https://doi.org/10.31449/inf.v46i3.3862 Informatica 46 (2022) 429-438 429 Design and Study of Urban Rail Transit Security System Based on Face Recognition Technology 1 Zhan Guo, 1 Zuming Xiao, 2 Roobaea Alroobaea, 3 Abdullah M. Baqasah, 4 Anas Althobaiti, 5* Harsimranjit Singh Gill 1 Mechanical and Electronic Department, Jingdezhen University, Jingdezhen Jiangxi, 333400, China 2 Department of Computer Science, College of Computers and Information Technology, Taif University, P. O. Box 11099, Taif 21944, Saudi Arabia 3 Department of Information Technology, College of Computers and Information Technology, Taif University, P. O. Box 11099, Taif 21944, Saudi Arabia 4 College of Environment and Technology, Bristol University, Bristol Bs1, United Kingdom 5 Guru Nanak Dev Engineering College, Ludhiana, India Emails: guozhan91@163.com, xiaozuming7@126.com, r.robai@tu.edu.sa, a.baqasah@tu.edu.sa, anasuk.tu@gmail.com, harsimran18@gmail.com Keywords: Face recognition technology; Urban rail; Traffic security system; Face recognition algorithm Received: December 11, 2021 In the modern world, it is difficult to prevent terrorism due to the relatively closed environment, dense personnel, large passenger flow, long line and wide coverage of urban rail transit. Identity recognition is a core element of security. The design and study of an urban rail transit security system based on face recognition technology are proposed in this paper. Through the study on the face recognition algorithm of intelligent security systems in urban rail transit, the related introduction of face recognition technology is done. The analysis of the main mode of face recognition is carried out utilizing the practical application design ideas. The results by experimental analysis show that if FAR is set to a very low range (such as 0.1% or even 0.01%) meanwhile FRR can reach a very low level (such as less than 1%). Such a system has practical value and otherwise, it may face a large number of passenger affairs and complaints to be handled. When FAR is set to 0.1% and N is 1.6 million, FRR can reach 2.1%. However, according to the test, when the picture quality deteriorates (during image captured by a webcam), the FRR will increase by 2 to 3 times. If a Webcam is used for recognition in Mugshot, the lowest FRR of the three top algorithms is only 5.21%. Povzetek: Tehnologija prepoznavanja obrazov je uporabljena za nadzor osumljencev - teroristov na vlakih. 1 Introduction At present, each urban rail transit is equipped with a video surveillance system. A large number of surveillance cameras are installed within the station area for subway operation and public security monitoring. Such a video monitoring system has almost become a tool to provide post-evidence recording and has lost the ability to prevent or stop criminal activities from occurring [1]. Urban rail transit has small space and large passenger flow, so safe operation is always the most concerning thing for government departments, operating units, and public security organs. How to identify the dangerous elements hidden in the crowd timely and accurately when entering the scope of rail transit is an urgent problem to be solved by the operation unit and the Ministry of public security. With the progress of society, some high technologies have been continuously used in industrial or civilian production and life. As the most advanced biometric technology and image processing technology in the world, face recognition technology is developing and improving day by day and is constantly applied to various fields in society. The technical development of the urban rail transit integrated monitoring system is shown in Figure 1 [2]. As a highly intelligent security monitoring means, face recognition technology is gradually applied in various fields of society. The introduction of a face recognition system in urban rail transit will certainly help to reduce the work pressure of public security personnel, provide good technical support for the safe operation of urban rail transit and criminal investigation and investigation, contribute to the personal and property safety of passengers, and maintain social stability. But due to the particularity of urban rail transit, the application of the technology in urban rail transit also needs to constantly improved. Many problems hinder face recognition technology in the further application of urban rail transit, so need continuous research to improve the technology. Based on the current research, this paper proposes the design and research of an urban rail transit security system based on face recognition technology, through the study of face recognition algorithm and face recognition technology and the analysis of the main mode of face recognition. Conduct the practical application of the design ideas. The results show that by setting False 430 Informatica 46 (2022) 429-438 Z. Guo et al. Acceptance Rate (FAR) to a very low range, such as 0.1% or even 0.01%, while False Rejection Rate (FRR) can reach a very low level, such as below 1%, the system has utility, otherwise, the application may face a large number of passenger affairs and complaints to be handled. When the FAR is set at 0.1%, and N = 1.6 million, the FRR can reach 2.1%. However, according to the test, when the picture quality deteriorates, such as Webcam (images collected by web camera), the FRR will increase by 2 to 3 times. If Webcam is identified in Mugshot, the lowest FRR of the three top algorithms is only 5.21%. The application of face recognition in the rail transit Automated Fare collection (AFC) system puts forward higher requirements for the integration, processing, and analysis ability of data. Figure 1: Technical development process of urban rail transit integrated monitoring system It also provides a good platform for the development and application of big data, cloud computing, Internet of Things, artificial intelligence and other technologies. To precisely acquire the concentration level of the travelers, it is necessary to precisely get the traveler stream of each key area. This paper plans identification gear for assessing traveler stream, which has incorporated the data acquisition equipment and algorithm process. From the assortment of image recognition, the most recent examination aftereffects of profound learning are utilized to recognize the traveler stream, and the blunder of the video traveler stream is rectified with the WIFI test hardware. To screen the early and late peak time of the megacities and the traveler stream in the vital place of the tram station, the abrupt occasions in the station and the activity span are decreased, while diminishing the unexpected occasions. The rest of this article is organized as: Section 2 presents the related works in various domains. Section 3 consists of methods comprising the concept. Results and analysis are discussed in Section 4 followed by concluding remarks in section 5. 2 Literature review In the literature, Yanpeng et al. analyzed the main technical indicators for the application of face recognition in urban rail transit AFC system. They investigated the technical level of the current top face recognition algorithms in the world, and proposed a design idea for the large-scale application of 1: N face recognition system in urban rail transit AFC system [3]. Feng et al. believe that the recognition server outputs the comparison results and alarms to the station’s local monitoring terminal and remote monitoring center according to the comparison results [4]. Chen et al. for rail transit industry in recent years rapid development needs of public security and intelligence operations proposes a face recognition system based on intelligent Design and Study of Urban Rail Transit Security System… Informatica 46 (2022) 429-438 431 subway design. The face recognition function design, front-end acquisition unit selection and deployment location, overall system architecture design is analyzed and discussed, and for other domestic embarks on a facial recognition system of the construction of the rail transit to provide reference [5]. Li et al. believe that the extraction of facial features is a key step in face recognition, which directly affects the accuracy of recognition. It is completed by the video analysis server set up in the station. The face recognition process is shown in Figure 2 [6]. Gao et al. present an article that puts forward the design and implementation method of the technology so that the school can complete the technical design and management of the student attendance system through the application of the equipment, and give full play to the supervision effect of the equipment [7]. Liu et al. present their view of the inability to obtain real-time learning status of learners in online learning. Their paper uses face recognition technology to monitor and analyze learners' learning status in front of the camera. Figure 2: Face recognition process The authors have designed and developed an intelligent supervising assistant system based on face recognition. The system collects students' images in real- time through the camera, establishes a mathematical model according to the main features of the face, extracts the relevant characteristic values, obtains the data such as face plane, three-dimensional rotation Angle, and eye closure state, and judges the learning state of learners. The experiment proves that the system can better assist managers to monitor the learning state of learners and improve learners' concentration during online learning [8]. Zhu et al. display information, such as the similarity ratio of face recognition, the location of the face, etc., through a graphical user interface, and switch the relevant video image to the monitoring terminal for easy tracking and monitoring [9]. Lu et al. study and analyze large-angle deflection face mainly by three-dimensional facial feature extraction. Three-dimensional facial feature extraction is a method of extracting facial feature points and head posture information from a given area [10]. Shen et al. believe that the location of facial features needs to adapt to the changes in various aspects of the face in different positions to the greatest extent, which can further improve the accuracy of the algorithm [11]. Shi and Lei believe that image preprocessing techniques include geometric ruler normalization, glasses extraction, and image gray-scale attribute correction. When normalizing the geometric size, the eyes and jaw points are automatically positioned, the two eyes are aligned by scaling and rotation, and the distance between the two centers of jaw points is a predefined constant, and then the image is cropped to a fixed size [12]. Although face recognition technology is in continuous development and improvement [13-20], the recognition rate, anti-counterfeiting, fingerprint, retina, and other has a large gap [21-25]. The large passenger flow and complex environmental characteristics of urban rail transit also affect the application of face recognition technology in urban rail transit. In this regard, the application of a face recognition system in urban rail transit should also pay attention to the following aspects: the uncertainty in the process of video image acquisition [26-29]. Due to the complex environment of urban rail transit, such as lighting, installation location, occlusion, and human posture, the acquired video and image quality are different, and the face acquisition is unclear or lacking [30-32], which affects the recognition rate. Therefore, in the installation and selection of the camera should pay attention to the installation position, light, lens exposure angle, and the selection of wide dynamic function, to improve the quality of video images [33-37]. Diversity of face patterns and uncertainty about-face plastic deformation. Because the same face has a diversity of faces such as beard, glasses, hairstyles, and the shape and deformation of different expressions, it affects the precise extraction of facial characteristics. The use of more advanced image processing technology and stable and accurate face expression method is the basis of the wide application of face recognition technology in urban rail transit. Urban rail transit has the characteristics of large passenger flow [38]. There are often many faces in a video image, and the task of video analysis and comparison is large. To realize real-time investigation, it is necessary to configure the number of video analysis 432 Informatica 46 (2022) 429-438 Z. Guo et al. servers and comparison servers reasonably [39], and the face detection and comparison technology are constantly optimized and improved. The face recognition system based on the original video monitoring system must establish a linkage with the original video monitoring system, make full use of the advantages of large-scale monitoring coverage, and realize the tracking of personnel through linkage control, to improve the case handling efficiency of police officers [40]. 3 Research method 3.1 Study on the Face Recognition Algorithm of Intelligent Security System in Urban Rail Transit The Adaboost face detection algorithm mainly uses the gray distribution characteristics of the face area to construct the classifier. The first step is to extract the Haar feature of the face gray distribution and use the integral diagram to quickly calculate the feature value. The second step is to use the weighted voting method to achieve the construction of the Adaboost strong classifier. The third step is to obtain a stronger joint classifier. The Haar feature represents a simple rectangular feature. Haar features can reflect a variety of image features, including horizontal, vertical, edge, center, linear and diagonal features [41]. Each Haar feature corresponds to a weak classifier, and the definition formula is:          other P x f P if x h i i i i i , 0 , , 1  (1) From equation 1: 𝑓 𝑖 (𝑥 ) represents the rectangular feature value of the i th rectangle, 𝑖 (𝑥 ) represents the classification result of the rectangular feature 𝑖 on 𝑥 . When the value is 0, it means that it is a non-human sample. If its value is 1, it means that it is a human face sample. 𝑃 𝑖 is used to determine the direction of the inequality, and 𝜃 𝑖 represents the optimal threshold of the rectangular feature 𝑖 [42]. 3.2 Overview of face recognition technology Face recognition technology is a technology that combines digital image processing, computer graphics, pattern recognition, visualization technology, human physiology, cognitive science, psychology, and other research fields to analyze the collected face graphics, determine the position, size, and posture of the face, and extract effective recognition information for face feature comparison, to realize identity recognition [43]. 3.3 Advantages of face recognition over other biometric technologies Biometrics is a technology that uses the inherent physiological or behavioral characteristics of the human body to perform identification. The physiological or behavioral characteristics require universality (covering a wide range of people) and differences (there should be identifiable differences between different individuals), stability (will not change within a certain period), and vitality (cannot be simulated by simulation) [44]. The comparison of their technical characteristics is shown in Tables 1 and 2 below. Project Face recognition Palm vein recognition Deployment cost Non-contact type High Data acquisition No feeling, no need to cooperate Non-contact type Accessibility Fast Palm extension fit Recognition speed High Common Safety Light and dress blocking; Age change Very high Possible interference Face recognition Age, physiological changes Table 1: Technical features of face recognition and palm vein recognition Project Fingerprint recognition Iris recognition Deployment cost Contact type medium Data acquisition Finger extension fit Non-contact type Accessibility Fast High fit required Recognition speed Common Slow Safety Dirt and skin wear Very high Possible interference Fingerprint recognition Contact lenses Table 2: Technical characteristics of fingerprint recognition and iris recognition It can be seen from Tables 1 and 2 that although palm vein recognition has advantages in terms of safety and stability, it is not easy to promote due to the cost of equipment and recognition speed. Fingerprint recognition is a contact check, which has problems of insanity and susceptibility to interference. Iris recognition is difficult to operate and slow to recognize [45-49]. 3.4 Face recognition training process Face tracking function was tested for different face conditions, including single face, multiple faces, and face in and out or interleaved [50]. The system automatically initializes a tracking window before the face tracking starts, and assigns a Camshift tracker to each face in the Design and Study of Urban Rail Transit Security System… Informatica 46 (2022) 429-438 433 video surveillance to realize multiple face recognition. Through many system function tests, it is found that the Camshift algorithm can have a good tracking effect and robustness in practical application, even for large Angle deflection or position change of face [51-54]. Even if there is a complex situation of people in and out of the area or face interleave, the system can normally start the face detection program and constantly correct the face tracking window, the system can quickly obtain all the face areas in each frame of the image, and is more rapid and stable than the face detection frame by frame. 4 Results and Analysis The following key indicators are used to replace traditional physical tickets with facial features to realize face-to-pay rides in urban rail transit. First, the face feature database N: the number of registered person images in the system's face database [55]. Second, the false acceptance rate (FAR): refers to the probability of identifying an unregistered user as a registered user. Suppose that in a face recognition test, the threshold is set to T. If the comparison value of a test object is greater than T, it is considered that the object should be recognized as true, as presented in Equation 2. One 1: N face recognition can be regarded as one face recognition performed N times, as presented in Equation 3. When FAR(1, T) is very small, FAR(N, T)=N(FAR(1, I), it can be seen that FAR will increase linearly as N increases [56-60]. Third, the false rejection rate (FRR): refers to the probability that registered users are rejected and is presented in Equation 4.   personnel registered - non by library in the s comparison of number Total T of number The personnel registered - non of value Comparison ,   T N FAR (2 )       N t FAR T N FAR , 1 1 1 ,    (3)   personnel registered - non by library in the s comparison of number Total T of number The personnel registered - non of value Comparison ,   T N FAR (4) Since a registrant is recognized as truly independent of other comparison data, it is generally considered that FRR (N, T) = FRR (1, T). But in actual tests, since the registration database is not a linear data structure, it may be an index or tree structure, causing FRR (N, T) to slowly increase as N increases. Fourth, the recognition time t: the time from the extraction of features from the face image to the completion of the feature comparison. In the AFC system, FAR determines the safety of the system, FRR determines the accessibility of the system, and t determines the passing speed of the gate. For most face recognition algorithms, the threshold T used by the computer for face recognition is different, and FAR and FRR are also different. FAR increases with the decrease of T (relaxation conditions), and FRR decreases with the decrease of T. Therefore, FAR and FRR are almost contradictory indicators. Different applications have different requirements for FAR and FRR indexes, as shown in Figure 3. High-security applications have a low tolerance for FAR. For public security agencies to find personnel, it is suitable to use a lower FN. Figure 3: Applications for different indicators For urban rail transit, due to the huge passenger flow and the money transaction involved, it should be classified as a high-security application. Therefore, the FAR needs to be set to a very low range, such as 0.1% or even 0.01%, and the FRR can reach a very low level, such as less than 1%. Such a system has practical value. Otherwise, once applied, it may face a large number of passenger affairs and complaints to be handled. Figures 4, 5, and 6 show the performance curves of the three top international algorithms in the latest Face Recognition Vendor Test (FRVT) evaluation. The processed images are all Mugshot (face photos). When FAR is set to 0.1% and N is 1.6 million, FRR can reach 2.1%. However, according to the test, when the picture 434 Informatica 46 (2022) 429-438 Z. Guo et al. quality deteriorates, such as when using a Webcam (image captured by a web camera), the FRR will increase by 2 to 3 times. If a Webcam is used for recognition in Mugshot, the lowest FRR of the three top algorithms is only 5.21%. Figure 4: Performance curve of microsoft4-2018 top algorithms Figure 5: Performance curve of yitu top algorithm Figure 6: Performance curve of microsoft1-2018 top algorithms Therefore, the actual application effect of face recognition is often different from the evaluation results. At present, domestic manufacturers focusing on intelligent face recognition, such as Yitu, Me gvii, Sense time, and other companies, have launched 1: N face recognition products, but at present, they can only meet the above requirements of FAR and FRR and achieve the face feature library N to the order of tens of thousands of yuan, and the recognition time t is controlled within 1s. It can be seen that with the current computer processing ability and face recognition algorithm performance, it cannot be directly applied on a large scale, and only small-scale pilot ideas can be carried out using specific personnel schemes, single-line commuter passenger schemes, and post-payment based on the third-party payment platform. 5 C o ncl u s i o n The urban rail transit should be classified as high- security application due to the huge passenger flow and the money transaction involved. This article presents the design idea of large application 1: N face recognition system in urban rail transit AFC system. The designed system can realize various functions, and has higher detection efficiency, lower error rate, and better application value. Therefore, the FAR needs to be set to a very low range, such as 0.1% or even 0.01%, and the FRR can reach a very low level, such as less than 1%. However, according to the test, when the picture quality deterior, such as Webcam (images collected by web camera), the FRR will increase by 2 to 3 times. If Webcam is identified in Mugshot, the lowest FRR of the three top algorithms is only 5.21%. The application of face recognition in the rail transit AFC system puts forward higher requirements for the integration, processing, and analysis ability of data. However, it also provides a good platform for the development and application of big data, cloud computing, Internet of Things, artificial intelligence and other technologies. It is believed that the future implications of this research Design and Study of Urban Rail Transit Security System… Informatica 46 (2022) 429-438 435 work lie in further exploration of face recognition technology which will certainly shine in the rail transit industry. Acknowledgments: The study was supported by “Science and technoloy research project of Jiangxi education department, China ( No. GJJ191175)” References [1] Li, S., Wu, S., Xiang, S., Zhang, Y ., Guerrero, J. M., & Vasquez, J. C. (2020). Research on synchronverter-based regenerative braking energy feedback system of urban rail transit. Energies, 13(17), 4418. [2] Nie, X. L., & Wei, Q. C. (2015). Research on the location of emergency rescue stations for urban rail transit based on the PSO [J]. Journal of Railway Engineering Society, 32(07), 100-105. [3] Yanpeng, Z., Jianwu, D., Xiaojuan, L., & Fan, Y . (2014). Optimized Handover Algorithm Based on Stackelberg Games in CBTC Systems for Urban Rail Transit. Journal of Engineering Science & Technology Review, 7(3). [4] Feng, X., Zhang, H., Gan, T., Sun, Q., Ma, F., & Sun, X. (2016). Random coefficient modeling research on short-term forecast of passenger flow into an urban rail transit station. Transport, 31(1), 94-99. [5] Chen, H., Wang, B., He, W., & Zheng, J. (2020). Research on passenger flow early warning of urban rail transit station based on system dynamics. In MATEC Web of Conferences (V ol. 308, p. 01003). EDP Sciences. [6] Li, J., Wang, J., Xu, N., Hu, Y ., & Cui, C. (2018). Importance degree research of safety risk management processes of urban rail transit based on text mining method. Information, 9(2), 26. [7] Gao, H., Liu, S., Cao, G., Zhao, P., Zhang, J., & Zhang, P. (2020). Big data analysis of beijing urban rail transit fares based on passenger flow. IEEE Access, 8, 80049-80062. [8] Liu, H., Xie, Y ., Liu, Y ., Nie, R., & Li, X. (2019). Mapping the knowledge structure and research evolution of urban rail transit safety studies. IEEE Access, 7, 186437-186455. [9] Zhu, K., Xun, P., Li, W., Li, Z., & Zhou, R. (2019). Prediction of passenger flow in urban rail transit based on big data analysis and deep learning. IEEE Access, 7, 142272-142279. [10] Lu, Q. C., Zhang, L., Xu, P. C., Cui, X., & Li, J. (2022). Modeling network vulnerability of urban rail transit under cascading failures: A Coupled Map Lattices approach. Reliability Engineering & System Safety, 221, 108320. [11] Shen, X., Wei, H., & Lie, T. T. (2020). Management and utilization of urban rail transit regenerative braking energy based on the bypass DC loop. IEEE Transactions on Transportation Electrification, 7(3), 1699-1711. [12] Shi, Z., Zhang, N., & Zhu, L. (2019). Understanding the Propagation and Control Strategies of Congestion in Urban Rail Transit Based on Epidemiological Dynamics Model. Information, 10(8), 258. [13] Amhoud, E. M., Chafii, M., Nimr, A., & Fettweis, G. (2021, April). OFDM with index modulation in orbital angular momentum multiplexed free space optical links. In 2021 IEEE 93rd Vehicular Technology Conference (VTC2021-Spring) (pp. 1- 5). IEEE. [14] Gill, H. S., Singh, T., Kaur, B., Gaba, G. S., Masud, M., & Baz, M. (2021). A metaheuristic approach to secure multimedia big data for IoT-based smart city applications. Wireless Communications and Mobile Computing, 2021. [15] Kumar, A., Sehgal, V . K., Dhiman, G., Vimal, S., Sharma, A., & Park, S. (2022). Mobile networks- on-chip mapping algorithms for optimization of latency and energy consumption. Mobile Networks and Applications, 27(2), 637-651. [16] Boguszewicz, C., Boguszewicz, M., Iqbal, Z., Khan, S., Gaba, G. S., Suresh, A., & Pervaiz, B. (2021). The fourth industrial revolution-cyberspace mental wellbeing: Harnessing science & technology for humanity. Global foundation for cyber studies and research. [17] Amhoud, E. M., Othman, G. R. B., & Jaouën, Y . (2017). Concatenation of space-time coding and FEC for few-mode fiber systems. IEEE Photonics Technology Letters, 29(7), 603-606. [18] Amhoud, E. M., Othman, G. R. B., Bigot, L., Song, M., Andresen, E. R., Labroille, G., & Jaouën, Y . (2017, September). Experimental demonstration of space-time coding for MDL mitigation in few-mode fiber transmission systems. In 2017 European Conference on Optical Communication (ECOC) (pp. 1-3). IEEE. [19] Singh, G. (2021). Privacy-preserving authentication and key exchange mechanisms in internet of things applications (Doctoral dissertation, Lovely Professional University Punjab). [20] Choudhary, K., Gaba, G. S., Miglani, R., Kansal, L., & Kumar, P. (2021). Artificial intelligence and machine learning aided blockchain systems to address security vulnerabilities and threats in the industrial Internet of things. Institution of Engineering and Technology. [21] Zerhouni, K., Amhoud, E. M., & Chafii, M. (2021). Filtered multicarrier waveforms classification: a deep learning-based approach. IEEE Access, 9, 69426-69438. [22] Gaba, G. S., Kumar, G., Monga, H., Kim, T. H., Liyanage, M., & Kumar, P. (2020). Robust and lightweight key exchange (LKE) protocol for industry 4.0. IEEE Access, 8, 132808-132824. [23] Sharma, A., & Kumar, N. (2021). Third eye: an intelligent and secure route planning scheme for 436 Informatica 46 (2022) 429-438 Z. Guo et al. critical services provisions in internet of vehicles environment. IEEE Systems Journal, 16(1), 1217- 1227. [24] Kumar, P., & Gaba, G. S. (2020). Biometric‐based robust access control model for industrial internet of things applications. IoT Security: Advances in Authentication, 133-142. [25] Hedabou, M. (2018). Cryptography for Addressing Cloud Computing Security, Privacy, and Trust Issues. In Computer and Cyber Security (pp. 281- 304). Auerbach Publications. [26] Igarramen, Z., & Hedabou, M. (2017, October). FADETPM: Novel approach of file assured deletion based on trusted platform module. In International Conference of Cloud Computing Technologies and Applications (pp. 49-59). Springer, Cham. [27] Azougaghe, A., Hedabou, M., & Belkasmi, M. (2015, December). An electronic voting system based on homomorphic encryption and prime numbers. In 2015 11th International Conference on Information Assurance and Security (IAS) (pp. 140- 145). IEEE. [28] Bentajer, A., Hedabou, M., Abouelmehdi, K., Igarramen, Z., & El Fezazi, S. (2019). An IBE- based design for assured deletion in cloud storage. Cryptologia, 43(3), 254-265. [29] Gaba, G. S., Kumar, G., Monga, H., Kim, T. H., & Kumar, P. (2020). Robust and lightweight mutual authentication scheme in distributed smart environments. IEEE Access, 8, 69722-69733. [30] Hedabou, M., Bénéteau, L., & Pinel, P. (2008). Some ways to secure elliptic curve cryptosystems. Advances in applied Clifford algebras, 18(3), 677-688. [31] Gaba, G. S., Kumar, G., Kim, T. H., Monga, H., & Kumar, P. (2021). Secure device-to-device communications for 5g enabled internet of things applications. Computer Communications, 169, 114- 128. [32] Sharma, A., Podoplelova, E., Shapovalov, G., Tselykh, A., & Tselykh, A. (2021). Sustainable smart cities: convergence of artificial intelligence and blockchain. Sustainability, 13(23), 13076. [33] Bentajer, A., Hedabou, M., Abouelmehdi, K., & Elfezazi, S. (2018). CS-IBE: a data confidentiality system in public cloud storage system. Procedia computer science, 141, 559-564. [34] Azougaghe, A., Oualhaj, O. A., Hedabou, M., Belkasmi, M., & Kobbane, A. (2016, October). Many-to-one matching game towards secure virtual machines migration in cloud computing. In 2016 International Conference on Advanced Communication Systems and Information Security (ACOSIS) (pp. 1-7). IEEE. [35] Masud, M., Gaba, G. S., Choudhary, K., Hossain, M. S., Alhamid, M. F., & Muhammad, G. (2021). Lightweight and anonymity-preserving user authentication scheme for IoT-based healthcare. IEEE Internet of Things Journal, 9(4), 2649-2656. [36] Sharma, A., Singh, P. K., Sharma, A., & Kumar, R. (2019). An efficient architecture for the accurate detection and monitoring of an event through the sky. Computer Communications, 148, 115-128. [37] Masud, M., Gaba, G. S., Choudhary, K., Alroobaea, R., & Hossain, M. S. (2021). A robust and lightweight secure access scheme for cloud based E-healthcare services. Peer-to-peer Networking and Applications, 14(5), 3043-3057. [38] Hedabou, M. (2006). A frobenius map approach for an efficient and secure multiplication on Koblitz curves. International Journal of Network Security, 3(3), 239-243. [39] Sharma, A., Georgi, M., Tregubenko, M., Tselykh, A., & Tselykh, A. (2022). Enabling smart agriculture by implementing artificial intelligence and embedded sensing. Computers & Industrial Engineering, 165, 107936. [40] Boukhriss, H., Azougaghe, A., & Hedabou, M. (2014, November). New technique of localization a targeted virtual machine in a Cloud Platform. In 2014 5th Workshop on Codes, Cryptography and Communication Systems (WCCCS) (pp. 124-127). IEEE. [41] Zhang, H., & You, J. (2018). An empirical study of transport efficiency of urban rail transit based on data envelopment analysis and tobit model. Journal of Tongji University (Natural Science), 46(09), 1306-1311. [42] Hu, H. (2017). Design and Research of English Self-study System Based on Computer Network. Revista de la Facultad de Ingenieria, 32(4), 534-540. [43] Ning, F. A. N. G. (2017). Design and research of multi axis motion control system based on plc. Academic Journal of Manufacturing Engineering, 15(1). [44] Lin, W. (2017). Research on the Urban Rail Transit Design between Subcenter and Center of City. Journal of railway engineering society, 12, 73-76. [45] Jin, H., Zhou, X., Sun, X., & Li, Z. (2021). Decay rate of rail with egg fastening system using tuned rail damper. Applied Acoustics, 172, 107622. [46] Wang, Y ., Wang, P., Li, Z., Chen, Z., & He, Q. (2020). Forecasting Urban Rail Transit Vehicle Interior Noise and Its Applications in Railway Alignment Design. Journal of Advanced Transportation, 2020. [47] Mehmood, R., Katib, S. S. I., & Chlamtac, I. (2020). Smart infrastructure and applications. Springer International Publishing. [48] Mo, Z., Zeng, M., & Guan, J. (2021, October). Analysis of the Dynamic Performance Between Pantograph and Rigid Conductor Rail Considering Construction Error of the Contact Wire Height. In International Conference on Electrical and Information Technologies for Rail Design and Study of Urban Rail Transit Security System… Informatica 46 (2022) 429-438 437 Transportation (pp. 502-514). Springer, Singapore. [49] Wang, X., Guo, Y ., Bai, C., Liu, S., Liu, S., & Han, J. (2020). The effects of weather on passenger flow of urban rail transit. Civil Engineering Journal, 6(1), 11-20. [50] Gill, H. S., Singh, T., Kaur, B., Gaba, G. S., Masud, M., & Baz, M. (2021). A metaheuristic approach to secure multimedia big data for IoT-based smart city applications. Wireless Communications and Mobile Computing, 2021. [51] Alghamdi, A., Al-Badi, A., Alroobaea, R., & Mayhew, P. (2013). A comparative study of synchronous and asynchronous remote usability testing methods. International Review of Basic and Applied Sciences, 1(3), 61-97. [52] Zhu, Z., Zhao, F., Yuan, D., Wang, H., Fei, X., & Author, J. K. (2018). Research and development of high-power and high-speed dc circuit breaker for urban rail transit. High Voltage Engineering, 44(2), 417-423. [53] Pan, D., Zhao, L., Luo, Q., Zhang, C., & Chen, Z. (2018). Study on the performance improvement of urban rail transit system. Energy, 161, 1154-1171. [54] Alhakami, W., ALharbi, A., Bourouis, S., Alroobaea, R., & Bouguila, N. (2019). Network anomaly intrusion detection using a nonparametric Bayesian approach and feature selection. IEEE Access, 7, 52181-52190. [55] Kaur, G., Singh, K., & Gill, H. S. (2021). Chaos- based joint speech encryption scheme using SHA- 1. Multimedia tools and applications, 80(7), 10927- 10947. [56] Krichen, M., Lahami, M., Cheikhrouhou, O., Alroobaea, R., & Maâlej, A. J. (2020). Security testing of internet of things for smart city applications: A formal approach. In Smart infrastructure and applications (pp. 629-653). Springer, Cham. [57] Xiaofeng, Y ., Hao, X., & Zheng, T. Q. (2019). Stray current and rail potential dynamic simulation system based on bidirectional variable resistance module. Transactions of China Electrotechnical Society, 34(13), 2793-2805. [58] Anh, A., Phuong, V ., Van Lien, N., & Hai, N. (2018). Braking energy recuperation for electric traction drive in urban rail transit network based on control super-capacitor energy storage system. Journal of Electrical Systems, 14(3). [59] Lawes, D., Ran, L., & Xu, Z. (2014, September). Design of a solid-state DC circuit breaker for light rail transit power supply network. In 2014 IEEE Energy Conversion Congress and Exposition (ECCE) (pp. 350-357). IEEE. [60] Li, P. (2017). Discussion on the construction cost control measures for urban rail transit. Journal of Railway Engineering Society, 34(8), 89-92. 438 Informatica 46 (2022) 429-438 Z. Guo et al.