Device-Free Location-Independent Human Activity Recognition using Transfer Learning Based on CNN

Ding, Xue; Jiang, Ting; Li, Yanan; Xue, Wenling; Zhong, Yi

doi:10.1109/iccworkshops49005.2020.9145092

Cited by 18 publications

(10 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similarly, a limited number of labeled samples from the target domain were used for investigating such issues. In study [13] , a fine-tuning approach was adopted to refine the depth positioning model using labeled samples from the target domain. This entails training initially on the source domain followed by adapting it to match characteristics specific to the target domain through a small set of labeled samples.…”

Section: Related Workmentioning

confidence: 99%

Indoor positioning based on channel state information and deep learning domain adaptation

Zhan,

2024

International Conference on Computer Graphics, Artificial Intelligence, and Data Processing (ICCAID 2023)

View full text Add to dashboard Cite

This article discusses the importance of indoor positioning technology and the challenges it faces, with a particular focus on fingerprint positioning technology based on Channel State Information (CSI). It delves into the rich channel feature information provided by CSI and its applications in indoor environments. However, over time or due to variations in the signal itself, indoor localization deployments may experience a decrease in accuracy or even become ineffective based on the initially collected data. This paper addresses the issue of decreasing model localization effectiveness due to temporal or environmental factors using knowledge from the domain adaptation area of deep learning. The aim of this paper is to minimize the domain discrepancy by reducing the distance in the feature space between the source and target domains, as well as the distance in the semantic feature centroid. As a result, it achieves an improvement of 32.70% and 33.59% over the performance degradation of the original model on the two target domains, respectively. When using only unlabeled data from the target domain, this approach outperforms three other basic deep domain adaptation methods.

show abstract

Section: Related Workmentioning

confidence: 99%

Indoor positioning based on channel state information and deep learning domain adaptation

Zhan,

2024

International Conference on Computer Graphics, Artificial Intelligence, and Data Processing (ICCAID 2023)

View full text Add to dashboard Cite

show abstract

“…Unfortunately, although the knowledge of CSI information after feature extraction can be transferred, the required accuracy cannot be met only by directly transferring the features. Ding et al [24] proposed a semi-supervised WiFi location-independent HAR, called WiLISensing. CNN architecture is used to identify activities in locations that do not require training or have few training samples through transfer learning methods by using a small number of sample transfer datasets to train the fully connected layer behind the network.…”

Section: Related Workmentioning

confidence: 99%

“…MatNet uses one-shot learning technology (one shot learning) to efficiently transfer the environment, but the high-efficiency transfer makes the model's recognition accuracy low; compared with DADA-AD using a small amount of data for fine-tuning, the recognition accuracy of DADA-AD transfer will be improved. Ding et al [24] proposed WiLISensing, which uses supervised transfer learning to improve the efficiency of extracting transferable features, freezes the feature extraction layer, and learns from the label data brought into the target domain. They use label data to improved the recognition accuracy of MatNet, but in the process of learning transferable features, only the results of classification can be used to measure the result of learning classification features.…”

Section: Motivationmentioning

confidence: 99%

“…DADA-AD can more accurately measure the difference between the two domains, and use the embedder to map, so that the time to calculate the similarity is linear. Therefore, this paper proposes the dynamic associate domain adaptation learning using attention-based DenseNet (DADA-AD) scheme to improve the generalization ability, maximize domain confusion, and minimize classification loss for source and target domains, as shown in Figure 2, combining the advantage of [19,24,25] to achieve the identification result that is not affected by the environment.…”

Section: Motivationmentioning

confidence: 99%

“…The data augmentation expanded the training data 1.5 times. In the pre-training result, the red line represents as our attention-based DenseNet (AD) scheme, the blue line represents the HAR-MN-EF scheme [19], the green line represents the DANDR scheme [25], and the orange line represents the WiLlSensing scheme [24]. Figure 18a illustrates that the PTA of AD is better than that of other schemes due to the advantage of the feature attention and reuse strategy.…”

Section: Performance Analysismentioning

confidence: 99%

See 2 more Smart Citations

A Semi-Supervised Transfer Learning with Dynamic Associate Domain Adaptation for Human Activity Recognition Using WiFi Signals

Chen

Chang

2021

Sensors

View full text Add to dashboard Cite

Human activity recognition without equipment plays a vital role in smart home applications, freeing humans from the shackles of wearable devices. In this paper, by using the channel state information (CSI) of the WiFi signal, semi-supervised transfer learning with dynamic associate domain adaptation is proposed for human activity recognition. In order to improve the CSI quality and denoising of CSI, we carried out missing packet filling, burst noise removal, background estimation, feature extraction, feature enhancement, and data augmentation in the data pre-processing stage. This paper considers the problem of environment-independent human activity recognition, also known as domain adaptation. The pre-trained model is trained from the source domain by collecting a complete labeled dataset of all of the CSI of human activity patterns. Then, the pre-trained model is transferred to the target environment through the semi-supervised transfer learning stage. Therefore, when humans move to different target domains, a partial labeled dataset of the target domain is required for fine-tuning. In this paper, we propose a dynamic associate domain adaptation called DADA. By modifying the existing associate domain adaptation algorithm, the target domain can provide a dynamic ratio of labeled dataset/unlabeled dataset, while the existing associate domain adaptation algorithm only allows target domains with the unlabeled dataset. The advantage of DADA is that it provides a dynamic strategy to eliminate different effects on different environments. In addition, we further designed an attention-based DenseNet model, or AD, as our training network, which is modified by an existing DenseNet by adding the attention function. The solution we proposed was simplified to DADA-AD throughout the paper. The experimental results show that for domain adaptation in different domains, the accuracy of human activity recognition of the DADA-AD scheme is 97.4%. It also shows that DADA-AD has advantages over existing semi-supervised learning schemes.

show abstract

Towards Cross Domain CSI Action Recognition Through One-Shot Bimodal Domain Adaptation

Zhang

Zhou

Luo

et al. 2023

Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering

View full text Add to dashboard Cite

Device-Free Location-Independent Human Activity Recognition using Transfer Learning Based on CNN

Cited by 18 publications

References 14 publications

Indoor positioning based on channel state information and deep learning domain adaptation

Indoor positioning based on channel state information and deep learning domain adaptation

A Semi-Supervised Transfer Learning with Dynamic Associate Domain Adaptation for Human Activity Recognition Using WiFi Signals

Towards Cross Domain CSI Action Recognition Through One-Shot Bimodal Domain Adaptation

Contact Info

Product

Resources

About