The paper considers a multi-beam LEO satellite network scenario for cooperative user terminal (UT) positioning, where each satellite can generate multiple beams using a planar antenna array, and each UT can receive positioning signals sent by beams from numerous satellites.
The key highlights and insights are:
The problem of joint satellite beam scheduling and beamforming design is formulated, aiming at optimizing UT TDOA positioning accuracy under per-beam transmission power constraint and the constraint on the number of serving beams for each UT.
To deal with the non-convex mixed-integer problem, it is decomposed into an inner beamforming design problem and an outer beam scheduling problem.
For the beamforming design, a UT SINR threshold adjustment based beamforming algorithm with the semidefinite relaxation (SDR) technique is proposed, which is inspired by the monotonic relationship between the positioning accuracy of a single UT and its perceived SINR.
For the beam scheduling, a fast and efficient greedy heuristic beam scheduling algorithm is developed, considering the trade-off between channel correlation and UT-satellite topology geometry.
Extensive numerical evaluations show that the proposed joint design can improve the average user positioning accuracy by 17.1% and 55.9% compared to conventional beamforming and beam scheduling schemes, respectively.
翻譯成其他語言
從原文內容
arxiv.org
深入探究