An image hash is a randomized compact representation of image content and finds applications in image authentication, image and video watermarking, and image similarity comparison. Usually, an image-hashing scheme is required to be robust and secure, and the security issue is particularly important in applications, such as multimedia authentication, watermarking, and fingerprinting. In this paper, we investigate the security of image hashing from the perspective of unicity distance, a concept pioneered by Shannon in one of his seminal papers. Using two recently proposed image-hashing schemes as representatives, we show that the concept of unicity distance can be adapted to evaluate the security of image hashing. Our analysis shows that the secret hashing key, or its equivalent form, can be estimated with high accuracy when the key is reused several dozen times. The estimated unicity distance determines the maximum number of key reuses in the investigated hashing schemes. A countermeasure of randomized key initialization is discussed to avoid key reuse and strengthen the security of robust image hashing.

An image hash is a content-based compact representation of an image and finds applications in image authentication, watermarking, and image similarity comparison. This paper reviews representative approaches developed by the research community in constructing robust and secure image hash. A unified framework and case studies on representative image hash schemes are presented to demonstrate important roles of signal processing and controlled randomization in constructing and evaluating robust and secure hashing for visual data.

This paper introduces a new framework for confidentiality preserving rank-ordered search and retrieval over large document collections. The proposed framework not only protects document/query confidentiality against an outside intruder, but also prevents an untrusted data center from learning information about the query and the document collection. We present practical techniques for proper integration of relevance scoring methods and cryptographic techniques, such as order preserving encryption, to protect data collections and indices and provide efficient and accurate search capabilities to securely rank-order documents in response to a query. Experimental results on the W3C collection show that these techniques have comparable performance to conventional search systems designed for non-encrypted data in terms of search accuracy. The proposed methods thus form the first steps to bring together advanced information retrieval and secure search capabilities for a wide range of applications including managing data in government and business operations, enabling scholarly study of sensitive data, and facilitating the document discovery process in litigation.

A cooperative communication system explores a new dimension of diversity in wireless communications to combat the unfriendly wireless environment. While this emerging technology is promising in improving communication quality, some security problems inherent to cooperative relay also arise. This paper investigates the security issues in cooperative communications under the context of multiple relay nodes using decode-and-forward strategy, where one of the relay nodes is adversarial and tries to corrupt the communications by sending garbled signals. We show that the conventional physical-layer signal detector will lead to a high error rate in signal detection in such a scenario, and the application-layer cryptography alone will not be able to distinguish the adversarial relay from legitimate ones. To trace and identify the adversarial relay, we propose a cross-layer tracing scheme that uses adaptive signal detection at the physical layer, coupled with pseudorandom tracing symbols at the application layer. Analytical results for tracing statistics as well as experimental simulations are presented to demonstrate the effectiveness of the proposed tracing scheme

Title of Dissertation: SECURING MULTI-LAYER COMMUNICATIONS: A SIGNAL PROCESSING APPROACH Yinian Mao, Doctor of Philosophy, 2006 Dissertation directed by: Professor Min Wu Department of Electrical and Computer Engineering Security is becoming a major concern in this information era. The development in wireless communications, networking technology, personal computing devices, and software engineering has led to numerous emerging applications whose security requirements are beyond the framework of conventional cryptography. The primary motivation of this dissertation research is to develop new approaches to the security problems in secure communication systems, without unduly increasing the complexity and cost of the entire system. Signal processing techniques have been widely applied in communication systems. In this dissertation, we investigate the potential, the mechanism, and the performance of incorporating signal processing techniques into various layers along the chain of secure information processing. For example, for application-layer data confidentiality, we have proposed atomic encryption operations for multimedia data that can preserve standard compliance and are friendly to communications and delegate processing. For multimedia authentication, we have discovered the potential key disclosure problem for popular image hashing schemes, and proposed mitigation solutions. In physical-layer wireless communications, we have discovered the threat of signal garbling attack from compromised relay nodes in the emerging cooperative communication paradigm, and proposed a countermeasure to trace and pinpoint the adversarial relay. For the design and deployment of secure sensor communications, we have proposed two sensor location adjustment algorithms for mobility-assisted sensor deployment that can jointly optimize sensing coverage and secure communication connectivity. Furthermore, for general scenarios of group key management, we have proposed a time-efficient key management scheme that can improve the scalability of contributory key management from O(log n) to O(log(log n)) using scheduling and optimization techniques. This dissertation demonstrates that signal processing techniques, along with optimization, scheduling, and beneficial techniques from other related fields of study, can be successfully integrated into security solutions in practical communication systems. The fusion of different technical disciplines can take place at every layer of a secure communication system to strengthen communication security and improve performance-security tradeoff. SECURING MULTI-LAYER COMMUNICATIONS: A SIGNAL PROCESSING APPROACH

In recent years, there has been an increasing trend for multimedia applications to use delegate service providers for content distribution, archiving, search, and retrieval. These delegate services have brought new challenges to the protection of multimedia content confidentiality. This paper discusses the importance and feasibility of applying a joint signal processing and cryptographic approach to multimedia encryption, in order to address the access control issues unique to multimedia applications. We propose two atomic encryption operations that can preserve standard compliance and are friendly to delegate processing. Quantitative analysis for these operations is presented to demonstrate that a good tradeoff can be made between security and bitrate overhead. In assisting the design and evaluation of media security systems, we also propose a set of multimedia-oriented security scores to quantify the security against approximation attacks and to complement the existing notion of generic data security. Using video as an example, we present a systematic study on how to strategically integrate different atomic operations to build a video encryption system. The resulting system can provide superior performance over both generic encryption and its simple adaptation to video in terms of a joint consideration of security, bitrate overhead, and friendliness to delegate processing.

Cooperative communication system explores a new dimension of diversity in wireless communications to combat unfriendly wireless environment through strategic relays. While this emerging technology is promising in improving communication quality, some security problems inherent to cooperative relay also arise. In this paper we investigate the security issues in cooperative communications that consist of multiple relay nodes using decode-and-forward strategy. In particular, we consider the situation where one of the relay nodes is adversarial and tries to corrupt the communications by sending garbled signals. We show that the conventional physical-layer signal detection will not be effective in such a scenario, and the application-layer cryptography alone is not sufficient to identify the adversarial relay. To combat adversarial relay, we propose a cross-layer scheme that uses pseudo-random tracing symbols, with an adaptive signal detection rule at the physical layer, and direct sequence spread spectrum symbol construction at the application layer for tracing and identifying adversarial relay. Our experimental simulations show that the proposed tracing scheme is effective and efficient

Image hash functions find extensive applications in content authentication, database search, and watermarking. This paper develops a novel algorithm for generating an image hash based on Fourier transform features and controlled randomization. We formulate the robustness of image hashing as a hypothesis testing problem and evaluate the performance under various image processing operations. We show that the proposed hash function is resilient to content-preserving modifications, such as moderate geometric and filtering distortions. We introduce a general framework to study and evaluate the security of image hashing systems. Under this new framework, we model the hash values as random variables and quantify its uncertainty in terms of differential entropy. Using this security framework, we analyze the security of the proposed schemes and several existing representative methods for image hashing. We then examine the security versus robustness tradeoff and show that the proposed hashing methods can provide excellent security and robustness.

In secure group communications, the time cost associated with key updates in the events of member join and departure is an important aspect of quality of service, especially in large groups with highly dynamic membership. To achieve better time efficiency, we propose a join-exit-tree (JET) key management framework. First, a special key tree topology with join and exit subtrees is introduced to handle key updates for dynamic membership. Then, optimization techniques are employed to determine the capacities of join and exit subtrees for achieving the best time efficiency, and algorithms are designed to dynamically update the join and exit trees. We show that, on average, the asymptotic time cost for each member join/departure event is reduced to O(log(logn)) from the previous cost of O(logn), where n is the group size. Our experimental results based on simulated user activities as well as the real MBone data demonstrate that the proposed JET scheme can significantly improve the time efficiency, while maintaining low communication and computation cost, of tree-based contributory key management

Security and robustness are two important requirements for image hash functions. We introduce "differential entropy" as a metric to quantify the amount of randomness in image hash functions and to study their security. We present a mathematical framework and derive expressions for the proposed security metric for various common image hashing schemes. Using the proposed security metric, we discuss the trade-offs between security and robustness in image hashing.

Sensor network has a great potential in applications such as habitat monitoring, wildlife tracking, building surveillance, and military combat. The design of a sensor network system involves several important issues, including the sensing coverage, node-to-node or node-to-base-station communications, and the security in information gathering and relay by the sensors. In this paper, we show that the system performance on these aspects depends closely on how the sensors are deployed in the field, and on how the sensor locations can be adjusted after the initial deployment. For static sensor deployment, we investigate the hexagon and square lattice topology and analyze their impact on secure connectivity and sensing coverage. For advanced sensing devices that allow for location adjustment after deployment, we have established a new framework for coordinated updates of sensor locations. We propose two new sensor location updating algorithms, the VFSec and the Weighted Centroid algorithm, to jointly optimize sensing coverage and secure connectivity. Simulation results show that these new algorithms provide superior tradeoff over the existing approaches that do not take security into considerations.

A powerful class of attacks in multimedia fingerprinting is known as collusion attacks, where a clique of colluders, each having a copy of the same multimedia content with different fingerprint, combine their copies to form a colluded copy. In this paper, we propose a countermeasure against collusion attacks for digital video: pseudo-random intentional de-synchronization techniques. Each user's copy of video is slightly pseudo-randomly changed (de-synchronized) in such a way that these changes will not be noticeable for an individual copy, but will be significant enough to produce perceptual artifacts when multiple copies are combined (e.g., via averaging, replacement attacks, etc.). To achieve this task, we propose several novel effective techniques, including constrained random temporal and spatial sampling. We discuss feasibility issues and limitations of video de-synchronization, and present several examples.

Information hiding has been studied in many security applications such as authentication, copyright management and digital forensics. In this work, we introduce a new application where successful information hiding in compiled program binaries could bring system-wide performance improvements. Our goal is to enhance computer system performance by providing additional information to the processor, without changing the instruction set architecture. We first analyze the statistics of typical programs to demonstrate the feasibility of hiding data in them. We then propose several techniques to hide a large amount of data in the operand fields with very low computation and storage requirements during the extraction process. The data embedding is made reversible to recover the original instructions and to ensure the correct execution of the computer program. Our experiments on the SPEC CPU2000 benchmark programs show that up to 110K bits of information can be embedded in large programs with as little as 3K bits of additional run-time memory in the form of a simple look-up table.

This paper addresses the access control issues unique to multimedia, by using a joint signal processing and cryptographic approach to multimedia encryption. Based on three atomic encryption primitives, we present a systematic study on how to strategically integrate different atomic operations to build a video encryption system. We also propose a set of multimedia-specific security metrics to quantify the security against approximation attacks and to complement the existing notion of generic data security. The resulting system can provide superior performance to both generic encryption and its simple adaptation to video in terms of a joint consideration of security, bitrate overhead, and communication friendliness.

Image hash functions provide compact representations of images. which is useful for search and authentication applications. In this work. we have identified a general three step framework and proposed a new image hashing scheme that achieves a better overall performance than the existing approaches under various kinds of image processing distortions. By exploiting the properties of discrete polar Fourier transform and incorporating cryptographic keys, the proposed image hash is resilient to geometric and filtering operations, and is secure against guessing and forgery attacks.

We propose a time-efficient contributory key agreement framework for secure communications in dynamic groups. The proposed scheme employs a special join-tree/exit-tree topology in the logical key tree and effectively exploits the efficiency of amortized operations. We derive the optimal parameters and design an activation algorithm for the join and exit trees. We also show that the asymptotic average time cost per user join and leave event is /spl theta/(log (log n)), where n is the group size. Our experiment results on both simulated user activities and the real MBone data have shown that the proposed scheme outperforms the existing tree-based schemes.