Probabilistic Cascading Classifier for Energy-Efficient Activity Monitoring in Wearables
Probabilistic Cascading Classifier for Energy-Efficient Activity Monitoring in Wearables
复制标题
用于可穿戴设备节能活动监测的概率级联分类器
DOI：
10.1109/jsen.2022.3175881
复制
发表时间：
2022-07
影响因子：
4.3
通讯作者：
Mahdi Pedram;Ramesh Kumar Sah;Seyed Ali Rokni;Marjan Nourollahi;H. Ghasemzadeh
中科院分区：
文献类型：
--
作者：
Mahdi Pedram;Ramesh Kumar Sah;Seyed Ali Rokni;Marjan Nourollahi;H. Ghasemzadeh
嵌入式系统的进步促使人们在日常生活中集成多种小型健康监测设备。这一趋势导致可穿戴传感器在广泛的应用中不断扩展。可穿戴技术与人体紧密相连，利用传感器和机器学习，通过活动识别和人体运动来描述个人的身体或心理习惯。由于可穿戴设备全天使用，因此这些系统的功耗需要相当低。目前的研究认为，此类机器学习方法是通过固定属性进行训练的，包括传感器采样率和根据时间序列数据计算的统计特征。然而，实际上，由于人类步态和运动的个性化本质，可穿戴设备需要不断重新配置其计算算法。此外，由于这些嵌入式传感器的功率和内存有限，计算算法必须变得节能和内存高效。在本文中，我们提出了一种用于实时、连续和节点上人类活动识别的资源高效框架。通常，活动识别问题是多类分类问题。然而，我们建议将这个基于MET（任务代谢当量）的问题转化为层次分类模型，为每个个体提供个性化的结构。我们讨论了这种新的可配置分类范例的设计和构造。我们的结果表明，针对不同的个性化场景，所提出的概率级联系统在使用有限内存检测活动时的准确度在 94.5% 到 96.9% 之间变化，而与传统方法相比，系统的功耗降低了高达 17.2%。
Advances in embedded systems have given rise to integrating several small-size health monitoring devices within daily human life. This trend led to an ongoing extension of wearable sensors in a broad range of applications. Wearable technologies, which are firmly connected with the human body, utilize sensors and machine learning to describe individuals’ physical or psychological routines through activity recognition and human movement. Since wearables are used all day long, the power consumption of these systems needs to be reasonably low. Current research considers that such machine learning methods are trained with fixed properties, including sensor sampling rate and statistical features computed from the time series data. However, in reality, wearables require continuous reconfiguration of their computational algorithms due to the personalized nature of human gait and movement. Furthermore, computational algorithms must become energy- and memory-efficient due to these embedded sensors’ limited power and memory. In this paper, we propose a resource-efficient framework for real-time, continuous, and on-node human activity recognition. Typically activity recognition problem is a multi-class classification problem. However, we suggest transforming this problem based on MET (Metabolic Equivalent of Task) into a hierarchical classification model, providing personalized structure for each individual. We discuss the design and construction of this new configurable classification paradigm. Our results demonstrate that the proposed probabilistic cascading system accuracy for different personalized scenarios varies between 94.5% and 96.9% in detecting activities using a limited memory, while power usage of the system is reduced by as high as 17.2% compared to the traditional methods.