Robust privacy preservation and authenticity of the collected data in cognitive radio network—Walsh–Hadamard based steganographic approach

Cognitive Radio Networks have recently attracted attention because of high efficiency and throughput performance. They transmit (1) repetitively collected readings (e.g. monitoring) and (2) highly confidential data (e.g. geometric location). However, the privacy and the authenticity of the transmitted data are major challenges. This paper proposes a novel steganographic technique that guarantees (1) strong end-to-end protection of the confidential information by hiding them randomly inside the normal readings using a generated key, and (2) robust evidence of authenticity for the transported readings. To expand hiding, the Walsh–Hadamard Transformation (WHT) is used to decompose normal readings into a set of coefficients. To achieve minimum distortion, only the least featured coefficients are used. To achieve high security, a key is used to reshape the coefficients into a random 2D MM-by-NN matrix and to generate a randomly selected order used in the hiding process. To accurately measure the distortion after hiding and extracting the confidential data, Percentage Residual Difference (PRD) has been used. It is obvious from experiments that our technique has little effect on the original readings (<1%<1%). Also, our security evaluation proves that unauthorized retrieval of the intended confidential information within a reasonable time is highly improbable.