|
| 基于RFID和LSTM的固定资产智能感知方法 |
| An Intelligent Perceptual Method of Fixed Assets based on RFID and LSTM |
| 投稿时间：2023-12-08 修订日期：2024-03-14 |
| DOI： |
| 中文关键词: 长短期记忆神经网络 射频识别 感知识别 时间序列 固定资产实时监测 |
| 英文关键词: LSTM RFID perceptual recognition time sequences real time fixed assets monitoring |
| 基金项目:安徽省高校学科(专业)拔尖人才学术资助项目(gxyq2022030);安徽省高校学科(专业)拔尖人才学术资助项目(gxbjZD2021067)、安徽省高校自然科学研究重点项目(KJ2020A0470)、安徽省特支计划创新领军人才项目(皖组办[2022]21号)、安徽建筑大学智能建筑与建筑节能安徽省重点实验室主任基金(IBES2022ZR01) |
|
| 摘要点击次数: 397 |
| 全文下载次数: 0 |
| 中文摘要: |
| 针对传统基于射频识别的固定资产(如贵重仪器仪表)感知方法存在射频信号易受环境影响、多个标签互干扰、感知识别精度低等问题,本文提出一种基于射频识别(RFID)和长短期记忆(LSTM)神经网络的固定资产状态智能感知方法。为克服RFID信号接收时间非连续性导致的识别精度低问题,引入序列分析思想,使用LSTM对一段时间内接收到的RFID信号序列进行模式学习,构建基于RFID和LSTM的固定资产感知模型,实现固定资产位置的二分类辨识。所提方法在高校实验室条件下开展实测实验进行性能验证。结果表明,使用序列分析的固定资产状态感知模型辨识准确率可达99.26%,比传统基于离散时刻点的RFID感知模型辨识准确率高11.05%,且对多标签的识别准确率可达93.0%以上,可满足智能建筑安防场景中对固定资产实时监测的应用需求。 |
| 英文摘要: |
| Considering that the traditional RFID-based fixed asset (such as valuable instruments) perception methods have many disadvantages, such as vulnerability of radio signals to environmental interference, multiple tag interferences with each other, and low recognition accuracy, in this paper an intelligent perceptual method of fixed assets based on long and short-term memory neural networks (LSTM) and Radio Frequency Identification (RFID) is proposed. To deal with the low recognition accuracy caused by the discontinuity of RFID signal reception, with the idea of sequence analysis introduced, a LSTM-based model is developed to learn the pattern of RFID signal sequences received over a period of time, which can realize the two-category identification of the location of fixed assets. The proposed model has been validated by real experiments carried out in a university laboratory environment. Results show that the recognition accuracy of the fixed asset state using time sequence analysis can reach 99.26%, which is 11.05% higher than the recognition accuracy of the traditional RFID perception model based on discrete moments, and the recognition accuracy of multiple tags can reach 93.0%, satisfying the actual requirements of real time fixed asset monitoring in the smart building security scenarios. |
|
View Fulltext
查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|
