检索结果-内蒙古大学图书馆

2015 International Conference on Computer Information Systems and Industrial Applications(CISIA2015)

作者： J.Guo S.X.Wu Z.Q.Wang X.L.Li Beijing Information Technology College

ISBN: (纸本)9781510805750

The digital fountain is a network encoding technology of an essentially infinite of encoded symbols,and LT codes is a class of codes that are useful in implementing an ideal digital *** on analysis of existing LT decoding algorithm,an improved method is *** each encoded symbol arrivals,the new method performs both the triangular and backsubstitution steps of Gaussian elimination on the fly,thus distributing the decoding work during all symbols reception and obtaining a short delay.

关键词： erasure channel digital fountain LT codes decoding method

来源：评论

学校读者我要写书评

暂无评论

A Signal decoding Framework Combining Fundamental and Harmonic Components for Transcranial Acoustoelectric Brain Imaging

引用

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 2025年 74卷

作者： Chen, Xinrui Wang, Xiao Song, Xizi Liu, Xiuyun Ming, Dong Tianjin Univ Acad Med Engn & Translat Med Tianjin 300072 Peoples R China Tianjin Univ State Key Lab Adv Med Mat & Devices Tianjin 300072 Peoples R China

Transcranial acoustoelectric (AE) brain imaging (tABI) is a potential neuroimaging technique with high spatial (similar to mm) and temporal (similar to ms) resolution. decoding precision is critical for decoded signal quality and imaging performance. However, the traditional method is based on the fundamental component of the AE signal, overlooking the valuable information of its harmonics. Here, we present a signal decoding framework combining fundamental and harmonic components. Adopting filter bank analysis to efficiently integrate the fundamental and harmonic components, a weighted filter bank Hilbert transform (wFBHT) decoding method is implemented. With steady-state visual stimulation, both brain electric signal steady-state visual evoked potential (SSVEP) and AE signal of live rats were recorded using tABI. First, the signal-to-noise ratios (SNRs) characteristics of harmonic were analyzed. The decoded AE (decAE) signals of the second-sixth harmonic exhibited a highly consistent spectral response with SSVEP, which provided a theoretical basis for the framework. Then, using the grid search and nonlinear fitting, the impacts of harmonic number N and weight w on the change rate of correlation coefficient (CRCC) between the decAE signal and SSVEP were further explored. Results demonstrated that CRCC increased with harmonic number N significantly superior to the traditional method ( p = 0.0004 ). With equal weight of first-fourth harmonics, the highest correlation coefficient increased by 47%. What is more, the feasibility and performance were validated in live rats. Compared with that of the traditional method, the SNR of the decAE signal with wFBHT was 20.20 dB and increased by 72.1%.

关键词： decoding method filter bank harmonics of acoustoelectric (AE) signal transcranial AE brain imaging (tABI) weighted decoding method filter bank harmonics of acoustoelectric (AE) signal transcranial AE brain imaging (tABI) weighted

来源：评论

学校读者我要写书评

暂无评论

Network Coding with Parallel Path Protection for Multiple Link Failure in WOBAN

引用

RECENT ADVANCES IN ELECTRICAL & ELECTRONIC ENGINEERING 2024年第5期17卷 466-474页

作者： Solanki, Sangita Upadhyay, Raksha Bhatt, Uma Rathore Bhat, Vijay Devi Ahilya Univ Inst Engn & Technol Dept Elect & Telecommun Engn Indore 452017 India SAGE Univ Inst Adv Comp Indore India

Background WOBAN is a high-speed network and hence any kind of failure results in huge data loss. Using the proposed network coding technique with parallel path protection can handle multiple link failures, the performance of the network can be *** This study aims to improve network performance using network coding with parallel path protection routing algorithm (NC-PPR) for multiple link failures in *** We investigated the multiple link failures in WOBAN by the proposed approach namely Coded Path Protection Algorithm, which enhances the survivability of the WOBAN against multiple link failures in the front end and eliminates the need of backup *** Extensive simulation is carried out to implement proposed work. A simulation model and code is developed in MATLAB to get the performance enhancement of *** We compared the performance of proposed algorithm with existing algorithm. The obtained results show that the proposed algorithm has superior performance than the existing *** In this paper, a new routing approach, which works in three phases namely path finding, encoding, and decoding, using random linear network coding (RLNC) is introduced to address the survivability issue of the WOBAN. The proposed approach also enhances the network performance in terms of PDR, overhead, and delay.

关键词： Network coding encoding method decoding method WOBAN generation-based group coding survivability

来源：评论

学校读者我要写书评

暂无评论

Mitigating Hallucination Issues in Small-Parameter LLMs through Inter-Layer Contrastive decoding

Mitigating Hallucination Issues in Small-Parameter LLMs thro...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Li, Fan Zhang, Xiaofeng Zhang, Peng Zhejiang Univ Polytech Inst Hangzhou Peoples R China Shanghai Jiao Tong Univ Shanghai Peoples R China Zhejiang Univ Sch Math Sci Hangzhou Peoples R China

ISBN: (纸本)9798350359329;9798350359312

In this paper, we introduce a new decoding method to mitigate the issue of hallucinations in Large Language Models (LLMs). Specifically, our method dynamically selects appropriate internal layers(DiLa) of the model and compares their logic to enhance the authenticity and accuracy during the decoding process. Experimental results show that DiLa significantly improves performance across multiple tasks without fine-tuning the model and without substantially reducing decoding efficiency. Overall, our approach is a simple yet efficient solution to alleviate the hallucination problem in LLMs.

关键词： Hallucinations decoding method Large Language Models Natural Language Processing

来源：评论

学校读者我要写书评

暂无评论

decoding methods for the flow shop scheduling with peak power consumption constraints

引用

INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH 2019年第10期57卷 3200-3218页

作者： Wang, Jing-jing Wang, Ling Tsinghua Univ Dept Automat Beijing Peoples R China

The permutation flow shop scheduling problem (PFSP) is of wide application backgrounds and plays an important role in the manufacturing systems. With the serious energy concerns in manufacturing enterprises, peak power consumption is considered one of the significant issues. For the PFSP with peak power consumption constraints (PFSPP), the real-time power consumption cannot exceed a given peak power at any time. Since the classical first-come first-serve scheduling method is not suitable for the PFSPP, this paper addresses the decoding methods to obtain feasible schedules based on the permutation encoding scheme. First, an earliest processing rule (EPR) is designed to determine the starting time of each operation, satisfying the power consumption constraints. Then, five decoding methods based on EPR are proposed to determine the suitable priority between the operations to yield feasible schedules with high quality. After analysing the complexity of the proposed decoding methods and comparing the performances via extensive numerical tests, some suggestions are provided for solving the PFSPP with different scales and power constraints.

关键词： Flow shop scheduling energy consumption peak power constraints decoding method

来源：评论

学校读者我要写书评

暂无评论

k-order Gaussian Fibonacci polynomials and applications to the coding/decoding theory

引用

JOURNAL OF DISCRETE MATHEMATICAL SCIENCES & CRYPTOGRAPHY 2022年第5期25卷 1399-1416页

作者： Asci, Mustafa Aydinyuz, Suleyman Pamukkale Univ Fac Sci & Arts Dept Math TR-20160 Denizli Turkey

In this paper we define k-order Gaussian Fibonacci polynomials with boundary conditions and give the generating function, explicit formula and some identities for k-order Gaussian Fibonacci polynomials. We introduce the matrix represent and we obtain the k-order Gaussian Fibonacci Polynomials matrix. We define a new coding theory called k-order Gaussian Fibonacci Polynomials coding theory and establish the code elements for values of k. This coding/decoding method bound to the Q(k) (x), R-k (x) and E-k,E-n (x) matrices. So, this method is different from the classical algebraic coding. Consequently, with this method, we move the coding theory onto a complex space which is a different field. Therefore, new working areas are created.

关键词： Fibonacci numbers Gaussian Fibonacci numbers k-order Gaussian Fibonacci numbers Fibonacci polynomials k-order Gaussian Fibonacci Polynomials Coding decoding method k-order Gaussian Fibonacci polynomial coding decoding theory

来源：评论

学校读者我要写书评

暂无评论

Codes with burst distance and periodical burst errors

引用

JOURNAL OF COMPUTATIONAL AND APPLIED MATHEMATICS 2022年第0期411卷

作者： Das, Pankaj Kumar Haokip, Letminthang Tezpur Univ Dept Math Sci Napaam Sonitpur 784028 Assam India

Burst-b distance introduced by Wainberg and Wolf (1972) has been found to be useful for correction of multiple burst errors and multiple erasures. Villalba et al. (2016) have derived extended Reiger and Singleton bound for linear code with minimum burst-b distance d(b) and then present a class of Maximum Distance Separable (MDS) codes (named as C-b code). In this paper, we derive an upper bound on d(b) for any linear code and a lower bound on d(b) for constant burst-b weight linear codes. We also present the existence of linear code with burst-b distance d(b) - 1 from code with burst distance d(b). The cardinality of a linear code and the connection of linearly independent columns of the parity check matrix of any MDS code with the distance d(b) are also given. Further, we consider periodical burst error which is found in many communication channels and investigate periodical burst-detection and -correction capability of linear codes having distance d(b). Then, we do the same investigation for C-b and its dual code C-b(perpendicular to). Finally, we give decoding procedure for the code C-b in case of periodical burst errors. (C) 2022 Elsevier B.V. All rights reserved.

关键词： Error detection and correction Periodical burst error Burst weight and distance Bound decoding method

来源：评论

学校读者我要写书评

暂无评论

Reproduction accuracy of Higher-Order Ambisonics with Max-rE and/or least norm solution in decoding

引用

ACOUSTICAL SCIENCE AND TECHNOLOGY 2019年第1期40卷 23-28页

作者： Otani, Makoto Shigetani, Haruki Kyoto Univ Grad Sch Engn Nishikyo Ku Kyoto 6158540 Japan

In sound field reproduction with Higher-Order Ambisonics (HOA), a sweet spot is formed around a reproduction point that is generally a center of spherical loudspeaker array. The conventional HOA decoding with a unique solution yields a spherical sweet spot, in which the reproduction error is less than 4% and whose radius is theoretically defined in the literature. On the other hand, it is known that Max-rE decoding or a least norm solution derived when using a larger number of loudspeakers would lead to an enhancement of reproduction accuracy. However, it is yet to be revealed how Max-rE decoding and the least norm solution affects the reproduction accuracy and the size and shape of the sweet spot, when they are individually or simultaneously applied to HOA decoding. This paper numerically investigates the reproduction accuracy of HOA with such different HOA decoding methods. The numerical results suggest that, compared to the conventional decoding method, ones with Max-rE and/or the least norm solution result in deformed or expanded sweet spot and suppression of reproduction errors outside the sweet spot.

关键词： Higher-Order Ambisonics decoding method Least-norm solution Max-rE

来源：评论

学校读者我要写书评

暂无评论

Accuracy of binaural signal in Higher-Order Ambisonics reproduction with different decoding approaches

引用

ACOUSTICAL SCIENCE AND TECHNOLOGY 2019年第2期40卷 144-147页

作者： Shigetani, Haruki Otani, Makoto Kyoto Univ Grad Sch Engn Nishikyo Ku Kyoto 6158540 Japan

来源：评论

学校读者我要写书评

暂无评论

A novel decoding method for the hybrid flow-shop scheduling problem with multiprocessor tasks

引用

INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY 2012年第9-12期59卷 1113-1125页

作者： Wang, Ling Xu, Ye Zhou, Gang Wang, Shengyao Liu, Min Tsinghua Univ Tsinghua Natl Lab Informat Sci & Technol TNList Beijing 100084 Peoples R China

As the strongly NP-hard problem, the hybrid flow-shop problem with multiprocessor tasks (HFSPMT) has gained increasing attention due to its academic significance and wide application background. For many heuristics and meta-heuristics to solve the HFSPMT, decoding method is the key element to decode sequences to schedules, which has great effect on the solution quality. To narrow the idle time between the consecutive operations in the processor and to increase the flexibility in selecting processors to schedule the following operations, several rules are proposed to adjust sequences to minimize the makespan of the HFSPMT. Based on the rules, a novel and effective decoding method named forward scheduling (FS) is proposed in this paper for solving the HFSPMT. At each stage, it first decodes the solution according to the non-decreasing order of the jobs' completion times at the previous stage, and then it adjusts the processing order of the jobs dynamically according to the rules to obtain potentially better schedule. Numerical results based on the well-known benchmarks and comparisons with some existing decoding methods demonstrate the effectiveness of the proposed FS decoding method.

关键词： Hybrid flow-shop scheduling Multiprocessor tasks decoding method Forward scheduling Dynamic adjustment

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：