Scholarlink

Improving the power efficiency of battery-powered wearable medical devices is highly important. This paper proposes an approximate squarer that aims to enhance power efficiency in the Pan-Tompkins algorithm, a key component in arrhythmia detection. It combines (1) the DRUM architecture, (2) the recursive structure, (3) an error recovery technique, and (4) folding and merged techniques for squaring operations to improve power efficiency while maintaining satisfactory arrhythmia detection accuracy. From our experimental evaluations, we find that the proposed approximate squarer achieves a 93.6% power reduction and a 79.1% area reduction, while maintaining the target recall of 93%, defined as detection accuracy.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A Low-Power and Small-Area Squarer for Wearable ECG Devices Utilizing Pan-Tompkins Algorithm

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Toshinori Sato,
  • Taiki Nagatomo,
  • Hiroyuki Hama

摘要

Improving the power efficiency of battery-powered wearable medical devices is highly important. This paper proposes an approximate squarer that aims to enhance power efficiency in the Pan-Tompkins algorithm, a key component in arrhythmia detection. It combines (1) the DRUM architecture, (2) the recursive structure, (3) an error recovery technique, and (4) folding and merged techniques for squaring operations to improve power efficiency while maintaining satisfactory arrhythmia detection accuracy. From our experimental evaluations, we find that the proposed approximate squarer achieves a 93.6% power reduction and a 79.1% area reduction, while maintaining the target recall of 93%, defined as detection accuracy.