growing ubiquity of location-based service (LBS) technology has increased the likelihood of users’ privacy breaches due to the exposure of their real-life information to untrusted third parties. Extensive use of such LBS applications allows untrusted third-party adversarial entities to collect large quantities of information regarding users’ locations over time, along with their identities. Due to the high risk of private information leakage using resource-constrained smart mobile devices, most LBS users may not be adequately encouraged to access all LBS applications. In this paper, we study the use of game theory to protect users against private information leakage in LBSs due to malicious or selfish behavior of third-party observers. In this study, we model a scenario of privacy protection gameplay between a privacy protector and an outside visitor and then derive the situation of the prisoner’s dilemma game to analyze the traditional privacy protection problems. Based on the analysis, we determine the corresponding benefits to both players using a point of view that allows the visitor to access a certain amount of information and denies further access to the user’s private information when exposure of privacy is forthcoming. Our proposed model uses the collection of private information about historical access data and current LBS access scenario to effectively determine the probability that the visitor’s access is an honest one. Moreover, we present the procedures involved in the privacy protection model and framework design, using game theory for decision-making. Finally, by employing a comparison analysis, we perform some experiments to assess the effectiveness and superiority of the proposed game-theoretic model over the traditional solutions.

Privacy-Preserving Multi-Keyword Top- Similarity Search Over Encrypted Data

Cloud computing provides individuals and enterprises massive computing power and scalable storage capacities to support a variety of big data applications in domains like health care and scientific research, therefore more and more data owners are involved to outsource their data on cloud servers for great convenience in data management and mining. However, data sets like health records in electronic documents usually contain sensitive information, which brings about privacy concerns if the documents are released or shared to partially untrusted third-parties in cloud. A practical and widely used technique for data privacy preservation is to encrypt data before outsourcing to the cloud servers, which however reduces the data utility and makes many traditional data analytic operators like keyword-based top- $k$ k document retrieval obsolete. In this paper, we investigate the multi-keyword top- $k$ k search problem for big data encryption against privacy breaches, and attempt to identify an efficient and secure solution to this problem. Specifically, for the privacy concern of query data, we construct a special tree-based index structure and design a random traversal algorithm, which makes even the same query to produce different visiting paths on the index, and can also maintain the accuracy of queries unchanged under stronger privacy. For improving the query efficiency, we propose a group multi-keyword top- $k$ k search scheme based on the idea of partition, where a group of tree-based indexes are constructed for all documents. Finally, we combine these methods together into an efficient and secure approach to address our proposed top- $k$ k similarity search. Extensive experimental results on real-life data sets demonstrate that our proposed approach can significantly improve the capability of defending the privacy breaches, the scalability and the time efficiency of query processing over the state-of-the-art methods.

Privacy Preserving Computation of Scalar Product and Sign of Scalar Product

Scalar product is repeatedly used in a large family of applications. It is a useful building block of many scientific and business applications. Secure scalar product computation is widely used in many privacy-preserving computation domains. In this work, we have proposed privacy preserving protocols for computing scalar product or only its sign. These protocols work on two basic principles - random data perturbation and algebraic methods, and thus avoid standard encryption/decryption overheads. We have designed the protocols to get sign of the scalar product, which can act as a basic building block in privacy preserving database queries over distributed databases and many other applications. We have developed two sets of protocols for the two-party computation with and without using an untrusted third-party. Experimental results on generated random datasets show that the proposed protocols indeed involve low communication overhead.

SMC Protocol for Naïve Bayes Classification over Grid Partitioned Data using Multiple UTPs

The case where data is distributed horizontally as well as vertically, it refers as grid partitioned data. SMC protocol for Naive Bayes classification over grid partitioned data is offered in this paper. Also present a solution of the Secure Multi-party Computation (SMC) problem in the form of a protocol that preserves privacy. In this system, a protocol with several Untrusted Third Parties (UTPs) is used, where there is almost impossible of privacy leakage. Multiple UTPs will calculate the model parameters for integrating the horizontally partitioned data. After that secure multiplication protocol will apply on vertical partitioned data (multiple UTPs) to classify the new tuple. The main contribution of this paper is that it shows a simple and easy calculation for developing Naive Bayes classifier for grid partitioned data. The evaluation method is simple and more efficient. General Terms SMC, Naive Bayes Classification.

Privacy Preserving Naïve Bayes Classifier for Horizontally Distribution Scenario Using Un-trusted Third Party

The aim of the classification task is to discover some kind of relationship between the input attributes and the output class, so that the discovered knowledge can be used to predict the class of a new unknown tuple. The problem of secure distributed classification is an important one. In many situations, data is split between multiple organizations. These organizations may want to utilize all of the data to create more accurate predictive models or classifier while revealing neither their training data nor the tuples to be classified. The Naive Bayes classifier is a simple but efficient baseline classifier. In this paper, we present a privacy preserving Naive Bayes classifier for horizontally partitioned data. Our three layer protocol uses an Un-trusted Third Party (UTP). We study how to calculate model parameters for privacy preserving three-layer Naive Bayes classifier for horizontally partitioned databases and communicate their intermediate results to the UTP not their private data. In our protocol, an UTP allows to meet privacy constraints and achieve acceptable performance.

A new watermarking protocol against conspiracy

A trusted third party introduced in watermarking protocols would decrease the security and affect the implementation of the protocols. In this paper, a new watermarking protocol with an un-trusted third party (UTTP) was proposed. Based on the idea of all-or-nothing disclosure of secret (ANDOS), all of the buyer, the seller and the third party didn't know the exact watermark, which was embedded in a digital content for tracing piracy. The proposed protocol provided mechanisms to trace piracy and protect customer's right. In addition, the problem that a seller colluded with UTTP to frame the buyer, namely, the conspiracy problem, could be avoided.