Classification was carried out by calculating the Euclidean distance between the stored templates and feature vectors extracted from the test signal. So as to combine the matching distances or scores of the three feature vectors, score normalisation was carried out. The normalisation of matching distances was done using Min-Max normalisation, such that given a set of matching distances {dk}, к = 1, 2, Ъ...п, the normalised distances were obtained by utilising the expression mentioned below:

Comparison of feature vectors for two subjects

FIGURE 13.5 Comparison of feature vectors for two subjects.

where sk is the matching score after normalisation, and min and max are the minimum and maximum values of [dk], respectively. Score-level fusion by employing simple weighted fusion was adopted for the final identification or verification decision. The combined score was computed as follows:

where sh s2, and s3 are the normalised scores of matching distances for the corresponding three feature vectors; vvt, vv2, and vv3 are their weights; and S is the combined score. Experiments were conducted to compute identification rate as well as the verification rate.


The approach was tested on a dataset consisting of samples from eleven subjects recorded in two sessions. Evaluations were carried out for both within-session and across-session settings. The template for each subject was created from the first part of the recording of first session, and the same template was also used for across- session testing. The feature vectors for the within-session testing were extracted from the last 10 seconds of the input signal. The across-session testing was carried out by extracting the features vectors for the signal samples taken from 10-to 20-second segment of the second set of recordings.

The results achieved for the identification and verification tasks have been listed in Table 13.3. For the within-session experiments, the rank 1 identification rate of 81.82% (9 out of 11) was achieved for the method described in this section. On the other hand, the verification rate for this setting was 90.91%. When the testing was done for across-session setting, the rank 1 identification rate dropped to 54.55% (6 out of 11). The verification performance for across session also dipped to 63.64%. Further analysis of the results for identification accuracy can be done by viewing the

TABLE 13.3

Accuracy for Within- and Across-Session Testing



Within Session (%)

Across Session (%)







CMC plot for within and across session testing

FIGURE 13.6 CMC plot for within and across session testing.

Cumulative Match Curve (CMC) plot shown in Figure 13.6. It can be seen that for within session, 100% subjects are recognised by rank 3, and for the across-session scenario, 8 out of the 11 subjects are identified by rank 5. This shows that PPG signal contains content that is characteristic of a particular human being. In this case, the recognition tasks have been performed using simple Euclidean distance classifier. As in the reported literature, machine learning and deep learning techniques have been found to have better classification abilities, so it is expected that using them as classifiers should provide higher accuracy.


A review of evolution of PPG for biometric application, since its inception in this area, has been presented in this chapter. An approach using multiple feature vectors for PPG biometric has also been explained. Based upon the survey of literature and results obtained for the method described here, it can be stated that PPG signal has a potential to be utilised for the biometric applications. Moreover, the accuracy for the across-session testing is expected to improve by using robust classifiers offered by the new age machine and deep learning techniques. However, the studies published in this area have been carried out on small datasets with less than fifty subjects. In order to establish the efficacy of the methods, experiments need to be carried out on datasets having multiple session recordings and population size of few hundreds. The future work in this direction will be undertaken by keeping these points in mind.


  • 1. Chauhan, S., Arora, A.S., &Kaul, A. (2010) A survey of emerging biometric modalities. Proceedings of the International Conference and Exhibition on Biometrics Technology.
  • 2. Kaul, A. (2016). Hybrid Biometric Approach for Human Identification (Doctoral Thesis). National Institute of Technology Hamirpur, India.
  • 3. Elgendi, M. (2012). Standard terminologies for photoplethysmogram signals. Current Cardiology Reviews, 8(3). 215-219.
  • 4. Gu. Y. Y.. Zhang, Y., & Zhang, Y. T. (2003, April). A novel biometric approach in human verification by photoplethysmographic signals. 4th International IEEE EMBS Special Topic Conference on Information Technology Applications in Biomedicine, 2003. (pp. 13-14). IEEE.
  • 5. Gu, Y. Y., & Zhang, Y. T. (2003, October). Photoplethysmographic authentication through fuzzy logic. IEEE EMBS Asian-Pacific Conference on Biomedical Engineering, 2003. (pp. 136-137). IEEE.
  • 6. Bao, S. D„ Zhang, Y. T, & Shen, L. F. (2006, January). Physiological signal based entity authentication for body area sensor networks and mobile healthcare systems. 2005 27th Annual IEEE Conference on Engineering in Medicine and Biology (pp. 2455-2458). IEEE.
  • 7. Yao, J., Sun, X., & Wan, Y. (2007, August). A pilot study on using derivatives of photoplethysmographic signals as a biometric identifier. 2007 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (pp. 4576-4579). IEEE.
  • 8. Wan, Y., Sun, X., &Yao, J. (2007, October). Design of a photoplethysmographic sensor for biometric identification. 2007 International Conference on Control, Automation and Systems (pp. 1897-1900). IEEE.
  • 9. Spachos, P., Gao, J., & Hatzinakos, D. (2011, July). Feasibility study of photoplethysmographic signals for biometric identification. 2011 17tli International Conference on Digital Signal Processing (DSP) (pp. 1-5). IEEE.
  • 10. Singh. M., & Gupta, S. (2012). Correlation studies of PPG finger pulse profiles for Biometric system. International Journal of Computer Science and Knowledge Management, 5(1), 1-3.
  • 11. Bonissi, A., Labati, R. D„ Perico, L„ Sassi, R„ Scotti, F.. & Sparagino. L. (2013, September). A preliminary study on continuous authentication methods for photoplethysmographic biometrics. 2013 IEEE Workshop on Biometric Measurements and Systems for Security and Medical Applications (pp. 28-33). IEEE.
  • 12. Kavsaoglu, A. R., Polat, K., & Bozkurt, M. R. (2014). A novel feature ranking algorithm for biometric recognition with PPG signals. Computers in Biology and Medicine, 49, 1-14.
  • 13. da Silva Dias, J.. Traore, I., Ferreira, V. G.. & David. J. (2015). Exploratory use of PPG signal in continuous authentication. The Brazilian Symposium on Information and Computational Systems Security. Florianopolis, Brazil: The Brazilian Symposium on Information and Computational Systems Security.
  • 14. Lee, A., &Kim, Y. (2015, November). Photoplethysmography as a form of biometric authentication. In 2015 IEEE Sensors (pp. 1-2). IEEE.
  • 15. Nadzri, N. I. M., & Sidek, K. A. (2016). Photoplethysmogram based biometric identification for twins incorporating gender variability. Journal of Telecommunication, Electronic and Computer Engineering (JTEC), #(12), 67-72.
  • 16. Chakraborty, S., & Pal, S. (2016, January). Photoplethysmogram signal based biometric recognition using linear discriminant classifier. 2016 2nd International Conference on Control, Instrumentation, Energy & Communication (CIEC) (pp. 183-187). IEEE.
  • 17. Choudhary, T., &Manikandan, M. S. (2016, March). Robust photoplethysmographic (PPG) based biometric authentication for wireless body area networks and m-health applications. 2016 Twenty Second National Conference on Communication (NCC) (pp. 1-6). IEEE.
  • 18. Sarkar, A., Abbott, A. L., &Doerzaph, Z. (2016, September). Biometric authentication using photoplethysmography signals. 2016 IEEE 8th International Conference on Biometrics Theory, Applications and Systems (BTAS) (pp. 1-7). IEEE.
  • 19. Namini, S. P. M., & Rashidi, S. (2016, October). Implementation of artificial features in improvement of biometrics based PPG. 2016 6th International Conference on Computer and Knowledge Engineering (ICCKE) (pp. 342-346). IEEE.
  • 20. Karimian, N.. Tehranipoor. M„ & Forte. D. (2017, February). Non-fiducial PPG-based authentication for healthcare application. 2017 IEEE EMBS International Conference on Biomedical & Health Informatics (BHI) (pp. 429-432). IEEE.
  • 21. Karimian, N.. Guo, Z., Tehranipoor, M., & Forte, D. (2017, March). Human recognition from photoplethysmography (PPG) based on non-fiducial features. 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (pp. 4636-4640). IEEE.
  • 22. Nowara, E. M., Sabharwal, A., & Veeraraghavan, A. (2017, May). PPG secure: Biometric presentation attack detection using photopletysmograms. 201712th IEEE International Conference on Automatic Face & Gesture Recognition (FG 2017) (pp. 56-62). IEEE.
  • 23. Reddy, V. R., Deshpande, P, & Pal, A. (2017, November). Simultaneous measurement and correlation of PPG signals taken from two different body parts for enhanced biometric security via two-level authentication. Proceedings of the 1st ACM Workshop on the Internet of Safe Things (pp. 32-37).
  • 24. Sidek, K. A., Kamaruddin. N.K., & Ismail. A.F. (2018). The study of PPG and APG signals for biometric recognition. Journal of Telecommunication, Electronic and Computer Engineering, 10, 17-20.
  • 25. Yadav, U„ Abbas, S. N.. & Hatzinakos, D. (2018, February). Evaluation of PPG biometrics for authentication in different states. 2018 International Conference on Biometrics (ICB) (pp. 277-282). IEEE.
  • 26. Sancho, J., Alesanco, A., & Garcia, J. (2018). Biometric authentication using the PPG: a long-term feasibility study. Sensors, 18(5), 1525.
  • 27. Al-Sidani, A., Ibrahim, B., Cherry, A., & Hajj-Hassan. M. (2018, April). Biometric identification using photoplethysmography signal. 2018 Third International Conference on Electrical and Biomedical Engineering, Clean Energy and Green Computing (EBECEGC) (pp. 12-15). IEEE.
  • 28. Zhao, T, Wang, Y„ Liu, J.. & Chen, Y. (2018, October). Your heart won't lie: PPG- based continuous authentication on wrist-worn wearable devices. Proceedings of the 24th Annual International Conference on Mobile Computing and Networking (pp. 783-785).
  • 29. Lee. S. W„ Woo, D. K„ Son, Y. K„ & Mah. P. S. (2019). Wearable Bio-Signal (PPG)- based personal authentication method using random forest and period setting considering the feature of PPG signals. JCP. 14(4), 283-294.
  • 30. Wang, K., & Chen. X. (2019, October). PPG signal identification method based on CSASVM. In 2019 IEEE 10th International Conference on Software Engineering and Service Science (ICSESS) (pp. 1-4). IEEE.
  • 31. Walia, A., & Kaul, A. (2019, October). Human recognition via PPG signal using temporal correlation. 2019 5th International Conference on Signal Processing, Computing and Control (ISPCC) (pp. 144-147). IEEE.
  • 32. Cheng, S., Chou, Y„ Liu, J., Gu, Y., & Huang, X. (2019, October). A novel identity authentication method by modeling Photoplethysmograph waveform. 2019 International Conference on Control, Automation and Information Sciences (ICCAIS) (pp. 1-5). IEEE.
  • 33. Al Sidani. A.. Cherry, A., Hajj-Hassan, H., & Hajj-Hassan. M. (2019, October). Comparison between К-nearest neighbor and support vector machine algorithms for PPG biometric identification. 2019 Fifth International Conference on Advances in Biomedical Engineering (ICABME) (pp. 1-4). IEEE.
  • 34. Khan. M. U„ Aziz, S„ Naqvi, S. Z. H., Zaib. A.. & Maqsood. A. (2020, March). Pattern analysis towards human verification using Photoplethysmograph signals. 2020 International Conference on Emerging Trends in Smart Technologies (ICETST) (pp. 1-6). IEEE.
  • 35. Jindal, V., Birjandtalab, J., Pouyan, M. B., & Nourani, M. (2016, August). An adaptive deep learning approach for PPG-based identification. 2016 38th Annual International Conference of the IEEE on Engineering in Medicine and Biology Society (EMBC) (pp. 6401-6404). IEEE.
  • 36. Everson, L., Biswas, D., Panwar, M., Rodopoulos, D., Acharyya, A., Kim, С. H., & Van Helleputte, N. (2018, May). BiometricNet: Deep Learning based biometric identification using wrist-worn PPG. 2018 IEEE International Symposium on Circuits and Systems (ISCAS) (pp. 1-5). IEEE.
  • 37. Luque, J., Cortes, G., Segura, C., Maravilla, A., Esteban, J., & Fabregat, J. (2018, September). End-to-End Photoplethysmography (PPG) based biometric authentication by using convolutional neural networks. 2018 26th European Signal Processing Conference (EUSIPCO) (pp. 538-542). IEEE.
  • 38. Patil, O. R.. Wang. W„ Gao, Y., Xu. W., & Jin, Z. (2018, October). A non-contact PPG biometric system based on deep neural network. 2018 IEEE 9th International Conference on Biometrics Theory, Applications and Systems (BTAS) (pp. 1-7). IEEE.
  • 39. Biswas, D., Everson, L., Liu, M., Panwar, M., Verhoef, В. E., Patki, S., & Van Helleputte, N. (2019). CorNET: Deep learning framework for PPG-based heart rate estimation and biometric identification in ambulant environment. IEEE Transactions on Biomedical Circuits and Systems, 13(2), 282-291.
  • 40. Hwang, D. Y., & Hatzinakos, D. (2019, May). PPG-based personalized verification system. In 2019 IEEE Canadian Conference of Electrical and Computer Engineering (CCECE) (pp. 1-4). IEEE.
  • 41. Sebastia, G. C. (2018). End-to-End Photoplethysmography-based Biometric Authentication System by Using Deep Neural Networks (Doctoral dissertation, Universitat Politecnica de Catalunya. EscolaTecnica Superior d'Enginyeria de Telecomunicacio de Barcelona).
< Prev   CONTENTS   Source   Next >