Yian Zhou

A dynamic Proof of Retrievability (PoR) scheme with O(logn) complexity

Cloud storage has been gaining popularity because its elasticity and pay-as-you-go manner. However, this new type of storage model also brings security challenges. This paper studies the problem of ensuring data integrity in cloud storage. In the Proof of Retrievability (PoR) model, after outsourcing the preprocessed data to the server, the client will delete its local copies and only store a small amount of meta data. Later the client will ask the server to provide a proof that its data can be retrieved correctly. However, most of the prior PoR works apply only to static data. The existing dynamic version of PoR scheme has an efficiency problem. In this paper, we extend the static PoR scheme to dynamic scenario. That is, the client can perform update operations, e.g., insertion, deletion and modification. After each update, the client can still detect the data losses even if the server tries to hide them. We develop a new version of authenticated data structure based on a B+ tree and a merkle hash tree. We call it Cloud Merkle B+ tree (CMBT). By combining the CMBT with the BLS signature, we propose a dynamic version of PoR scheme. Compared with the existing dynamic PoR scheme, our worst case communication complexity is O(logn) instead of O(n).

Privacy-Preserving Transportation Traffic Measurement in Intelligent Cyber-physical Road Systems

Traffic measurement is a critical function in transportation engineering. We consider privacy-preserving point-to-point traffic measurement in this paper. We measure the number of vehicles traveling from one geographical location to another by taking advantage of capabilities provided by the intelligent cyberphysical road systems (CPRSs) that enable automatic collection of traffic data. The challenge is to allow the collection of aggregate point-to-point data while preserving the privacy of individual vehicles. We propose a novel measurement scheme, which utilizes bit arrays to collect “masked” data and adopts maximum-likelihood estimation (MLE) to obtain the measurement result. Both mathematical proof and simulation demonstrate the practicality and scalability of our scheme.

Privacy-Preserving Multi-Point Traffic Volume Measurement Through Vehicle-to-Infrastructure Communications

Traffic volume measurement is critical in vehicular networks. Existing research on traffic volume measurement mainly focuses on single-point traffic statistics. In this paper, we switch our view from single-point to multi-point and study the important problem of privacy-preserving multi-point traffic volume measurement in vehicular cyber-physical systems (VCPSs), which complements the state of the art. While embracing automatic traffic data collection, which the VCPS provides through vehicle-to-infrastructure communications, we also need to accept the accompanying challenges: First, the privacy of all participating vehicles should be preserved as an inherent requirement of a VCPS; second, the measurement scheme should be efficient enough to fit todays large-scale vehicular networks. In this paper, we start from a novel scheme that measures traffic volume between two arbitrary points (locations) through variable-length bit array masking. Then, we extend the idea of variable-length bit array masking to address the more challenging problem of three-point traffic measurement and present a general framework to measure traffic among three or more locations. We also perform extensive simulations to demonstrate the superior performance, applicability, and scalability of our schemes.

Temporally or Spatially Dispersed Joint RFID Estimation Using Snapshots of Variable Lengths

Radio-frequency identification (RFID) technology has been widely used in applications such as inventory control, object tracking, supply chain management. An important research is to estimate the number of tags in a certain area covered by readers. This paper extends the research in both temporal and spatial dimensions to provide much richer information for monitoring the dynamics of distributed RFID systems. More specifically, we are interested in estimating the joint properties of any two snapshots taken at arbitrary locations and arbitrary times in a system. With many practical applications, there is however little prior work on this problem. We propose a joint RFID estimation protocol based on a simple yet versatile snapshot construction. Given the snapshots of any two tag sets, although their sizes may be very different, we design a way to combine their information and more importantly derive formulas to extract the joint properties of the two tag sets from the combined information, with an accuracy that can be arbitrarily set. Through formal analysis, we determine the optimal system parameters that minimize the execution time of taking snapshots, under the constraints of a given accuracy requirement. Our simulation results show that the proposed protocol can reduce the execution time by multifold when comparing with the best alternative approach in the literature.

Point-to-Point Traffic Volume Measurement through Variable-Length Bit Array Masking in Vehicular Cyber-Physical Systems

In this paper, we consider an important problem of privacy-preserving point-to-point traffic volume measurement in vehicular cyber physical systems (VCPS), whose focus is utilizing VCPS to enable automatic traffic data collection, and measuring point-to-point traffic volume while preserving the location privacy of all participating vehicles. The novel scheme that we propose tackles the efficiency, privacy, and accuracy problems encountered by previous solutions. Its applicability is demonstrated through both mathematical and numerical analysis. The simulation results also show its superior performance.

On Deletion of Outsourced Data in Cloud Computing

Data security is a major concern in cloud computing. After clients outsource their data to the cloud, will they lose control of the data? Prior research has proposed various schemes for clients to confirm the existence of their data on the cloud servers, and the goal is to ensure data integrity. This paper investigates a complementary problem: When clients delete data, how can they be sure that the deleted data will never resurface in the future if the clients do not perform the actual data removal themselves? How to confirm the non-existence of their data when the data is not in their possession? One obvious solution is to encrypt the outsourced data, but this solution has a significant technical challenge because a huge amount of key materials may have to be maintained if we allow fine-grained deletion. In this paper, we explore the feasibility of relieving clients from such a burden by outsourcing keys (after encryption) to the cloud. We propose a novel multi-layered key structure, called Recursively Encrypted Red-black Key tree (RERK), that ensures no key materials will be leaked, yet the client is able to manipulate keys by performing tree operations in collaboration with the servers. We implement our solution on the Amazon EC2. The experimental results show that our solution can efficiently support the deletion of outsourced data in cloud computing.

Enabling Non-repudiable Data Possession Verification in Cloud Storage Systems

After clients outsource their data to the cloud, they will lose physical control of their data. Many schemes are proposed for clients to verify the integrity of their data. This paper considers a complementary problem: When a client claims that the server has lost their data, how can we be sure that the client is correct and honest about the loss? It is possible that the clients meta data is corrupted or the client is lying in order to blackmail the server. In addition, most previous work relies on sequential indices. However, the indices bring significant overhead to bind an index to each block. We propose to replace sequential indices with much flexible non-sequential {\it coordinates}. The binding of coordinates to data blocks is performed through a Coordinate Merkle Hash Tree (CMHT). Based on CMHT, we can improve both the average and the worst-case update overhead by simplifying the updating algorithm.

MVP: An Efficient Anonymous E-Voting Protocol

Thanks to the Internet, voters can cast their ballots over the electronic voting (E-voting) systems conveniently and efficiently without going to the polling stations. However, existing E- voting protocols suffer from anonymity issues and/or high deployment overhead. In this paper, we design a practical anonymous E-voting protocol (referred to as MVP) based on a novel data collection technique called \emph{dual random matrix masking} (DRMM), which guarantees anonymity with low overhead of computation, and achieves the security goals of receipt-freeness, double voting detection, fairness, ballot secrecy, and integrity. Through extensive analyses on correctness, efficiency, and security properties, we demonstrate our proposed MVP protocol can be applied to E-voting in a variety of situations with accuracy and anonymity.

Highly Compact Virtual Counters for Per-Flow Traffic Measurement through Register Sharing

Per-flow traffic measurement is a fundamental problem in the era of big network data, providing critical information for many practical applications including capacity planning, traffic engineering, data accounting, resource management, and scan/intrusion detection in modern computer networks. It is challenging to design highly compact data structures for approximate per-flow measurements. In this paper, we show that a highly compact virtual counter architecture can achieve fast processing speed (slightly more than 1 memory access per packet) and provide accurate measurement results under tight memory allocation. Extensive experiments based on real network trace data demonstrate its superior performance over the best existing work.

Privacy preserving origin-destination flow measurement in vehicular cyber-physical systems

Traffic volume measurement is one of the most basic functions of road planning and management. In this paper, we investigate an important problem of privacy preserving “point-to-point” traffic volume measurement. We formalize “point-to-point” traffic as an origin-destination (O-D) flow, which represents the set of vehicles traveling from one geographical location (origin) to another location (destination). We take advantage of vehicular cyber-physical systems (VCPS) to exploit the potential for a fundamental shift in the way how O-D data are collected. The challenge is to allow the collection of statistical O-D flow information, yet protect identities of individual vehicles. To address that, we design two novel schemes which utilize both the latest technological advance in VCPS and the nice properties of a family of commutative one-way hash functions. Furthermore, we adopt statistical methodology and use sampling to achieve far better efficiency with graceful degradation in measurement accuracy. We perform simulations to demonstrate the feasibility and scalability of our schemes.