Field programmable gate array implementation of an adaptive filtering based noise reduction and enhanced compression technique for healthcare applications,Transactions on Emerging Telecommunications Technologies

当前位置： X-MOL 学术 › Trans. Emerg. Telecommun. Technol. › 论文详情

Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)

Field programmable gate array implementation of an adaptive filtering based noise reduction and enhanced compression technique for healthcare applications
Transactions on Emerging Telecommunications Technologies ( IF 2.5 ) Pub Date : 2022-09-25 , DOI: 10.1002/ett.4654
Viswanadham Baby Koti Lakshmi Aruna ₁ , Chitra Ekambaram ₁ , Mididoddi Padmaja ₂

Affiliation

Nowadays, the development of e-health monitoring devices is getting more demand for remote health applications. In addition, cardiovascular diseases is considered the most chronic disease in recent years. So, this requires a fast diagnosis and therapy to clear this issue and requires an information transformation from the patient to the distant hospital. But, it faces many challenges like more energy consumption, data security, storage, and transmission. In addition, the electrocardiogram (ECG) data must be processed in real-time, and the information must have high predictability and lossless compression. So, in this article, an enhanced set partitioning in hierarchical tree decoder with deep belief neural network is developed for ECG compression. Due to an inappropriate channel in telemedicine applications, the ECG signals transmitted through a wireless network are affected by noise. So, fast normalized least mean square (FNLMS) algorithm based adaptive filter is deliberated to eliminate the unwanted noise from the ECG signal. Also, the structure of the FNLMs based adaptive filter is modified by developing a systolic structure to increase the speed of filtering process. Moreover, the parallel and pipelined architecture is provided for the proposed compression unit to reduce the processing time and the area overhead as compared to the existing methods. A publicly available database called the MIT-BIH database and real-time ECG data records are used for simulation purposes. The proposed design was implemented on the Xilinx platform and the MATLAB tool. The performance of the proposed design is estimated by the parameters like compression ratio, signal to noise ratio, root mean square error, and percentage root mean square difference (PRD), PRD normalized, and quality-score. It is compared with existing designs to show the efficiency of our proposed design.

中文翻译：

用于医疗保健应用的基于自适应滤波的降噪和增强压缩技术的现场可编程门阵列实现

如今，电子健康监测设备的发展对远程健康应用的需求越来越大。此外，心血管疾病被认为是近年来最为慢性的疾病。因此，这需要快速诊断和治疗来解决这个问题，并且需要从患者到远程医院的信息转换。但是，它面临着更多的能源消耗、数据安全、存储和传输等挑战。此外，心电图 (ECG) 数据必须实时处理，信息必须具有高可预测性和无损压缩。因此，在本文中，针对 ECG 压缩开发了具有深度信念神经网络的层次树解码器中的增强集划分。由于远程医疗应用中的渠道不当，通过无线网络传输的心电信号会受到噪声的影响。因此，设计了基于快速归一化最小均方 (FNLMS) 算法的自适应滤波器来消除 ECG 信号中不需要的噪声。此外，通过开发收缩结构以提高滤波过程的速度，修改了基于 FNLM 的自适应滤波器的结构。此外，与现有方法相比，为所提出的压缩单元提供了并行和流水线架构，以减少处理时间和面积开销。称为 MIT-BIH 数据库的公开可用数据库和实时 ECG 数据记录用于模拟目的。建议的设计在 Xilinx 平台和 MATLAB 工具上实现。所提出的设计的性能是通过压缩比等参数来估计的，信噪比、均方根误差和均方根差百分比 (PRD)、PRD 归一化和质量得分。将其与现有设计进行比较，以显示我们提出的设计的效率。

更新日期：2022-09-25

点击分享查看原文

点击收藏

阅读更多本刊最新论文本刊介绍/投稿指南11