Research On Commutative Encryption-watermarking Models And Algorithm For Geographical Vector Data

As the important information resource and data used in geography,the security protection for geographical vector data needs to be urgently solved.Digital watermarking technology has provided an important and effective method for copyright protection of geographical vector data,however,it is not able to prevent data leakage in the process of data distribution.On the other hand,the encryption technology for geographical vector data is able to prevent data leakage in distribution but unable to track the usage and protect the copyright after distribution.Thus,cryptography and digital watermarking should work together in security protection for geographical vector data.Unfortunately,the simple and direct combination of watermarking and cryptography would bring shortcomings such as lack of security,requirements on operations sequence,problems on key management and so on.So it is necessary to study the security technology,which combines cryptography and watermarking together without mutual interference,in order to both protect the geographical vector data security in distribution and track the data usage after distribution.Commutative encryption and watermarking(CEW)provides a brand new solution and thorough method for secure distribution and usage tracking on geographical vector data.CEW represents the technology making the sequence of encryption and watermark embedding exchangeable,and the sequence of decryption and watermark extraction exchangeable.It ensures the close combination and mutual independence of cryptography and watermarking together based on operation mechanisms.CEW eliminates the restriction on operation sequence of cryptography and watermarking and overcomes the drawbacks brought by the direct combination of this two technologies.Hence it is able to provide more flexible and comprehensive security protection for geographical vector data.In this paper,the theory,model and algorithm related to CEW for geographical vector data have been studied,with the work and results mainly clarified as follows:(1)Based on the organization structure,features and visual format of geographical vector data,the concept,properties and evaluation indexes of CEW for geographical vector data are proposed.Furthermore,several classical implementation methods of CEW are summarized.Taking the characteristics of geographical vector data into consideration,the mechanism and model of CEW for geographical vector data have been proposed.(2)From the perspective of CEW robustness,the embedding strategy and watermark synchronization have been analyzed to deduce the quantitative robustness analysis principles.For common attacks on geographical vector data,the robustness analysis principles are also proposed after properties and features of data deletion attack,data addition attack,data update attack and geometric attack have been discussed.In addition,the robustness index based on probability and repetition experiments have been proposed,aiming to evaluate robustness more scientifically and effectively.Then the robustness computation model is proposed based on the probability-based robustness index,combinatorial mathematics and inclusion-exclusion principle together with robustness analysis principles.Experiments have been conducted to verify the effectiveness and reliability of the proposed model by comparing the model results and watermarking algorithm results.(3)From the perspective of CEW watermark capacity,the watermark capacity prediction model is established by considering the features of geographical vector.Based on the robustness computation model proposed previously,the relationship between watermark capacity and robustness is verified by experiments.To increase the robustness and watermark capacity simultaneously,the capacity optimization model under constraints has been proposed and solved according to the spatial precision constraint and numerical precision constraint.The optimization model is verified by experiments to demonstrate its capabilities of improving watermark embedding capacity for small-scale vector data and increasing robustness against data deletion,data addition and data update attack.(4)The domains where CEW performed have been studied with the spatial and geometric features of geographical vector data.As the distance ratio and angle generated by adjacent vertexes have been selected as the CEW domain,the security,robustness,capacity and other properties of the CEW domain have been quantitatively analyzed.Based on the CEW domain,the CEW mechanism on congruence relation for geographical vector data is proposed.Then the encryption and decryption algorithms considering for characteristics of geographical vector data are then studied.Besides,the corresponding watermarking algorithm is also proposed and optimization is performed according to the robustness computation model and capacity optimization model.The experiments have been conducted to demonstrate that the proposed CEW algorithm for geographical vector data satisfies the essential properties of CEW,such as the commutativity of cryptography and watermarking,high security and high watermark imperceptibility.Comparing with other watermarking algorithms for geographical vector data,the proposed CEW algorithm also outperforms in watermark capacity and robustness.