Electronics Letters最新文献_第6页

Pseudorandom-Noise-Modulated Ultra-Wideband Magnetless Nonreciprocal Transmission Line

IF 0.7 4区工程技术

Electronics Letters Pub Date : 2025-02-24 DOI: 10.1049/ell2.70186

Haoran Wang, Tao Zhang, Xiaoyi Wang, Guo-min Yang, Zhaohui Chen

引用次数: 0

Forward jamming mitigation using partially occupied DDMA-MIMO waveform

IF 0.7 4区工程技术

Electronics Letters Pub Date : 2025-02-24 DOI: 10.1049/ell2.70169

Xuan Fang, Dehua Zhao, Liang Zhang

引用次数: 0

Categorization Method of Power Supply Quality Events Using Rule-Based Decision Trees, Combining Traditional Indices With Higher-Order Statistics

IF 0.7 4区工程技术

Electronics Letters Pub Date : 2025-02-22 DOI: 10.1049/ell2.70189

Paula Remigio-Carmona, Olivia Florencias-Oliveros, Juan-José González-de-la-Rosa

引用次数: 0

A Note of Non-Existence of Strongly Regular and Deza Graphs

IF 0.7 4区工程技术

Electronics Letters Pub Date : 2025-02-22 DOI: 10.1049/ell2.70179

Fernando Andres Benavides, Carlos Alberto Trujillo Solarte

{"title":"A Note of Non-Existence of Strongly Regular and Deza Graphs","authors":"Fernando Andres Benavides, Carlos Alberto Trujillo Solarte","doi":"10.1049/ell2.70179","DOIUrl":"https://doi.org/10.1049/ell2.70179","url":null,"abstract":"In this letter, we prove that strongly regular graphs and Deza digraphs do not exist with parameters <math>\u0000 <semantics>\u0000 <mrow>\u0000 <mo>(</mo>\u0000 <msup>\u0000 <mi>p</mi>\u0000 <mi>m</mi>\u0000 </msup>\u0000 <mo>,</mo>\u0000 <mi>k</mi>\u0000 <mo>,</mo>\u0000 <msup>\u0000 <mn>2</mn>\u0000 <mi>m</mi>\u0000 </msup>\u0000 <mo>,</mo>\u0000 <msup>\u0000 <mn>2</mn>\u0000 <mi>m</mi>\u0000 </msup>\u0000 <mo>)</mo>\u0000 </mrow>\u0000 <annotation>$(p^m,k,2^m,2^m)$</annotation>\u0000 </semantics></math> where <math>\u0000 <semantics>\u0000 <mrow>\u0000 <mi>m</mi>\u0000 <mo>=</mo>\u0000 <mn>3</mn>\u0000 <mo>,</mo>\u0000 <mn>4</mn>\u0000 </mrow>\u0000 <annotation>$m=3,4$</annotation>\u0000 </semantics></math>. As a consequence, we provide many parameters for which there is no a difference set.","PeriodicalId":11556,"journal":{"name":"Electronics Letters","volume":"61 1","pages":""},"PeriodicalIF":0.7,"publicationDate":"2025-02-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1049/ell2.70179","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143471980","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A GGIW PMBM Smoother for Multiple Extended Object Tracking

IF 0.7 4区工程技术

Electronics Letters Pub Date : 2025-02-22 DOI: 10.1049/ell2.70188

Wenhui Wang, Ye Xu, Kejie Zhang, YouPeng Sun, Peng Li

引用次数: 0

Enhanced Segment Anything Model for Accurate White Blood Cell Segmentation

IF 0.7 4区工程技术

Electronics Letters Pub Date : 2025-02-21 DOI: 10.1049/ell2.70185

Yu Zang, Yang Su, Jun Hu

引用次数: 0

BLADE: Energy-efficient attention accelerator with fused kernel and bit-level redundancy elimination

IF 0.7 4区工程技术

Electronics Letters Pub Date : 2025-02-20 DOI: 10.1049/ell2.70137

Zhiwei Lin, Yubin Qin, Jiachen Wang, Yang Wang, Huanyu Wang, Zhe Zheng, Wenpeng Cui, Shaojun Wei, Yang Hu, Shouyi Yin

{"title":"BLADE: Energy-efficient attention accelerator with fused kernel and bit-level redundancy elimination","authors":"Zhiwei Lin, Yubin Qin, Jiachen Wang, Yang Wang, Huanyu Wang, Zhe Zheng, Wenpeng Cui, Shaojun Wei, Yang Hu, Shouyi Yin","doi":"10.1049/ell2.70137","DOIUrl":"https://doi.org/10.1049/ell2.70137","url":null,"abstract":"Attention-based transformer model has achieved remarkable performance in various artificial intelligence fields, while the attention computation, being a combination of matrix multiplication and softmax function, is still sub-optimized in terms of hardware implementation. Normally, it needs 3 pass of input memory access to compute attention, and the on-chip storage requirement is coupled with the input length, both of which pose significant memory issues. Further, the computation burden is heavy for long inputs. This paper proposes an algorithm-hardware co-design for attention. On algorithm side, it uses a linear-softmax fused kernel to fuse the matrix multiplication and non-linear functions, which enables high on-chip memory source utilization. On hardware side, it shows an accelerator named BLADE with identical partial product removing, which eliminates unnecessary computation by exploiting math feature of softmax. Experiment on ViT, Swin transformer, GPT-2, and LLaMA2 show that the proposed design achieves <math>\u0000 <semantics>\u0000 <mrow>\u0000 <mn>10.6</mn>\u0000 </mrow>\u0000 <annotation>$10.6$</annotation>\u0000 </semantics></math>–<math>\u0000 <semantics>\u0000 <mrow>\u0000 <mn>18.7</mn>\u0000 <mo>%</mo>\u0000 </mrow>\u0000 <annotation>$18.7%$</annotation>\u0000 </semantics></math> of energy efficiency improvement compared to state-of-the-art FlashAttention implementations.","PeriodicalId":11556,"journal":{"name":"Electronics Letters","volume":"61 1","pages":""},"PeriodicalIF":0.7,"publicationDate":"2025-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1049/ell2.70137","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143447038","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

LET-NLM-Decoder: A WFST-based asynchronous lazy-evaluation token-group decoder for first-pass neural language model decoding

IF 0.7 4区工程技术

Electronics Letters Pub Date : 2025-02-20 DOI: 10.1049/ell2.70145

Fangyi Li, Hang Lv, Yiming Wang, Lei Xie

{"title":"LET-NLM-Decoder: A WFST-based asynchronous lazy-evaluation token-group decoder for first-pass neural language model decoding","authors":"Fangyi Li, Hang Lv, Yiming Wang, Lei Xie","doi":"10.1049/ell2.70145","DOIUrl":"https://doi.org/10.1049/ell2.70145","url":null,"abstract":"Neural language models (NLMs) have been shown to outperform n-gram language models in automatic speech recognition (ASR) tasks. NLMs are usually used in the second-pass lattice rescoring rather than the first-pass decoding, since its encoded infinite history virtually cannot be compiled into static decoding graphs. However, the modeling power of NLMs is not fully leveraged due to the constraints imposed by the lattice, leading to accuracy loss. To improve this, on-the-fly composition decoders were proposed to utilize NLMs in first-pass decoding with increased computational cost. In this paper, an asynchronous lazy-evaluation token-group decoder with exact lattice generation is proposed to reduce the computational cost of the on-the-fly composition decoder, achieving significant decoding speedup. More specifically, having a novel token-group with a representative element data structure, the proposed decoder performs lazy-evaluation which expands the tokens until a word boundary is reached. Furthermore, based on the score of the representative element in a token-group, the decoder prunes unpromising tokens by an A* algorithm. The experiments show that the proposed decoder can accelerate the vanilla on-the-fly composition decoder by up to 6.9 times, and get paths with even better average likelihoods than lattice rescoring approaches.","PeriodicalId":11556,"journal":{"name":"Electronics Letters","volume":"61 1","pages":""},"PeriodicalIF":0.7,"publicationDate":"2025-02-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1049/ell2.70145","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143447010","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Time-Based Wireless Synchronous Detection of Bioimpedance Signals

IF 0.7 4区工程技术

Electronics Letters Pub Date : 2025-02-19 DOI: 10.1049/ell2.70181

Virgilio Valente, Hanieh Ashrafirad

引用次数: 0

Non-Linearity Incremental Ensemble Learning Based Depth Explorations of Latent Individual Contributions

IF 0.7 4区工程技术

Electronics Letters Pub Date : 2025-02-19 DOI: 10.1049/ell2.70178

Xiaozheng Deng, Yuanyuan Zhang, Shasha Mao, Peng Liu

{"title":"Non-Linearity Incremental Ensemble Learning Based Depth Explorations of Latent Individual Contributions","authors":"Xiaozheng Deng, Yuanyuan Zhang, Shasha Mao, Peng Liu","doi":"10.1049/ell2.70178","DOIUrl":"https://doi.org/10.1049/ell2.70178","url":null,"abstract":"In ensemble learning, individual accuracy and diversities are two key factors for improving the ensemble performance, and most existing methods have been designed from the view of boosting individual diversities. Whereas stronger individual diversities inevitably result in decreasing accuracies of partial individuals, it brings a great challenge for improving the ensemble performance. In fact, we observe that the contribution of individuals should be a crucial factor for ensemble learning, which can effectively alleviate the problem of balancing diversity and accuracy, but it is ignored in most existing studies. Based on this, we propose a new incremental non-linearity deep ensemble learning method that effectively combines multiple individuals based on the exploration of individual contributions by utilizing deep learning. In the proposed method, a characterization matrix is first constructed to represent the original individuals. Then, a deep ensemble network is constructed to explore the potential contribution of individuals in conjunction with the optimization objective of ensemble learning. Interestingly, a special layer is designed to achieve the minimization of ensemble error. Finally, experimental results on public datasets illustrate that the proposed method achieves the average 0.66–4.8% performance improvements compared to existing typical ensemble methods.","PeriodicalId":11556,"journal":{"name":"Electronics Letters","volume":"61 1","pages":""},"PeriodicalIF":0.7,"publicationDate":"2025-02-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1049/ell2.70178","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143438737","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0