No abstract

No abstract

Deep learning, such as convolutional neural networks, has been achieved great success in image processing, computer vision task, and image compression, and has achieved better performance. This paper designs a multiple description coding frameworks based on symmetric convolutional auto-encoder, which can achieve high-quality image reconstruction. First, the image is input into the convolutional auto-encoder, and the extracted features are obtained. Then, the extracted features are encoded by the multiple description coding and split into two descriptions for transmission to the decoder. We can get the side information by the side decoder and the central information by the central decoder. Finally, the side information and the central information are deconvolved by convolutional auto-encoder. The experimental results validate that the proposed scheme outperforms the state-of-the-art methods. INDEX TERMS Convolutional auto-encoder (CAE), multiple description coding (MDC), predictive coding, quality metric.

No abstract

Cognitive radio (CR) is regarded as a powerful technology to solve the problem of spectrum shortage and underutilization. As a key function of CR technology, cooperative spectrum sensing (CSS) allows secondary users (SUs) to detect the primary user (PU)'s signal so that they identify and opportunistically access the available spectrum. However, the openness of CSS paradigm makes cognitive radio networks (CRNs) suffer from Byzantine attack, thereby undermining the premise of CR framework. To this aim, we formulate a probabilistic hard Byzantine attack model, in which malicious users (MUs) can conduct various attack strategies, and make an in‐depth investigation on the blind scenario. On the one hand, in order to ensure the robustness of CSS, a method to evaluate the reliability of the secondary user (SU)'s sensing result based on the channel status detection is proposed and an innovative weight coefficient is considered to selectively utilize the sensing information from MUs. On the other hand, we design a sequential fusion method based on reputation value (RV) and differential mechanism, with the aim of improving the efficiency of CSS. According to above methods and mechanism, the weighted differential sequential symbol (WDS2) algorithm is designed, which integrates the weight evaluation into sequential method to make the global decision for CSS. Finally, compared to the existing various data fusion algorithms, simulation results show that the proposed WDS2 not only defends against various Byzantine attacks to secure the robustness of CSS, but also requires less samples in support of an accurate global decision to improve the efficiency of CSS.