Abstract
With the fast increase in the demand for location-based services and the proliferation of smartphones, the topic of indoor localization is attracting great interest. In indoor environments, users’ performed activities carry useful semantic information. These activities can then be used by indoor localization systems to confirm users’ current relative locations in a building. In this paper, we propose a deep-learning model based on a Convolutional Long Short-Term Memory (ConvLSTM) network to classify human activities within the indoor localization scenario using smartphone inertial sensor data. Results show that the proposed human activity recognition (HAR) model accurately identifies nine types of activities: not moving, walking, running, going up in an elevator, going down in an elevator, walking upstairs, walking downstairs, or going up and down a ramp. Moreover, predicted human activities were integrated within an existing indoor positioning system and evaluated in a multi-story building across several testing routes, with an average positioning error of 2.4 m. The results show that the inclusion of human activity information can reduce the overall localization error of the system and actively contribute to the better identification of floor transitions within a building. The conducted experiments demonstrated promising results and verified the effectiveness of using human activity-related information for indoor localization.
Highlights
Human activities have been commonly used to define a vast number of human behavioral patterns
The proposed human activity recognition (HAR) model was integrated in a fingerprinting-based indoor positioning system based on the work of Guimarães et al [10] and on the improvements of step detection and particle filter algorithms from the work of Santos et al [27]
To train and evaluate the proposed model, two different datasets were considered: (1) an HAR dataset containing diverse motion data recorded by volunteers performing nine activities, and (2) an indoor positioning dataset containing multiple floor transitions for assessing HAR model integration within the Indoor Positioning Systems (IPS) scenario
Summary
Human activities have been commonly used to define a vast number of human behavioral patterns. A CNN is usually composed of two parts: convolution and pooling operations. These operations are applied consecutively and act as a deep feature extractor of the raw data. These features are connected to a fully connected layer that produces the output classification (see Figure 1). The convolutional layer performs convolution operations on the data of the preceding layer with a given convolution filter and activation function, generating a set of different feature maps. The pooling layer performs a downsampling operation with a given kernel/size that reduces the size of the generated feature maps by representing its values by its average or maximum value according to the kernel [21,22,23]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
