Attacking cloud-enabled storage is becoming increasingly lucrative as more personal and enterprise data moves to the cloud. Traditional security mechanisms temporarily limit such attacks, but over a long period of time attackers will eventually find vulnerabilities; this can lead to compromising large amounts of valuable data and lead to large-scale privacy breaches. This paper addresses this problem by incorporating proactive security guarantees into cloud-enabled storage. Proactive security deals with an adversary's ability to eventually compromise all involved servers in a distributed storage or computation system. While there are several proactively secure secret sharing protocols that can be used to improve confidentiality of data stored in the cloud, their high overhead has traditionally limited them to less than ten parties and to only 100s of bytes typical for cryptographic keys. Realizing proactively secure cloud storage for larger data (e.g, MBs) requires careful design and calibration of system parameters, and faces several challenges. In this paper we design, implement and assess performance of the first system for Proactively Secure Cloud-Enabled Storage (PiSCES) of data larger than cryptographic keys. Based on our practical performance results we advocate that the high level of resilience and long-term security and confidentiality guarantees enabled by proactive security should be considered in future distributed and cloud-based storage and computing services.

Private Set Intersection (PSI) and other private set operations have many current and emerging applications. Numerous PSI techniques have been proposed that vary widely in terms of underlying cryptographic primitives, security assumptions as well as complexity. One recent strand of PSI-related research focused on an additional privacy property of hiding participants’ input sizes. Despite some interesting results, only one practical size-hiding PSI (SH-PSI) has been demonstrated thus far [1]. One legitimate general criticism of size-hiding private set intersection is that the party that hides its input size can attempt to enumerate the entire (and possibly limited) domain of set elements, thus learning the other party’s entire input set. Although this “attack” goes beyond the honest-but-curious model, it motivates investigation of techniques that simultaneously hide and limit a participant’s input size. To this end, this paper explores the design of bounded size-hiding PSI techniques that allow one party to hide the size of its input while allowing the other party to limit that size. Its main contribution is a reasonably efficient (quasi-quadratic in input size) b- protocol based on bounded keyed accumulators. This paper also studies the relationships between several flavors of the “Strong Diffie-Hellman” (SDH) problem.

Private Set Intersection (PSI) is a cryptographic primitive that allows two network connected parties with hidden inputs to jointly compute the intersection of these inputs while keeping their specific inputs secret. PSI can be used as a building block for a variety of applications, most notably querying a remote relational database without revealing the query or the database. Many constructions of PSI exist, each building off of a subset of an assortment of cryptographic primitives such as: oblivious transfer, hash functions, garbled circuits, public key encryption and signature schemes, and basic number theoretic hardness assumptions. In this dissertation, we study variations of PSI and their practicality to modern day applications. Specifically, we study new security constraints for PSI that arise from applications such as genomics, consumer applications, and inter-agency information sharing. These constraints lead to several novel secure computation protocols. Through actualized prototypes of these schemes we conclude that, specialized PSI protocols are fast enough for use today, even on resource constrained hardware.

There are many realistic settings where two mutually suspicious parties need to share some specific information while keeping everything else private. Various privacy-preserving techniques (such as Private Set Intersection) have been proposed as general solutions. Based on timely real-world examples, this paper motivates the need for a new privacy tool, called Private Set Intersection with Projection (PSI-P). In it, Server has (at least) a two-attribute table and Client has a set of values. At the end of the protocol, based on all matches between Client's set and values in one (search) attribute of Server's database, Client should learn the set of elements corresponding to the second attribute, and nothing else. In particular the intersection of Client's set and the set of values in the search attribute must remain hidden. We construct several efficient (linear complexity) protocols that approximate privacy required by PSI-P and suffice in many practical scenarios. We also provide a new construction for PSI-P with full privacy, albeit it is slightly less efficient. Its key building block is a primitive called Existential Private Set Intersection (PSI-X) which yields a binary flag indicating whether the intersection of two private sets is empty or non-empty.

An Oblivious RAM (ORAM) protocol [13] allows a client to retrieve N -th element of a data array 𝖣 stored by the server s.t. the server learns no information about N . A related notion is that of an ORAM for Secure Computation (SC-ORAM) [17], which is a protocol that securely implements a RAM functionality, i.e. given a secret-sharing of both 𝖣 and N , it computes a secret-sharing of 𝖣[N] . SC-ORAM can be used as a subprotocol for implementing the RAM functionality for secure computation of RAM programs [7, 14, 17]. It can also implement a public database service which hides each client’s access pattern even if a threshold of servers colludes with any number of clients. Most previous works used two-party secure computation to implement each step of an ORAM client algorithm, but since secure computation of many functions becomes easier in the three-party honest-majority setting than in the two-party setting, it is natural to ask if the cost of an SC-ORAM scheme can be reduced if one was willing to use three servers instead of two and assumed an honest majority. We show a 3-party SC-ORAM scheme which is based on a variant of the Binary Tree Client-Server ORAM of Shi et al. [20]. However, whereas previous SC-ORAM implementations used general 2PC or MPC techniques like Yao’s garbled circuits, e.g. [14, 22], homomorphic encryption [11], or the SPDZ protocol for arithmetic circuits [15], our techniques are custom-made for the three-party setting, giving rise to a protocol which is secure against honest-but-curious faults using bandwidth and CPU costs which are comparable to those of the underlying Client-Server ORAM.

The recent decade has witnessed a rapid increase in popularity of mobile personal devices (notably, smartphones) t hat function as all-purpose personal communication portals. Concurrently, On-line Social Networks (OSNs) have continued their impressive proliferation. Meanwhile, the notion of "OSN privacy" remains elusive and even self-contradictory. Centralized nature of prominent OSNs is unlikely to change, which does not bode well for OSN users' privacy. However, some user privacy can be gained from making certain OSN functionality available off-line, such as discovering common contacts and other features, as well as establishing affinity- based connections. OSN providers stand to gain from this, since users could avail themselves of OSN functionality in scenarios where none currently exists, e.g., whenever Internet connectivity is unavailable, expensive or insufficient. At the same time, OSN users benefit from increased privacy because off-line interactions can be made opaque to OSN providers. This paper explores off-line private proximity-based use of OSNs. Although our approach is quite general, the proposed system (called UnLinked) is grafted atop a specific and popular OSN -- LinkedIn. One key challenge is how to ensure authenticity and privacy of users' information (e.g., connections and other profile data) when they engage in off-line interactions. This is addressed by designing an efficient technique for authorized two-way private set intersection (ATW-PSI), which allows two OSN users to jointly learn only the intersection of their input sets, while being assured of the authenticity of each others' input. The paper describes and evaluates a practical prototype that allows physically proximate LinkedIn users to commit to a connection if they have a mutually acceptable number of common connections.

We extend the searchable symmetric encryption (SSE) protocol of [Cash et al., Crypto’13] adding support for range, substring, wildcard, and phrase queries, in addition to the Boolean queries supported in the original protocol. Our techniques apply to the basic single-client scenario underlying the common SSE setting as well as to the more complex Multi-Client and Outsourced Symmetric PIR extensions of [Jarecki et al., CCS’13]. We provide performance information based on our prototype implementation, showing the practicality and scalability of our techniques to very large databases, thus extending the performance results of [Cash et al., NDSS’14] to these rich and comprehensive query types.

Balancing security and privacy concerns with information sharing is a top priority for corporations, law enforcement agencies, governments, and other organizations. Secure pattern matching (SPM) addresses some of the challenges faced in sharing and searching private data.

Recent progress in genomics and bioinformatics is bringing complete and on-demand sequencing of human (and other) genomes closer and closer to reality. Despite exciting new opportunities, affordable and ubiquitous genome sequencing prompts some serious privacy and ethical concerns, owing to extreme sensitivity and uniqueness of genomic information. At the same time, new medical applications, such as personalized medicine, require testing genomes for specific markers that themselves represent sensitive (e.g., proprietary) material. This paper focuses on privacy challenges posed by such genetic tests. It presents a secure and efficient protocol called: Size- and Position-Hiding Private Substring Match- ing (SPH-PSM). This protocol allows two parties -- one with a digitized genome and the other with a set of DNA markers -- to conduct a test, such that the result is only learned by the former, and no other information is learned by either party. In particular, the genome owner does not even learn the size or the position of the markers, which makes SPH-PSM the first of its kind. Finally, we report on a prototype of the proposed technique which attests to its practicality.

As fast and accurate sequencing of human genomes becomes affordable, it is expected that individuals will soon be able to carry around copies of their sequenced DNA, using it for medical, identification, and social purposes. This will undoubtedly prompt a wide range of new and interesting genomic applications. However, the very same progress raises some worrisome privacy issues, since a genome represents a treasure trove of highly personal and sensitive information. Some recent research explored privacy-preserving personal genomic operations by applying (or customizing) cryptographic protocols based on techniques such as: conditional oblivious transfer, garbled circuits, and homomorphic encryption. In this paper, we take this line of work a step further by investigating real-world practicality and usability of (as well as interest in) some of these methods. Motivated by both medical and social applications, we aim to test viability of privacy-agile computational genomic tests in a portable and pervasive setting of modern smartphones. We design a personal genomic toolkit (called GenoDroid), implement it on the Android platform, assess its performance, and conduct a pilot usability study that yields some interesting results.

As fast and accurate sequencing of human genomes becomes affordable, it is expected that individuals will soon be able to carry around copies of their sequenced DNA, using it for medical, identification, and social purposes. This will undoubtedly prompt a wide range of new and interesting genomic applications. However, the very same progress raises some worrisome privacy issues, since a genome represents a treasure trove of highly personal and sensitive information. Some recent research explored privacy-preserving personal genomic operations by applying (or customizing) cryptographic protocols based on techniques such as: conditional oblivious transfer, garbled circuits, and homomorphic encryption. In this paper, we take this line of work a step further by investigating real-world practicality and usability of (as well as interest in) some of these methods. Motivated by both medical and social applications, we aim to test viability of privacy-agile computational genomic tests in a portable and pervasive setting of modern smartphones. We design a personal genomic toolkit (called GenoDroid), implement it on the Android platform, assess its performance, and conduct a pilot usability study that yields some interesting results.

The mammalian brain is best understood as a multi-scale hierarchical neural system, in the sense that connection and function occur on multiple scales from micro to macro. Modern genomic-scale expression profiling can provide insight into methodologies that elucidate this architecture. We present a methodology for understanding the relationship of gene expression and neuroanatomy based on correlation between gene expression profiles across tissue samples. A resulting tool, NeuroBlast, can identify networks of genes co-expressed within or across neuroanatomic structures. The method applies to any data modality that can be mapped with sufficient spatial resolution, and provides a computation technique to elucidate neuroanatomy via patterns of gene expression on spatial and temporal scales. In addition, from the perspective of spatial location, we discuss a complementary technique that identifies gene classes that contribute to defining anatomic patterns.