Efficient Transmission of an Encrypted Image Through a MIMO–OFDM System with Different Encryption Schemes

摘要

In today’s world, broadband multimedia services are growing at a rapid pace. It is very much essential to secure data in financial, defense, health care, government, and marketing services. To ensure the data from eavesdroppers, the physical layer security (channel coding, modulation schemes, and diversity techniques, etc.) is one of the best solutions. The security attributes at this level reduce maintenance, operational cost, and complexity without affecting the quality of service. Physical layer security with cryptographic techniques makes it well suited for transferring multimedia data through unsecure or potentially risky communication (wired and wireless) path. To provide security attributes to an image/data, one needs to protect the transmitted data from unauthorized access. Several encryption algorithms have been proposed to ensure confidentiality and security for multimedia data over an insecure wireless channel. Cryptography techniques do not involve any system complexity but add a little computation complexity. Use of advanced processors mitigates the computational burden. In this paper, efficient transmission of an encrypted image through a multiple-input multiple-output orthogonal frequency division multiplexing (MIMO–OFDM) system over an additive white Gaussian noise channel with different encryption schemes is presented. The advanced encryption standard (AES), data encryption standard (DES), and Rubik’s Cube encryption algorithms are incorporated to improve security aspects. The comparison aims to select the most appropriate wireless communication systems (2 × 2 MIMO–OFDM and 4 × 4 MIMO–OFDM), and the most suitable image encryption algorithm for efficient transmission of the encrypted image. The quality of the received image is evaluated using the peak signal-to-noise ratio between the original and decrypted image at different signal-to-noise (SNR) ratio. Performance estimates of the cryptosystems are also provided by computing the bit-error-rate at different SNR. The Performance estimates of encryption algorithms (irrespective of the MIMO–OFDM system) are also included, which consists of various performance metrics. The metrics are histogram, entropy, correlation coefficient, number of changing pixel rate, and unified averaged changed intensity. From the MATLAB simulation experiments, it is observed that the performance of 4 × 4 Crypto-MIMO–OFDM system configurations outperformed 2 × 2 Crypto-MIMO– OFDM system configurations. The performance of the Crypto-MIMO–OFDM system is compared with the original MIMO–OFDM system performance without encryption algorithm. For all variants of Crypto-MIMO–OFDM systems, the use of Rubik’s Cube encryption algorithm shows significant improvement over DES and AES.