Braided convolutional codes (BCCs) form a class of iteratively decodable convolutional codes that are constructed from component convolutional codes. In braided code division multiple access (BCDMA), these very efficient error correcting codes are combined with a multiple access method and inherent interleaving for channel diversity exploitation into one single scheme. In this paper, we describ...

Lung nodule detection is a class imbalanced problem because nodules are found with much lower frequency than nonnodules. In the class imbalanced problem, conventional classifiers tend to be overwhelmed by the majority class and ignore the minority class. We therefore propose cascaded convolutional neural networks to cope with the class imbalanced problem. In the proposed approach, cascaded conv...

In this letter, we propose a new UHF RFID channel coding method, which improves the reliability of the system by using the excellent error correcting performance of the convolutional code. We introduce the coding principle of convolutional code, and compare with the cyclic codes used in the past. Finally, we analyze the error correcting performance of convolutional codes. The analysis results s...

This paper proposes a new convolutional neural architecture based on treestructures, called the tree-based convolutional neural network (TBCNN). Two variants take advantage of constituency trees and dependency trees, respectively, to model sentences. Compared with traditional “flat” convolutional neural networks (CNNs), TBCNNs explore explicitly sentences’ structural information; compared with ...

We propose a retrieval-augmented convolutional network and propose to train it with local mixup, a novel variant of the recently proposed mixup algorithm. The proposed hybrid architecture combining a convolutional network and an off-theshelf retrieval engine was designed to mitigate the adverse effect of off-manifold adversarial examples, while the proposed local mixup addresses on-manifold one...

Convolutional codes are preferred types of error control codes which can achieve low BERs at signal to noise ratio (SNR) very close to Shannon limit. Here, a new method of convolutional encoding is proposed using the general Booth algorithm for multiplication. This algorithm follows a fast multiplication process and achieves a significantly less computational complexity over its conventional co...

A convolutional code has memory over a short block length. This memory results in encoded output symbols that depend not only on the present input, but also on past inputs. An (n,k,m) convolutional code is implemented using k-input, n-output linear sequential system with a shift-register having m stages. In practice k and n are small and m is large to achieve low error probabilities. In the par...

This paper applies a deep convolutional/highway MLP framework to classify genomic sequences on the transcription factor binding site task. To make the model understandable, we propose an optimization driven strategy to extract “motifs”, or symbolic patterns which visualize the positive class learned by the network. We show that our system, Deep Motif (DeMo), extracts motifs that are similar to,...