Persistent Watermarking of Relational Databases Raju Halder, Agostino Cortesi Department of Computer Science Universita` Ca’ Foscari di Venezia, Italy {halder, cortesi}@unive.it

CNC’2010, Calicut, Kerala, India

Raju Halder, Agostino Cortesi (Ca’ Foscari Univ.)

Persistent Watermarking of Relational Databases

CNC’2010

1 / 19

Outline 1

Introduction to Database Watermarking

2

Notion of Persistency in Watermarking

3

Invariants of database states

4

Way to improve the existing schemes in terms of persistency

5

Proposed persistent watermarking technique Partitioning Watermarking

6

Time and Space Complexity

7

Discussions

Raju Halder, Agostino Cortesi (Ca’ Foscari Univ.)

Persistent Watermarking of Relational Databases

CNC’2010

2 / 19

Introduction Information systems are shared by many people around the world. - Example: “Database as a Service model” Content of the databases faces serious challenges: - Illegal redistribution, ownership claims, forgery, theft, etc. Encryption/Decryption? so restrictive. Alternative way is Digital watermarking. Digital watermarking embeds some kind of information into the underlying data of the databases. Purpose: ownership proof, traitor tracing, tamper detection, localization, etc.

Raju Halder, Agostino Cortesi (Ca’ Foscari Univ.)

Persistent Watermarking of Relational Databases

CNC’2010

3 / 19

Introduction Key (K) Original Database

Watermark Embedding

Watermarked Database

Watermark W Figure: Watermarking Embedding Phase

Key (K) Suspicious Database

Watermark Verification

Watermark W

Compare

Claim as true or false

Watermark W’ Figure: Watermark Verification Phase

Raju Halder, Agostino Cortesi (Ca’ Foscari Univ.)

Persistent Watermarking of Relational Databases

CNC’2010

4 / 19

Introduction

Two types of Watermarking: Distortion-based: introduce distortion to the underlying data Distortion-free: introduce no distortion to the underlying data

Performed at bit/character/attribute/tuple-level over attribute values of the types numeric/string/categorical/any

Raju Halder, Agostino Cortesi (Ca’ Foscari Univ.)

Persistent Watermarking of Relational Databases

CNC’2010

5 / 19

Persistent Watermarking

Existing watermarking schemes rely on the content of the database. Benign Updates or any intensional processing may change database content, and may damage/distort the existing watermark. Notion of persistency comes into the context.

Raju Halder, Agostino Cortesi (Ca’ Foscari Univ.)

Persistent Watermarking of Relational Databases

CNC’2010

6 / 19

Persistent Watermarking Let hdB, Qi be a database model, where Q is the set of queries bounded with the database dB. Suppose the initial state of the dB is d1 . Set of valid states d2 , . . . , dn of DB obtained by executing queries of Q.

Definition (Persistent Watermark) A watermark W embedded in the state d1 is called persistent w.r.t. Q if ∀ i ∈ [2 . . . n] : verify(d1 , W ) = verify(di , W ) where ”verify(d, W )” is a boolean function such that probability of ”verify(d, W ) = true” is negligible when W , watermark embedded in d.

Raju Halder, Agostino Cortesi (Ca’ Foscari Univ.)

Persistent Watermarking of Relational Databases

CNC’2010

7 / 19

Invariants of database states (1) Stable Cells: Data cells in database states that are not affected by the associated queries. (2) Semantics-based Properties of the data: Three types: Intra-cell (IC) property: - Represents specific properties of individual cells. - Ex: Integer attribute values c represented by Interval [a, b], a ≤ c ≤ b. Intra-tuple (IT) property: - Property based on the inter-relation between two or more attribute values. - Ex: Inter-relation between basic and gross salary: represented by relational abstract domain. Intra-attribute among-tuples (IA) property: - Property based on relationship among the set of tuples. - Ex: # male employee = # female employee + 1. - Represented by relational or non-relational abstract domain. Raju Halder, Agostino Cortesi (Ca’ Foscari Univ.)

Persistent Watermarking of Relational Databases

CNC’2010

8 / 19

improvement in terms of persistency 1

Let Q be the set of queries bounded with database state d.

2

Identify the set of stable cells STBdQ .

3

Extract the set of semantics-based properties PRdQ .

4

p s and gencode Encode STBdQ and PRQd by suitable functions fencode respectively. Ex. - Chinese Remainder Theorem for intervals in IC property, or p Minimal perfect hash function as gencode .

5

Exploit these invariant encoded values in the existing schemes to obtain a persistent watermarking.

Raju Halder, Agostino Cortesi (Ca’ Foscari Univ.)

Persistent Watermarking of Relational Databases

CNC’2010

9 / 19

Outline 1

Introduction to Database Watermarking

2

Notion of Persistency in Watermarking

3

Invariants of database states

4

Way to improve the existing schemes in terms of persistency

5

Proposed persistent watermarking technique Partitioning Watermarking

6

Time and Space Complexity

7

Discussions

Raju Halder, Agostino Cortesi (Ca’ Foscari Univ.)

Persistent Watermarking of Relational Databases

CNC’2010

10 / 19

Partitioning Generates m non-overlapping partitions {S1 , . . . , Sn } of tuples. Si ∩ Sj = ∅ where i , j. K= Secret Key. PK= Primary Key. {A1s , A2s . . . Aqs }= Attributes in stable part. Partitioning Steps: Step 1: ν = (hash(K ◦ PK ◦ A1s ◦ A2s ◦ · · · ◦ Aqs ) mod m) + 1. Step 2: Assign the tuple into the partition Sν .

Raju Halder, Agostino Cortesi (Ca’ Foscari Univ.)

Persistent Watermarking of Relational Databases

CNC’2010

11 / 19

Outline 1

Introduction to Database Watermarking

2

Notion of Persistency in Watermarking

3

Invariants of database states

4

Way to improve the existing schemes in terms of persistency

5

Proposed persistent watermarking technique Partitioning Watermarking

6

Time and Space Complexity

7

Discussions

Raju Halder, Agostino Cortesi (Ca’ Foscari Univ.)

Persistent Watermarking of Relational Databases

CNC’2010

12 / 19

Watermarking Algorithm Step 1: Determine Intra-attribute among-tuples property (IAdQ ), and obtain its encode value (say, k3 ). Step 2: For each partition Si do Step 3. Step 3: FOR each tuple r in each partition Si do following: 1

Determine Intra-cell (IAdQ ) and Intra-tuple (IAdQ ) property of r, and obtain their encoded value (say, k1 and k2 resp.).

2

Generate a binary tuple b(PK , b1 , . . . , bn , p1 , p2 , p3 ): b.PK = r.PK .

3

b.p1 =k1 , b.p2 =k2 , b.p3 =k3 .

4

Randomize the attribute values of r based on primary key PK , secret key K , and the stable attribute values.

5

For each j th attribute in randomized r do following: if(r.aj = stable) b.aj = (MSBs of r.aj ) ⊗ (i th signature bit). else b.aj = 0.

Raju Halder, Agostino Cortesi (Ca’ Foscari Univ.)

Persistent Watermarking of Relational Databases

CNC’2010

13 / 19

Watermarking Algorithm Q = increase the basic and gross salary of the employees by at most 30% PK

Unstable part

Stable part th



Tuple belongs to 5 Partition

< E002; Alice; 900; 1685; 29; 1 > Q

Randomize based on hash value HASH(stable part, K, PK) < E002; 1685; 1; 29; 900; Alice > Extract MSB < E002; 0; 0; 0; 0; 0 >

Intra-cell Property (ICd ) : Basic Sal: [900, 1170] Gross Sal: [1685, 2191]

Encoded by k1 Q

Intra-tuple Property (ITd ) :

Gross Sal ≥

165 × BasicSal +200 100

Encoded by k1

th

XOR 5 signature bit =1 < E002; 0; 1; 1; 0; 1 >

Append sementic-based Properties

Intra-attribute among-tuples Q Property (IAd ) : “The number of employees in every department is more than 2” represented by [3;+1]

Encoded by k3

< E002; 0; 1; 1; 0; 1; k1; k2; k3 >

Raju Halder, Agostino Cortesi (Ca’ Foscari Univ.)

Persistent Watermarking of Relational Databases

CNC’2010

14 / 19

Time and Space Complexity

Lemma Let η= total number of tuples in a database state d and µ = complexity of the abstraction operation applied on each tuple. The time and space complexity of the watermark generation algorithm is O(η × µ) and O(η) respectively.

Raju Halder, Agostino Cortesi (Ca’ Foscari Univ.)

Persistent Watermarking of Relational Databases

CNC’2010

15 / 19

Discussions and Conclusions

We strictly improve the scheme of Li and Deng [1]. The watermark detection algorithm of Li and Deng [1] is probabilistic. In our case, it is deterministic. Let λ = available MSBs of the attribute A . We can choose j th MSB using: j = hash(PK % A) mod λ. We can also use an watermark generation parameter that controls the number of attributes in the watermark. The proposed scheme is not publicly verifiable, since it depends on a secret key.

Raju Halder, Agostino Cortesi (Ca’ Foscari Univ.)

Persistent Watermarking of Relational Databases

CNC’2010

16 / 19

References

Yingjiu Li and Robert Huijie Deng, Publicly verifiable ownership protection for relational databases. In Proceedings of the 2006 ACM Symposium on Information, computer and communications security (ASIACCS 06), Taipei, Taiwan: ACM Press, 2006, pp. 78-89. Rakesh Agrawal, Peter J. Haas, and Jerry Kiernan, Watermarking relational data: framework, algorithms and analysis. The VLDB Journal, vol. 12, no. 2, pp. 157-169, 2003.

Raju Halder, Agostino Cortesi (Ca’ Foscari Univ.)

Persistent Watermarking of Relational Databases

CNC’2010

17 / 19

Thank you for your attention !

Raju Halder, Agostino Cortesi (Ca’ Foscari Univ.)

Persistent Watermarking of Relational Databases

CNC’2010

18 / 19

Suggestions Please !!!!

Raju Halder, Agostino Cortesi (Ca’ Foscari Univ.)

Persistent Watermarking of Relational Databases

CNC’2010

19 / 19

Persistent Watermarking of Relational Databases

A watermark W embedded in the state d1 is called persistent w.r.t. Q if. ∀ i ∈ [2 ... n] ... watermark embedded in d. .... b.aj = (MSBs of r.aj ) ⊗ (ith signature bit).

185KB Sizes 1 Downloads 321 Views

Recommend Documents

A Persistent Public Watermarking of Relational ...
Compare. Original Signature S ... Watermark W is embedded in initial state d0. Definition ... Watermark W embedded in the state d0 is called persistent w.r.t. Q if.

A Persistent Public Watermarking of Relational ... - Springer Link
Halder, R., Dasgupta, P., Naskar, S., Sarma, S.S.: An internet-based ip protection ... Multimedia and Security Workshop (ACM Multimedia 1999), Orlando, pp.

Best BOOKDownload Information Modeling and Relational Databases ...
Best BOOKDownload Information Modeling and. Relational Databases: From Conceptual Analysis to. Logical Design (The Morgan Kaufmann Series in Data.

Multilevel Security for Relational Databases - IT Today
CHAPTER 2 BASIC CONCEPT OF MULTILEVEL DATABASE. SECURITY. 17 ...... every year. 2.5.2 Impact of ... of the teaching staff of the Department of Computer Science and. Engineering at ... an M.Sc. degree in communication systems.

Exploiting the Power of Relational Databases for ...
Mar 26, 2009 - last years that lead to the development of specialized stream ... tions, financial, web applications, etc. .... functionality of advanced DBMSs.

Download Sas/Access Software for Relational Databases
Download Sas/Access Software for Relational. Databases: Reference, Version 6 (DB2/2 Chapter. Full Books. Books detail. Title : Download Sas/Access ...

Persistent Systems - Rakesh Jhunjhunwala
Dec 22, 2014 - Free float (%). 61.1 ... Free cash flow (a+b) ... IDFC SEC is a full-service, integrated investment banking, and institutional broking group.

Comparative Study of Reversible Image Watermarking: Fragile ...
Status and Key Issues", International Journal of Network Security, Vol.2, No.3, PP.161–171, May 2006. [9] Ingemar J. Cox, Matthew L. Miller, Jeffrey A. Bloom, ...

Comparative Study of Reversible Image Watermarking: Fragile ...
1 PG Student, Department of Electronics and Communication, Gujarat ... this paper is to define the purpose of reversible watermarking, reflecting recent progress ...

Non-blind watermarking of network flows - arXiv
Mar 10, 2012 - scalability constraints. We confirm our analysis by simulating the detec- tors and testing them against large traces of real network flows. Index Terms—Traffic analysis, flow watermarking, non-blind water- marking, hypothesis testing

Treatment of Mild Persistent Asthma
May 17, 2007 - In childhood, the pa- tient visited her local hospital for treatment of acute asthma ... µg per puff) twice a day, and gave her an albuterol inhaler to use as an as-needed rescue treatment. With this .... also increased the number of

Rejection of persistent bounded disturbances - Semantic Scholar
compensation does not improve the optimal rejection of finite-energy (i.e., £ 2) disturbances. .... solutions which gives the same value at k = O. [] .... One alternative is to consider the worst case performance over a given class of disturbances.

Persistent Blessings of Luck
Kaplan, and Stucke (2014) confirm the phenomenon with more recent data. Korteweg and ... Rampell, aptly puts it, “You only score home runs if you swing HARD at pitches. .... ity could come from luck as well as innate differences.4 ...... In period

Self-Sizing of Clustered Databases
(1) Institut National Polytechnique de Grenoble, France. (2) Université Joseph Fourier, Grenoble, France. (3) Institut National Polytechnique de Toulouse, France.

Persistent Systems - Rakesh Jhunjhunwala
Dec 22, 2014 - leave limited upside in the near term even as we like the business model. INSTITUTIONAL SECURITIES. INDIA RESEARCH. EVENT UPDATE.

Self-Sizing of Clustered Databases
We used Jade to implement self-sizing in a cluster of replicated databases. Here, self-sizing consists in dynamically increasing or decreasing the number of database replica in order to accommodate load peaks. The remainder of the paper is organized

Natural Language Watermarking
Watermark Testing. Watermark Selecting. ○ Stylistic concerns. ○ Security concerns. Watermark Embedding. 13:38. The 1st Workshop on Info. Hiding. 16 ...