The practice of block symmetric encryption for a secure Internet connection⋆ Tetiana Korobeinikova1,*,†, Ihor Zhuravel1,†, Lesya Mychuda1,† and Axel Sikora2,† 1 Lviv Polytechnic National University, 12 Stepana Bandery str., 79000 Lviv, Ukraine 2 University of Applied Sciences Offenburg, 24 Badstrasse, 77652 Offenburg, Germany Abstract The paper discusses the process of block parallelization in the Advanced Encryption Standard (AES) cipher, focusing on the Counter (CTR) mode. It details the benefits of this process, including increased data processing performance and effective resource utilization; emphasizes the independent encryption of each data block in CTR mode, which allows for effective parallelization, especially when handling large data volumes. This work outlines the steps involved in the AES operation scheme in CTR mode, from splitting data into blocks to generating the final ciphertext. It further explains the concept of a unique “counter” or “initialization vector” for each block, which, combined with the key, generates a unique encryption key, enabling parallel processing. The idea implementation delves into the programming of the block parallelization algorithm using services on the Java Spring Boot platform. It describes the roles of the purposed Client Service and Server Service in encrypting and transmitting messages and files and decrypting received messages. This work presents an experiment that tests the hypothesis that blocks parallelization in AES cipher using CTR mode increases performance during the processing of large data volumes. The experiment involves different data volumes and compares the processing speeds of the AES algorithm with and without parallelization. The results confirm the hypothesis, showing that block parallelization in AES for large data volumes can double the data processing speed compared to the non- parallel approach. The paper concludes that block parallelization might be effective not only for the AES algorithm but also for any block symmetric algorithm. It also suggests that parallelization allows for more efficient use of multi-core systems and reduces the execution time to complete the encryption operation. Keywords block parallelization, AES cipher, CTR mode, data processing performance, encryption optimization1 1. Introduction The subject is methods and means of protecting connections on the Internet, including the analysis of encryption In a world of increasing digital connectivity and Internet efficiency. interaction, the issue of connection security is becoming one of the most crucial problems [1–9]. Ensuring the 2. Related works confidentiality, integrity, and availability of information is becoming a priority [9–11]. This work is focused on It is known that the increasing number of digital connections reviewing and improving block symmetric encryption during Internet interaction becomes the issue of connection methods aimed at solving the problems of connection security. Cristina Del-Real et al. [4] say that the design of security on the Internet. New technologies and encryption software systems plays a crucial role in mitigating algorithms are important tools for maintaining data cybersecurity incidents; Sood et al. [5] and Jang et al. [9] are confidentiality and ensuring resilience to modern cyber paying attention to DoS attacks through HTTP and IP. threats [12–17]. Gentile et al. In [1] showed the algorithms to ensure Thus, there is a need to analyze, improve, and expand the suitable data transmission and encryption ratios and used methodological base of methods and tools that use block Transport Layer Security (TLS) tunnels for local sensor data symmetric ciphers. and secure socket layer tunnels to transmit TLS-encrypted The goal is to improve the methods of block symmetric data to a cloud-based central broker. Also, Kampourakis et encryption through the process of block parallelization to al. in [3] o er a framework that can accurately detect all solve the problem of secure connections on the Internet. anomalous enterprise network activities. The object is the processes associated with the use of Nowadays connections to the Internet are connections block encryption algorithms, in the context of their to the Cloud. Cloud storage is popular among syudy and application on the Internet and the HTTPS protocol. businesses because of cost reduction, performance CSDP-2024: Cyber Security and Data Protection, June 30, 2024, Lviv, 0000-0003-2487-8742 (T. Korobeinikova); 0000-0003-1114-0124 Ukraine (I. Zhuravel); 0000-0001-8266-1782 (L. Mychuda); 0000-0003-0878-2919 ∗ Corresponding author. (A. Sikora) † These authors contributed equally. © 2024 Copyright for this paper by its authors. Use permitted under tetiana.i.korobeinikova@lpnu.ua (T. Korobeinikova); Creative Commons License Attribution 4.0 International (CC BY 4.0). ihor.m.zhuravel@lpnu.ua (I. Zhuravel); lesia.z.mychuda@lpnu.ua (L. Mychuda); axel.sikora@hs-offenburg.de (A. Sikora) CEUR Workshop ceur-ws.org ISSN 1613-0073 114 Proceedings improvement, productivity enhancement, and security. of resources. Modern security standards recommend using However, Alqahtani et al. in [10] are focused on security keys of at least 80 bits. risks since data is stored with third-party providers, and Encryption and data transmission are important in internet access can limit visibility and control. Compared to wireless networks. There is potential for encrypting test traditional on-premise computing, data security and packets with lightweight encryption algorithms if the protection are critical concerns in cloud computing. security of the device is not compromised. There are Ensuring data security in the cloud involves various numerous encryption algorithms for wired networks, divided methods, with cryptography being the most crucial into two categories: symmetric key encryption and (Sasikumar and Nagarajan in [11]). asymmetric key encryption. Symmetric key encryption So, cryptography provides several security features, requires the distribution of a key between the parties before including authentication, con dentiality, integrity, and data is transmitted. availability. Asymmetric key encryption solves the problem of key Today, it is important to analyze, improve, and expand distribution but is computationally intensive. In wireless the methodological base of methods and tools that use block devices, symmetric key encryption, such as RC4, is symmetric ciphers. predominant, as it is fast and efficient. However, the Boura et al. in [14] offer tools and algorithms to search discovered vulnerabilities in RC4 led to the introduction of a for related-key distinguishers and attacks of a differential new security standard for WLANs—IEEE 802.11i, based on nature against the AES. Yevseiev et al. [15] offer the AES. AES is known for its speed, flexibility, and robust Niederreiter-modified crypto-code structure with additional security features. initialization vectors that require an increase in the speed of Block ciphers, such as AES, are used in WLANs to encrypt cryptographic transformation of the system as a whole. data, ensuring its confidentiality. Security protocols such as While Opirskyy et al. [16] are working on standardizing WPA2 and WPA3 use AES-CCMP to ensure data integrity. post-quantum cryptography. Block ciphers are also used to authenticate devices and users, So, there is a critical need to work on improving the and to manage encryption keys. They are the core methods of block symmetric encryption through the process components of WLAN security. of block parallelization to solve the problem of secure Block ciphers are used in WLANs to ensure data connections on the Internet. confidentiality and integrity, as well as device and user authentication. They are standardized, efficient, and scalable, 3. Initial information that making them a reliable choice for securing wireless communications. Overall, they enhance WLAN security by establishes research meeting security standards and requirements. Stream encryption algorithms. A stream cipher is a type Rationale for choosing the AES to solve the problem of symmetric encryption that encrypts each digit of the of secure Internet connection. DES, while a popular plaintext separately using a key stream. Stream ciphers are encryption standard in the past, has fallen out of favor due to faster than block ciphers and require less hardware but can its limited key length and vulnerability to modern attacks. be vulnerable to attacks [17–19]. AES is known for its high security, efficiency, and wide There are two types of stream ciphers: synchronous and support, making it the best choice for secure Internet self-synchronizing. Synchronous stream ciphers generate a connections. The Twofish algorithm, although it has its sequence of pseudorandom digits independently of the advantages, is not as widespread as AES and is not a standard plaintext and ciphertext messages. Self-synchronizing stream for general use. ciphers use the last N digits of the ciphertext to generate a key The rationale for choosing AES is based on its stream [20]. standardization, wide support, high level of security, and Block encryption algorithms. A block cipher is an efficiency. Its status as a global standard recommended by key algorithm that encrypts fixed-length data blocks to protect security institutions such as NIST emphasizes its reliability. information. Electronic Code Book mode encrypts blocks Its ability to work efficiently in real time makes it an ideal independently, but patterns can be detected. To increase choice for secure Internet connection tasks where speed and security, modes that introduce randomization through an security of data transmission are important. Initialization Vector (IV) are used. Various modes of block This choice provides strong encryption, meets modern cipher operation have been developed and defined in national requirements, and helps to create secure connections in a and international standards [21–22]. The basic idea is to virtual environment. introduce randomization of the plaintext data using an IV to AES can be slow when processing large amounts of data achieve probabilistic encryption. for several reasons. First, the AES block size is fixed (typically The rationale for using block symmetric encryption 128 bits), and a large amount of data requires many blocks, for secure Internet connections. Encryption algorithms which can affect the overall speed. In addition, some AES are known for their high computational requirements, modes of operation may include padding operations and especially for wireless devices with limited resources [23]. other additional operations that can also increase the time Encryption guarantees data confidentiality and consumption. protection against interception and plays an important role in A large amount of data can lead to additional authentication, data integrity, and access control. computational costs and increased execution time, in cases Encrypting even a small amount of data, such as 13.6 where enhanced security or authentication modes are used. kilobytes, using a 32-bit Blowfish key can consume about 75% 115 To improve the algorithm, a block parallelization mechanism  Block independence (CTR mode uses encryption was chosen. with a key and unpredictable counter values for each The decision to optimize the AES algorithm using block block. This allows each block of data to be encrypted parallelization for secure Internet connections can be independently, which is ideal for parallel processing). justified:  Ease of implementation and computation (CTR mode is noted for its ease of implementation and  Large volumes of data: large amounts of data are minimized computation, as each block is encrypted commonly processed on the Internet, especially independently of the others. This makes it effective for use in environments where computing resources are when transferring files or in other data-intensive limited). scenarios. Block parallelization allows the  Parallel processing capability (since each block is efficient distribution of encryption tasks among encrypted independently, CTR mode lends itself easily different computing resources, ensuring fast to parallel computing. Different blocks can be processed processing. by different computing units or even on different  Increased performance: performance is a key devices, contributing to speed and efficiency). factor in determining the quality of an Internet  Data Structure Preservation (CTR mode allows connection. Parallelization allows you to use you to preserve the structure of the original data during parallel resources to simultaneously process encryption. Each block is replaced by an encrypted multiple blocks of data, increasing the speed of block of the same length, which avoids data expansion encryption and decryption operations. or compression during encryption). Note that while CTR mode has many advantages for Data security: AES is a secure algorithm, but parallel processing, it is important to manage the counter and improving performance should not compromise security. key properly to avoid vulnerabilities. When using CTR mode, Block parallelization allows for optimal performance you should avoid reusing counter values for the same key. without compromising encryption security. The AES operation scheme in the CTR mode is shown in Fig. 1. 4. The block parallelization process Parallelization is a key strategy for encryption optimization aimed at increasing data processing performance in this work. It will show the main aspects of the block parallelization process, its benefits, and possible challenges, and consider how the implementation of block parallelization helps to optimize the AES cipher for high- performance information processing tasks. 4.1. Detailed description of the process When parallelizing blocks and using AES encryption, CTR (Counter) [24] mode is the most effective strategy, especially in situations where parallel processing and high performance are important. Optimized parallelization allows to maintenance of high performance in real-time processing areas where Figure 1: Diagram of the AES algorithm in CTR mode responsiveness to data is key. At the same time, efficient resource utilization and security ensure that large amounts of The AES operation scheme in the CTR mode contains the information are effectively handled, providing an optimal following steps: combination of performance and data privacy [25]. 1. Splitting Data into Blocks: the input data is divided The choice of AES ciphers in CTR mode for block into blocks of fixed size. Each block has a length due to the parallelization is based on their combination of security, AES block size (128 bits or 16 bytes). efficiency, and standardization, namely, keys of different 2. Generate Counters: a unique counter is created for lengths (including 256 bits) and implementation speed. each block. It generates a unique key for each block. In CTR mode, each block of data is encrypted 3. Encrypting Blocks with AES: the resulting counters independently, which allows to effectively parallelize the are used as input for the AES block used in the encryption encryption process. This is especially important when mode. The generated result is XOR'd with the corresponding processing large amounts of data, where parallel use of block of input data, which ensures block encryption. resources can significantly improve performance. 4. Counter increment: after the block encryption is The combination of AES and CTR modes provides a completed, the counter is incremented to generate a new robust encryption mechanism that can work effectively in a value for the next block. parallelized environment, ensuring a high level of security 5. Repeat the Process for Each Block: All input blocks and data processing efficiency. are processed similarly, generating unique keys for each. The main advantages of using CTR mode when 6. Generate Ciphertext: the encrypted blocks are parallelizing blocks: combined into the final ciphertext. 116 4.2. Block parallelization process diagram The main idea of the AES block parallelization algorithm in CTR mode is to create independent keys for each data block, which allows to effectively parallelize the encryption process. Each block is processed independently, which improves the speed of processing large amounts of information. The AES block parallelization scheme in CTR mode is shown in Fig. 2. Figure 2: Block parallelization scheme In this mode, a unique “counter” or “initialization vector” is created for each block, which is combined with the key to generate a unique encryption key. Thus, each block uses a different key, and they can be processed in parallel. Figure 3: Block diagram of the AES encryption algorithm The AES block parallelization algorithm in CTR mode with block parallelization (Fig. 3) involves partitioning into blocks, generating unique counters for each block, encrypting each block independently  Storing of encrypted blocks: when a block of the others, and combining the encrypted blocks to produce completes its work, the encrypted block is saved to a the final ciphertext. separate thread that functions as a container for The key stream is an important part of the encryption collected encrypted blocks. This encrypted block process, especially when considering parallel computing to collection stream can be used for further processing or encrypt multiple blocks of data simultaneously. The key transmission. stream is used to create unique keys for each block of data.  Creating of the ciphertext: whet the computation of all blocks is complete, the encrypted blocks are This is achieved by using a counter and key values. The formed into ciphertext. This text can be used for storage, counter value is determined by the stream number that points transmission, or other purposes, ensuring the data to a specific data block. The key is formed using the result of confidentiality and integrity. the Diffie-Hellman algorithm [26], which ensures the security  Using the Diffie-Hellman algorithm: the DH of key exchange. algorithm guarantees the security of key exchange by defining a shared secret key between the parties, which  Parallel computation: all data blocks compute ensures the confidentiality of information. The result of their keys in a parallel way, which significantly speeds this algorithm is used as the basis for creating unique up the encryption process. Each block uses its key keys for each block of data in the stream. stream to generate a unique key. This increases the efficiency of encryption because the blocks can be In general, this process ensures a high level of security processed simultaneously without unnecessary delays. and encryption efficiency for parallel data block computation. 117 5. Programming the block messages are saved to a file. Server Service is responsible for encrypting files and sending encrypted files to the client. parallelization algorithm To implement this functionality, Java Spring Boot, a This section describes the process of developing and framework for developing web applications and programming the AES block parallelization algorithm using microservices, is used. It ensures the configuration and developed services on the Java Spring Boot platform. These efficiency of service deployment. services will interact to securely exchange and process The proposed approach ensures a secure and efficient encrypted messages and files. mechanism for exchanging encrypted messages and files Client Service is responsible for messages encrypting between the client and the server. and messages transmitting to the server for further So, two services, the client and the server, interact with a processing. It uses AES and keys to encrypt messages. The secure exchange of secret keys over an unreliable channel encrypted messages are sent to the server for decryption and using the Diffie-Hellman algorithm. The structure of the saving to a file. server is shown in Fig. 4a and the structure of the client is Server Service is responsible for receiving encrypted shown in Fig. 4b. messages and decrypting those messages. The decrypted a) b) Figure 4: The structures of developed services The Server Service diagram of classes is shown in Fig. 5. The developed Server Service consists of such classes:  ServerController: handles HTTP requests; receives and transmits data between the client and other services; initiates calls to other services for data processing.  AESService: implements the AES encryption and decryption algorithm. It is used by the ServerController to ensure the confidentiality of the exchanged data.  DataUtils: contains general methods for data operations.  KeyService: handles keys (generation, storage, and processing) necessary for data encryption and decryption. The ServerController and other services can interact with the KeyService to ensure the security of the exchanged data.  ServerService: receives and processes data from the ServerController; uses the AESService for encryption and decryption, DataUtils for data processing, and KeyService for key management. Figure 5: The Server Service diagram of classes The developed Client Service consists of (the diagram of classes is shown in Fig. 6). 118 6. Conclusions 6.1. Detailed description of the process It is assumed that the implementation of block parallelization in AES cipher using CTR mode increases performance during processing a large volume of data. The main idea is: that dividing into parallel blocks allows more efficient use of resources and reduces encryption and decryption time. The hypothesis assumes that because of parallelization implementation: (1) processing speed increases; (2) system scalability grows. The experiment contains a comparative analysis of speed and productivity between systems with and without block parallelization in the AES cipher during the processing of a real data volume. 6.2. Input data and its analysis For the implementation of an improved AES algorithm, the text data was used. Each data block was defined by a size that corresponds to the AES standard, 128 bits. The considered data volume was determined by the number of blocks that were chosen for encryption. The chosen data volume value corresponded to specific experimental conditions to study the efficiency of parallelization in the algorithm. Input data for the AES algorithm with improved block parallelization came from user input. This considered the variety of algorithm usage scenarios and subjected it to realistic conditions. Algorithm parameters, including block size and data volume, were chosen based on the efficiency check of the improvement. 6.3. Working with data Different data volumes were taken and checked by the AES algorithm without parallelization and with it to determine the efficiency of the improvement. To start, check the connection between the two services and generate a key for the algorithm (Fig. 7). Figure 6: The Client Service diagram of classes  ClientController: handles HTTP requests from the server; receives and transmits data between the server and other services; initiates calls to other services for data processing. Implementation of the proposed process of block parallelization in the AES cipher in CTR mode based on two developed services significantly optimizes and improves the operation of the encryption system. The increase in Figure 7: Connecting two services performance is the result of the efficient use of computing resources due to the process of parallelizing the processing For the request_1 in the first test, small data volumes (less of data blocks. Splitting the tasks between two services 500 blocks) and standard AES algorithm in CTR mode allows better management of encryption and decryption without parallelization were used (Fig. 8a). The result is 3 processes, contributing to the system's scalability needs. milliseconds to encrypt the request_1. Then a request_1 is created with the same data sets, but the algorithm with parallelization is used. (Fig. 8b). 119 a) b) Figure 8: Test 1. Requests execution result with small data volumes The parallelization does not play a significant role for small CTR mode without (Fig.9a) and with (Fig. 9b) parallelization data (it was processed in 3 milliseconds). was used. The request_2 was executed in 94 milliseconds, For the request_2 in the second test, bigger data volumes which is 38 milliseconds less than the previous result. were used (10000 blocks) and the standard AES algorithm in a) b) Figure 9: Test 2. Requests execution result with bigger data volumes For the request_3 in the third test, big data volumes were request_3 was processed for 10.39 seconds (Fig.10a) and 6.05 used (1000000 blocks) and the standard AES algorithm in CTR seconds (Fig. 10b), which is twice as fast as the previous result. mode without and with parallelization was used. The a) b) Figure 10: Test 3. Requests execution result with big data volumes 120 6.4. Implementation results times to complete the encryption operation compared to the regular algorithm. In this section, the results of implementing the AES block parallelization process are shown. The data processing speeds with (Table 1) and without (Table 2) parallelization process 7. Conclusions are compared. In this work, a detailed exploration of the block parallelization process when using the Advanced Table 1 Encryption Standard (AES) cipher is presented. The focus is Processing time of the AES algorithm (without on the Counter (CTR) mode, which is identified as the most parallelization) effective strategy for parallelizing blocks, particularly in contexts where parallel processing and high performance are crucial. The benefits of block parallelization, including Number of Average Time (s) blocks Processing Time of the Request (s) block independence, ease of implementation and computation, parallel processing capability, and data structure preservation are shown. It also emphasizes the 0. importance of proper management of the counter and key 500 0.003 0.004 0.004 0.003 0.003 to avoid vulnerabilities. 0034 0. The process of block parallelization is described in 10000 0.122 0.117 0.124 0.122 0.119 121 detail, from splitting data into blocks, generating counters, 10. encrypting blocks with AES, incrementing the counter, 1000000 10.90 10.50 10.54 10.42 10.32 384 repeating the process for each block, to generating the ciphertext. The paper highlights that each block is processed Table 2 independently, which enhances the speed of processing Processing Time of the Improved AES Algorithm (with large volumes of information. The development and parallelization) programming of the AES block parallelization algorithm were performed by developing services on the Java Spring Average Time Boot platform. These services interact to securely exchange Number of blocks Processing Time of the Request (s) and process encrypted messages and files. The structure and (s) functions of the developed Server Service and Client Service are explained. The paper presents an experiment that tests the 500 0.003 0.004 0.003 0.003 0.004 0.0034 hypothesis that the implementation of block parallelization 10000 0.094 0.089 0.087 0.093 0.090 0.0906 in the AES cipher in CTR mode increases performance 1000000 5.10 4.89 5.03 4.95 4.10 4.98 during the processing of large volumes of data. The experiment involves testing two algorithms (with and without parallelization) for operation speed and efficiency The speed of encryption and decryption decreases with the using different volumes of data. The results of the data increase. experiment confirm the hypothesis. Block parallelization in AES for big data volumes increases data processing speed 6.5. Hypothesis Confirmation by two times compared to the non-parallel approach. The hypothesis about the implementation of block Parallelization allows for more efficient use of multi-core parallelization in the AES cipher in CTR mode will lead to a systems and accelerates the block processing by 50%. significant improvement in speed and encryption efficiency Without parallelization, each block processing is sequential in the context of processing large volumes of data. An and takes a longer execution time. experiment was conducted that involved testing two The work demonstrates that block parallelization is not algorithms (with and without parallelization) for operation only effective for the AES algorithm but for any block speed and efficiency. symmetric algorithm. It provides optimal resource usage, For this experiment, different volumes of data were reducing the execution time by two times to complete the used, which could be used in the real world. The experiment encryption operation compared to the regular algorithm. showed that the hypothesis was confirmed, so block This process ensures a high level of security and encryption parallelization might be effective not only for the AES efficiency for parallel data block computation. The proposed algorithm but for any block symmetric algorithm. The approach ensures a secure and efficient mechanism for experiment showed that block parallelization in AES for big exchanging encrypted messages and files between the client data volumes increases data processing speed by 2 times and the server. The increase in performance is the result of compared to the non-parallel approach. Parallelization the efficient use of computing resources due to the process allows to use of multi-core systems more efficiently and of parallelizing the processing of data blocks. Splitting the accelerates the block processing by 50%. Without tasks between two services allows better management of parallelization, each block processing is sequential and it encryption and decryption processes, contributing to the takes a longer execution time (10.38 seconds) with system’s scalability needs. processing over 100,000 blocks. Parallelization provides optimal resource usage, reducing the execution time by 2 121 References Future Networks and Distributed Systems (ICFNDS '23) (2024) 548–554. doi: 10.1145/3644713.3644794. [1] A. Gentile, et al., A Performance Analysis of Security [14] C. Boura, P. Derbez, M. Funk, Related-Key Differential Protocols for Distributed Measurement Systems Based Analysis of the AES, IACR Transactions on Symmetric on Internet of Things with Constrained Hardware and Cryptology 2023(4) (2023) 215–243. doi: Open Source Infrastructures, Sensors 24 (2024) 2781. 10.46586/tosc.v2023.i4.215-243. doi: 10.3390/s24092781. [15] S. Yevseiev, et al., Development of Niederreiter Hybrid [2] M. Baee, L. Simpson, W. Armstrong, Anomaly Crypto-code Structure on Flawed Codes, Eastern- Detection in the Key-Management Interoperability European J. Enterp. Technol. 1(9(97) (2019) 27–38. doi: Protocol Using Metadata, in IEEE Open Journal of the 10.15587/1729-4061.2019.156620. Computer Society 5 (2024) 156–169. doi: [16] A. Horpenyuk, I. Opirskyy, P. Vorobets, Analysis of 10.1109/OJCS.2024.3386715. Problems and Prospects of Implementation of Post- [3] V. Kampourakis, G. Makrakis, C. Kolias, From Seek- Quantum Cryptographic Algorithms, in: CEUR and-Destroy to Split-and-Destroy: Connection Workshop Proceedings, vol. 3504 (2023) 39–49. Partitioning as an Effective Tool against Low-Rate [17] V. Luzhetsky, et al., Adaptive Compression Methods DoS Attacks, Future Internet 16 (2024) 137. doi: of Data Based on Fibonacci Linear Forms, Proc. SPIE 10.3390/fi16040137. 10445, Photonics Applications in Astronomy, [4] C. Del-Real, E. De Busser, B. van den Berg, Shielding Communications, Industry, and High Energy Physics Software Systems: A Comparison of Security by Experiments (2017). doi: 10.1117/12.2280944. Design and Privacy by Design Based on a Systematic [18] F.-H. Hsiao, S.-W. Chang, Integrating RC6 Stream Literature Review, Computer Law & Security Review Cipher to a Chaotic Synchronization System, in IEEE 52 (2024) 105933. doi: 10.1016/j.clsr.2023.105933. Access 12 (2024) 26308–26323. doi: [5] S. Sood, N. Hubballi, SlowTrack: Detecting Slow Rate 10.1109/ACCESS.2024.3364702. Denial of Service Attacks Against HTTP with [19] N. Zidarič, et al., The Welch-gong Stream Cipher— Behavioral Parameters, J. Supercomput. 80 (2024) Evolutionary Path, Cryptogr. Commun. 16 (2024) 129– 1788–1817. doi: 10.1007/s11227-023-05453-3. 165. doi: 10.1007/s12095-023-00656-0. [6] O. Harasymchuk et al., Generator of Pseudorandom [20] Y. He, et al., Improved Cube Attacks on Some Bit Sequence with Increased Cryptographic Security, Authenticated Encryption Ciphers and Stream Metallurgical and Mining Industry: Sci. Tech. J. 6(5) Ciphers in the Internet of Things, IEEE Access 8 (2020) (2014) 24–28. 20920–20930. doi: 10.1109/ACCESS.2020.2967 070. [7] O. Yudin, et al., Efficiency Assessment of the [21] O. Kara, Lower Data Attacks on Advanced Encryption Steganographic Coding Method with Indirect Standard, Turkish J. Electrical Eng. Comput. Sci. Integration of Critical Information (2019) 2019 IEEE 32(2(8)) (2024). doi: 10.55730/1300-0632.4072. International Conference on Advanced Trends in [22] J. Qiu, Ciphertext Database Audit Technology Under Information Theory, ATIT 2019 - Proceedings, pp. 36- Searchable Encryption Algorithm and Blockchain 40. Technology, J. Global Inf. Manag. 30(11) (2022) 1–17. [8] V. Maksymovych, M. Shabatura, O. Harasymchuk, M. doi: 10.4018/JGIM.315014. Karpinski, et al., Development of Additive Fibonacci [23] A. Sahun, et al., Devising a Method for Improving Generators with Improved Characteristics for Crypto Resistance of the Symmetric Block Cybersecurity Needs. Appl. Sci. (Basel) 2022, 12, 1519, Cryptosystem RC5 Using Nonlinear Shift Functions, doi:10.3390/app12031519. Eastern-European J. Enterp. Technol. 5(9(113)) (2021) [9] M. Jang, K. Lee, An Advanced Approach for Detecting 17–29. doi: 10.15587/1729-4061.2021.240344. Behavior-Based Intranet Attacks by Machine [24] J. Liang, H. Shahrzad, R. Miikkulainen, Asynchronous Learning, in IEEE Access 12 (2024) 52480–52495. doi: Evolution of Deep Neural Network Architectures, 10.1109/ACCESS.2024.3387016. Appl. Soft Comput. 152 (2024) 111209. doi: [10] F. Alqahtani, M. Almutairi, F. Sheldon, Cloud Security 10.1016/j.asoc.2023.111209. Using Fine-Grained Efficient Information Flow [25] V. Luzhetsky, et al., Adaptive Compression Methods Tracking, Future Internet 16 (2024) 110. doi: of Data Based on Fibonacci Linear Forms, Proc. SPIE 10.3390/fi16040110. 10445, Photonics Applications in Astronomy, [11] K. Sasikumar, S. Nagarajan, Comprehensive Review Communications, Industry, and High Energy Physics and Analysis of Cryptography Techniques in Cloud Experiments (2017). doi: 10.1117/12.2280944. Computing, in IEEE Access 12 (2024) 52325–52351. [26] A. May, C. Schneider, Dlog is Practically as Hard (or doi: 10.1109/ACCESS.20 24.3385449. Easy) as DH – Solving Dlogs via DH Oracles on EC [12] T. Luong, N. Cuong, B. Vo, AES Security Improvement Standards, IACR Transactions on Cryptographic by Utilizing New Key-Dependent XOR Tables, in IEEE Hardware and Embedded Systems 2023(4) (2023) 146– Access 12 (2024) 53158–53177. doi: 166. doi: 10.46586/tches.v2023.i4.146-166. 10.1109/ACCESS.2024.3387268. [13] K. Rakhimberdiev, A. Bozorov, M. Berdimurodov, Round Key Generation Algorithm Used in Symmetric Block Encryption Algorithms to Ensure the Security of Economic Systems, 7th International Conference on 122