International Journal of Electrical and Computer Engineering (IJECE) Vol. 12, No. 4, August 2022, pp. 4391~4399 ISSN: 2088-8708, DOI: 10.11591/ijece.v12i4.pp4391-4399  4391 Journal homepage: http://ijece.iaescore.com Principal coefficient encoding for subject-independent human activity analysis Pang Ying Han 1 , Sarmela Anak Perempuan Raja Sekaran 1 , Ooi Shih Yin 1 , Tan Teck Guang 2 1 Faculty of Information Science and Technology, Multimedia University, Melaka, Malaysia 2 DRSoft Sdn Bhd, Melaka, Malaysia Article Info ABSTRACT Article history: Received Apr 30, 2021 Revised Jan 19, 2022 Accepted Feb 3 2022 Tracking human physical activity using smartphones is an emerging trend in healthcare monitoring and healthy lifestyle management. Neural networks are broadly used to analyze the inertial data of activity recognition. Inspired by the autoencoder neural networks, we propose a layer-wise network, namely principal coefficient encoder model (PCEM). Unlike the vanilla neural networks which apply random weight initialization and back- propagation for parameter updating, an optimized weight initialization is implemented in PCEM via principal coefficient learning. This principal coefficient encoding allows rapid data learning with no back-propagation intervention and no gigantic hyperparameter tuning. In PCEM, the most principal coefficients of the training data are determined to be the network weights. Two hidden layers with principal coefficient encoding are stacked in PCEM for the sake of deep architecture design. The performance of PCEM is evaluated based on a subject-independent protocol where training and testing samples are from different users, with no overlapping subjects in between the training and testing sets. This subject-independent protocol can better assess the generalization of the model to new data. Experimental results exhibit that PCEM outperforms certain state-of-the-art machine learning and deep learning models, including convolutional neural network, and deep belief network. PCEM can achieve ~97% accuracy in subject- independent human activity analysis. Keywords: 1D inertial motion data Autoencoder Deep analytic model Principal coefficient encoding Subject-independent This is an open access article under the CC BY-SA license. Corresponding Author: Pang Ying Han Faculty of Information Science and Technology, Multimedia University 75450 Ayer Keroh, Melaka, Malaysia Email: yhpang@mmu.edu.my 1. INTRODUCTION Human activity recognition (HAR) becomes more prominent with the increasing and advancement of the smart home concept, healthcare monitoring and healthy lifestyle management [1]. Using remote healthcare monitoring tools on smartphones is very common and prevalent in remote mobile health monitoring (RMHM) systems. The adoption of smartphones in HAR is made competent with its ability to capture motion data through its multiple inertial sensors as well as its significant attachment in our daily life [2]. With a simple and straightforward installation process, HAR app can be activated on the smartphone and running in the background to track our physical activity. Smartphone-based HAR is usually developed in four phases: data acquisition, data segmentation, feature extraction and classification. During data acquisition, various factors such as the position of the smartphone, and collection frequency are considered. In literature, different positions’ placements of the smartphone have been investigated, e.g.: in front pocket, back pocket, in hand, on waist, typing, and phoning