Safwan El Assad

A new chaos-based image encryption system

During the last decade, a variety of chaos-based cryptosystems has been introduced to protect the content of the transmitted images. In this paper, we propose a new fast, simple, and robust chaos-based cryptosystem structure and we analyze its performances. The cryptosystem uses a diffusion layer followed by a bit-permutation layer, instead of byte-permutation, to shuffle the positions of the image pixels. Moreover, the permutation layer is achieved by a new proposed formulation of the 2D cat map that allows an efficient implementation, measured by the time complexity, in terms of arithmetic and logic operations, and also, in terms of clock cycles, of the key-dependent permutation process in comparison with the standard one. Hence, it provides a very fast diffusion process to spread the influence of a single bit over the others. The new cryptosystem includes a robust and uniform chaotic pseudo-random generator (a very simplified version of a generator published in our patent) to change the control parameters in each round of the encryption/decryption processes. The generator is highly nonlinear and produces robust sequences of discrete values having very long orbits. The proposed cryptosystem is defined on finite numbers, and its speed is faster than many chaos-based cryptosystems, while having a very high security level. The security analysis and the obtained simulation results show that the proposed cryptosystem is resistant to various types of attacks and it is efficient for hardware and software implementation. HighlightsThe proposed chaos-based cryptosystem is a new and efficient structureIt is based on a binary diffusion layer, followed by a bit-permutation layerThe permutation is achieved by an efficient proposed formulation of the 2D cat map.It has a shorter encryption, while having a very high level of security.It is adequate for a software and hardware implementations (FPGA card or an ASIC

Improvement of the Security of ZigBee by a New Chaotic Algorithm

The security protocols used in ZigBee rely on an advanced encryption standard-counter mode (AES-CTR) algorithm to encrypt data before transmission. This algorithm is very robust, but it is time consuming. For some industrial and medical applications, it does not meet the real-time requirement. When the AES is used in counter mode CTR, it becomes like a stream cipher that aims to generate pseudorandom bits. Also, to encrypt data, the latter are combined with the plaintext using the XOR operation. New fast stream ciphers were proposed for the eStream project, but these ciphers have shown some weakness. On the other hand, ciphers based on chaotic functions seem to be more promising. Detailed analyses have shown that chaotic functions have very good cryptographic properties and can be used to construct high speed and strong stream ciphers. In this paper, a new robust and fast chaotic encryption algorithm RFCA is presented. This consists of a chaotic cipher composed of two perturbed maps piecewise linear chaotic map. This algorithm is, in particular, adequate for data encryption in ZigBee networks where robustness and real time are both essential. A comparison between our algorithm (RFCA) and the AES-CTR, the simplified AES, and the eStream finalist candidates, is presented with regard to speed and robustness. This is done using correlation coefficients, unified average changing intensity, number of pixels change rate, and test of randomness for the generated bit sequences using the National Institute of Standards and Technology statistical test suite.

Fast and Secure Chaos-Based Cryptosystem for Images

Non-linear dynamic cryptosystems or chaos-based cryptosystems have been attracting a large amount of research since 1990. The critical aspect of cryptography is to face the growth of communication and to achieve the design of fast and secure cryptosystems. In this paper, we introduce three versions of a chaos-based cryptosystem based on a similar structure of the Zhang and Fridrich cryptosystems. Each version is composed of two layers: a confusion layer and a diffusion layer. The confusion layer is achieved by using a modified 2-D cat map to overcome the fixedpoint problem and some other weaknesses, and also to increase the dynamic key space. The 32-bit logistic map is used as a diffusion layer for the first version, which is more robust than using it in 8-bit. In the other versions, the logistic map is replaced by a modified Finite Skew Tent Map (FSTM) for three reasons: to increase the nonlinearity properties of the diffusion layer, to overcome the fixed-point problem, and to increase the dynamic key space. Finally, all versions of the proposed cryptosystem are more resistant against known attacks and faster than Zhang cryptosystems. Moreover, the dynamic key space is much larger than the one used in Zhang cryptosystems. Performance and security analysis prove that the proposed cryptosystems are suitable for securing real-time applications.

Design of a fast and robust chaos-based crypto-system for image encryption

In this paper, we propose a fast and robust scheme for image encryption including a new chaotic generator. The proposed crypto-system uses a variable block cipher length (8, 128, and 512) with different modes: cipher block chaining (CBC), output feedback (OFB), cipher feedback (CFB) and counter (CTR). The fastness and robustness properties of our crypto-system are determined by the chaotic generator using a 32-bits finite precision with integer representation to facilitate hardware implementation, and by a keystream permutation using a 2D-cat map with chaotic control parameters. The chaotic generator is constructed with two non linear (the skew tent map) IIR filters. To quantify the security level of the proposed cryptosystem, we analyze the global dynamical properties of the chaotic generator using the NIST (National Institute of Standards and Technology) test, and we show that, the algorithm can resist the statistical and differential attacks; it also passed the key sensitivity test. Moreover, the algorithm has a large key space. The experimental results indicate that the scheme is secure, efficient, and faster than conventional advanced encryption standard (AES).

Secure TCP/IP communications over DVB-S/DVB-RCS using chaotic sequences

The DVB-S (Digital Video Broadcasting - Satellite) was originally developed as a TV broadcasting protocol. Later on two encapsulation protocols were developed in order to allow network level TCP/IP encapsulation: Multi Protocol Encapsulation — MPE and Unidirectional Lightweight Encapsulation — ULE. This provided a high quality internet access via satellite, which can be multiplexed even with radio and TV transmissions. In this paper, we propose a security system for the ULE based on chaotic sequences for the key management and data encryption. The key management system is based on a multilayer key model. The proposed system provides security services responding to the security requirements of IP over satellite DVB. To quantify the efficiency of the proposed system, we have done a comparative study (theoretically and by simulation) of the data overhead for different transport protocols and different sets of security parameters. Obtained results confirm the efficiency of the proposed system.

Chaotic Digital Encoding for 2D Trellis-Coded Modulation

In this paper, nonlinear digital filters with finite precision are analyzed as recursive systematic convolutional (RSC) encoders. An infinite impulse response (IIR) digital filter with finite precision (wordlength of N bits) is a rate 1 RSC encoder over a Galois field GF(2N). The Frey chaotic filter is analyzed for two different wordlengths N, and it is demonstrated that this encoder can be used for trellis-coded modulation (TCM) schemes. A definition for the encoding rate is provided in the context of the new structure. The Frey encoder scheme is modified in order to reduce the encoding rate from 1 to 1/2. In fact, this modification consists in increasing the number of encoder outputs, using the same wordlength as for the input. The resulted encoders are used for two-dimensional (2D) TCM schemes. Also, the signal sets are partitioned following the Ungerboeck’s rules. The symbol error rate (SER) is estimated for all proposed structures and the results show the expected coding gains as compared to their equivalent non-encoded and linear versions.

Optimum PAM-TCM schemes using left-circulate function over GF(2 N )

In this paper, optimum recursive systematic convolutional (RSC) encoders over Galois field GF(2N) are designed using a nonlinear function, i.e., left-circulate function (LCIRC). The LCIRC function performs a bit left circulation over the representation word; it is used in microprocessors as an accumulator operation, and in chaotic sequence generators working in finite precision. Different encoding rates are obtained for these encoders when using different representation wordlengths at the input and the output, denoted as Nin and N, respectively. A generalized 1-delay GF(2N) RSC encoder scheme using LCIRC is proposed for performance analysis and optimization, for any possible encoding rate, Nin/N. The minimum Euclidian distance is estimated for these optimum encoders and a general expression is found as a function of the wordlengths Nin and N. The symbol error rate (SER) is estimated by simulation for a quaternary pulse amplitude modulation - trellis-coded modulation (PAM-TCM) transmission over an additive white Gaussian noise (AWGN) channel. The simulation results confirm the expected coding gains determined theoretically.

ROI encryption for the HEVC coded video contents.

In this paper we investigate privacy protection for the HEVC standard based on the tile concept. Tiles in HEVC enable the video to be split into independent rectangular regions. Two solutions are proposed to encrypt the tiles containing the Region Of Interest (ROI). The first solution performs encryption at the bitstream level by encrypting all HEVC syntax elements within the ROI tiles. The second solution enables a selective encryption of the ROI tiles under constant bitrate and format compliant requirements. To avoid temporal propagation of the encryption outside the ROI boundaries caused by inter prediction, the motion vectors of non ROI regions are restricted inside the non encrypted tiles in the reference frames. Simulation results show that the proposed solutions perform secure and adaptive encryption of ROI in the HEVC video. Moreover, the bitrate overhead caused by the MVs restriction window varies between 1%-2.5% depending on both the video content and the number of tiles within the frame.

A new ML detector for trellis-coded spatial modulation using hard and soft estimates

In this paper we propose a new detector for the spatial modulation (SM) receiver. In trellis coded spatial modulation (TCSM) schemes, the joint detection is used for identifying both the antenna index and the transmitted symbol. We define a hybrid maximum-likelihood (ML) SM detector, which determines the transmit antenna index soft estimate and the transmitted symbol hard estimate. The antenna indexes soft estimates are decoded using the logarithmic maximum a posteriori probability (log-MAP) algorithm. It is shown that for at least four receive antennas, the hybrid ML-SM detector offers a coding gain of at least 2 dB over the hard-output solutions in spatially correlated (SC) channels. The proposed detector is less complex than soft-output one, with a negligible bit error rate (BER) performance decrease. The BER is estimated by simulation for QPSK-TCSM transmissions over stationary Rayleigh and SC fading channels with additive white Gaussian noise (AWGN).

Efficient multicore implementation of an advanced generator of discrete chaotic sequences

This paper details the design and implementation performances of an efficient generator of chaotic discrete integer valued sequences. The generator exhibits orbits having very large lengths compared to those given in the literature. It is implemented in C language and parallelized using the Parameterized and Interfaced Synchronous Dataflow Model of Computation (PiSDF MoC). The proposed structure is shown to be scalable, parallel and time efficient. The resulting implementation combines a very long minimal chaotic sequence omin > 7*2128 32-bit samples and a very high throughput of 173Mbps on 4 cores of a General Purpose Processor.