Abstract:Device-to-Device (D2D) communication has now become one of the most promising technologies in wireless communications of the Internet of Things (IoT). In view of the requirements on data security, response speed, signal decryption quality and storage cost of mobile devices in D2D networks, we propose chaotic deep network (CDN) to achieve secure transmission which is swifter, higher quality and lower cost. The proposed scheme consists of a prepositive nonrepetitive training procedure, a well-designed parallel encryption process, and a set of pretrained chaotic deep neural decryption networks. Benefiting from the utilization of deep learning methods, CDN achieves the swift and accurate decryption at a much lower sampling rate, which brings huge dimension reduction of both measurement matrices and ciphertext signals. Also, chaotic initial values and parameters are applied to matrix generation and network training, leading to great time reduction, storage saving and security improvement. In addition, CDN incorporates the frameworks of semitensor product (STP) and block-based image processing (BIP), which not only breaks through the dimension matching limitation of matrix multiplication by using STP but also maintains the parallelizable block-cipher mode of BIP. Proved by experiments, CDN at the sampling ratio of 25% achieves almost the same or even better peak signal to noise ratio compared to other 8 most frequently used methods at that of 50% for the same test images, and obtains better visual effects. When using a large-sized image of 2(10) x2(10) pixels for the experiments, CDN is dozens and even hundreds of times faster than those methods, while reduces the size of measurement matrix from a 500-kB level to a 3-kB level, and the size of compressed data to be transmitted can be reduced by more than 50% as well. Besides, the total key space is approximately 1093. The adjacent pixel correlation is less than 0.01.

Unveil the Time Delay Signature in Delayed Chaotic Communication System Via CNN.

Time delay estimation from the time series for optical chaos systems using deep learning

Unveil the Time Delay Signature of Optical Chaos Systems with a Convolutional Neural Network

An enhanced electro-optic chaos secure communication system immune to time delay signature extraction

Machine Learning Approach to Unveil Time Delay Signature of Electro-optical Chaotic System with Strong Nonlinearity

QTFN: A General End-to-End Time-Frequency Network to Reveal the Time-Varying Signatures of the Time Series

A Time-Delay Signature Elimination and Broadband Electro-Optic Chaotic System with Enhanced Nonlinearity by Deep Learning

Chain-structure time-delay reservoir computing for synchronizing chaotic signal and an application to secure communication

Time-delay signature concealing electro-optic chaotic system with multiply feedback nonlinear loops

Chaotic Deep Network for Mobile D2D Communication

Security-enhanced Chaos Communication with Time-Delay Signature Suppression and Phase Encryption

A Dynamic Threshold Value Control Method for Chaotic Neural Networks

A Security-Enhanced Image Communication Scheme Using Cellular Neural Network

SE-NDEND: A Novel Symmetric Watermarking Framework with Neural Network-Based Chaotic Encryption for Internet of Medical Things

Enhancing Security of Electro-optical Delay Chaotic System Based on Optoelectronic Cascade

Security-Enhanced Electro-Optic Mutual Injection Secure Communication Scheme with Time-Delay Signature Suppressing

A Novel Chaotic System with Suppressed Time-Delay Signature Based on Multiple Electro-Optic Nonlinear Loops

Time Delay Signature Extraction of Optical-Feedback-Induced Chaos with Reservoir Computing

Cryptography Based on Delayed Chaotic Neural Networks

A Research On Dsss Based Cnn

A Novel Symmetric Cryptography Based on Chaotic Signal Generator and a Clipped Neural Network