Query-preserving watermarking of relational databases and XML documents

Abstract

Watermarking allows robust and unobtrusive insertion of information in a digital document. Very recently, techniques have been proposed for watermarking relational databases or XML documents, where information insertion must preserve a specific measure on data (e.g. mean and variance of numerical attributes.)In this paper we investigate the problem of watermarking databases or XML while preserving a set of parametric queries in a specified language, up to an acceptable distortion.We first observe that unrestricted databases can not be watermarked while preserving trivial parametric queries. We then exhibit query languages and classes of structures that allow guaranteed watermarking capacity, namely 1) local query languages on structures with bounded degree Gaifman graph, and 2) monadic second-order queries on trees or tree-like structures. We relate these results to an important topic in computational learning theory, the VC-dimension. We finally consider incremental aspects of query-preserving watermarking.

On the Properties of Non-Media Digital Watermarking: A Review of State of the Art Techniques

“…The authors have argued the possibility of using the same key for all nodes that deters inter-node collision attacks if the same node content is watermarked with different keys. 5) Gross-Amblard studied the capacity of watermarking protocols that preserve a set of predefined queries [57]. He showed that for a scalable watermarking protocol, marks should be hidden into weights of active tuples of the dataset, that is those tuples that are part of query answers.…”

“…right protection/ privacy preservation of social network data, XMLs, graph coloring/partitioni ng solutions limited bandwidth in order to preserve graph structure; complexity of wm detection due to NPcomplete subgraph matching problem in particular for large graph datasets a constraint-based wm for graph coloring [53] and graph partitioning [54]; a querypreserving wm for graphs generated form map data [55] and XML documents [57]; XML wm by taking both the structure and content into account using a resilient canonical labelling [56]; overlaying a random-looking graph wm on a sub-graph extracted from the social network graph [58]; convolving a heapordered tree with control/data flow graph of a program [59] preserving key properties of graphs such as shortest path, transparency to graph partitioning tools, preserving high level semantics such as number of added colors to the original graph coloring solution often robustness against attacks such as graph surgery attacks or unauthorized detection is of importance Spatial data right protection / authentication of maps, shapes, geospatial data robust features selection is the most difficult part of watermarking schemes for spatial data; for a comprehensive list of challenges forn watermarking of geospatial data, readers are referred to [61] watermarking map data based on modification of B-spline control points of curves (using the IAM algorithm for aligning sample points) [62], or vertex coordinates of map shapes selected by the 'buffer-zone' algorithm [64], or key points of three feature layers [65], or spatial topological relations among polygons identified by Hilbert space-filling curve mapping [66], or formed intervals by interpolated virtual coordinates [69]; an optimum blind wm detector based on modelling Fourier descriptors of a vector map by a Rayleigh distribution [63]; a group-based wm using no visual distortion, preserving polygon shapes, maintaining geographical precision such as common boarders of countries, application-specific transparency zooming, data format change such as vector/rasterraster/vector conversions, etc interpolated vertices and the difference expansion method [68], or by modifying the polyline length and polygon area of feature layers through GMMbased clustering [67]; Spatiotemp oral data right protection/ authentication/dat a obfuscation of trajectories rare data redundancy limits wm capacity; nonsensitivity of trajectory shapes against spatial transformations, surviving common trajectory operations such as segmentation and compression embedding wm into distance ratio of selected coordinates of trajectories [72]; an additive wm with preserving NN structures [70]; a multiplicative with HC preservation …”

Blind and squaring-resistant watermarking of vectorial building layers

“…It is easy to alter least significant bits of points of the map but the angular quality is not taken into account (on the contrary, least significant bits methods may perform well on simple points databases, like point-of-interest data sets in use in GPS viewers). Methods described in [8,25] allow for the description of usability queries to be preserved by watermarking. But they either focus on basic numerical aggregates like SUM queries [8], which are not rich enough to represent angular constraints, or based on a trial and error method to handle generic black-box queries [25].…”

“…Methods described in [8,25] allow for the description of usability queries to be preserved by watermarking. But they either focus on basic numerical aggregates like SUM queries [8], which are not rich enough to represent angular constraints, or based on a trial and error method to handle generic black-box queries [25]. Using the latter method, it might not be possible to reach a valid set of alterations since no search strategy is defined.…”

Robust and Blind Watermarking of Relational Database Systems

“…Attribute selection is based on the value of a hash function. For each selected attribute, some bit positions will be marked amongst a predetermined number of least significant bits of the attribute [1,2,4,7,11,16,28] .…”