检索结果-内蒙古大学图书馆

1st Workshop on Distributed Machine Learning (DistributedML)

作者： Gajjala, Rishikesh R. Banchhor, Shashwat Abdelmoniem, Ahmed M. Dutta, Aritra Canini, Marco Kalnis, Panos KAUST Thuwal Saudi Arabia Indian Inst Technol Delhi India

ISBN: (纸本)9781450381826

Distributed stochastic algorithms, equipped with gradient compression techniques, such as codebook quantization, are becoming increasingly popular and considered state-of-the-art in training large deep neural network (DNN) models. However, communicating the quantized gradients in a network requires efficient encoding techniques. For this, practitioners generally use Elias encoding-based techniques without considering their computational overhead or data-volume. In this paper, based on huffman coding, we propose several lossless encoding techniques that exploit different characteristics of the quantized gradients during distributed DNN training. Then, we show their effectiveness on 5 different DNN models across three different data-sets, and compare them with classic state-of-the-art Elias-based encoding techniques. Our results show that the proposed huffman-based encoders (i.e., RLH, SH, and SHS) can reduce the encoded data-volume by up to 5.1x, 4.32x, and 3.8x, respectively, compared to the Elias-based encoders.

关键词： Distributed training Gradient compression Quantization huffman coding Elias and Run-length Encoding

来源：评论

学校读者我要写书评

暂无评论

Reversible Data Hiding in Encrypted Images Based on Multi-MSB Prediction and huffman coding

引用

IEEE TRANSACTIONS ON MULTIMEDIA 2020年第4期22卷 874-884页

作者： Yin, Zhaoxia Xiang, Youzhi Zhang, Xinpeng Anhui Univ Key Lab Intelligent Comp & Signal Proc Minist Educ Sch Comp Sci & Technol Hefei 230601 Peoples R China Fudan Univ Sch Comp Sci Shanghai 200433 Peoples R China

With the development of cloud storage and privacy protection, reversible data hiding in encrypted images (RDHEI) has attracted increasing attention as a technology that can: embed additional data in the image encryption domain, ensure that the embedded data can be extracted error-free, and the original image can be restored losslessly. In this paper, a high-capacity RDHEI algorithm based on multi-MSB (most significant bit) prediction and huffman coding is proposed. At first, multi-MSB of each pixel was predicted adaptively and marked by huffman coding in the original image. Then, the image was encrypted by a stream cipher method. At last, the vacated space can be used to embed additional data by multi-MSB substitution. Experimental results show that our method achieved higher embedding capacity while comparing with the state-of-the-art methods.

关键词： Data mining Encryption Image restoration Image coding Receivers huffman coding Reversible data hiding encrypted images privacy protection huffman coding

来源：评论

学校读者我要写书评

暂无评论

Hybrid PAPR reduction scheme with huffman coding and DFT-spread technique for direct-detection optical OFDM systems

引用

OPTICAL FIBER TECHNOLOGY 2018年 40卷 1-7页

作者： Peng, Miao Chen, Ming Zhou, Hui Wan, Qiuzhen Jiang, LeYong Yang, Lin Zheng, Zhiwei Chen, Lin Hunan Normal Univ Coll Phys & Informat Sci Changsha 410081 Hunan Peoples R China China Mobile Commun Grp Hunan Co Ltd Changsha Branch Changsha 410007 Hunan Peoples R China Hunan Univ Sch Informat Sci & Engn Key Lab Micro Nano Optoelect Devices Minist Educ Changsha 410082 Hunan Peoples R China

High peak-to-average power ratio (PAPR) of the transmit signal is a major drawback in optical orthogonal frequency division multiplexing (OOFDM) system. In this paper, we propose and experimentally demonstrate a novel hybrid scheme, combined the huffman coding and Discrete Fourier Transmission-Spread (DFT-spread), in order to reduce high PAPR in a 16-QAM short-reach intensity-modulated and direct-detection OOFDM (IMDD-OOFDM) system. The experimental results demonstrated that the hybrid scheme can reduce the PAPR by about 1.5, 2, 3 and 6 dB, and achieve 1.5, 1, 2.5 and 3 dB receiver sensitivity improvement compared to clipping, DFT-spread and huffman coding and original OFDM signals, respectively, at an error vector magnitude (EVM) of -10 dB after transmission over 20 km standard single-mode fiber (SSMF). Furthermore, the throughput gain can be of the order of 30% by using the hybrid scheme compared with the cases of without applying the huffman coding.

关键词： Discrete Fourier Transmission-Spread (DFT-spread) huffman coding Optical orthogonal frequency division multiplexing (OOFDM) Peak-to-average power ratio (PAPR)

来源：评论

学校读者我要写书评

暂无评论

A Dynamic Programming Approach to Length-Limited huffman coding: Space Reduction With the Monge Property

引用

IEEE TRANSACTIONS ON INFORMATION THEORY 2010年第8期57卷 3918-3929页

作者： Golin, Mordecai Zhang, Yan Hong Kong UST Dept Comp Sci & Engn Kowloon Hong Kong Peoples R China

The "state-of-the-art" in length-limited huffman coding (LLHC) algorithms is the Theta(nD)-time, Theta(n)-space one of Hirschberg and Larmore, where is the size of the code and D <= n is the length restriction on the codewords. This is a very clever, very problem specific, technique. This paper presents a simple dynamic-programming (DP) method that solves the problem with the same time and space bounds. The fact that there was an Theta(nD) time DP algorithm was previously known;it is a straightforward DP with the Monge property (which permits an order of magnitude speedup). It was not interesting, though, because it also required Theta(nD) space. The main result of this paper is the technique developed for reducing the space. It is quite simple and applicable to many other problems modeled by DPs with the Monge property. This is illustrated with examples from web-proxy design and wireless mobile paging.

关键词： Dynamic programming huffman coding Monge property prefix-free codes web-proxies wireless paging

来源：评论

学校读者我要写书评

暂无评论

Anomaly Detection in Audio With Concept Drift Using Dynamic huffman coding

引用

IEEE SENSORS JOURNAL 2022年第17期22卷 17126-17138页

作者： Kumari, Pratibha Saini, Mukesh Indian Inst Technol IIT Ropar Rupnagar 140001 India

When detecting anomalies in audio, it can often be necessary to consider concept drift: the distribution of the data may drift over time because of dynamically changing environments, and anomalies may become normal as time elapses. We propose to use dynamic huffman coding for anomaly detection in audio with concept drift. Compared with the existing method of adaptive Gaussian mixture modeling (AGMM), dynamic huffman coding does not require a priori information about the clusters and can adjust the number of clusters dynamically depending on the amount of variation in the audio. To control the size of the huffman tree, we propose to merge clusters that are close to each other instead of replacing rare clusters with new data. This reduces redundancy in the huffman tree while ensuring that it never forgets past information. On audio datasets with concept drift which we have curated ourselves, our proposed method achieves a higher area under the curve (AUC) compared with AGMM and fixed-length huffman trees.

关键词： Anomaly detection huffman coding Hidden Markov models Encoding Data models Adaptation models Sensors Anomaly detection concept drift unsupervised modelling dynamic huffman coding long term audio surveillance

来源：评论

学校读者我要写书评

暂无评论

Improved near-lossless technique using the huffman coding for enhancing the quality of image compression

引用

MULTIMEDIA TOOLS AND APPLICATIONS 2022年第20期81卷 28509-28529页

作者： Otair, Mohammed Abualigah, Laith Qawaqzeh, Mohammed K. Amman Arab Univ Fac Comp Sci & Informat Amman 11953 Jordan Univ Sains Malaysia Sch Comp Sci George Town 11800 Malaysia

Digital data compression aims to reduce the size of digital files in line with technological development. However, most data is distinguished by its large size, which requires a large storage capacity, and requires a long time in transmission operations via the Internet. Therefore, a new compress files method is needed to reduce the image size, maintain its quality, utilize storage spaces, and minimize time. This paper aims to improve digital image compression's compression rates by dividing the image into several blocks. Thus, a new near-lossless method using the huffman coding technique is proposed. Digital image compression techniques are classified as lossless and lossy. huffman coding is a lossless-based technique used in the proposed method to maintain image quality during compression. The proposed method consists of several steps, which are dividing the image into blocks, finding the lowest value in each block and subtracting it from the rest of the values in the same block, then subtracting one from the odd numbers, dividing all the values on two, and finally applying the huffman coding technique to the block. The proposed method is applied to a well-known gray and color set with different types and different dimensions. Standard evaluation measures are used (i.e., PSNR, MSE, and CR) to evaluate the proposed method's performance. When compressing images using the proposed method, the results demonstrated 0.11% enhancement when used two by two blocks. It also got high compression rates (25%).

关键词： Near-lossless technique huffman coding Image quality Image compression

来源：评论

学校读者我要写书评

暂无评论

MEMORY EFFICIENT AND HIGH-SPEED SEARCH huffman coding

引用

IEEE TRANSACTIONS ON COMMUNICATIONS 1995年第10期43卷 2576-2581页

作者： HASHEMIAN, R Department of Electrical Engineering Northern Illinois University DeKalb IL USA

Code compression is a key element in high-speed digital data transport. A major compression is performed by converting the fixed-length codes to variable-length codes through a (semi-)entropy coding scheme. huffman coding is shown to be a very efficient coding scheme. To speed up the process of search for a symbol in a huffman tree and to reduce the memory size we have proposed a tree clustering algorithm to avoid high sparsity of the tree. The method is shown to be extremely efficient in memory requirement, and fast in searching for the symbol. For an experimental video data with huffman codes extended up to 13 hits in length, the entire memory space is shown to be 122 words, compared to 2(13) = 8192 words in a normal situation.

关键词： huffman coding Clustering algorithms Video compression Random access memory HDTV Digital video broadcasting TV broadcasting Vector quantization Discrete cosine transforms Bit rate

来源：评论

学校读者我要写书评

暂无评论

Low-Complexity huffman coding-Based Adaptive Spatial Modulation

引用

IEEE COMMUNICATIONS LETTERS 2024年第11期28卷 2468-2472页

作者： Hu, Gangfei Yang, Yukun Shang, Pingping Peng, Yuyang Selvaprabhu, Poongundran Hai, Han Donghua Univ Coll Informat Sci & Technol Shanghai 201620 Peoples R China Hubei Univ Technol Sch Elect & Elect Engn Wuhan 430068 Peoples R China Macau Univ Sci & Technol Sch Comp Sci & Engn Macau Peoples R China Vellore Inst Technol Sch Elect Engn Dept Commun Engn Vellore 632014 India

Adaptive spatial modulation (ASM) is a novel technology in multiple-input-multiple-output (MIMO) systems, which dynamically adjusts the modulation order of the transmit antenna based on the constantly changing channel state information (CSI). The combination of ASM and huffman coding, using the huffman method, allows for flexible spatial symbols to provide different activation probabilities for the transmit antenna. However, the exhaustive search conducted by huffman-ASM requires a large number of candidate combinations to identify the optimal modulation combination. This results in high computational complexity and a heavy feedback burden. To address the challenge, this letter proposes a low-complexity algorithm based on the huffman-ASM scheme. This scheme does not include all candidate combinations in the calculation but directly selects several optimal candidate combinations based on channel gain sorting;therefore, it reduces system complexity and alleviates the burden on the feedback path. The simulation results indicate that, compared to the conventional scheme, the proposed scheme exhibits only a slight decrease in performance while significantly reducing computational complexity.

关键词： Modulation Transmitting antennas Vectors huffman coding Computational complexity Symbols Sorting Spatial modulation channel gain adaptive modulation antenna selection

来源：评论

学校读者我要写书评

暂无评论

huffman coding Analysis of XOR Filtered Images 15

Huffman Coding Analysis of XOR Filtered Images

引用

IEEE International Symposium on Signal Processing and Information Technology (ISSPIT)

作者： Cooper, Takiyah K. Desoky, Ahmed H. Univ Louisville CECS Dept Louisville KY 40292 USA

ISBN: (纸本)9781509004805

The field of lossless image compression studies the various ways to represent image data in the most compact and efficient manner possible that also allows the image to be reproduced without any loss. There are many ways that this is accomplished and the most efficient strategy is to leverage the fundamental concepts of entropy reduction through decorrelation. The application of decorrelation methodologies prior to entropy coding is a modeling-coding strategy that produces significant results. This study focuses on the usage of the exclusive-or logic operator as a decorrelation filter. The exclusive-or logic operator is simply and reversibly applied to continuous-tone images for the purpose of extracting differences between neighboring pixels. The results of the exclusive-or operator based decorrelation filter are then processed through entropy encoding using huffman coding. Experiments indicate that the exclusive-or operator based decorrelation filter has a significant impact on reducing entropy and therefore compression results.

关键词： Difference coding Exclusive-Or huffman coding Image Compression

来源：评论

学校读者我要写书评

暂无评论

Green computing for wireless sensor networks: Optimization and huffman coding approach

引用

PEER-TO-PEER NETWORKING AND APPLICATIONS 2017年第3期10卷 592-609页

作者： Aanchal Kumar, Sushil Kaiwartya, Omprakash Abdullah, Abdul Hanan Jawaharlal Nehru Univ Sch Comp & Syst Sci Comp Sci New Delhi India UTM Fac Comp Johor Baharu 81310 Johor Malaysia

Lifetime maximization has witnessed continuous attention from academia as well as industries right from the inception of Wireless Sensor Networks (WSNs). Recently, mobile sink, trajectory based forwarding and energy supply based node selection have been suggested in literature for optimizing residual energy of nodes. In the most of these approaches, energy consumption has been minimized focusing on the optimization of one particular parameter. The consideration of impact of more than one parameters on energy consumption is lacking in literature. In this context, this paper proposes huffman coding and Ant Colony Optimization based Lifetime Maximization (HA-LM) technique for randomly distributed WSNs. In particular, ACO based multiple paths exploration and huffman based optimal path selection consider the impact of two network parameters on energy consumption. The parameters include path length in terms of hop count and residual energy in terms of load of nodes of the path and the least energy node. The construction of multiple paths from source to the sink is mathematically derived based on the concept of two types of ants;namely, Advancing Ant (A-ANT) and Regressive Ant (R-ANT) in ACO. The optimal path is identified from the available multiple paths using huffman coding. Analytical and simulation results of HA-LM are comparatively evaluated with the state-of-the-art techniques considering four performance metrics;namely, average residual energy, energy consumption, number of alive sensors and standard deviation of energy. The comparative performance evaluation attests the superiority of the proposed technique to the state-of-the-art techniques.

关键词： Green computing Lifetime maximization Energy consumption Optimization ACO huffman coding WSNs

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：