In this paper we provide a state space approach for constructing convolutional codes of rate 1/n and complexity delta, whose free distance is n(delta + 1), the maximal possible free distance.
ISBN:
(纸本)0780344081
In this paper we provide a state space approach for constructing convolutional codes of rate 1/n and complexity delta, whose free distance is n(delta + 1), the maximal possible free distance.
作者：
Jiang, Qing-YuanLi, Wu-JunNanjing Univ
Dept Comp Sci & Technol Collaborat Innovat Ctr Novel Software Technol & I Natl Key Lab Novel Software Technol Nanjing Jiangsu Peoples R China
Due to its low storage cost and fast query speed, crossmodal hashing (CMH) has been widely used for similarity search in multimedia retrieval applications. However, most existing CMH methods are based on hand-crafted ...
详细信息
ISBN:
(纸本)9781538604571
Due to its low storage cost and fast query speed, crossmodal hashing (CMH) has been widely used for similarity search in multimedia retrieval applications. However, most existing CMH methods are based on hand-crafted features which might not be optimally compatible with the hash-code learning procedure. As a result, existing CMH methods with hand-crafted features may not achieve satisfactory performance. In this paper, we propose a novel CMH method, called deep cross-modal hashing (DCMH), by integrating feature learning and hash-code learning into the same framework. DCMH is an end-to-end learning framework with deep neural networks, one for each modality, to perform feature learning from scratch. Experiments on three real datasets with image-text modalities show that DCMH can outperform other baselines to achieve the state-of-the-art performance in cross-modal retrieval applications.
In this paper, we prove a MacWilliams identity for the weight adjacency matrices based on the constraint codes of a convolutional code (CC) and its dual. Our result improves upon a recent result by Gluesing-Luerssen a...
详细信息
ISBN:
(纸本)9781479959990
In this paper, we prove a MacWilliams identity for the weight adjacency matrices based on the constraint codes of a convolutional code (CC) and its dual. Our result improves upon a recent result by Gluesing-Luerssen and Schneider, where the requirement of a minimal encoder is assumed. We can also establish the MacWilliams identity for the input-parity weight adjacency matrices of a systematic CC and its dual. Most importantly, we show that a type of Hamming weight enumeration functions of all codewords of a CC can be derived from the weight adjacency matrix, which thus provides a connection between these two very different notions of weight enumeration functions in the convolutional code literature. Finally, the relations between various enumeration functions of a CC and its dual are summarized in a diagram. This explains why no MacWilliams identity exists for the free-distance enumerators.
convolutional code is one of important channel codes to combat fading and noise in 3GPP 3.84/1.28Mcps TDD systems. However, the design of the convolutional codes used in current specifications just obtain the best per...
详细信息
ISBN:
(纸本)0780389662
convolutional code is one of important channel codes to combat fading and noise in 3GPP 3.84/1.28Mcps TDD systems. However, the design of the convolutional codes used in current specifications just obtain the best performance for BPSK modulation and AWGN propagation channel. In 3GPP 3.84/1.28Mcps TDD systems, QPSK modulation is adopted. Moreover, multipath fading channels are often encountered in the practical communication environments. Therefore, convolutional codes used in the specifications cannot always achieve the best performance in the practical 3GP 3.84/1.28Mcps TDD systems. In this paper, we proposed an improved convolutional encoder for 3GPP 3.84/1.28Mcps TDD downlink system by analyzing the integration effects of QPSK modulation and multipath fading channels on the code design. Performance analysis and simulation show that better system performance can be achieved by using the proposed encoder.
In this paper, a novel setpoint-based design approach for Irregular Repeat Accumulate (IRA) codes in iterative detection and decoding structures is presented. In contrast to conventional IRA code design in which the c...
详细信息
ISBN:
(纸本)9781479944828
In this paper, a novel setpoint-based design approach for Irregular Repeat Accumulate (IRA) codes in iterative detection and decoding structures is presented. In contrast to conventional IRA code design in which the convolutional decoder is combined with the detector, the goal behind this approach is to keep the IRA decoding structure consisting of convolutional decoder and repetition decoder intact, i.e. to consider it as an inner loop of the overall detection structure. The outer loop is then composed of the IRA decoder and the system specific detector. This approach requires to adapt the irregular repetition code jointly to the convolutional decoder as well as to the detector which is achieved by formulating setpoints for the inner and outer code characteristic. As will be shown, the presented code design approach, although starting from a completely different viewpoint as the conventional approach, leads to an irregular repetition code with a very similar transfer characteristic and code rate than the conventional approach.
Recursive spatial multiplexing (RSM), a closed loop multiple-input multiple-output (MIMO) structure for achieving the capacity offered by MIMO channels with a low-complexity detector, is investigated in measured indoo...
详细信息
ISBN:
(纸本)9781479980307
Recursive spatial multiplexing (RSM), a closed loop multiple-input multiple-output (MIMO) structure for achieving the capacity offered by MIMO channels with a low-complexity detector, is investigated in measured indoor MIMO transmission channels limited by different types of interference. It is seen that RSM can deal effectively with interferences following either a Gaussian mixture or Middleton distribution. In cases of interferences which are symmetric alpha stable distributed, however, RSM suffers from a performance degradation. Performance improvements resulting from the use of a convolutional coding at the single-input single-output (SISO) encoder of the RSM architecture are investigated for the different interference distributions.
Analyzing and extracting features and variability from different artifacts is an indispensable activity to support systematic integration of single software systems and Software Product Line (SPL). Beyond manually ext...
详细信息
ISBN:
(纸本)9781538649695
Analyzing and extracting features and variability from different artifacts is an indispensable activity to support systematic integration of single software systems and Software Product Line (SPL). Beyond manually extracting variability, a variety of approaches, such as feature location in source code and feature extraction in requirements, has been proposed for automating the identification of features and their variation points. While requirements contain more complete variability information and provide traceability links to other artifacts, current techniques exhibit a lack of accuracy as well as a limited degree of automation. In this paper, we propose an unsupervised learning structure to overcome the abovementioned limitations. In particular, our technique consists of two steps: First, we apply Laplacian Eigenmaps, an unsupervised dimensionality reduction technique, to embed text requirements into compact binary codes. Second, requirements are transformed into a matrix representation by looking up a pre-trained word embedding. Then, the matrix is fed into CNN to learn linguistic characteristics of the requirements. Furthermore, we train CNN by matching the output of CNN with the pre-trained binary codes. Initial results show that accuracy is still limited, but that our approach allows to automate the entire process.
This paper focuses on presenting an efficient underwater acoustic communication system for image transmission. Water nature slows propagation speed down, and corrupts the transmitted signal. Moreover, the channel capa...
详细信息
ISBN:
(纸本)9781509057214
This paper focuses on presenting an efficient underwater acoustic communication system for image transmission. Water nature slows propagation speed down, and corrupts the transmitted signal. Moreover, the channel capacity is limited. The proposed system aims at effectively withstanding channel effects, minimizing the bit error rate and obtaining images with acceptable quality at the receiver. This system includes four parts;image source coding, channel coding, modulation scheme and image denoising. Set partitioning in hierarchical trees (SPIHT) is used for efficient image compression. To detect transmission errors, reduce the decoder complexity and increase the efficiency of transmission, convolutional coding is employed. Quadrature amplitude modulation (4-QAM) is used with orthogonal frequency division multiplexing (OFDM) to provide a reliable spectrally efficient system and enhance its resistivity to channel effects. Differential coding is used with 4-QAM to handle phase errors, while avoiding complex carrier tracking. To improve the image quality, 2-D double-density dual-tree complex discrete wavelet transform (DWT) is used to denoise the received decompressed image. Simulation results show that the proposed system is capable of transmitting and receiving underwater images with acceptable quality.
In this paper, we study the design and the delay-exponent of anytime codes over a three terminal relay network. We propose a bilayer anytime code based on anytime spatially coupled low-density parity-check (LDPC) code...
详细信息
ISBN:
(纸本)9781479959990
In this paper, we study the design and the delay-exponent of anytime codes over a three terminal relay network. We propose a bilayer anytime code based on anytime spatially coupled low-density parity-check (LDPC) codes and investigate the anytime characteristics through density evolution analysis. By using mathematical induction technique, we find analytical expressions of the delay-exponent for the proposed code. Through comparison, we show that the analytical delay-exponent has a close match with the delay-exponent obtained from numerical results.
Recent work has shown that convolutional networks can be substantially deeper, more accurate, and efficient to train if they contain shorter connections between layers close to the input and those close to the output....
详细信息
ISBN:
(纸本)9781538604571
Recent work has shown that convolutional networks can be substantially deeper, more accurate, and efficient to train if they contain shorter connections between layers close to the input and those close to the output. In this paper, we embrace this observation and introduce the Dense convolutional Network (DenseNet), which connects each layer to every other layer in a feed-forward fashion. Whereas traditional convolutional networks with L layers have L connections-one between each layer and its subsequent layer-our network has L(L+1)/2 direct connections. For each layer, the feature-maps of all preceding layers are used as inputs, and its own feature-maps are used as inputs into all subsequent layers. DenseNets have several compelling advantages: they alleviate the vanishing-gradient problem, strengthen feature propagation, encourage feature reuse, and substantially reduce the number of parameters. We evaluate our proposed architecture on four highly competitive object recognition benchmark tasks (CIFAR-10, CIFAR-100, SVHN, and ImageNet). DenseNets obtain significant improvements over the state-of-the-art on most of them, whilst requiring less computation to achieve high performance. Code and pre-trained models are available at https://***/liuzhuang13/DenseNet.
暂无评论