Enhancing Sensor Network Efficiency Through Optimized Flooding Mechanism Nadiia Dovzhenko1, 2, Oleg Barabash1, Andrii Musienko1, Yevhen Ivanichenko2, and Iryna Krasheninnik3 1 National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute”, 37 Peremogy ave., Kyiv, 03056, Ukraine 2 Borys Grinchenko Kyiv Metropolitan University, 18/2 Bulvarno-Kudriavska str., Kyiv, 04053, Ukraine 3 Bogdan Khmelnitsky Melitopol State Pedagogical University, 59 Naukove Mistechko str., Zaporizhzhia, 69000, Ukraine Abstract Sensor networks play a crucial role in modern technologies, especially with the widespread implementation of the Internet of Things, where they are used for collecting data from the physical world and transmitting it for analysis and further processing. Therefore, data security in sensor networks is a key aspect, as it affects confidentiality, integrity, and availability of information. Sensor networks employ various encryption and authentication methods to protect the transmitted and processed data. Additionally, the issue of securing the sensors and devices themselves from unauthorized access and attacks, such as Denial of Service, is becoming increasingly prominent. Naturally, security standards and protocols, specifically adapted for sensor networks and IoT, are being developed and implemented to minimize security risks. The development of machine learning and artificial intelligence technologies is also gaining popularity, as it enhances threat detection mechanisms and anomalies in network traffic, thereby more effectively protecting sensor networks. In the context of resource management and energy consumption, it is also important to consider security aspects, as attacks on sensor networks can lead to unjustified resource expenditure and, consequently, a reduction in the lifespan of devices and sensors. Keywords 1 Network, sensor, nodes, efficiency, flooding, anomalies, IoT, network traffic, routing, protection, security 1. Introduction This allows for efficient data collection and exchange about various physical and Sensor networks today are one of the most environmental conditions that these sensors promising technologies, having found track and monitor. widespread application in areas such as the Integration with next-generation networks, creation of “smart” cities, industrial such as 5G and the Internet of Things (IoT), automation systems, environmental significantly expands the application monitoring, healthcare, and many others [1]. possibilities of sensor networks, enhancing The basis of their popularity lies in the use their efficiency and opening access to a wide of relatively inexpensive components—sensor range of services. nodes, which are combined in large numbers Advancements in computing and into wireless networks [2, 3]. communication technologies have enabled the integration of sensing functions and the CPITS-2024: Cybersecurity Providing in Information and Telecommunication Systems, February 28, 2024, Kyiv, Ukraine EMAIL: nadezhdadovzhenko@gmail.com (N. Dovzhenko); bar64@ukr.net (O. Barabash); mysienkoandrey@gmail.com (A. Musienko); y.ivanichenko@kubg.edu.ua (Y. Ivanichenko); irina_kr@mdpu.org.ua (I. Krasheninnik) ORCID: 0000-0003-4164-0066 (N. Dovzhenko); 0000-0003-1715-0761 (O. Barabash); 0000-0002-1849-6716 (A. Musienko); 0000-0002- 6408-443X (Y. Ivanichenko); 0000-0001-6689-3209 (I. Krasheninnik) ©️ 2024 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0). CEUR Workshop Proceedings (CEUR-WS.org) CEUR ceur-ws.org Workshop ISSN 1613-0073 Proceedings 465 development of wireless communication efficient communication protocols and interfaces. The use of microprocessors in energy management algorithms is key to miniature devices allows for the processing of extending the life of the network; large volumes of data in various environments. • Sensor nodes must respond promptly, be Incorporating numerous nodes into sensor unobtrusive, convenient to use, and low- networks boosts their functionality but may cost. The integration of technologies compromise the overall network reliability, such as microelectromechanical systems owing to a higher probability of failures in allows the creation of miniature, highly individual nodes. efficient sensor nodes at an affordable Additionally, the distance limitations for price. wireless information transmission can restrict Today, many sensor networks are limited in the network's range and efficiency in terms of coverage area and the number of tasks distributed applications. To mitigate the risks they can perform. They are capable of associated with node failures and ensure transmitting only certain types of information network reliability, connectivity strategies and with limited bandwidth. approaches, redundancy, and routing in the However, the continuous development of network are applied. This enhances its technologies and innovations allows for the resilience to failures, ensuring service expansion of sensor network capabilities, continuity and connectivity between nodes particularly through the improvement of data even in the event of isolation of individual encoding and transmission systems, making network elements [4]. them more flexible and efficient in various Owing to the close integration between application conditions. smart sensors and sensor nodes, sensor networks acquire unique characteristics that 2. Features of Effective Resource require a meticulous approach to their design and implementation. Management The main advantage of such an approach is Effective resource management and security in the ability to use energy efficiently and modern sensor networks require a improve the quality of monitoring through comprehensive approach that includes not data processing directly on sensor nodes with only the development of the latest protection the help of intelligent algorithms. mechanisms and traffic management but also a The main criteria describing sensor focus on optimizing energy consumption and networks include the following: resource utilization. This is because sensor • The sensor network must be wireless; networks typically consist of a large number of this minimizes environmental devices located in diverse conditions, requiring interference and simplifies deployment adaptability and high efficiency in in various locations; management to ensure their long-term • The sensor network consists of operation. thousands of sensors (network nodes) For example, in sensor networks, there are with any coverage area and performs often identical scenarios of abnormal use of any tasks assigned to it; scalability is signaling and bandwidth, which can lead to critical for adapting to different inefficient use of energy and network applications and territory sizes; resources. • Sensors within the network must self- • Abnormal use of signaling. If the wireless organize into a wireless network capable network’s idle mode timer is set to ten of transmitting arbitrary information seconds, establishing a session involves between any two sensors in the network, additional signaling and sending a single with the necessary transmission speed; packet every 11 seconds. technologies such as mesh networks • This results in sending 330 packets or allow sensors to dynamically reorganize about 13 KB of data during one operating to optimize communication paths; period, necessitating at least 54 minutes • Sensor nodes must consume a minimal of the mobile device’s battery life and amount of energy, as they operate over a airtime for sending 330 signaling events. significant period; the use of energy- 466 • Abnormal use of airtime. For example, connectivity between the constituent elements when a node transmits data for five of the network is crucial for several reasons. seconds, it leads to the continuous active use of network resources. In this case, approximately 720 packets or 28.8 KB are transmitted over one hour, requiring 60 minutes of battery life and only sending one signal message. • Anomalous bandwidth usage demonstrates the significant resource requirements for downloading large files, such as videos larger than 1 GB, which necessitates at least 1.5 hours of continuous high-frequency communication sessions at a speed of 1.5 Mbps. The mentioned scenarios confirm the need for developing and implementing effective traffic and resource management mechanisms, as well as security methods, including encryption and authentication algorithms, Figure 1: An example of the connectivity of protection algorithms against anomaly nodes in a sensor network detection from DoS attacks, and other threats First of all, there is efficient use of energy [5–7]. resources. This is because components, which Additionally, significant attention must be often have limited energy resources, are used paid to the development of standards and in sensor networks. Therefore, rational and protocols to ensure equipment compatibility, effective connectivity can significantly which simplifies the integration of new minimize energy consumption by optimizing technologies and the scaling of existing data transmission routes [9]. networks. Second, there is less attention paid to issues This also includes the development of of scalability. Clear and logical connectivity comprehensive security systems that protect positively affects the incorporation of new data from unauthorized access and cyber- approaches to expand the constituent attacks, using advanced methods of components of the sensor network without encryption, authentication, and anomaly significant changes to the existing infrastructure detection. Thirdly, it’s notable that the reliability of data In conclusion, effective resource transmission increases with rational management and security assurance in sensor connectivity of network components. networks require an integrated approach that Connectivity between nodes facilitates reliable combines the latest data management data transmission from sensors to the central technologies, energy optimization, data collection and processing node, which is cybersecurity, and artificial intelligence. Such especially important in critical applications such an approach will ensure high reliability, as medical, military, or security systems [10]. efficiency, and security of sensor networks, Fourthly, there is minimization of data loss. adapted to complex and dynamic application Reliable connectivity reduces the risk of data conditions [8]. loss during transmission between nodes, ensuring more accurate and dependable 3. Connectivity in Sensor collection and priority processing of data for further retransmission within the network. Networks Factors such as coverage, flexibility, mobility, and response speed of sensor When designing the sensor network network nodes are also crucial. Thus, ensuring infrastructure, it is necessary to highlight the effective connectivity is fundamental to the issue of connectivity. After all, establishing successful operation of sensor networks [11]. 467 4. Optimization of the Flooding possible to markedly decrease the transmission count. Mechanism Should each initiating node relay messages solely to its direct neighbors, and Flooding is a basic message propagation subsequently, these neighbors transmit only to mechanism in sensor networks that, despite its nodes in the subsequent layer, the total simplicity, can be optimized to reduce transmissions could be curtailed to redundancy and enhance efficiency. approximately 200–300, contingent on the One of the key disadvantages of flooding is network’s structural configuration and the a significant number of redundant messages, nodes’ positioning. which can quickly exhaust the energy An increase in the tree depth, 𝑘, further resources of nodes, especially in conditions of diminishes the requisite number of restriction or economy. transmissions. Therefore, it is advisable to optimize the In Fig. 2, the number of transmissions for flooding process using a forwarding tree, different numbers of nodes (10, 50, 100) is which allows for limiting the number of compared between traditional flooding and transmissions by selectively sending messages optimized flooding using a forwarding tree at through a structured approach [12]. 𝑘 = 2. The flooding protocol in sensor networks It is evident that as the number of nodes in operates by having each node, upon receiving the sensor network increases, the number of a message, transmit it to all its neighbors, necessary transmissions with traditional except the source node. A node only uses flooding rises linearly and at a much faster rate information about its nearest neighbors for than with optimized flooding. transmission. To improve efficiency and reduce redundancy, a “forwarding tree” structure is 1000 used to optimize message distribution by 900 selectively transmitting authentication codes 800 not to all neighbors, but only to selected nodes, 700 which helps reduce the total number of transmissions. 600 The creation of a “forwarding tree” begins 500 with an initiator that designates each of its 400 neighbors as the root of a subtree of depth 𝑘. 300 For each such root, the initiator transmits the authentication codes required by all nodes 200 in those subtrees. Further expansion of the tree 100 occurs by including nodes that meet two 0 criteria: they are 𝑘 hops away from the current 0 20 40 60 80 100 root and are reachable from any node at the Tr k=2 last level in the current forwarding tree. To illustrate the practicality of the Figure 2: Comparison of traditional flooding suggested method, envision a sensor network and optimized flooding using a forwarding tree comprising 100 nodes, with each node at 𝑘 = 2 connected to an average of 10 immediate Optimized flooding demonstrates significantly neighbors [13]. better efficiency by reducing the total number of Employing conventional flooding for transmissions, especially in larger networks [14]. message dissemination throughout this In Fig. 3, the number of transmissions across network would necessitate each node to different node counts (10, 50, 100) is compared broadcast the message 10 times, cumulatively between traditional flooding and optimized resulting in around 1000 transmissions, flooding utilizing a forwarding tree at various excluding additional redundancies. tree depths (𝑘 = 2, 𝑘 = 3, 𝑎𝑛𝑑 𝑘 = 5). However, through the application of a forwarding tree with a depth of 𝑘 = 2, it's 468 1000 active much longer in scenarios with optimized flooding, especially at larger values of 𝑘. 900 800 5. Conclusions 700 The optimization of the flooding mechanism 600 involves the implementation of strategies that 500 reduce the number of redundant messages caused by the traditional flooding method. This 400 is achieved by carefully selecting the nodes 300 that participate in data transmission to minimize the energy consumption of each 200 node and increase the overall efficiency of the 100 network. This approach allows for an increase in the 0 lifetime of the sensor network and improves 0 20 40 60 80 100 the quality of service by reducing the time of k=2 k=3 k=5 N, tr message delivery and increasing the reliability of data transmission. Figure 3: Comparison of traditional flooding It is also worth noting that the optimization and optimized flooding using forwarding tree of the flooding mechanism in sensor networks for 𝑘 = 2, 𝑘 = 3 𝑎𝑛𝑑 𝑘 = 5 not only reduces energy consumption and increases the efficiency of data distribution but As 𝑘 increases, the figure illustrates how also contributes to increasing their security. deeper forwarding trees can further reduce the Fewer transmissions reduce the risk of total number of transmissions, thereby interception and unauthorized access to data conserving the energy of the nodes [15]. This and make Denial-of-Service (DoS) attacks highlights the significance of optimizing more difficult because fewer active nodes need message propagation in sensor networks to to be attacked. Thus, optimized flooding helps enhance efficiency and conserve energy [16]. create a more attack-resistant sensor network Traditionally, flooding is calculated as structure. follows: 𝐸𝑡𝑜𝑡𝑎𝑙 = 𝑇 ∗ 𝑃, (1) References where 𝐸𝑡𝑜𝑡𝑎𝑙 is the total energy consumption of one node, 𝑇 is the number of transmission [1] H. Hulak, et al., Dynamic Model of cycles, 𝑃 is the number of messages Guarantee Capacity and Cyber Security transmitted in one cycle. Management in the Critical Automated In the traditional flooding scenario, each System, in: 2nd International Conference node conventionally transmits 10 messages on Conflict Management in Global per cycle, and each node exhausts its energy Information Networks, vol. 3530 (2023) after 100 cycles (assuming all energy is spent 102–111. on transmission only). [2] V. Sokolov, P. Skladannyi, H. Hulak, Optimized flooding is calculated as follows: Stability Verification of Self-Organized 𝐸𝑜𝑝𝑡 = 𝑇 ∗ 𝑃𝑘 , (2) Wireless Networks with Block where 𝐸𝑜𝑝𝑡 is the total energy consumption of Encryption, in: 5th International one node with optimized flooding, 𝑃𝑘 is the Workshop on Computer Modeling and number of messages transmitted in one cycle Intelligent Systems, vol. 3137 (2022) when condition 𝑘 changes. 227–237. In the scenario with optimized flooding for [3] V. Buriachok, et al., Invasion Detection 𝑘 = 2, each node will exhaust its energy in 200 Model using Two-Stage Criterion of cycles; for 𝑘 = 3—for 333 cycles; for 𝑘 = 5— Detection of Network Anomalies, in: for 500 cycles. This means that for 100 Workshop on Cybersecurity Providing in transmission cycles, nodes will be able to stay 469 Information and Telecommunication Components. IEEE 7th International Systems, vol. 2746 (2020) 23–32. Conference on Methods and Systems of [4] N. Dovzhenko, et al., Comprehensive Navigation and Motion Control Analysis of Efficiency and Security (MSNMC) (2023) 38–41. Challenges in Sensor Network Routing, [13] V. Mashkov, O. Barabash, Self-Checking in: Cybersecurity Providing in of Modular Systems under Random Information and Telecommunication Performance of Elementary Checks, Systems II Vol. 3550 (2023) 275–280. Engineering Simulation 12 (1995) 433– [5] V. Sokolov, P. Skladannyi, N. Korshun, 445. ZigBee Network Resistance to Jamming [14] L. Globa, et al., Approach to Uniform Attacks, in: IEEE 6th International Platform Development for the Ecology Conference on Information and Digital Environment of Ukraine, Telecommunication Technologies and Progress in Advanced Information and Radio Electronics (2023) 161–165. doi: Communication Technology and 10.1109/UkrMiCo61577.2023.10380360. Systems, LNNS 548 (2022) 83–100. doi: [6] V. Sokolov, P. Skladannyi, A. Platonenko, 10.1007/978-3-031-16368-5_4. Jump-Stay Jamming Attack on Wi-Fi [15] Y. Melnyk Yurii, et al., The Process of Systems, in: IEEE 18th International Con- Network Flows Distribution based on ference on Computer Science and Infor- Traffic Engineering Method, mation Technologies (2023) 1–5. doi: International Journal of Advanced 10.1109/CSIT61576.2023.10324031. Trends in Computer Science and [7] V. Sokolov, P. Skladannyi, V. Astapenya, engineering 8(6) (2019) 3036–304. doi: Bluetooth Low-Energy Beacon 10.30534/ijatcse/2019/60862019. Resistance to Jamming Attack, in: IEEE [16] O. Barabash, et al., Distribution of Values 13th International Conference on of Cantor Type Fractal Functions with Electronics and Information Specified Restrictions, Contemporary Technologies (2023) 270–274. doi: Approaches and Methods in Funda- 10.1109/ELIT61488.2023.10310815. mental Mathematics and Mechanics [8] N. Dovzhenko, et al., Method of Sensor (2021) 433–455. doi: 10.1007/978-3- Network Functioning under the 030-50302-4_21. Redistribution Condition of Requests between Nodes, in: Cybersecurity Providing in Information and Telecommunication Systems Vol. 3421 (2023) 278–283. [9] S. Dovgiy, O. Kopiika, O. Kozlov, Architectures for the Information Systems, Network Resources and Network Services, in: Cybersecurity Providing in Information and Telecommunication Systems II Vol. 3187 (2021) 293–301. [10] Z. Hu, et al., Analytical Assessment of Security Level of Distributed and Scalable Computer Systems, Int. J. Intell. Syst. Appl 8(12) (2016) 57–64. [11] V. Mukhin, et al., Method of Restoring Parameters of Information Objects in a Unified Information Space Based on Computer Networks, Int. J. Comput. Netw. Inf. Secur. 12(2) (2020) 11–21. [12] O. Barabash, et al., Development of a Hybrid Network Traffic Load Management Mechanism Using Smart 470