Development of Methods for Multithreaded Data Transmission of the Application Layer for Unmanned Aerial Vehicles
Students Name: Chornyi Bohdan Antonovych
Qualification Level: magister
Speciality: Information Communication Networks
Institute: Institute of Telecommunications, Radioelectronics and Electronic Engineering
Mode of Study: full
Academic Year: 2021-2022 н.р.
Language of Defence: ukrainian
Abstract: Unmanned aerial vehicles (UAVs) are actively used to solve problems of monitoring large land areas (eg, video monitoring of the area), taking various readings using sensors (eg, the Earth’s electromagnetic field). Most research in this area is currently being conducted for scenarios with a single flying source node that transmits data to a ground station via repeater nodes. Such a network is a flying self-organizing network (Flying Ad Hoc Networks - FANET) with one source. New solutions are echoed in the design of such networks; Thus, LoRa technology for inter-machine interaction (M2M) can be used to solve "smart city" problems. Self-organizing UAV networks solve the problem of increasing the coverage area of the study area and improving the quality of data transmission. However, the time of the mission remains significant given that the data source is only one node, which needs to fly over the entire planned area. All this can negatively affect the implementation of the mission in emergency situations (EM), when it is important to quickly and efficiently transfer data from the UAV to the ground station. The use of several flying source nodes in the network can divide the surveyed area into parts according to their routes and simultaneously record and transmit data in multi-stream mode from video cameras or sensors to the ground station in real time, which will significantly reduce mission time. An important task is the development of methods and algorithms to improve the characteristics of data transmission in the UAV network with multiple nodes-sources in multithreading. This task is new and in the first stage requires the development of the simplest scenario with a minimum number of active source nodes (two, three source nodes). Also, there are already requests from the industrial sector for video capture missions from two simultaneous source nodes (more than three nodes require increased material costs). To date, the following data transfer methods have been more widely used in single source node scenarios: Automatic Repeat Request (ARQ) and Forward Error Correction (FEC). These methods have demonstrated their effectiveness in wired and wireless networks. However, scenarios for the application of these methods in a self-organizing network with multiple sources have not been sufficiently studied. The use of ARQ at the channel level in the UAV network with multiple sources can cause overload of the input buffers of data of flying nodes, packet loss and delay, which can adversely affect the quality of the received data. The use of FEC involves the addition of redundancy, which can also have a negative impact on the amount and minimum required speed of transmitted data. Also important for UAV manufacturers is the economic costs of development, production and operation. When implementing data transmission methods at the channel and physical levels, it becomes necessary to purchase new equipment or significant costs for its development, in some cases it is also necessary to reconsider the design features of the UAV. In this regard, the best way to solve the problem of improving the quality of data transmission is to focus on research and development of new methods and algorithms for data transmission at higher levels (network, transport, application level), where new algorithms are implemented programmatically without any changes to the hardware. When implementing data transmission methods at the channel and physical levels, it becomes necessary to purchase new equipment or significant costs for its development, in some cases it is also necessary to reconsider the design features of the UAV. In this regard, the best way to solve the problem of improving the quality of data transmission is to focus on research and development of new methods and algorithms for data transmission at higher levels (network, transport, application level), where new algorithms are implemented programmatically without any changes to the hardware. When implementing data transmission methods at the channel and physical levels, it becomes necessary to purchase new equipment or significant costs for its development, in some cases it is also necessary to reconsider the design features of the UAV. In this regard, the best way to solve the problem of improving the quality of data transmission is to focus on research and development of new methods and algorithms for data transmission at higher levels (network, transport, application level), where new algorithms are implemented programmatically without any changes to the hardware. The object of the study is process of transmitting multi-stream video data in selforganized UAV networks, characterized by the loss of fragments of application layer data from several flying source nodes. Subject of study - methods of multithreading transmission with recovery of lost fragments of application layer data in a self-organizing UAV network, providing improved transmission quality from several flying source nodes. The aim of the work is to improve the quality of multithreaded video transmission in self-organized UAV networks by recovering lost fragments of application layer data based on retransmission from several flying source nodes. Key words: 5G; MIMO; NOMA; LTE-U; OMA; throughput.