YAYGIN VERİ SINIFLANDIRMA ALGORİTMALARININ FreeRTOS İŞLETİM SİSTEMİ ÜZERINDE DENEYSEL PERFORMANS DEĞERLENDİRMESİ

Abstract

Bu araştırmanın temel motivasyonu, çeşitli Makine Öğrenimi ve Sinir Ağı algoritmaları gibi veri sınıflandırma algoritmalarının gerçek zamanlı senaryolar özelinde hayati kritik (safety critical) sistemlerde maliyet-etkinliğini değerlendirmektir. Bunun gerçekleştirilmesi için geleneksel veri sınıflandırma algoritmaları ayrı parçalara bölünmüş ve her bir parça Gerçek Zamanlı İşletim Sistemi'nin (RTOS) belirli bir iş parçacığına (thread) atanmıştır. Algoritmalar, dört farklı orta büyüklükteki kaggle veri kümesinde K katlı çapraz doğrulama kullanılarak eğitilmiş ve test edilmiştir. Gerçek zamanlı uygulama, C++ 20 programlama dili kullanılarak FreeRTOS üzerinde gerçekleştirilmiştir. Deney, ARM Cortex M4 işlemcili FreeRTOS platformu ve Linux platformu üzerinden simüle edilmiştir. Güvenli veri haberleşmesini sağlamak adına algoritmalardan faydalanılmıştır. Çıktılar, FreeRTOS tarafından bir karışıklık matrisinde (confusion matrix) toplanmıştır. Tüm algoritmalara ait performanslar değerleri tablolar ve grafiklerle sunulmuş olup, Naif Bayes algoritması ona en yakın algoritmadan 13 kat daha hızlı ve daha doğru sonuç vererek gerçek zamanlı uygulamalar için ideal bir algoritma olarak belirtilmiştir. Rastgele Orman algoritması parametrelerinden biri olan tahmin edici karar ağacı sayısı 5 seçilmesine rağmen performans metriklerinde yetersizlik görülmediği sunulmuştur. Bu çalışmada benimsenen yaklaşım, planlanabilir analiz temelinde veri sınıflandırmasını analiz etmek için umut vadeden potansiyele sahiptir. Ayrıca, farklı veri sınıflandırma algoritmalarının gerçek zamanlı olarak karşılaştırılmasını mümkün kılmaktadır.

Keywords

• : Gerçek zamanlı işletim sistemi
• veri sınıflandırması
• makine öğrenmesi
• sinir ağları
• C++20

An Experimental Performance Evaluation of Common Data Classification Algorithms on FreeRTOS

Abstract

The primary motivation of this research is to evaluate the cost-effectiveness of data classification algorithms—such as various Machine Learning and Neural Network methods—in safety-critical systems under real-time conditions. To achieve this, traditional data classification algorithms were modularized, with each component assigned to a specific thread within a Real-Time Operating System (RTOS). The algorithms were trained and tested using K-fold cross-validation on four medium-sized kaggle datasets. The real-time application was developed on FreeRTOS using the C++20 programming language. Experiments were simulated both on the FreeRTOS platform and on a Linux platform equipped with an ARM Cortex-M4 processor. The algorithms were employed to ensure secure data communication, and the output results were captured in a confusion matrix generated by FreeRTOS. Performance metrics for all algorithms are presented in tables and graphs. Among them, the Naive Bayes algorithm emerged as the most suitable for real-time applications, delivering results that were 13 times faster and more accurate than the next best algorithm. Notably, even when the number of decision trees in the Random Forest algorithm was limited to five, performance metrics showed no significant degradation. The approach adopted in this study demonstrates promising potential for analyzing data classification through schedulability analysis. It also enables effective real-time comparisons between different classification algorithms.

Keywords

• Real-time operating system
• data classification
• machine learning
• neural network
• C++20

References

1. Ampomah, E. K., Nyame, G., Qin, Z., Addo, P. C., Gyamfi, E. O., & Gyan, M. (2021). Stock market prediction with gaussian naïve bayes machine learning algorithm. Informatica, 45(2).
2. Armano, G., & Manconi, A. (2023). Devising novel performance measures for assessing the behavior of multilayer perceptrons trained on regression tasks. Plos one, 18(5), e0285471.
3. Bastide, R., Navarre, D., Palanque, P., Schyn, A., & Dragicevic, P. (2004, October). A model-based approach for real-time embedded multimodal systems in military aircrafts. In Proceedings of the 6th international conference on Multimodal interfaces (pp. 243-250).
4. Blanc, G., Lange, J., Qiao, M., & Tan, L. Y. (2022). Properly learning decision trees in almost polynomial time. Journal of the ACM, 69(6), 1-19.
5. Carvalho, D. V., Pereira, E. M., & Cardoso, J. S. (2019). Machine learning interpretability: A survey on methods and metrics. Electronics, 8(8), 832.
6. Clements, A. A., Almakhdhub, N. S., Saab, K. S., Srivastava, P., Koo, J., Bagchi, S., & Payer, M. (2017, May). Protecting bare-metal embedded systems with privilege overlays. In 2017 IEEE Symposium on Security and Privacy (SP) (pp. 289-303). IEEE.
7. Demirezen, M. U., Civrizoğlu, A., & Yavanoğlu, U. (2021). Sualtı objelerinin makine öğrenmesi yöntemleri ile tespitinde zaman serisi-görüntü dönüşümü tabanlı yeni yaklaşımlar. Gazi Üniversitesi Mühendislik Mimarlık Fakültesi Dergisi, 36(3), 1399-1416.
8. Dwivedi, U. K., Kumar, A., & Sekimoto, Y. (2023). Real-time classification of longitudinal conveyor belt cracks with deep-learning approach. PloS one, 18(7), e0284788.

9. Fischer, J., Wirtz, S., & Scherer, V. (2023). Random forest classifier and neural network for fraction identification of refuse-derived fuel images. Fuel, 341, 127712.
10. Fisher, R. A., & Marshall, M. (1936). Iris data set. RA Fisher, UC Irvine Machine Learning Repository, 440, 87. FreeRTOS. (n.d.). Retrieved from https://www.freertos.org/
11. Hambarde, P., Varma, R., & Jha, S. (2014, January). The survey of real time operating system: RTOS. In 2014 International Conference on Electronic Systems, Signal Processing and Computing Technologies (pp. 34-39). IEEE.
12. Hrgarek, N. (2012, June). Certification and regulatory challenges in medical device software development. In 2012 4th International Workshop on Software Engineering in Health Care (SEHC) (pp. 40-43). IEEE.
13. Ibrahim, S. W., Djemal, R., Alsuwailem, A., & Gannouni, S. (2017). Electroencephalography (EEG)-based epileptic seizure prediction using entropy and K-nearest neighbor (KNN). Communications in Science and Technology, 2(1).
14. Jeddi, A. B., Shafieezadeh, A., & Nateghi, R. (2023). PDP-CNN: A Deep Learning Model for Post-hurricane Reconnaissance of Electricity Infrastructure on Resource-constrained Embedded Systems at the Edge. IEEE Transactions on Instrumentation and Measurement.
15. Joachims, T. (1998). Making large-scale SVM learning practical (No. 1998, 28). Technical report. Kaggle. (n.d.). Retrieved from https://www.kaggle.com/datasets/anjanisona/bill-authentication,
16. https://www.kaggle.com/code/siddheshkadam/classification-in-asteroseismology/input,
17. https://www.kaggle.com/datasets/denisadutca/customer-behaviour,
18. https://www.kaggle.com/datasets/fedesoriano/heart-failure-prediction Karri, M., & Annavarapu, C. S. R. (2023). A real-time embedded system to detect QRS-complex and arrhythmia classification using LSTM through hybridized features. Expert Systems with Applications, 214, 119221. Lee, D., Subramanian, L., Ausavarungnirun, R., Choi, J., & Mutlu, O. (2015, October). Decoupled direct memory access: Isolating CPU and IO traffic by leveraging a dual-data-port DRAM. In 2015 International Conference on Parallel Architecture and Compilation (PACT) (pp. 174-187). IEEE.
19. Li, Q., & Yao, C. (2003). Real-time concepts for embedded systems. CRC press.
20. Liao, T., Lei, Z., Zhu, T., Zeng, S., Li, Y., & Yuan, C. (2021). Deep Metric Learning for K Nearest Neighbor Classification. IEEE Transactions on Knowledge and Data Engineering, 35(1), 264-275.
21. Michelucci, U. (2022). An introduction to autoencoders. arXiv preprint arXiv:2201.03898.
22. Mooney, V. J., & Blough, D. M. (2002). A hardware-software real-time operating system framework for SoCs. IEEE Design & Test of Computers, 19(6), 44-51.
23. Murikipudi, A., Prakash, V., & Vigneswaran, T. (2015). Performance analysis of real time operating system with general purpose operating system for mobile robotic system. Indian Journal of Science and Technology, 8(19), 1-6.
24. Nuratch, S. (2018, May). Design and Implementation of Real-time Embedded Data Acquisition and Classification with Web-based Configuration and Visualization. In 2018 International Conference on Embedded Systems and Intelligent Technology & International Conference on Information and Communication Technology for Embedded Systems (ICESIT-ICICTES) (pp. 1-4). IEEE.
25. Okas, Paweł, Łukasz Krzak, & Cezary Worek. (2015). C++ 14 concurrency on ARM Cortex-M using FreeRTOS and GCC. IFAC-PapersOnLine, 48(4), 262-267.
26. Prisaznuk, P. J. (2008, October). ARINC 653 role in integrated modular avionics (IMA). In 2008 IEEE/AIAA 27th Digital Avionics Systems Conference (pp. 1-E). IEEE.
27. Regehr, J. (2005, September). Random testing of interrupt-driven software. In Proceedings of the 5th ACM International Conference on Embedded software (pp. 290-298).
28. Reinders, J., Ashbaugh, B., Brodman, J., Kinsner, M., Pennycook, J., & Tian, X. (2021). Data parallel C++: mastering DPC++ for programming of heterogeneous systems using C++ and SYCL (p. 548). Springer Nature.
29. Shafiullah, G. M., Gyasi-Agyei, A., & Wolfs, P. (2007, August). Survey of wireless communications applications in the railway industry. In The 2nd International Conference on Wireless Broadband and Ultra Wideband Communications (AusWireless 2007) (pp. 65-65). IEEE.
30. Shiau, Y. H., & Liang, S. J. (2007, July). Real-time network virtual military simulation system. In 2007 11th International Conference Information Visualization (IV'07) (pp. 807-812). IEEE.
31. Stan, O., Cohen, A., Elovici, Y., & Shabtai, A. (2019). Intrusion detection system for the mil-std-1553 communication bus. IEEE Transactions on Aerospace and Electronic Systems, 56(4), 3010-3027.
32. Stankovic, J. A. (1996). Real-time and embedded systems. ACM Computing Surveys (CSUR), 28(1), 205-208. Suprayitno, S., Fauzi, W. A., Ain, K., & Yasin, M. (2023). Real-time military person detection and classification system using deep metric learning with electrostatic loss. Bulletin of Electrical Engineering and Informatics, 12(1), 338-354.
33. Wang, K. C. (2017). Embedded real-time operating systems. Embedded and Real-Time Operating Systems. Springer, Cham. pp. 401–475.
34. Xia, Y., Li, Q., Huang, R., & Zhao, X. (2022, June). Design of Intelligent Medical Service Robot based on Raspberry Pi and STM32. In 2022 IEEE 10th Joint International Information Technology and Artificial Intelligence Conference (ITAIC) (Vol. 10, pp. 1577-1581). IEEE.
35. Zhang, Y., Luo, X., Zhu, X., Li, Z., & Bors, A. G. (2020). Enhancing reliability and efficiency for real-time robust adaptive steganography using cyclic redundancy check codes. Journal of Real-Time Image Processing, 17, 115-123.