**6. Limitations**

Although the proposed iSTCA system realizes quantitative representation for exposed virus concentrations with the help of the landmark-calibrated PDR technique, there are some challenges that need to be overcome. First of all, there are some strict restrictions in the data acquisition process. The participants are required to hold the smartphone, specifically the Pixel 4a, at chest level. As a result, except for the diversity of users considered, other factors that affect the motion sensor readings are not seriously taken into account, such as the mobile device heterogeneity (e.g., different types or various vendors) and the device's status variation (e.g., putting in a pocket or handbag). In addition, a large amount of power of the smartphone is consumed during the indoor positioning process, resulting in the smartphone being overheated.

### **7. Conclusions and Future Work**

Technology-assisted virus exposure tracking approaches are increasingly being adopted to mitigate and tame the epidemic. In view of the complexity of quantifying virus exposure due to human movement and airborne dispersion of virus particles, we propose iSTCA, a self-containing contact awareness approach that exploits PDR-based techniques. Quantitative information support directly concerned with risk assessment is provided for self-protection and epidemic control. More precisely, to reduce and calibrate the accumulative errors of trajectories based on landmarks, we apply Bi-LSTM and multi-head CNN with residual concatenation to long-term dependency in forward and backward directions and extract local correlations at various resolutions for landmark identification. The proposed method exploits the trajectories of people with viral-laden droplets exhaled and the transmission and attenuation of viruses in the air to quantify the virus quanta concentration in an indoor environment via spatiotemporal analytics for prevention and sanitization. In future work, we will continue studying the landmark identification model with different devices and various attitudes of the device and conduct further research on the exploration of other advanced deep neural networks and fusion algorithms. We will consider employing wearable devices, such as smartwatches and smart bands, to replace mobile phones for power saving in indoor positioning. Moreover, we plan to apply the proposed techniques for the development of services in developing communities without reliable digital infrastructure.

**Author Contributions:** Conceptualization, L.G. and S.K.; methodology, L.G. and S.K.; software, L.G.; validation, L.G.; formal analysis, L.G.; investigation, L.G.; resources, L.G. and S.K.; data curation, L.G.; writing—original draft preparation, L.G.; writing—review and editing, S.K.; visualization, L.G.; supervision, S.K.; project administration, S.K.; funding acquisition, S.K. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was funded by Japan Society for the Promotion of Science (JSPS), Grants-in-Aid for Scientific Research (KAKENHI), Japan: Grant Number JP20H00622 and supported in part by China Scholarship Council (CSC), China: Grant Number 202008050086.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Informed consent was obtained from all subjects involved in the study.

**Data Availability Statement:** The data presented in this study are available upon request from the corresponding author. The data are not publicly available due to privacy reasons.

**Conflicts of Interest:** The authors declare no conflict of interest.
