Research in steganalysis is motivated by the concern that communications associated with illicit activity could be hidden in seemingly innocent electronic transactions. By developing defensive tools before steganographic communication grows, computer security professionals may be better prepared for the threat. This research investigated an artificial immune system approach to novel steganography detection for digital images. The results show promising detection and false positive rates.

We investigate central issues such as invertibility, stability, synchronization, and frequency characteristics for nonlinear wavelet transforms built using the lifting framework. The nonlinearity comes from adaptively choosing between a class of linear predictors within the lifting framework. We also describe how earlier families of nonlinear filter banks can be extended through the use of prediction functions operating on a causal neighborhood of pixels. Preliminary compression results for model and real-world images demonstrate the promise of our techniques.

A recently proposed wavelet domain communication system (WDCS) using transform domain processing demonstrates enhanced interference avoidance capability under adverse environmental conditions. This work extends previous results by incorporating a wavelet packet based decomposition technique that permits demonstration of an M-ary orthogonal signaling capability and provides increased adaptability over a larger class of interference signals. The newly proposed WDCS and its response to various interference scenarios are modeled and simulation results obtained using MATLAB/sup /spl reg//. Bit error rate is the key metric for analysis and performance comparisons. Relative to the non-packet based system, the packet based WDCS provides improved/comparable bit error performance in several interference scenarios - single-tone, multiple-tone, swept-tone, and partial-band interference. The system was evaluated using an E/sub b//N/sub 0/ of 4.0 dB and interference energy-to-signal energy (I/E) ratios ranging from 0.0 dB to 16.0 dB. For binary, 4-ary, and 8-ary CSK data modulations, the packet based WDCS exhibited average interference suppression capabilities of 6.7, 9.2, and 12.0 dB, respectively.

This paper develops two new adaptive wavelet transforms based on the lifting scheme. The lifting construction exploits a spatial-domain, prediction-error interpretation of the wavelet transform and provides a powerful framework for designing customized transforms. We use the lifting construction to adaptively tune a wavelet transform to a desired signal by optimizing data-based prediction error criteria. The performances of the new transforms are compared to existing wavelet transforms, and applications to signal denoising are investigated.