Source-linked AI summary
Beam Tracking for UAV Mounted SatCom on-the-Move with Massive Antenna Array
Jianwei Zhao, Feifei Gao, Qihui Wu, Shi Jin, Yi Wu, Weimin Jia
TL;DR
UAV navigation destabilizes satellite beam pointing in Ka-band UAV-satellite communication, motivating a blind tracking method for massive antenna arrays. The method combines mechanical stabilization and dynamic isolation with electrical simultaneous-perturbation refinement, and simulations show precise pointing and convergence under the evaluated conditions.
Problem
UAV navigation continually changes beam pointing in SatCom on-the-move, while attitude and measurement errors complicate maintaining alignment with the target satellite.
Method
The method combines mechanical beam stabilization and dynamic isolation with electrical array-weight adjustment using an array-structure simultaneous perturbation algorithm.
Results
The proposed tracking reaches final azimuth and elevation angle errors on the order of 10^-2, while requiring 4 iterations at SNR=20dB and 7 at SNR=10 dB.
Takeaways & Limitations
Joint mechanical and electrical adjustment can precisely point the spatial beam toward the target satellite while tracking satellite DOA instead of the full spatial beam.
Abstract
from arXiv · showhide
Unmanned aerial vehicle (UAV)-satellite communication has drawn dramatic attention for its potential to build the integrated space-air-ground network and the seamless wide-area coverage. The key challenge to UAV-satellite communication is its unstable beam pointing due to the UAV navigation, which is a typical SatCom on-the-move scenario. In this paper, we propose a blind beam tracking approach for Ka-band UAVsatellite communication system, where UAV is equipped with a large-scale antenna array. The effects of UAV navigation are firstly released through the mechanical adjustment, which could approximately point the beam towards the target satellite through beam stabilization and dynamic isolation. Specially, the attitude information can be realtimely derived from data fusion of lowcost sensors. Then, the precision of the beam pointing is blindly refined through electrically adjusting the weight of the massive antennas, where an array structure based simultaneous perturbation algorithm is designed. Simulation results are provided to demonstrate the superiority of the proposed method over the existing ones.
I. INTRODUCTION
UAV-satellite communication supports integrated space-air-ground networking, but UAV navigation continually disrupts beam alignment. The paper proposes blind hybrid mechanical-electrical tracking for Ka-band massive arrays to maintain and refine satellite pointing.
- Motivation: UAV-satellite communication can relay between terrestrial and satellite networks to serve users in infrastructure-limited areas or assist crowded communications.The paper motivates UAVs as relays for disaster-area coverage, service recovery, and base-station offloading.
- Motivation: Ka-band enables dense antenna deployment on UAVs, providing spatial gain against path loss while avoiding expensive high-gain directional antennas.Its short wavelength permits many small antennas in a compact area, reducing overall system cost.
- Challenges: Continuous UAV navigation changes the channel, while limited RF chains make exhaustive beam sweeping resource-intensive and linear movement models unsuitable for nonlinear motion.These constraints motivate beam tracking tailored to the UAV-satellite link.
- Challenges: Successful UAV-satellite communication requires the spatial beam to lock onto the satellite despite yaw, pitch, and roll variations.The beam is preferably steered near the antenna-plane normal because array gain depends on the projected aperture area.
- Proposed approach: The proposed blind hybrid method first uses beam stabilization and dynamic isolation for mechanical coarse alignment, then electrically refines pointing with an array-structure simultaneous perturbation algorithm.The method targets a massive uniform plane array and combines navigation compensation with signal-based refinement.
- System model: The channel model uses a massive M × N uniform plane array and a single satellite antenna, with limited RF chains and a dominant line-of-sight path.The model represents array responses through azimuth and elevation directions and includes propagation attenuation factors.
B. Problem Formulation
The UAV-satellite link requires accurate beam pointing despite navigation-induced attitude changes. The proposed system combines mechanical adjustment, low-cost attitude sensing, and coordinate-frame transformations to obtain coarse alignment.
- B. Problem Formulation: Beam-pointing error degrades communication quality, and UPA array gain declines with the sine of the satellite elevation angle as the beam moves away from the array normal.
- B. Problem Formulation: UAV navigation, including heading, pitching, and rolling, is a major source of beam-pointing variation, while satellite perturbation can be neglected under position-keeping.
- B. Problem Formulation: The system uses perception sensors, an arithmetic and control unit, and mechanical or electrical adjustment to stabilize the beam according to measured UAV attitude.
- B. Problem Formulation: Low-cost MEMS gyros, accelerometers, and GPS are proposed to derive UAV attitude information instead of using costly high-precision inertial navigation systems.
- III. COARSE BEAM ALIGNMENT WITH MECHANICAL ADJUSTMENT: Mechanical adjustment provides coarse beam tracking through two steps: beam stabilization and dynamic isolation.
- A. Beam Stabilization: Beam stabilization jointly uses UAV attitude and satellite location information, represented through transformations among n-frame, b-frame, a-frame, f-frame, and t-frame.
- A. Beam Stabilization: The n-frame is geodetic, the b-frame is UAV-fixed, the a-frame is azimuth-rotary, the f-frame is UPA-based, and the t-frame represents the spatial beam.
- A. Beam Stabilization: The b-frame-to-t-frame transformation is constructed through continuous rotations involving azimuth α, elevation β, and polarization γ, yielding the beam-alignment angles.
B. Dynamic Isolation
Dynamic isolation models the beam’s physical angular rate as the sum of UAV-navigation and control-monitor contributions. The method uses coordinated azimuth, elevation, and polarization adjustments to cancel navigation-induced motion.
- B. Dynamic Isolation: The UAV-navigation angular rate ω_ub is measured by gyros and is coupled to beam motion through geometric and friction constraints.
- B. Dynamic Isolation: Dynamic isolation compensates ω_ut by jointly controlling azimuth, elevation, and polarization so the spatial beam points toward the target satellite.
- B. Dynamic Isolation: The control-monitor contribution ω_mt combines the transformed azimuth, elevation, and polarization angular rates.
- B. Dynamic Isolation: The beam’s physical angular rate ω is the sum of UAV-navigation rate ω_ut and control-monitor rate ω_mt.
- B. Dynamic Isolation: Maintaining ω = 0 is the design condition for keeping the spatial beam pointed at the target satellite despite UAV attitude variation.
- B. Dynamic Isolation: Mechanical adjustment is blind because it uses attitude information rather than training symbols for beam tracking.
IV. FINE BEAM ALIGNMENT WITH ELECTRICAL ADJUSTMENT
Mechanical adjustment alone cannot provide sufficiently accurate beam pointing because of system and measurement errors. Electrical adjustment is therefore proposed to improve pointing precision.
- IV. FINE BEAM ALIGNMENT WITH ELECTRICAL ADJUSTMENT: System and measurement errors make purely mechanical adjustment insufficient for accurate beam pointing, motivating a subsequent electrical adjustment method.
A. Electrical Adjustment with Simultaneous Perturbation
The electrical adjustment refines coarse beam alignment by using received-signal power and array-structure-based simultaneous perturbation, enabling blind fine tracking with a single RF chain.
- Motivation: Single-RF-chain beamforming requires direction-of-arrival information, while sequential phase-shifter perturbation is time-consuming and scales poorly with massive arrays.The proposed approach instead uses the antenna-array structure to accelerate convergence.
- Algorithm: Electrical adjustment uses two noisy received-power measurements to estimate the gradient, with random Bernoulli perturbations and scheduled step parameters.The beamforming vector is updated iteratively until received power increases only marginally.
- Algorithm: The method is named array structure based simultaneous perturbation because it exploits antenna-array structure during electrical adjustment.The design is presented as Algorithm 1.
- Algorithm: The hybrid method first stabilizes the beam mechanically, then performs array-structure-aided DOA tracking and electrical fine alignment.The algorithm initializes the beam, applies stabilization and dynamic isolation, perturbs the beam, estimates a gradient from power measurements, and updates the beam direction.
- Convergence: The proposed algorithm’s convergence speed is independent of the number of phase shifters, or equivalently the number of array antennas, and is faster than traditional techniques.The paper attributes this speed advantage to using the array structure for beam tracking.
B. Complex Gains Estimation
The experiments validate attitude estimation and the two-stage beam-tracking process: mechanical adjustment provides coarse alignment, while electrical adjustment reduces the remaining pointing error.
- Attitude determination: Kalman-filter-based data fusion integrates low-cost sensor measurements and produces attitude estimates that coincide with the practical attitude values.The fused attitude is compared with measurements from the XW-ADU7612 reference system.
- Attitude determination: The maximum attitude error from data fusion reaches 0.5°.This residual error is consistent with the paper’s analytical results.
- Mechanical adjustment: Mechanical adjustment keeps azimuth and elevation beam-pointing errors within 0.5° for most of the time but only approximately aligns the beam.The residual attitude error is coupled into beam pointing and causes deviation from the target satellite.
- Electrical adjustment: The prior normalized received signal power is 0.952, confirming that mechanical adjustment approximately points the beam toward the target satellite.Electrical perturbation methods then converge toward maximum received power, with the proposed method compared against methods and.
- Electrical adjustment: At SNR=20dB, the proposed method converges in 4 iterations, increasing to 7 iterations at SNR=10dB.Lower SNR increases noise dominance and slows electrical-adjustment convergence.
- Electrical adjustment: At SNR=10dB, the eventual azimuth and elevation angle errors remain on the order of 10^-2 as iterations increase.The paper concludes that joint mechanical and electrical adjustment precisely points the beam to the target satellite.
VI. CONCLUSIONS
The paper proposes blind beam tracking for Ka-band UAV-satellite communication using hybrid massive antennas, combining mechanical coarse alignment with electrical fine alignment.
- Mechanical adjustment counters UAV navigation through beam stabilization and dynamic isolation, providing coarse beam alignment.
- Electrical adjustment further improves tracking precision and provides fine beam alignment.
- Tracking the satellite’s spatial beam through its direction of arrival reduces training overhead and improves system efficiency.
LOW-COST SENSORS
The UAV attitude is measured using low-cost gyros, an accelerometer, and dual-GPS carrier-phase measurements, producing the sensor inputs needed for attitude determination.
- MEMS Gyros: Three MEMS gyros measure UAV attitude variation through roll, pitch, and yaw angular-rate outputs.
- MEMS Gyros: Gyro integration provides rough measurements of the UAV’s three-dimensional attitude angles.
- Accelerometer: An accelerometer derives UAV attitude from local gravity acceleration, including measured pitch and roll angles.
- GPS: Two GPS antennas provide the measured yaw angle through carrier-phase differential technology using their baseline coordinates.
APPENDIX B LOW COST ATTITUDE DETERMINATION WITH DATA
The attitude-determination procedure represents UAV orientation with normalized quaternions and fuses low-cost sensor measurements through a Kalman filter to track attitude accurately.
- Low-cost sensors can introduce gyro drift and sideslip errors, motivating Kalman-filter data fusion for accurate attitude estimation.
- Quaternion representation uses four parameters to describe the UAV’s three-dimensional attitude and supports attitude updating over the sampling interval.
- The quaternion is maintained in normalized form during the attitude-estimation procedure.
- Sensor-derived coarse attitude values provide quaternion measurements for the measurement equation.
- The Kalman filter combines the system and measurement equations, updating and normalizing the quaternion across time steps.
- The tracked quaternion is transformed into attitude angles for subsequent attitude adjustment.