Abstract: Aiming at the difficulty of tracking traitors and low coding efficiency about fingerprint for vector map data, a fingerprint algorithm using GD-PBIBD code is proposed in this paper. Firstly, the fingerprint codes are generated by using the GD-PBIBD, also the fingerprint is shuffled by using the Logistic map before embedding. Then, the feature points of the vector map data are extracted by using the D-P algorithm, and the phase coefficients and amplitude coefficients are obtained by performing DFT transformation on the feature points. The QIM method is used to embed the fingerprint into the amplitude coefficient of the DFT transform domain, and the DFT inverse transform is implemented to obtain the vector map data containing the fingerprint. The experiment selected some Chinese road network and a certain area data in green space. After the fingerprint was embedded in the algorithm, linear and nonlinear simulation attacks were carried out. The fingerprint sequence was successfully extracted and the collaborators were tracked by Hamming distance. The algorithm has a simple coding structure. When the code length is fixed, it can accommodate more users than the BIBD code, and the efficiency is high. The algorithm is robust against single-user fingerprint attacks and multiple attacks, and can resist multi-user minimum and the maximum and minimum attacks, which track all traitors, can be used for vector map data copyright protection, provide a basis for vector space data traitor tracking.

Key words: vector spatial data, fingerprint encoding, GD-PBIBD, anti-collusion attack, DFT