Two-dimensional (2-D) turbo product code (TPC) has been investigated for almost ten years and commonly applied in recent years. In this paper, we propose improved schemes about three-dimensional (3-D) TPC decoder. Thr...
详细信息
ISBN:
(纸本)9781424410910
Two-dimensional (2-D) turbo product code (TPC) has been investigated for almost ten years and commonly applied in recent years. In this paper, we propose improved schemes about three-dimensional (3-D) TPC decoder. Three different types of decoding methods are proposed, which do not simply decode by dividing the 3-D TPC into 2-D TPC, but exchange all the extrinsic information among three directions in order to achieve better performance. We bring up a parallel optimal decoder, which efficiently reduce the 3-D TPC decoding complexity. Simulation results show that the performance of the improved parallel 3-D TPC decoder is substantially improved. Various rates can be achieved by using different component codes or by applying shortening and puncturing techniques to meet diversified applications in modern communication systems. We also give the laws for code rate regulation and an example of 3-D TPC frame structure.
Super-resolution reconstruction is an essential task of seismic inversion due to the low resolution and strong noise of field data. Popular deep networks derived from U-Net lack the ability to recover detailed edge fe...
详细信息
Super-resolution reconstruction is an essential task of seismic inversion due to the low resolution and strong noise of field data. Popular deep networks derived from U-Net lack the ability to recover detailed edge features and weak signals. In this article, we propose a dual decoder U-Net (D(2)UNet) to explore both the detail and edge information of the data. The encoder inputs the low-resolution image and the edge image obtained through the Canny algorithm. Edge images can provide rich shape and boundary information, which is helpful to generate more accurate and high-quality data. The dual decoder consists of a main decoder for high-resolution recovery and an edge decoder for edge contour detection. These two decoders interact with a texture-warping module (TWM) with deformable convolution. TWM aims to distort realistic edge details to match the fidelity of low-resolution inputs, especially the location of edges and weak signals. The loss function is a combination of L-1 loss and multiscale structural similarity loss (MS-SSIM) to ensure perception quality. Results on synthetic and field seismic images show that D(2)UNet not only improves the resolution of noisy seismic images, but also maintains the image fidelity.
暂无评论