Strong Privacy Research Articles

Efficient Data Access Control With Fine-Grained Data Protection in Cloud-Assisted IIoT

The Industrial Internet of Things (IIoT) has provided a promising opportunity to build digitalized industrial systems. A fundamental technology of IIoT is the radio-frequency identification (RFID) technique, which allows industrial participants to identify items and anchor time-series IoT data for them. They can further share the IoT data through the cloud service to enable information exchange and support critical decisions in production operations. Storing IoT data in the cloud, however, requires a data access control mechanism to protect sensitive business issues. Unfortunately, using traditional cryptographic access control schemes for time-series IoT data face severe efficiency and key leakage problems. In this article, we design a secure industrial data access control scheme for cloud-assisted IIoT. Our scheme enables participants to enforce fine-grained access control policies for their IoT data via ciphertext policy-attribute-based encryption (CP-ABE) scheme. Our scheme adopts a hybrid cloud infrastructure for participants to outsource expensive CP-ABE tasks to the cloud service with strong privacy guarantees. Importantly, our scheme guarantees a new privacy notion named item-level data protection for IoT data to prevent key leakage problem. We achieve these goals via several encryption and optimization techniques. Our performance assessments combine system implementation with large-scale emulations and confirm the security and efficiency of our design.

SPCTR: Sealed Auction-Based Procurement for Closest Pre-Tender with Range Validation

Over the past decades, there have existed extensive research works on the designs of the closest pre-tender procurement bidding. However, most solutions for the closest pre-tender only target at economic benefits while omitting the problem of bid privacy leakage. Moreover, existing works fail to provide approaches with adequate security and high efficiency. In this paper, for the first time, we propose SPCTR, a sealed-price auction-based procurement bidding system for the closest pre-tender with range validation. SPCTR allows a range validation for a supplier’s bid without leaking the secret bid. Besides, SPCTR achieves a sealed-price comparison with the pre-tender to find the closest pre-tender bid. Compared with previous works, SPCTR provides strong privacy protection for the bids of suppliers without sacrificing high efficiency. SPCTR is constructed based on carefully designed cryptographic tools with generality and simplicity which enable various operations on the encrypted values, and these tools can be easily applied to other contexts. We not only formally prove that SPCTR is secure against semihonest adversaries but also comprehensively analyze the efficiency. Experimental results validate that SPCTR achieves procurement bidding with light computation time and communication cost in practice.

Computing value of spatiotemporal information

Location data from mobile devices is a sensitive yet valuable commodity for location-based services and advertising. We investigate the intrinsic value of location data in the context of strong privacy, where location information is only available from end users via purchase. We present an algorithm to compute the expected value of location data from a user, without access to the specific coordinates of the location data point. We use decision-theoretic techniques to provide a principled way for a potential buyer to make purchasing decisions about private user location data. We illustrate our approach in three scenarios: the delivery of targeted ads specific to a user's home location, the estimation of traffic speed, and the prediction of location. In all three cases, the methodology leads to quantifiably better purchasing decisions than competing approaches.

Differential privacy protection method based on published trajectory cross-correlation constraint.

Aiming to solve the problem of low data utilization and privacy protection, a personalized differential privacy protection method based on cross-correlation constraints is proposed. By protecting sensitive location points on the trajectory and their affiliated sensitive points, this method combines the sensitivity of the user's trajectory location and user privacy protection requirements and privacy budget to propose a (R,Ɛ) -extended differential privacy protection model. Using autocorrelation Laplace transform, specific Gaussian white noise is transformed into noise that is related to the user's real trajectory sequence in both time and space. Then the noise is added to the user trajectory sequence to ensure spatio-temporal correlation between the noise sequence and the user trajectory sequence. This defines the cross-correlation constraint mechanism of the published trajectory sequence. By superimposing the real trajectory sequence on the user’s noise sequence that satisfies the autocorrelation, a published trajectory sequence that satisfies the cross-correlation constraint condition is established to provide strong privacy guarantees against adversaries. Finally, the feasibility, effectiveness and rationality of the algorithm are verified by simulation experiments, and the proposed method is compared with recent studies in the same field on basis of merits and weakness and so on.

Privacy-preserving obfuscation for distributed power systems

This paper considers the problem of releasing privacy-preserving load data of a decentralized operated power system. The paper focuses on data used to solve Optimal Power Flow (OPF) problems and proposes a distributed algorithm that complies with the notion of Differential Privacy, a strong privacy framework used to bound the risk of re-identification. The problem is challenging since the application of traditional differential privacy mechanisms to the load data fundamentally changes the nature of the underlying optimization problem and often leads to severe feasibility issues. The proposed differentially private distributed algorithm is based on the Alternating Direction Method of Multipliers (ADMM) and guarantees that the released privacy-preserving data retains high fidelity and satisfies the AC power flow constraints. Experimental results on a variety of OPF benchmarks demonstrate the effectiveness of the approach.

Constitutional Pandemic Surveillance

Non-law enforcement governmental information gathering – special needs searches in the Fourth Amendment parlance – must be “reasonable” to pass constitutional scrutiny. We review the case law on special needs searches to shed light on the contours of this reasonableness inquiry. Then, to aid in this analysis, we measure the perceived intrusiveness of pandemic surveillance through two nationally representative surveys of Americans conducted in the Spring of 2020, during the COVID-19 crisis. Our results show that, even at the height of an international pandemic, people find surveillance for public health purposes to be more intrusive than surveillance for traditional law enforcement purposes. To account for these strong privacy concerns, we propose safeguards that we believe would make cell phone location tracking and other similar digital monitoring regimes constitutionally reasonable.

Differentially private 1R classification algorithm using artificial bee colony and differential evolution

Classification is an important topic in data mining field. Privacy preserving classification is a substantial subtopic that aims to perform classification of private data with satisfactory accuracy, while allowing sensitive information leakage at minimal level. Differential privacy is a strong privacy guarantee that determines privacy leakage ratio by using ϵ parameter; and enables privacy of individuals whose sensitive data are stored in a database. There exist some differentially private implementations of well-known classification algorithms such as ID3, random tree, random forests, Naïve Bayes, SVM, logistic regression, k-NN etc. Although One Rule (1R) is a simple but powerful classification algorithm, any implementation of differentially private 1R classification algorithm has not been proposed in the literature to our best knowledge. Motivated by this gap, first we propose a differentially private 1R classification algorithm (DP1R), then improve its performance by using metaheuristics that are differential evolution (DE) and artificial bee colony (ABC) in this study. Additionally, we also apply DE and ABC to improve performance of differentially private Naïve Bayes classifier and compare with DP1R. Moreover, DP1R is compared with the state-of-the-art differentially private algorithms such as differentially private SVM, differentially private logistic regression, differentially private ID3, and differentially private random tree on nine publicly available UCI datasets. The experimental results demonstrate that DP1R is an efficient classifier that has very similar accuracy to differentially private SVM which has the best accuracy results, however with respect to running time comparison of the methods, DP1R has the best performance among the all methods.

Differentially Private Tensor Deep Computation for Cyber–Physical–Social Systems

In the recent past, deep learning has received remarkable acceptance in real-world applications. Social computing expands the existing notion of cyber space and physical space to a more advance cyber-physical-social system (CPSS). Therefore, deep learning provides a propitious technique for accurate mining of information from CPSS, thus facilitates CPSS to offer services of exceptional quality efficiently. However, most of the current deep learning methods are struggling to keep up with the ever-increasing heterogeneous and highly nonlinear dissemination of data. Furthermore, the advancement of deep learning presents privacy concerns. This article proposes a deep private tensor autoencoder (dPTAE), where tensors are used for data representation, and differential privacy guarantees strong privacy. The core idea of our work is to enforce differential privacy through noise injection into the objective functions instead of the results they produce. In addition, the proposed method preserves the privacy of information shared amongst CPSS in smart environments. We applied dPTAE on three representative data sets. Rigorous experimental evaluations and theoretical analysis demonstrate that dPTAE is significantly effective and efficient.

Deep Robust Cramer Shoup Delay Optimized Fully Homomorphic For IIOT secured transmission in cloud computing

Several sensors obtain data during industrial outturn and send this collected data to the cloud server via Internet communication. Due to the reason that a cloud server is not an entirely trusted entity, data authenticity has to be provided prior to outsource to the cloud server, so that only authorized users or devices can access those authentic data from distinct topography areas. Hence, a strong privacy preservation mechanism is required during data collection. In addition, the data latency and network delay involved in communication should also be observed. In order to robust privacy preservation, Robust Cramer Shoup Delay Optimized Fully Homomorphic (RCS-DOFH) is proposed. This method includes three steps. First to minimize the communication overhead and time, Kullback–Leibler divergence is used in the Robust Cramer Shoup Decryption (RCSD) mechanism. Next, to minimize the data latency and network delay, Delay Optimized Fully Homomorphic Encryption (DOFHE) mechanism is designed. In this mechanism, delivery delay is calculated between the base station and IIoT device signal Finally, privacy preserving deep learning using RCSD and DOFHE is presented for privacy preserved secure data transmission. At first, RCSD mechanism is utilized to decrypt private generated signals along with its weight parameters. Then, the encryption is performed by using DOFHE mechanism. After that, activation function over the encryption region is determined. By this way, the proposed RCS-DOFH method achieves secure data transmission with minimum latency and network delay. The comparison of the RCS-DOFH method is provided and experiments conducted on SECOM dataset showed that the proposed method outperforms other conventional methods.

FedMEC: Improving Efficiency of Differentially Private Federated Learning via Mobile Edge Computing

Federated learning is a recently proposed paradigm that presents significant advantages in privacy-preserving machine learning services. It enables the deep learning applications on mobile devices, where a deep neural network (DNN) is trained in a decentralized manner among thousands of edge clients. However, directly apply the federated learning algorithm to the mobile edge computing environment will incur unacceptable computation costs in mobile edge devices. Moreover, among the training process, frequent model parameters exchanging between participants and the central server will increase the leakage possibility of the users’ sensitive training data. Aiming at reducing the heavy computation cost of DNN training on edge devices while providing strong privacy guarantees, we propose a mobile edge computing enabled federated learning framework, called FedMEC, which integrating model partition technique and differential privacy simultaneously. In FedMEC, the most complex computations can be outsourced to the edge servers by splitting a DNN model into two parts. Furthermore, we apply the differentially private data perturbation method to prevent the privacy leakage from the local model parameters, in which the updates from an edge device to the edge server is perturbed by the Laplace noise. To validate the proposed FedMEC, we conduct a series of experiments on an image classification task under the settings of federated learning. The results demonstrate the effectiveness and practicality of our FedMEC scheme.

Enabling Strong Privacy Preservation and Accurate Task Allocation for Mobile Crowdsensing

Mobile crowdsensing engages a crowd of individuals to use their mobile devices to cooperatively collect data about social events and phenomena for special interest customers. It can reduce the cost on sensor deployment and improve data quality with human intelligence. To enhance data trustworthiness, it is critical for service provider to recruit mobile users based on their personal features, e.g., mobility pattern and reputation, but it leads to the privacy leakage of mobile users. Therefore, how to resolve the contradiction between user privacy and task allocation is challenging in mobile crowdsensing. In this paper, we propose SPOON, a strong privacy-preserving mobile crowdsensing scheme supporting accurate task allocation from geographic information and credit points of mobile users. In SPOON, the service provider enables to recruit mobile users based on their locations, and select proper sensing reports according to their trust levels without invading user privacy. By utilizing proxy re-encryption and BBS+ signature, sensing tasks are protected and reports are anonymized to prevent privacy leakage. In addition, a privacy-preserving credit management mechanism is introduced to achieve decentralized trust management and secure credit proof for mobile users. Finally, we show the security properties of SPOON and demonstrate its efficiency on computation and communication.

Privacy and Ethics Are Fundamental to Tech Development

For years, privacy regulators have said that privacy is good for business. Strong privacy management programs and accountability mechanisms build trust with consumers. In the public sector, privacy regulators have seen massive information sharing projects fail when public input or consultation, or independent oversight is not considered. After a sequence of events in 2018, society as a whole began asking questions about what is being done with personal information and questioned whether it is in our best interests. This presentation made at the University of Alberta’s Kule Institute’s event on “AI, Ethics and Society” in May 2019 provides an overview of the shifts that have taken place and how privacy regulators internationally have incorporated discussions about ethical assessments, in addition to traditional privacy impact assessments, as a way to guide current and future tech developments involving personal information in a way that is legal, fair and just.

Highly efficient federated learning with strong privacy preservation in cloud computing

Federated learning is a new machine learning framework that allows mutually distrusting clients to reap the benefits from the joint training model without explicitly disclosing their private datasets. However, the high communication cost between the cloud server and clients has become the main challenge due to the limited network bandwidth. Moreover, the model parameters it shares may be utilized to perform model inversion attacks. Aimed at these problems, a new scheme for highly efficient federated learning with strong privacy preservation in cloud computing is presented. We design a lightweight encryption protocol to provide provably privacy preservation while maintaining desirable model utility. Additionally, an efficient optimization strategy is employed to enhance the training efficiency. Under the defined threat model, we prove the proposed scheme is secure against the honest-but-curious server and extreme collusion. We evaluate the effectiveness of our scheme and compare it with existing related works on MNIST and UCI Human Activity Recognition Dataset. Results show that our scheme reduces the execution time by 20% and transmitted ciphertext size by 85% on average while achieving similar accuracy as the compared secure multiparty computation (SMC) based methods.

Privacy-Preserving Efficient Verifiable Deep Packet Inspection for Cloud-Assisted Middlebox

With the increasing traffic volume, enterprises choose to outsource their middlebox services, such as deep packet inspection, to the cloud to acquire rich computational and communication resources. However, since the traffic is redirected to the public cloud, information leakages, such as packet payload and inspection rules, arouse privacy concerns of both middlebox owner and packet senders. To address the concerns, we propose an efficient verifiable deep packet inspection (EV-DPI) scheme with strong privacy guarantees. Specifically, a two-layer architecture is designed and deployed over two non-collusion cloud servers. The first layer fast filters out most of legitimate packets and the second layer supports exact rule matching. During the inspection, the privacy of packet payload and the confidentiality of inspection rules are well preserved. To improve the efficiency, only fast symmetric crypto-systems, such as hash functions, are used. Moreover, the proposed scheme allows the network administrator to verify the execution results, which offers a strong control of outsourced services. To validate the performance of the proposed EV-DPI scheme, we conduct extensive experiments on the Amazon Cloud. Large-scale dataset (millions of packets) is tested to obtain the key performance metrics. The experimental results demonstrate that EV-DPI not only preserves the packet privacy, but also achieves high packet inspection efficiency.

A Practical System for Privacy-Aware Targeted Mobile Advertising Services

With the prosperity of mobile application markets, mobile advertising is becoming an increasingly important economic force. In order to maximize revenue, ads are recommended to be delivered to potentially interested users, which requires user targeting, i.e., analyzing users’ profiles and exploring users’ interests. However, collecting user personal information for targeted mobile advertising services raises critical privacy concerns. Although some solutions like anonymization and obfuscation have been proposed for privacy-aware targeted advertising, they undesirably face the issues of security, efficiency, and/or ad relevance. In this paper, we propose a practical system enabling secure and efficient targeted mobile advertising services. It allows the ad network to perform accurate user targeting, while ensuring strong privacy protection for mobile users. Specifically, we show how to properly leverage a cryptographic primitive called private stream searching to support secure, accurate, and practical targeted mobile ad delivery. Moreover, we propose secure billing schemes to enable the ad network to charge advertisers in a privacy-preserving manner. The security strength of our system is thoroughly analyzed. Through extensive experiments, we show that our system achieves practical efficiency on mobile devices.

Fisheye Lens: Data Stewardship and Privacy Rights Under the Northeast Multispecies Fishery Management Plan's Amendment 23's Proposed Electronic Monitoring

The Atlantic Cod, Gadus morhua, once held a prime ecological and economic presence in the North Atlantic Ocean, and European colonizers capitalized on this resource by establishing a cod fishery in Newfoundland as far back as 1517. In 1992, Newfoundland’s fishery nearly eliminated its cod stocks, and the Gulf of Maine’s stocks were dying three times as fast as managers predicted. The fishery became one of the world’s most concrete examples of the tragedy of the commons, where harvesters exhaust an open-access resource. Guiding the New England Fisheries Management Council (NEFMC) cod fishery rebuilding effort is a tremendous amount of fisheries modeling aimed at estimating a requisite fishery’s stock size, using landings, discards, and survey data. The Northeast Fisheries Science Center (NEFSC) sources and verifies data partly through the Northeast Fisheries Observer Program (NEFOP). Among the observers’ duties is physically boarding industry vessels to record data. The observers’ physical bias, however, could produce inaccurate data, leading the NEFMC to make inaccurate decisions and hinder the northeastern U.S. cod fishery’s recovery. Inaccurate data would thus hinder managers’ ability to rebuild this historically profitable fishery and the lack of cod would continue to deprive the Northeast’s ecosystem of a key predator. In response to the concern over bias, the NEFMC is currently considering several alternatives to the traditional human observer program in Amendment 23 to the Northeast Multispecies Fisheries Management Plan. One such alternative is electronic monitoring (EM). EM is a system of cameras that captures video of fishing activities onboard vessels, a crucial tool to produce high-quality and low-cost fisheries data to better inform management decisions. However, Northeast fishers have several concerns about EM. EM is a seemingly simple solution to human observers but produces highly detailed video and image data instead of pure numbers, which could identify private individuals, business, or their trade secrets. EM thus presents uncertainty in fishers’ private information, leading to apprehension of EM systems. The opportunity to design a program that is amenable to fishers could speed up implementing EM and its benefits to the northeast’s ecological and economic marine communities. Amending the Magnuson Stevens Act, promulgating strong privacy regulations and internal rules, and ensuring privacy guarantees in third-party contracts would best protect fishers’ privacy, which the National Oceanic and Atmospheric Administration should consider in light of their broad authority to collect EM data. Strong fisheries management requires strong cooperation and mutual respect; guaranteeing strong privacy rights for EM data furthers everyone’s interests. The goal of this Note is to provide a framework to promote the recovery of New England’s marine environment through coastal stakeholder and federal management cooperation.

Pairwise Learning with Differential Privacy Guarantees

Pairwise learning has received much attention recently as it is more capable of modeling the relative relationship between pairs of samples. Many machine learning tasks can be categorized as pairwise learning, such as AUC maximization and metric learning. Existing techniques for pairwise learning all fail to take into consideration a critical issue in their design, i.e., the protection of sensitive information in the training set. Models learned by such algorithms can implicitly memorize the details of sensitive information, which offers opportunity for malicious parties to infer it from the learned models. To address this challenging issue, in this paper, we propose several differentially private pairwise learning algorithms for both online and offline settings. Specifically, for the online setting, we first introduce a differentially private algorithm (called OnPairStrC) for strongly convex loss functions. Then, we extend this algorithm to general convex loss functions and give another differentially private algorithm (called OnPairC). For the offline setting, we also present two differentially private algorithms (called OffPairStrC and OffPairC) for strongly and general convex loss functions, respectively. These proposed algorithms can not only learn the model effectively from the data but also provide strong privacy protection guarantee for sensitive information in the training set. Extensive experiments on real-world datasets are conducted to evaluate the proposed algorithms and the experimental results support our theoretical analysis.

A Tale of Two Trees: One Writes, and Other Reads

Abstract The Bitcoin network has offered a new way of securely performing financial transactions over the insecure network. Nevertheless, this ability comes with the cost of storing a large (distributed) ledger, which has become unsuitable for personal devices of any kind. Although the simplified payment verification (SPV) clients can address this storage issue, a Bitcoin SPV client has to rely on other Bitcoin nodes to obtain its transaction history and the current approaches offer no privacy guarantees to the SPV clients. This work presents T 3, a trusted hardware-secured Bitcoin full client that supports efficient oblivious search/update for Bitcoin SPV clients without sacrificing the privacy of the clients. In this design, we leverage the trusted execution and attestation capabilities of a trusted execution environment (TEE) and the ability to hide access patterns of oblivious random access machine (ORAM) to protect SPV clients’ requests from potentially malicious nodes. The key novelty of T 3 lies in the optimizations introduced to conventional ORAM, tailored for expected SPV client usages. In particular, by making a natural assumption about the access patterns of SPV clients, we are able to propose a two-tree ORAM construction that overcomes the concurrency limitation associated with traditional ORAMs. We have implemented and tested our system using the current Bitcoin Unspent Transaction Output (UTXO) Set. Our experiment shows that T 3 is feasible to be deployed in practice while providing strong privacy and security guarantees to Bitcoin SPV clients.

An empirical examination of voluntary profiling: Privacy and quid pro quo

Evidence suggests that firms which use customer data analytics perform better than those that do not. However, the current policy of voluntary profiling allows firms to collect and use customer information only if customers voluntarily disclose information with them. Further, surveys and literature show that many customers are not comfortable with firms collecting their information due to privacy concerns. A vast literature has examined customer information disclosing behavior using the privacy calculus. The primary premise of the privacy calculus is that despite strong privacy concerns, customers disclose information if the benefits they can get from disclosure justify the costs of losing privacy, or privacy costs. Based on the privacy calculus, firms and marketers believe that customers voluntarily disclose information in exchange for monetary benefits such as discount coupons or cash rewards. Using transaction data that we collected from a firm that sells skin care cosmetic products on its website, we investigate if there is statistical evidence that shows customers disclose information in exchange for monetary benefits. In line with the privacy calculus, we find that customers with low privacy costs and that expect high benefits from personalized services such as product recommendations are more likely to disclose information. Monetary incentives only work as an effective means to elicit information from customers in the age range of 12–21 years old. Customers, on average, and especially customers in the age range of 22–54 years old are not likely to disclose information simply as a tradeoff for monetary benefits. Personalized services outweigh monetary benefits in enticing these customers to disclose information.

Top-K search scheme on encrypted data in cloud

<span>A Secure and Effective Multi-keyword Ranked Search Scheme on Encrypted Cloud Data. Cloud computing is providing people a very good knowledge on all the popular and relevant domains which they need in their daily life. For this, all the people who act as Data Owners must possess some knowledge on Cloud should be provided with more information so that it will help them to make the cloud maintenance and administration easy. And most important concern these days is privacy. Some sensitive data exposed in the cloud these days have security issues. So, sensitive information ought to be encrypted earlier before making the data externalized for confidentiality, which makes some keyword-based information retrieval methods outdated. But this has some other problems like the usage of this information becomes difficult and also all the ancient algorithms developed for performing search on these data are not so efficient now because of the encryption done to help data from breaches. In this project, we try to investigate the multi- keyword top-k search problem for encryption against privacy breaks and to establish an economical and secure resolution to the present drawback. we have a tendency to construct a special tree-based index structure and style a random traversal formula, which makes even identical question to supply totally different visiting ways on the index, and may additionally maintain the accuracy of queries unchanged below stronger privacy. For this purpose, we take the help of vector area models and TFIDF. The KNN set of rules are used to develop this approach.</span>