Abstract: To achieve dynamic tracking of ground directional antennas for relay drones, ensuring that the main beam of the antenna is always aligned with the drone and that the relay communication signal strength remains optimal, we designed an automatic tracking platform for the ground end of the drone based on RSSI blind estimation tracking. By extracting the antenna RSSI signal through Telnet and optimizing it using the Kalman filtering algorithm, we improved the initial capture speed through rectangular scanning and completed dynamic tracking of the antenna using a stepping ‘cross’ tracking scheme. The software and hardware of the automatic tracking platform were designed, and the physical platform was tested. The test results show that the tracking platform has good tracking speed and accuracy, meeting the needs of drones for self-tracking of directional antennas.
0 Introduction
In both winning information warfare and carrying out non-war military operations, the communication support capability of the troops has been put forward with high demands. Currently, the requirements of command information systems for ‘broadband communication, dynamic communication, interference-resistant communication, and communication in mountainous areas’ are becoming increasingly urgent. Practice has proven that micro-unmanned aerial vehicle (MUAV) relay communication systems are one of the effective means to solve these problems. Micro-unmanned aerial vehicles, equipped with communication devices, fly in the air as communication relay nodes, establishing a broadband network within tactical range with ground communication nodes, enabling high-speed transmission of data, voice, and images among various nodes. Due to the limitations of the payload of micro-unmanned aerial vehicles, the onboard communication equipment must meet the requirements of miniaturization, lightweight, and low power consumption. At the same time, to meet the coverage requirements of relay communication, the onboard antenna can only use omnidirectional antennas with low gain and wide coverage. Under the condition that the transmitter power cannot be further increased, its EIRP (Effective Isotropic Radiated Power) value is constrained, and it can only rely on the ground receiving system to improve the antenna gain to compensate. The ground end uses a directional antenna with automatic tracking capability, ensuring that the main beam of the directional antenna is aligned with the onboard antenna with a certain degree of accuracy, leveraging the high gain advantage of the directional antenna to achieve reliable communication link connections, further expanding the coverage of the tactical network, improving communication bandwidth and quality, and reducing the technical requirements for onboard equipment of the drone.
To ensure the reliability of the communication link between the drone and the ground end, reference [6] developed a low-cost drone antenna tracking platform, composed of an inertial measurement unit, global positioning system, and servo motor. Reference [7] designed an array antenna tracking system using single pulse tracking technology. Jenvey et al. from Monash University in Australia designed a ground antenna tracking platform using single pulse tracking to maximize the link quality of the video connection between small drones and the ground end. As shown in Figure 1, when the distance between the drone and the ground station is within 700 m, this platform can implement stable and reliable tracking of the drone, enhancing video transmission quality. Daniel Stojcsics from Obuda University designed a drone ground station antenna tracking platform using program tracking technology to improve the drone’s flight control range. Huang Wei from the China Aerospace Science and Technology Corporation designed a drone measurement and control communication directional antenna tracking system based on GPS-guided tracking algorithms.
Among existing research results, program tracking technology based on GPS positioning is mature and reliable. However, in micro-unmanned aerial vehicle relay communication systems, if the ground directional antenna uses program tracking mode, it requires the micro-unmanned aerial vehicle to be equipped with GPS modules and wireless transmission devices, which will inevitably increase the payload and power consumption of the micro-unmanned aerial vehicle, greatly reducing its endurance and affecting the effectiveness of the relay communication system. Single pulse tracking undoubtedly has the highest tracking accuracy and unmatched advantages, but its complex design and high cost make it unsuitable for micro-unmanned aerial vehicle antenna tracking platforms. Currently, blind estimation antenna tracking technology is still in its infancy, and there are very few systems that have been put into application and are available for reference. Based on existing research results, this paper designs a new MUAV antenna automatic tracking platform based on RSSI (Received Signal Strength Indication) blind estimation.
1 Platform Working Principle and Composition
1.1 Platform Working Principle
To reduce platform complexity, we selected the stepping tracking technology based on RSSI blind estimation as the tracking control scheme. The specific working principle is as follows: the ground end does not rely on the drone’s positioning information and directly extracts the ground end antenna RSSI as a reference under the premise of unobstructed line of sight, completing the initial capture by searching for the signal strength threshold through rectangular scanning. The stepping tracking algorithm adjusts the antenna direction to achieve dynamic tracking of the drone. The working principle of the platform is shown in Figure 2.
1.2 Platform Composition
The antenna automatic tracking platform mainly consists of the antenna, signal strength extraction unit, core control unit, execution unit, functional extension unit, and power supply. The antenna is the device for sending and receiving communication signals and is also the control object of the tracking system. The platform uses a grid antenna with a gain of 19 dBi. The signal strength extraction unit completes the extraction of the antenna RSSI. The core control unit processes the extracted antenna RSSI and issues control commands to the execution unit. After receiving the main control unit’s instructions, the execution unit drives the antenna to the specified position. The schematic diagram of the platform composition is shown in Figure 3.
2 Platform Hardware Design
2.1 Core Control and Signal Extraction Unit
The platform uses the Arduino UNO development board, which has open-source hardware resources, low cost, and strong scalability, as the core control unit of the antenna automatic tracking platform, as shown in Figure 4.
The signal strength extraction unit uses an Arduino Ethernet expansion board along with a dual LAN (Low Noise Amplifier) port POE (Power Over Ethernet) power supply. The Arduino Ethernet control module is an expansion board with an embedded W5100 chip, as shown in Figure 5.
It integrates a fully hardware-based TCP/IP (Transmission Control Protocol/Internet Protocol) protocol stack, Ethernet media transmission layer, and physical layer, which has been validated in the market for many years.
The Arduino Ethernet expansion board is connected to the Arduino UNO control board via a long pin header in a bus format, and the RJ-45 interface of the Arduino Ethernet expansion board is connected to one LAN port of the POE power supply. The other LAN port of the POE power supply is directly connected to the communication receiver, as shown in the connection diagram in Figure 6.
2.2 Execution Unit
The execution unit includes the turntable and the servo components inside the turntable. The turntable is the device that supports the antenna, while the servo components inside the turntable are controlled by the controller to rotate the turntable according to the tracking algorithm, driving the antenna to track the relay drone. Depending on its rotation characteristics, the turntable can be divided into horizontal turntables that can only rotate left and right and omnidirectional turntables that can rotate both left and right and up and down. The real-time tracking of the antenna for the relay drone requires the antenna to adjust its direction in both azimuth and elevation planes. Therefore, the platform uses an omnidirectional turntable structure. The schematic diagram of the turntable structure is shown in Figure 7.
The servo components inside the turntable are an important part of the entire execution unit and are the key to the design, mainly including the selection of servo motors and position detection elements. To reduce size and save costs, the platform uses servos as the servo motors for the turntable. The technical parameters are shown in Table 1.
3 Software Design
3.1 RSSI Extraction
Establish a Telnet connection between the controller and the communication receiver to extract the signal reception strength of the communication receiver. To establish a Telnet connection, it is necessary to initialize the Arduino Ethernet first, which includes releasing buffer data, setting the starting and ending addresses of the buffer, and setting the local IP address and MAC address. Secondly, to establish a Telnet session, authentication must be done through a username and password. In this design, both the username and password are set to admin by default. By calling the EthernetClient::connect(IPAddress ip, uint16_t port) function to establish a Telnet connection with the communication receiver, where the parameter ip is the IP address of the communication receiver and port is the port number, defaulting to 23. The process of establishing a Telnet connection is shown in Figure 8.
Once the Telnet connection is successfully established, the controller sends a status request command to the communication receiver, which is sent as a string. After receiving the status request command, the communication receiver will send all its status information to the W5100 module, which is also stored as a string in the W5100 module’s receive buffer. During the tracking process, the antenna automatic tracking platform only needs the RSSI information from the status information, as saving the remaining information would consume the system’s storage resources. Therefore, it is necessary to extract the string to obtain useful information.
3.2 RSSI Filtering Processing
The propagation of wireless signals in space is complex, often involving multipath, scattering, and electromagnetic interference, which can cause serious time-varying characteristics and large fluctuations in the received RSSI, often leading to mis-tracking or severe jitter of the tracking platform. Therefore, filtering algorithms need to be employed to filter and optimize the collected RSSI data, removing sudden data and noise fluctuations from the RSSI data. The optimized RSSI values are then used for tracking calculations. Figure 9 shows the model of the Kalman filter for RSSI filtering.
First, establish a measurement system based on the RSSI measurement environment. The established system model does not require extreme precision; it can use this system model to estimate the next state. Assuming the current state of the system is k, it can be estimated based on the previous state k-1. Assuming that the RSSI extracted at the current moment is the same as the previous moment, and since there is no control quantity in the system, the estimated result for the current state is as follows:
According to equation (3), we obtain the optimal estimate of the RSSI at state k. To continue the recursion, we update the covariance of the RSSI at state k (K|K), since the RSSI measurement is a single model and single measurement system. This gives us the covariance at state k:
When the system recurses to state k+1, P(K|K) becomes P(K-1|K-1) from equation (2), allowing the algorithm to continue recursively.
3.3 Initial Capture
Initial capture is to ensure that the drone enters the main beam range of the ground directional antenna to obtain a certain relay communication signal. This platform uses a rectangular scanning method for the initial capture of the drone, determining the success of the capture using a threshold judgment method. The scanning schematic is shown in Figure 10.
3.4 Dynamic Tracking
The dynamic tracking process adopts the stepping tracking system, which can be described as ‘cross’ tracking. It is a cyclical process of sampling, comparing, and stepping, allowing the antenna to dynamically search for the RSSI extreme values with smaller step sizes based on the initial capture. The flowchart is shown in Figure 11. The specific implementation process is as follows: the azimuth and elevation axes rotate in a right-left-up-down stepping order, during which the current RSSI is collected and compared with the previous RSSI. If the current RSSI > previous RSSI, the antenna continues to step forward in the previous direction; conversely, if the current RSSI < previous RSSI, it steps back in the opposite direction. After each right-left-up-down stepping sequence, an RSSI maximum value can be determined, and this process continues, dynamically searching for the RSSI extreme values until the RSSI value remains above the tracking threshold after multiple ‘cross’ tracking processes.
4 Platform Testing
Testing Location: An open area in a park
Testing Environment: No obstructions, communication signals meet line-of-sight propagation
Testing Steps:
(1) To simplify the testing model, set up a network with two communication devices, omitting the ground communication transmitter. A micro-unmanned aerial vehicle carrying a communication device flies as the ground transmitter for relay communication, while another communication receiver is connected to the antenna tracking platform, serving as the ground receiver. During testing, the antenna tracking platform is connected to a PC, allowing observation of RSSI changes through the Arduino IDE serial monitor.
(2) The micro-unmanned aerial vehicle is controlled by the flight control system, performing uniform circular motion at a distance of 500 m from the ground end and an altitude of 100 m within a radius of R=50 m.
(3) Set the RSSI sampling frequency to once per second, first testing the RSSI changes without the antenna tracking platform, manually adjusting the antenna’s direction while observing and recording the RSSI changes.
(4) Test the antenna tracking platform. After ensuring the connections of the antenna tracking platform are correct, turn on the platform, observe and record the RSSI changes. The testing diagram of the platform is shown in Figure 12.
As shown in Figure 13, this is the RSSI display in the Arduino IDE serial monitor on the PC.
As shown in Figure 14, this displays the changes in 45 sets of RSSI data under the same conditions during manual tracking and automatic tracking of the ground end antenna.
The test results indicate that manual tracking cannot align the antenna beam with the relay drone, resulting in the ground end antenna’s RSSI failing to reach the desired value, and the RSSI value fluctuates significantly, failing to meet the signal strength requirements for the micro-unmanned aerial vehicle relay communication. After multiple RSSI samplings, automatic tracking can maintain the RSSI value within an ideal range, indicating that the antenna tracking platform designed in this paper effectively achieves antenna tracking.
5 Conclusion
This paper presents a ground antenna automatic tracking platform based on RSSI blind estimation control scheme in the context of a specific micro-unmanned aerial vehicle relay communication system. Corresponding software and hardware designs were carried out, and testing was conducted, with results showing that the designed antenna automatic tracking platform exhibits good tracking performance and improves the communication quality of relay communications.
References
[1] Jia Pengwan, Feng Shoupeng, Zhang Aihui. A Virtual Teaching System for Drone Relay Communication Based on VR-Platform[J]. Armament Automation, 2012, 31(6): 93-96.
[2] He Yi, Jiang Fei, Zhang Yani. Research on Relay Communication of Command System Based on Multi-Rotor UAV and 4G[C]. Beijing: The 3rd China Command Control Conference, 2015.
[3] LIU Z Q, ZHANG Y S. A Novel Broad Beamwidth Conformal Antenna on Unmanned Aerial Vehicle[J]. Antennas and Wireless Propagation Letters, IEEE, 2012, 11: 196-199.
[4] MIN B C, MATSON E T, JUNG J W. Active Antenna Tracking System with Directional Antennas for Enhancing Wireless Communication Capabilities of a Networked Robotic System[J]. Journal of Field Robotics, 2015.
[5] GU Y, ZHOU M, FU S. Airborne WiFi Networks through Directional Antennae: An Experimental Study[C]. Wireless Communications and Networking Conference (WCNC), 2015 IEEE. IEEE, 2015: 1314-1319.
[6] GANTI S R, KIM Y. Design of Low-Cost On-Board Autotracking Antenna for Small UAS[C]. Information Technology-New Generations (ITNG), 2015 12th International Conference on. IEEE, 2015: 273-279.
[7] GEZER B L. Multi-Beam Digital Antenna for Radar, Communications, and UAV Tracking Based on Off-the-Shelf Wireless Technologies[D]. Monterey California. Naval Postgraduate School, 2006.
[8] JENVEY S. A Portable Monopulse Tracking Antenna for UAV Communications[C]. 22nd International Unmanned Air Vehicle Systems Conference. 2007: 1-8.
[9] STOJCSICS D, MOLN?魣R A. AirGuardian-UAV Hardware and Software System for Small Size UAVs[J]. Int J Adv Robotic Sy, 2012, 9(174).
[10] Huang Wei, Zhou Nain, Wang Cheng, et al. Construction of a UAV Communication Antenna Servo System[J]. Electronic Technology Application, 2009, 35(6): 98-101.
[11] Yang Run, Yan Kaiyin, Ma Shuwen. Design and Implementation of a Two-Axis Gimbal for Small Unmanned Aerial Vehicles[J]. Automation and Instrumentation, 2014(7): 165-168.
