Autonomous Hybrid Connectivity Management System for Aircraft
Patent Information
- Application Number
- TR202613393
- Authority / Receiving Office
- TR · TR
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2026-08-07
- Publication Date
- 2026-09-21
Smart Images

Figure 00000014_0000
Abstract
Description
1 TARIFF Autonomous Hybrid Connectivity Management System for Aircraft TECHNICAL AREA 5 The invention has implications for the civil and commercial aviation sector in general, and especially for in-flight wireless technology. connectivity systems, satellite communication infrastructures, and AI-powered connectivity. In-flight hybrid connectivity developed for use in the field of management technologies. It is related to the management system. 10 The invention is particularly useful for LEO (Low Earth Orbit) technology used in aircraft. internet / data), GEO (Geostationary Earth Orbit - Internet / data via satellite), ATG Different communication sources such as (Air-to-Ground) performance, usability and suitability are evaluated in a multidimensional way through artificial intelligence. analyzes and manages connection transitions seamlessly and optimally. It relates to the in-flight hybrid connectivity management system. STATE OF THE ART Today, the increasing number of passengers, especially in the airline industry, means that air travel... With the digitalization of flights and the rising demand for in-flight internet access, in-flights Connectivity systems are undergoing a major transformation. Commercial airlines In transportation, in-flight internet connectivity is primarily used to enhance the passenger experience, during in-flight... 25 to support operations and improve database services is used. The main connection technologies used in this context are; GEO These are satellites, LEO satellites and ATG systems. The aircraft operate in different geographical locations and Seamless and intelligent connectivity between various connection types while navigating at different altitudes. the ability to switch between modes, and high-quality internet access throughout the flight. This is of critical importance. Since GEO satellites are in geostationary orbit, 30 It offers wide coverage and is generally long-distance over the ocean. They are preferred for flights. However, due to their high altitude, there is a delay time. The risk is high. This is especially true for video calls, online meetings, or online gaming. This reduces service quality in time-sensitive applications such as those mentioned. LEO satellites, however... 2 It orbits at a lower trajectory and offers lower latency. However, because they are not fixed, they require a constant satellite pass-through and link renewal. This creates technical challenges in terms of ensuring connection continuity. ATG systems create connections with base stations on the ground. They provide services by establishing; however, these systems only work on landmasses, 5 They cannot be used for flights over seas and oceans. In current communication systems, handover decisions are crucial. mostly predefined fixed rules, prioritization mechanisms or This is done through simple threshold values. For example, “LEO connection 10 LEO is used as long as the connection is available; if the connection is unavailable, GEO is used. a fixed prioritization approach such as "connection is made" or "LEO signal" If the level falls below a certain threshold, an alternative connection will be used. Threshold-based decision-making mechanisms can be implemented, such as "it is passed". Together, these types of static decision-making mechanisms exhibit significant variability during flight. It may be insufficient for evaluating a large number of parameters simultaneously. In particular, the instantaneous traffic load on board the aircraft, and the types of connections and services passengers use, Priority and QoS requirements of applications, different communication methods such as LEO / GEO / ATG instantaneous capacity and density status of resources, signal quality, latency, packets Factors such as loss and weather conditions are not sufficiently taken into account in the decision-making process. 20 This situation means that although the current connection is technically usable, the user the most suitable connection in terms of experience or efficient use of system resources This leads to situations where things might not be possible. In current technological applications, artificial intelligence and machine learning algorithms have 25 Its use is extremely limited. Passengers' connectivity preferences include video viewing and online browsing. Application usage, such as holding meetings and using social media, is connected to the system. The effects on service quality are not being evaluated. Therefore, the current situation... The systems can only make decisions based on whether there is a connection or not; network Resources are being used inefficiently, and service quality varies from user to user. It is changing. Decision-making mechanisms based on static and fixed priorities are no longer the reality. It is unable to adapt to temporal variables. It lacks application level awareness; For example, a passenger watching a video and a passenger texting will have the same network priority. There is a risk of losing communication during transit; especially between LEO satellites. 3 Interruptions may occur during transitions. There is no proactive optimization mechanism. Transitions are usually made after a problem arises. Predictive learning and Adaptation is lacking; a network behavior analysis based on historical data is not performed. Based on research conducted under the known state of the art, US20220377640A1 5 Application number [number] was found. In the said application, the aircraft... LEO / MEO / GEO satellite links and Air-to-Ground (ATG / DA2GC) links. It addresses the ability to dynamically switch between them. The system; existing by evaluating the performance metrics of the connection, alternative to satellite connection It is possible to switch to satellite or ATG connection. Thus, only 10 Instead of a selection based on signal presence, it's a dynamic selection based on connection performance. Connection management is targeted. In the application in question, connection transition Although it offers a more dynamic approach by taking performance metrics into account, It is based on predefined criteria and performance metrics. This Therefore, passenger / application-based QoS requirements vary depending on the 15 users on board the aircraft. Factors such as connectivity needs, current network congestion, service priorities, and weather conditions shortcomings in terms of evaluating contextual variables across three dimensions together. It carries. All these shortcomings in current applications result in a 20% reduction in network quality during the flight. fluctuations, service disruptions, decreased user satisfaction, and in-flight issues. This leads to inefficient use of the connectivity infrastructure. Therefore, the existing The fundamental problem that cannot be solved technically is implementing different communication sources. real-time and AI-powered, with level of awareness The deficiency is the lack of an autonomous connectivity system to manage this. This deficiency is only 25 Not in terms of connection continuity; but in terms of the sustainability of connection quality, passenger This creates a need for improvement in terms of customer satisfaction and system efficiency. As a result, improvements are being made to in-flight hybrid connectivity management systems, Therefore, it will eliminate the disadvantages mentioned above and the existing 30 New structures are needed to provide solutions to these systems. 4 THE PURPOSE OF THE INVENTION The present invention meets the aforementioned requirements and overcomes all the disadvantages. in-flight hybrid connectivity management that eliminates and introduces some additional advantages It is related to the system. 5 The main purpose of the invention is to use LEO (Low Earth Orbit) technology (via satellite) inside aircraft. internet / data), GEO (Geostationary Earth Orbit - Internet / data via satellite), ATG Different communication sources such as (Air-to-Ground) performance, usability and suitability are evaluated in a multidimensional way through artificial intelligence. analyzes and manages connection transitions seamlessly and optimally. The goal is to provide an in-flight hybrid connectivity management system. One purpose of the invention is to meet the connectivity needs on board the aircraft; providing passengers with internet access. usage, media consumption, video conferencing and messaging, flight 15 personnel's data transmission and flight systems' interaction with ground control centers It can be addressed across a wide range of areas, including communication. Another aim of the invention is to improve the connectivity experience for airline passengers. Next in line are airline data management, in-cabin service automation, flight 20 security protocols operate uninterrupted and are resistant to connection interruptions. to offer systems. Another aim of the invention is to improve in-flight connectivity infrastructures in the aviation sector. enabling it to become smart, flexible and user-oriented, multiple communication 25 decision-making that utilizes technologies in an integrated manner and is supported by artificial intelligence. The goal is to provide an autonomous connectivity management system supported by its infrastructure. Another purpose of the invention is to ensure that connection transitions are made solely based on signal strength or connection type. not only based on availability, but also on connection quality, application type, user 30 evaluation according to multidimensional criteria such as behavior and system efficiency to provide. The structural and characteristic features and all the advantages of the invention are given in the figures below. And thanks to the detailed explanation written with references to these figures, it becomes clearer. This will be understood as such. Therefore, the evaluation should also be based on these forms and details. This should be done taking the explanation into consideration. BRIEF DESCRIPTION OF THE FIGURES The best way to utilize the advantages of the existing invention, together with its structure and additional elements. For understanding, it should be evaluated together with the figures explained below. is necessary. Figure 1 Block diagram of the in-flight hybrid connectivity management system, the subject of the invention. 10 It is the appearance. REFERENCE NUMBERS 1. Performance monitoring module 15 2. Multiple connection sources 3. Connection transition decision module 4. User profiling module 5. Connection orchestrator DETAILED EXPLANATION OF THE INVENTION This detailed description explains the invention, which refers to an in-flight hybrid connectivity management system. preferred structures are solely for the purpose of better understanding the subject. and is explained in a way that will not create any limiting effects. 25 The invention, whose block diagram view is given in Figure 1, is intended for use in the civil and commercial aviation sectors. used especially for providing in-flight wireless connectivity, different LEO has different coverage, latency and bandwidth values in its orbits. (Low Earth Orbit - Internet / data via satellite), GEO (Geostationary Earth Orbit - 30 (2) coming from different connection sources such as internet / data via satellite) a main unit for collecting and managing satellite and ground station links acting as a communication node, receiving and directing signals during flight. 6 and the performance of the in-flight connection terminal (1) which performs the connection transfer, usability and suitability analyzed in a multidimensional way through artificial intelligence. an in-flight hybrid that manages connection transitions seamlessly and optimally. It is related to the connection management system. The subject of the invention is a connection management system; Simultaneously utilize all available multi-connection resources (2) that the aircraft can access monitoring, signal strength, latency, bandwidth for each connection source. performance indicators such as width, packet loss, or coverage time a performance monitoring module that measures in real time (1), 10 By taking the data obtained by the mentioned performance monitoring module (1) passengers' internet usage, media consumption, video conferencing or By analyzing connection usage behaviors such as messaging, we can determine traffic types. Long Short-Term Memory (LSTM) Perceived Quality of Service (QoE - Quality of Experience) through the model 15 a user profiling module that estimates its value (4), Markov Decision Process (MDP) based situation past results obtained using the action model and the Q-learning algorithm By processing flight data, it analyzes the available connection conditions for the aircraft. As a result, the most suitable connection transition timing and transition 20 a connection transition decision module (3) that determines the type of connection to be made, Performance monitoring module (1), connection transition decision module (3) and user By combining the data obtained by the profiling module (4), the final link management decisions, online learning, real-time resources 25 Proactive link switching, caching, fault tolerance, and dual link switching. ensuring service continuity and connection quality by using its strategies an AI-powered connection orchestrator (5) It includes. In an example application of the connectivity management system that is the subject of the invention, onboard uninterrupted, efficient communication infrastructure through multiple connection sources (2) and the system works together to ensure that the application is managed in an conscious manner. It complements and works in high cooperation. The system, from the beginning of the flight... 7 It is kept continuously active by the performance monitoring module (1). Performance monitoring module (1), LEO satellites, GEO satellites, ground-based ATG all multiple connection resources (2) accessible to the aircraft, such as systems, simultaneously It monitors signal strength, latency, and bandwidth for each connection type. Performance indicators such as packet loss and coverage time are measured in real-time. These 5 The data shows which type of connection is more suitable at different phases of the flight. This forms the basis for determining the connection performance data obtained for the aircraft. It is evaluated together with user behavior within it. At this point, the user Profiling module (4) is activated. User profiling module (4) of passengers It analyzes what types of applications the connection is used for. For example, video streaming, 10 Different types of traffic include messaging, social media use, and gaming. They are classified and customized connection profiles are created for each passenger. This The profiling process categorizes the system's connectivity resources based solely on technical capacity. not only, but also according to users' experience and quality expectations. It enables evaluation. Performance monitoring module (1) and user 15 By processing the data obtained by the profiling module (4), the central decision of the system is made The connection transition decision module (3) is activated. module (3) determines which connection source to switch to at certain moments of the flight. decides. Transitions between multiple connection sources (2) are only possible via connection. This is done proactively, not when the connection is interrupted, but before the connection quality degrades. This 20 decisions are made by learning from past connection performance and making predictions. This is achieved through artificial intelligence models, particularly through reinforcement learning and decision-making processes. Connection transition decision module (3), Markov Decision Process (MDP - Markov Decision Obtained using a process-based state-action model and the Q-learning algorithm. By processing past flight data, it analyzes the current connection conditions for the aircraft. 25 This results in the most suitable connection transition timing and the transition process. It determines the type of connection. In the system covered by the invention, only which connection is present. It governs not only how it will be used, but also how this link will be shared. The structure is available. The connection transition decision module (3) used for this purpose is available. It decides how connectivity resources will be distributed among passengers. Here are 30 taking into account both individual user experience and the overall efficiency of the system Multi-objective optimization algorithms are used. For example, network congestion. by preventing multiple passengers from having their video viewing experience disrupted, using low bandwidth. This makes it possible to seamlessly maintain messaging services that require high bandwidth. 8 The outputs and interactions of all these elements form the intelligence and decision-making center of the system. The connection is assembled by an AI-powered connection orchestrator (5). The orchestrator (5) coordinates the data flows from other components and the final The link orchestrator (5) makes decisions to meet the changing needs of the system throughout the flight. an artificial intelligence system that can update itself, learn, and replan accordingly. It includes an intelligence infrastructure. It also provides predictability of connection transitions. to increase, maintain data continuity with caching strategies and prevent connection errors It also undertakes tasks such as increasing resistance to fire. The system described in this invention provides real-time performance of all communication sources. monitoring, in-depth analysis of user behavior, connection Managing transitions with learning algorithms, ensuring fair and efficient allocation of connection resources. the allocation in this way and the entire process being managed by a central artificial intelligence unit It is based on orchestration. This structure addresses connection disruptions in existing systems. 15 It offers a proactive, flexible, and user-oriented communication infrastructure. The invention Application areas include; optimizing in-flight internet access, passenger To enhance the user experience, connection quality will be improved using artificial intelligence. Personalization, bandwidth management of in-flight entertainment systems, flight Creating and implementing an uninterrupted communication infrastructure for security reasons 20 In line with layer awareness, different types of communication such as video, voice calls, and messaging are available. The implementation of quality-of-service focused connection allocation for traffic types is in place. It is receiving. The invention relates to a performance monitoring system in an in-flight hybrid connectivity management system. 25 In the module (1), delay, signal strength, packet loss, bandwidth for each link type. Metrics such as breadth and coverage duration estimates are continuously collected, and normalized: Latency: Signal Strength: 30 Package Loss: Bandwidth: Estimated Coverage Duration: 9 This data is used to calculate the overall performance score of the connection as follows: It is converted into a score function: Here, these are weighting coefficients and are dynamically adjusted to the flight scenario, up to 5. adjustable. In the user profiling module (4), the purpose for which the passenger uses the connection is determined. is detected. For this purpose, a multi-layered structure has been proposed. For traffic type recognition. Deep Packet Inspection (DPI) or a pre-trained neural network (e.g., 1D-10) Data packets are classified using CNN. Passenger usage history is used for user profiling. and based on in-flight behavior, K-means or Gaussian Mixture Model (GMM) Users are grouped using an LSTM (Learning, Technology, and Environment) system. The user's application experience is analyzed using this system. The model is used to make predictions from historical data. In the link transition decision module (3), link transitions are classic "yes-no" or signal transitions. not based on power, but on the Markov Decision Process (MDP) and It is determined by reinforcement learning. MDP Definition: Status(s): Current connection type + metrics + user density 20 Action (a): Link transition (LEO → GEO, GEO → ATG, etc.) Reward (r): Increased user QoE, no disconnections, bandwidth efficiency Here, the best decision is policy. 25 Reinforcement Learning Method: By using Q-Learning or Deep Q Network (DQN), the best over time Transition strategies are learned. In the system described in the invention, all outputs are integrated into a central decision-making system. This 30 The structure is modular and a continuously learning system, with policies that adapt to the situation. based on the principles of updating, real-time monitoring and rescheduling. It works. The system uses caching during connection transitions, short-term binary connections, It utilizes strategies such as fault tolerance and predictive redirection. The invention, Real-time analysis of connectivity resources during flight, predictive transition. Maximizing connection quality through management and implementation level awareness. It offers a multi-layered, AI-based solution that enables its extraction. 5 The technical infrastructure it offers includes machine learning, traffic analysis, QoE modeling, and enhanced systems. combining contemporary algorithms such as learning and multi-objective optimization It brings about these differences from existing systems, both in terms of passenger experience. and an innovative architecture that maximizes system resource efficiency It proposes that the invention only measures the connection quality at a technical level and enables switching. 10 No, it's about proactively managing connectivity resources and improving user experience. taking into account the level of application that directly affects and for each passenger It aims to provide a personalized connectivity experience. All these capabilities, These are unique and advanced features not found in existing systems. The system, connectivity... its infrastructure is efficient and intelligent at both technical and experiential levels 15 By enabling its management, it offers an innovative contribution to the aviation sector. The system described in this invention differs from traditional connection management systems in that... Connection transitions and resource allocation are managed not according to fixed rules, but through learning and predictive processes. through an artificial intelligence architecture that can perform actions and make adaptive decisions It accomplishes this. The most fundamental uniqueness of the system is that the connection infrastructure is only 20 User-based analysis at the application level, not based on physical metrics like signal quality. It is the system's assessment that takes into account a passenger's behavior. In this way, the system evaluates a passenger's behavior. whether they watched the video, whether they joined the online meeting, or just It can distinguish between text-based messaging and connection sources accordingly. It can manage. In the decision-making infrastructure of the invention, how and what connection transitions occur. The timing is managed by an AI-powered process. For this purpose, within the system Connection types are modeled as states, and possible transitions as actions. The system, Over time, we will find our own way to navigate between these situations and actions. They have the ability to learn from experience. This learning process is reinforced. Learning algorithms are used. Thus, the system finds the most suitable connection during flight. 30 By switching to this mode in advance, it prevents connection drops or sudden speed reductions. This structure, unlike classical systems, simply waits for the signal to weaken before switching. Instead of a logic that does this, it considers flight route, connection density, expected coverage time. It acts based on predictions by analyzing variables such as these. At the same time, the system... 11 the ability to analyze the communication behavior of each passenger on board an aircraft It determines which applications the passenger used the connection for and this A unique connection profile is created for that passenger based on the information. For example, a passenger watching a video. someone who only uses a messaging app and needs a certain level of connection quality Passengers have different needs. The system analyzes these differences and allocates network resources accordingly. It optimizes accordingly. In this process, artificial intelligence (AI) works on time series data. Learning models are used, and the immediate quality expectation is predicted for each passenger. The system not only makes transition decisions; it also manages connection resources for all passengers. It also decides how the total capacity of the network will be distributed among them. In this process, the total capacity of the network is determined. This creates a balance between the quality of service each passenger needs. Thus, 10 While some passengers are prevented from consuming excessive resources, others are also allowed to consume a minimum amount. needs are met. In this multi-objective optimization structure, evolutionary algorithms and Modern computational methods, such as meta-heuristic techniques, are used. [Link] The transitions themselves are also supported by specific strategies. The system, during the transition... It implements various measures to prevent data loss. These include two short-term 15-day measures. The connection must be active simultaneously, and critical data must be pre-cached during the transition. These methods include obtaining and smoothing connection delays. The invention One of its most unique aspects is that decision-making processes largely take place inside the aircraft, This means it happens at the endpoint units. This allows the connection to be interrupted. Even in those moments, the system can continue to function because the decision-making mechanisms are external to a 20-degree angle. It is not dependent on the center. This structure is important in terms of flight safety and continuity of connection. It provides a significant advantage.
Claims
12 REQUESTS 1. To provide in-flight wireless connectivity in the civil and commercial aviation sector. Used for this purpose, different coverage, delay and bandwidth at different orbits LEO (Low Earth Orbit - Internet / data via satellite), with width values of 5 GEO (Geostationary Earth Orbit - Internet / data via satellite) and different types of services performance, availability and suitability of connection resources (2) analyzing connection transitions in a multidimensional way through artificial intelligence an in-flight hybrid connectivity management system that manages everything seamlessly and optimally. and its feature is; 10 Simultaneously utilize all available multi-connection resources (2) that the aircraft can access monitoring, signal strength, latency, bandwidth for each connection source. performance indicators such as width, packet loss, or coverage time a performance monitoring module that measures in real time (1), By taking the data obtained by the mentioned performance monitoring module (1), 15 passengers' internet usage, media consumption, video conferencing or traffic by analyzing connection usage behaviors such as messaging. a user profiling module that determines the type (4), Obtaining historical flight data and processing this data to improve in-flight availability analyzes the connection conditions and as a result, determines the most suitable connection 20 a method that specifies the transition timing and the type of connection to be used for the transition connection transition decision module (3), Performance monitoring module (1), connection transition decision module (3) and By combining the data obtained by the user profiling module (4), the final making connection management decisions, ensuring service continuity and connection 25 an AI-powered link orchestrator that ensures its quality (5) It includes.
2. An in-flight hybrid connectivity management system according to Claim 1, characterized by its long-haul capability. Through the Long Short-Term Memory (LSTM) model, 30 An estimate of the perceived quality of service (QoE - Quality of Experience) value. It includes a user profiling module (4). 13 3. An in-flight hybrid connectivity management system according to Claim 1, featuring Markov State-action model based on Markov Decision Process (MDP) and A system that processes historical flight data obtained using the Q-learning algorithm. It includes the connection transition decision module (3).
4. According to Claim 1, it is an in-flight hybrid connectivity management system, the feature of which is; online learning, real-time resource reallocation, and policy adaptation. operating in accordance with its principles, proactive connection switching, caching, error correction. By using tolerance and dual-link transition strategies, service continuity and an AI-powered connection orchestrator that ensures connection quality (5) 10 It includes.