Page 10 - 2025S

UEC Int’l Mini-Conference No.54 3

5. Result However, the accuracy of the algorithm needs to be
The two signals obtained after processing by the ICA algorithm improved to be able to measure respiratory signals on real
are compared with the original signal. human in the near future.

References
[1] Lee, Y.S., Pathirana, P.N., Evans, R.J. et al. Separation of
Doppler radar-based respiratory signatures. Med Biol Eng
Comput 54, 1169–1179 (2016).
https://doi.org/10.1007/s11517-015-1379-3
[2] Yi, Y., Billor, N., Ekstrom, A. et al. CW_ICA: an efficient
dimensionality determination method for independent
component analysis. Sci Rep 14, 143 (2024).
https://doi.org/10.1038/s41598-023-49355-z.
[3] Song C, Yavari E, Gao X, Lubecke VM, Boric-Lubecke O.
Respiration Signal Pattern Analysis for Doppler Radar Sensor
(a) with Passive Node and Its Application in Occupancy Sensing
of a Stationary Subject. Biosensors (Basel). 2025 Apr
27;15(5):273. doi: 10.3390/bios15050273. PMID: 40422012;
PMCID: PMC12109633.
[4] S. M. M. Islam and V. M. Lubecke, "Extracting Individual
(b) Respiratory Signatures from Combined Multi-Subject
Mixtures with Varied Breathing Pattern Using Independent
Component Analysis with the JADE Algorithm," 2020 IEEE
Asia-Pacific Microwave Conference (APMC), Hong Kong,
Hong Kong, 2020, pp. 734-736,
(c) doi: 10.1109/APMC47863.2020.9331715.
[5] S. M. M. Islam, E. Yavari, A. Rahman, V. M. Lubecke and
Figure 4 . Result: (a) Signals of each object after using ICA, O. Boric-Lubecke, "Separation of Respiratory Signatures for
(b) First original signal, (c) Second original signal Multiple Subjects Using Independent Component Analysis
with the JADE Algorithm," 2018 40th Annual International
The results show that the signal after processing is similar to the Conference of the IEEE Engineering in Medicine and Biology
original signal in terms of frequency but has differences in Society (EMBC), Honolulu, HI, USA, 2018, pp. 1234-1237,
amplitude and waveform, signal phase, and the order of the doi: 10.1109/EMBC.2018.8512583
signals changes differently after each processing. [6] S. M. M. Islam, O. Boric-Lubecke and V. M. Lubekce,
6. Conclusion "Concurrent Respiration Monitoring of Multiple Subjects by
Phase-Comparison Monopulse Radar Using Independent
This result shows that the method of using ICA to separate
Component Analysis (ICA) With JADE Algorithm and
respiratory signals from multiple subjects is a very Direction of Arrival (DOA)," in IEEE Access, vol. 8, pp.
promising research direction. 73558-73569, 2020, doi: 10.1109/ACCESS.2020.2988038.

Figure 5. Future work

*The author is supported by (GECHA) MEXT Scholarship