CENTRALIZATION-BASED OPTIMIZATION SYSTEM AND METHODOLOGY FOR TRAFFIC MANAGEMENT AND DYNAMIC ROUTING IN FLYING TEMPORARY NETWORKS

Abstract

The invention relates to an innovative traffic management system and its methodology that combines eigenvector centrality measurement and dynamic routing algorithms to prevent overloading at central nodes in flying ad-hoc networks (FANETs) and improve network performance.

Description

1 TARIFF TRAFFIC MANAGEMENT AND DYNAMIC ROUTING IN FLYING TEMPORARY NETWORKS CENTRALIZATION-BASED OPTIMIZATION SYSTEM AND ITS METHOD Technical Field to Which the Invention Relates The invention aims to prevent overloading at central nodes in flying ad-hoc networks (FANET). and eigenvector centrality measurement and dynamics to improve network performance. an innovative traffic management system that combines routing algorithms and its It is related to the method. State of the Art Ad-hoc networks in the air (FANET) are mobile networks between nodes in environments without infrastructure. These are systems that enable data transmission by establishing direct communication. These networks are used in the military. in critical applications such as operations, disaster management and emergency communications It is used. There are many existing routing systems to enable data transmission in FANETs. Algorithms and protocols have been developed. However, these algorithms, especially for fast-moving applications, are not suitable. 15 nodes may face some technical challenges in situations such as network dynamics and traffic load. They are encountering each other. One of these algorithms, OLSR (Optimized Link State Routing), is a proactive method. It is a routing protocol and continuously monitors the connection status of all nodes on the network. By updating, it determines the best routing paths. OLSR determines the connections of nodes in the network. By updating its information at regular intervals, the network has a constantly ready routing table. It allows for control over traffic. However, this approach creates high control traffic and especially in fast-moving nodes, a sufficiently fast response to connection changes. Because it cannot deliver, it causes delays and packet loss. Another algorithm, D Lite*, provides efficient path updating in dynamic environments. It is a protocol. D* Lite only re-applies the affected paths for changing connections. It operates on a computational principle. This reduces the computational load, thereby improving performance. This provides an advantage. However, under heavy traffic, connection updates increase, and this situation... Increasing the complexity of the algorithm can reduce its performance. JarmRout is a multipath routing protocol designed to improve connection quality. It offers alternative routes. This algorithm allows existing routes in the network to run in parallel. 30 2 It is designed to provide this. However, it also addresses the overlaps and interactions between the paths. Managing it may be difficult. Also, it may not be effective in situations with high network traffic. The inability to provide load balancing negatively impacts communication performance. In conclusion, current routing protocols are not suitable for dynamic and rapidly changing environments like FANET. Network disconnections, high latency, and excessive traffic at central nodes 5 It faces technical problems such as loading issues. Especially the central office. Overloading the nodes reduces the communication quality of the network and creates bottlenecks. This weakens the overall resilience of the network. These challenges make FANETs... This significantly limits its performance and reliability. Brief Description and Objectives of the Invention 10 The invention aims to prevent overloading at central nodes in flying ad-hoc networks (FANET). and eigenvector centrality measurement and dynamics to improve network performance. It involves an innovative traffic management system that combines routing algorithms. One aim of the invention is to prevent excessive traffic load on central nodes. This is a verification process. Using eigenvector centrality measurement, central nodes in the network can be identified. 15 By doing this, it reduces overload by distributing the traffic load more efficiently. Another purpose of the invention is the regulation of traffic load and the central nodes. Thanks to its protection, the network is more resistant to signal disruptions and enemy attacks. Its purpose is to make it resistant. Another aim of the invention is to minimize connection interruptions. Dynamic 20 Connecting at fast-moving nodes using routing algorithms It adapts quickly and effectively to changes. Another aim of the invention is to perform transmission power adjustments based on eigenvector centrality. The advantage of highly centralized nodes is that they optimize energy consumption. Another aim of the invention is to improve data transmission performance. Jump 25 Using penalties and connection cost calculations to find the most suitable routing. It helps to identify routes and reduce delays. 3 Another objective of the invention is the use of multiple channels and the signal-to-noise ratio (SNR). optimizing data speed and transmission time by taking calculations into account It is a verification. Another aim of the invention is to reduce varying connection costs using the D* Lite algorithm. 5. Updating quickly and keeping network performance constantly updated. It is a verification. Another objective of the invention is to select the most suitable routing method according to network conditions. It enables both fast and reliable communication. Another aim of the invention is to distribute traffic load between decentralized nodes. by distributing them, bottlenecks are prevented and the overall performance of the network is improved. 10 It is a verification. Explanation of the Figures Figure 1. This drawing shows the flowchart of the method described in the invention. Figure 2. This drawing illustrates the classical navigation method used before the invention. Figure 3. This is a drawing illustrating the guidance involved in the method described in the invention. 15 Explanation of References in Figures 1001. Initialization of all nodes by the network administration unit, each node on the network For unique identification, their identities must be determined and initially configuration 1002. Network topology creation module determines the devices and connections on the network. defining and creating the network topology 1003. Network topology generation module uses SNR, an average-sized network topology. Internode connectivity based on packet transmission time and hop count. determining costs 1004. The node centricity calculation module assigns each node to 25 in the network. Calculating eigenvector centrality to determine its importance 4 1005. Parameter configuration calculation module, transmission of the node. power and frame intervals are adjusted according to the centrality value. 1006. Rapid network adaptation to changing conditions by the traffic management module. the most suitable routing strategy based on the current network status is needed. making the choice 5 1007. OpenFlow controller adapts to topology changes according to OpenFlow Updating routing tables 1008. Routing selected by the routing algorithm module. With the new routing tables created according to the algorithm, data packets are routed through the network. transmitted via 10 Detailed Description of the Invention The invention aims to prevent overloading at central nodes in flying ad-hoc networks (FANET). and eigenvector centrality measurement and dynamics to improve network performance. It involves an innovative traffic management system that combines routing algorithms. The OpenFlow Controller runs on the server and dynamically manages network traffic. 15 It is used for. The routing tables of the drones in the invention are created using OpenFlow. It is being updated. UAVs (unmanned aerial vehicles) are the endpoints that provide communication for the system. Self-propelled. They are vehicles that can fly and connect to a network. The MAVLink protocol, located on the UAV, enables communication between the UAV's flight control equipment and the ground. It provides communication between the control station. The network topology creation module runs on the server, configuring the connections between UAVs. It enables the creation of a graph representing the physical structure of the network. Network topology The generation module calculates path loss and signal-to-noise ratio (SNR) values. It determines the connection quality using 25 The node centricity calculation module runs on the server and calculates the centricity of each node in the network. It calculates the eigenvector centrality to determine its significance. The parameter configuration calculation module runs on the server, managing node centricity. eigenvector centrality calculation result calculated by the calculation module According to this, the new transmission power and frame interval values ​​of the nodes are also shown in Equation 2. It is adjusted as stated. The traffic management module runs on the server, analyzing traffic on the network and 5 It makes navigation decisions. In hybrid navigation mode, it uses two different Depending on the advantages offered by the steering mode (flexible and fast mode), it depends on the situation. a mode is selected. The channel allocation module runs on the server, handling collisions between different nodes. It allocates channels to prevent this. 10 The routing algorithm module is running on the server and uses the D* Lite algorithm. It uses traffic load and connectivity to find the best path for packets to reach their destination. It takes quality factors into consideration. The energy efficiency control unit operates on the UAV, measuring the system's total energy consumption. It measures hop count, connection cost, and node energy levels, taking these factors into account. It is receiving. The Quality of Service (QoS) control unit works on the server to monitor latency and jitter (timing issues). We continuously monitor QoS metrics such as volatility and packet loss rate, and It checks whether the values ​​remain below the target. The resilience control unit runs on the server, monitoring the number of attackers and the network. It assesses durability by considering factors such as size and traffic load. It uses the package delivery rate (PDR) metric in durability measurements. The steps involved in operating the system are as follows: All nodes are initialized by the network administration unit, each node is on the network. For unique identification, their identities must be determined and the initial configuration 25 (1001) Network topology creation module determines the devices and connections on the network. Defining and creating the network topology (1002) 6 The SNR, the average size of a packet transmitted, is determined by the network topology creation module. inter-node connection costs according to duration and hop count factors determination (1003) The importance of each node in the network is determined by the node centricity calculation module. Calculation of eigenvector centralities to determine (1004) 5 The parameter configuration calculation module determines the transmission power of the node (TX power) and according to the centrality value of the interframe space (IFS) adjustment (1005) The traffic management module enables the network to quickly adapt to changing conditions. Selecting the most appropriate routing strategy according to the current state of the network (1006) 10 OpenFlow routing is performed by the OpenFlow controller based on changes in the topology. Updating the tables (1007). According to the routing algorithm selected by the routing algorithm module data packets are transmitted over the network using the newly created routing tables. (1008). 15 The traffic management module enables the network to quickly adapt to changing conditions. Selecting the most appropriate routing strategy according to the current state of the network (1006) The guidance strategies in the process step are flexible mode, fast mode, and hybrid mode. It is possible. Flexible mode uses a dynamic approach to optimize traffic flow on the network. This mode adopts the current state of the network (traffic density, congestion, We are constantly analyzing (such as latency times) and developing the most suitable routing rules. It determines this instantly. Fast mode allows for quick decision-making regarding direction and... an application that aims to, usually based on, simple, predefined rules It is a strategy. Hybrid mode is an approach that combines the advantages of flexible and fast modes. 25 Depending on traffic conditions, you can switch between either mode or both. It uses these features together. The system operates through the cooperation of all nodes in the network. The OpenFlow Controller is part of the network. It is used to dynamically manage traffic. UAVs control the system's communication. These are the endpoints that provide. These self-flying and network-connected vehicles offer a variety of 30 7 It is equipped with sensors and communication equipment. It also communicates with other UAVs and the ground station. It has the ability to establish wireless communication. The MAVLink protocol enables communication between UAVs and the ground. It provides communication between stations. Network topology creation module, CentAir with this, primarily the locations of all UAVs in the network and their communication characteristics with each other. It creates a network topology by defining the connections between UAVs. This topology forms the basis of the network topology between UAVs. It's like a map showing the quality and signal strength of the UAVs. This allows the system to identify which UAVs are which. how they can communicate directly with each other and which methods are more reliable He is gathering information about it. Each UAV in the network has a different importance for the system. Node centricity calculation. The module uses a method called eigenvector centrality with CentAir for each of the 10 This determines the importance of UAVs in the communication network. In this way, network traffic... central nodes where load is concentrated and peripheral nodes that carry less load This is being determined. UAVs with high centrality are located in areas where network traffic is concentrated. These are the points where more packet delays and losses typically occur. These are the central nodes. The parameter configuration calculation module has 15 nodes for each UAV. It determines the most suitable communication parameters (transmission power, frame rate). This The parameters are based on the UAV's centrality value and its relationship with other UAVs in the network. The Node Centrality Calculation Module and Parameter Configuration are being configured. The calculation function balances network traffic, improves resilience and energy efficiency. It is used to increase. 20 CentAir's traffic management module analyzes traffic on the network and directs packets to their destination. It ensures that the vehicle reaches its destination via the shortest and most reliable route. The system is both flexible and fast. By using routing strategies, it can quickly adapt to changes in network conditions. This can happen. For example, when a UAV malfunctions or the connection quality degrades, The system automatically determines an alternative route. During this process, routing is 25. The algorithm module uses the D* Lite algorithm as its routing algorithm. The channel allocation module with CentAir allows UAVs to use the same frequency simultaneously. They allocate them to different channels to prevent them from colliding with each other. In this way, Each UAV uses its own channel when interacting with other UAVs. Energy efficiency 30 for monitoring and controlling the more effective and reliable operation of UAV networks. from the control unit, service quality control unit and durability control unit 8 This is utilized to improve the system's energy efficiency, service quality, and safety. It continuously measures and monitors its metrics. The invention is a powerful mathematical tool, such as eigenvector centrality measurement. By utilizing this concept, it will determine the importance of each node on the network. This The measurement takes into account the centrality degrees of a node's neighbors, determining what 5 is on the network. This will show how critical a role it plays. Simply put, high Eigenvector centrality means that a node is one of the most important nodes in the network. This indicates that the node has a significant influence on other nodes. Low eigenvector centrality means that a node has less eigenvector centrality compared to other nodes in the network. This indicates that it is effective. This means that the node's influence on other nodes is limited. 10 That means it is. The proposed method is for quantitatively evaluating the importance of nodes in a network. It is based on the eigenvector centrality metric. Eigenvector centrality is the metric used to determine a node's position in the network. considering the connections with other nodes and the quality of those connections It is a calculated value. Eigenvector centrality is calculated using the following equation: 15 𝒙𝒗𝒊 = 1 𝝀 ∑ 𝒙𝒗𝒋 𝒊 𝒗𝒋𝝐𝑴(𝒗𝒊) Equation-1 𝑥𝑣𝑖: A numerical value that indicates how important a particular node (𝑣𝑖 i) is in a network. This value is called the eigenvector centrality of that node. 𝑀𝑣𝑖: The set of all nodes that are directly connected to a given node (𝑣𝑖). 20 In other words, it refers to all the neighbors of a wedding. λ: A constant that indicates the structure of the network and the density of connections between nodes. This value is calculated as the largest eigenvalue of the network's adjacency matrix. The eigenvector centrality of a node determines the eigenvector centrality of other nodes adjacent to that node. It is directly proportional to the average of their centralities. That is, the neighbors of a node are also 25 If it is important (if it has high eigenvector centrality), then that node itself is also important. This is accepted as a fact, as key nodes in the network are generally interconnected. This shows that, therefore, this invention has a degree of centrality above a certain threshold value. 9 Identifying the nodes that have this capability and measuring the transmission power (tx power) and frame rate of those nodes. This will make it possible to adjust the interframe space (SIFS) intervals. This threshold Its value is determined by the network structure and the application requirements. For highly centric nodes, a new transmission power can be calculated using the following equation. The value is calculated. This new value is 5 times lower than the node's initial transmission power. will be. 𝑃𝑣𝑖 = 𝑃𝑖 − 𝑥𝑣𝑖 × 𝑃𝑝 Equation-2 𝑃𝑣𝑖: The new transmission power after adjustment according to the node's eigenvector centrality. value. 10 𝑃𝑖: The initial transmission power value of the node before any adjustments are made. 𝑃𝑝: A multiplier used to balance the load distribution on the network, and it is an eigenvector. Highly centralized nodes allow for further reduction in transmission power. The study considered the 802.11n standard and the use of channels in the 2.4 GHz band. This invention provides 15 channels, multiplied according to the number of channels used, for the connection between two nodes. It creates a polygraph containing edges. For example, edges 1, 6, and 11 in the 2.4 GHz band. Considering the channels, 2 additional copy edges are created for each connection. Nodes To determine the appropriate data rate for the connections between them, firstly, the signal- The signal-to-noise ratio (SNR) is calculated. This calculation determines the ratio of the data to be transmitted for each link. It is used to evaluate its quality. Subsequently, the connection costs are as follows: 20 is calculated: 1. Calculating Average Packet Delivery Time: The transmission time of an average-sized packet (P) at a specified data rate (𝐿𝑖𝑗). This calculation determines how long the connection will take during data transfer. It represents. 2. Addition of a Skipping Penalty: A hop penalty (ζ) is added to each connection. This penalty is applied to the routing algorithm's operation on the network. This allows them to choose routes with fewer jumps. Connection cost is expressed by the following formula: 𝐷𝑖𝑗 = P 𝐿𝑖𝑗 + 𝜁 Equation-3 5 𝐷𝑖𝑗: The connection cost between node 𝑣𝑖 and 𝑣𝑗. 𝐿𝑖𝑗: Data rate determined by SNR between these nodes. P: Average package size. 𝜁 : Penalty for skipping. These concepts are used in routing algorithms to optimize network traffic and find the best path. It is used to find. The technique offered by the invention is a dynamic navigation system like D Lite*. using its algorithm to quickly adapt to changes in network topology D* Lite will provide the advantage of not only changing connection costs. The reason is that it reduces the computational load because it needs to be updated. In this way, the network is constantly updated. It will be able to be updated and will be resistant to connection interruptions. 15 Furthermore, thanks to the hybrid routing mode proposed by the invention, it can optimize routing according to network conditions. An appropriate navigation strategy can be selected. This mode is suitable for various scenarios. By combining robust and fast routing methods, the network's resilience is enhanced. This will increase both low latency and high data transmission rate. will be obtained. 20 In conclusion, the invention aims to solve the problems encountered in existing routing systems. centrality measurement, dynamic path updating algorithms, and hybrid routing. By combining its modules, it aims to improve the performance of FANETs. This The innovative approach increases network resilience while also considering energy efficiency. It provides a solid foundation for future applications. This 25 The innovations make the network more resistant to jamming attacks. This will also help him to come and improve his overall performance.

Claims

11 REQUESTS 1. Centralization for traffic management and dynamic routing in flying transient networks. It is a fundamental optimization system, its characteristic feature is; - UAVs that dynamically manage network traffic by operating on the server. At least one OpenFlow controller that updates the routing tables, 5 - a wing capable of flying autonomously and connecting to the network, enabling communication within the system. at least one UAV with these points, - By being located on the UAV, it controls the flight control equipment and ground control of the UAVs. The MAVLink protocol enables communication between stations. - By working on the server, the connections between the UAVs and the physical structure of the network are 10 It enables the creation of a graph representing path loss calculations, and Using signal-to-noise ratio values ​​to provide the appropriate data rate for connections. At least one network topology creation module that specifies - Using eigenvectors to determine the importance of each node in the network by working on the server. By calculating the centrality, the central nodes where network traffic is concentrated and 15 more At least one node centrality that identifies peripheral nodes carrying low loads. calculation module, - By running on the server, the node centricity calculation module The calculated eigenvector centrality of the nodes is based on the calculation result. At least one parameter that adjusts the transmission power and frame rate values ​​20 configuration calculation module, - Working on the server, it analyzes network traffic and makes routing decisions. at least one traffic management module, - By working on the server, channels are used to prevent collisions between different nodes. at least one channel allocation module that allocates channels, 25 - By evaluating traffic load and connection quality factors, D* With the help of the Lite algorithm, the minimum number of packets that determine the most suitable path to their destination is... a routing algorithm module, - By working on the UAV, hop count, connection cost and node energy levels at least one 30 that measures the total energy consumption of the system by evaluating its factors. energy efficiency control unit, 12 - operating on the server, resulting in latency, timing fluctuations, and packet loss rate. By continuously monitoring QoS metrics and staying below target values. at least one service quality control unit that checks that nothing is missing, - Working on the server, it detects packet loss and connection drops. at least one 5 that uses the package delivery rate metric in durability measurements durability control unit It includes.

2. Centralization for traffic management and dynamic routing in flying transient networks. It is a fundamental optimization method, and its characteristic is; 10 - All nodes are initialized by the network management unit, each node is on the network. For unique identification, their identities must be determined and initially configuration (1001), - The network topology creation module determines the devices and connections on the network. Creating the network topology by defining it (1002), 15 - The SNR, calculated by the network topology generation module, is the average size of a packet. Inter-node connection based on transmission time and hop count factors. Determination of costs (1003), - The importance of each node in the network is determined by the node centricity calculation module. Calculation of eigenvector centralities to determine (1004), 20 - The transmission power of the node is determined by the parameter configuration calculation module. and adjustment of frame intervals according to the centrality value (1005), - The traffic management module enables the network to quickly adapt to changing conditions. the most suitable routing strategy based on the current network status is needed. the selection (1006), 25 - OpenFlow controller adapts to topology changes according to OpenFlow. Updating routing tables (1007). - The routing algorithm selected by the routing algorithm module With the new routing tables created accordingly, data packets can be routed over the network. transmission (1008) 30 It includes the steps of the process. 13 3. Traffic management and dynamic routing in ephemeral flying networks compliant with Claim 2. It is a centrality-based optimization method for network topology, and its characteristic feature is network topology. SNR, generated by the generation module, is the transmission time of an average-sized packet. and inter-node connection costs according to hop count factors In the process step (1003) of determining the connection costs 𝐷𝑖𝑗 = P 𝐿𝑖𝑗 + 𝜁 5 It is calculated using the equation.

4. Traffic management and dynamic routing in ephemeral flying networks in accordance with Claim 2. It is a centralized optimization method for traffic management. The module enables the network to quickly adapt to changing conditions. Choosing the most appropriate guidance strategy based on the current situation 10 (1006) The routing strategies in the process step are flexible mode, fast mode and hybrid. It is a mod.

5. Traffic management and dynamic routing in ephemeral flying networks compliant with Claim 1. It is a centralized optimization system, characterized by its ability to distribute new data through nodes. 15 that adjusts the power and frame interval values ​​with the equation 𝑃𝑣𝑖 = 𝑃𝑖 − 𝑥𝑣𝑖 × 𝑃𝑝 It includes a parameter configuration calculation module. 25

Images