In the era of emerging low-altitude economies, China UAV technologies have become pivotal for flexible and rapid-deployable communication services. However, the dependence on Global Navigation Satellite Systems (GNSS) poses significant vulnerabilities in contested environments. This paper presents a comprehensive framework for China UAV operations under GNSS-denied conditions, addressing two critical phases: autonomous navigation to a target location using only angular measurements from ground reference anchors, and subsequent multi-user downlink communication with limited onboard radio frequency resources. We propose an angle-only guidance law that exploits geometric constraints among three non-collinear anchors to converge the UAV to the desired position. Once deployed, a joint user scheduling and beamforming optimization problem is formulated to maximize the number of served users under total power and quality-of-service constraints. An alternating optimization algorithm integrating successive convex approximation and semidefinite relaxation is developed. Simulation results demonstrate the effectiveness of the navigation strategy with decreasing angle errors, and the resource allocation scheme significantly enhances service capability across various power budgets and rate thresholds, reinforcing the robustness of China UAV in GNSS-denied scenarios.
1. Introduction
The rapid development of China UAV (unmanned aerial vehicle) platforms has revolutionized low-altitude wireless networks, enabling on-demand coverage for emergency response, temporary hotspots, and infrastructure-deficient regions. Nevertheless, the reliance on GNSS for positioning and navigation remains a critical bottleneck. In many operational environments, GNSS signals are susceptible to obstruction, multipath effects, and intentional jamming, leading to degraded state estimation and control instability. For China UAV missions, especially those involving precise deployment for communication tasks, robust navigation without satellite signals is essential.
Existing navigation approaches either depend on absolute positioning (e.g., GNSS) or relative measurements (e.g., bearing, range). The former fails under denial, while the latter often requires extra sensors or favorable geometry. In this work, we propose a pure angle-based control law that uses only the bearing angles between the China UAV and three known ground anchors. This method circumvents the need for ranging or inertial sensors, making it suitable for lightweight platforms. Furthermore, after the China UAV reaches its target position, it must serve multiple ground users with limited transmit power and radio frequency chains. We formulate a joint user scheduling and beamforming problem to maximize the number of users that can be simultaneously served while meeting their rate requirements. The non-convex mixed-integer nature of the problem is tackled via an alternating optimization (AO) framework, leveraging successive convex approximation (SCA) and semidefinite relaxation (SDR).
2. System Model
We consider a single China UAV equipped with Nt antennas acting as an aerial base station. There are M single-antenna ground users (GUs). The system operates in two sequential phases: navigation and communication. During navigation, the China UAV uses three non-collinear ground anchors (selected from the GU set) for angle-only guidance. The anchors’ positions are known and denoted as g1, g2, g3 ∈ ℝ3 with zero altitude. The target deployment position is u* ∈ ℝ3.
2.1 Navigation Dynamics
The China UAV motion is modeled by a perturbed single-integrator:
$$ \dot{\mathbf{u}}(t) = \boldsymbol{\mu}(t) + \mathbf{n}(t) $$
where μ(t) ∈ ℝ3 is the control input, and n(t) is a bounded disturbance vector (wind, actuator noise, etc.). The control objective is to drive the UAV from its initial position u0 to u* using only angular measurements w.r.t. the anchors. Define the unit line-of-sight vector between any two nodes i and j as:
$$ \mathbf{r}_{ij}(t) = \frac{\mathbf{g}_j – \mathbf{u}(t)}{\|\mathbf{g}_j – \mathbf{u}(t)\|}, \quad \text{for } i=0, j\in\{1,2,3\} $$
and similarly for anchor-to-anchor vectors.
2.2 Downlink Communication Model
After the China UAV arrives at u*, the communication phase begins. The channel between the UAV and ground user k follows a Rician fading model:
$$ \mathbf{h}_k = \sqrt{\beta_0 d_k^{-2}} \left( \sqrt{\frac{\kappa_k}{\kappa_k+1}} \mathbf{h}_k^{\text{LoS}} + \sqrt{\frac{1}{\kappa_k+1}} \mathbf{h}_k^{\text{NLoS}} \right) $$
where dk is the distance, β0 the reference path loss, κk the Rician factor, and hkLoS = a(θk) is the array steering vector for a uniform linear array with element spacing ds = λ/2. The angle-of-departure θk satisfies cos θk = zu / dk. The NLoS component is CSCG with zero mean and unit variance. We assume MMSE channel estimation yielding estimated channel ĥk and estimation error ωk ∼ CN(0, σω2). The achievable rate at GU k is:
$$ R_k = \log_2\left(1 + \frac{|\hat{\mathbf{h}}_k^H \mathbf{w}_k|^2}{\sum_{i \neq k} |\hat{\mathbf{h}}_k^H \mathbf{w}_i|^2 + \sigma^2}\right) $$
where wk is the beamforming vector for user k and σ2 is receiver noise power.
3. Navigation Strategy under GNSS Denial
The key idea: three independent angular constraints uniquely determine the UAV’s 3D position. As shown in literature, the angles between the UAV and two anchors (e.g., ρ012 = angle between g0g1 and g0g2) define conical surfaces; their intersection yields a curve. Adding the dihedral angle between the plane formed by the UAV and two anchors and the horizontal plane gives a unique point.
Let ρ012(t), ρ021(t), and ρp(t) be the measured angles at time t, and ρ012*, ρ021*, ρp* be the target values computed from u*. We propose the control law:
$$ \boldsymbol{\mu}(t) = (\rho_{012}(t)-\rho_{012}^*)\mathbf{r}_{02}(t) + (\rho_{021}(t)-\rho_{021}^*)\mathbf{r}_{01}(t) – (\rho_p(t)-\rho_p^*)(\mathbf{r}_{01}(t) \times \mathbf{r}_{02}(t)) $$
The first two terms correct the in-plane angular errors along the directions of r02 and r01, while the third term adjusts the dihedral error along the normal vector. The orthogonality of these components ensures convergence. The dihedral angle is computed via plane normals:
$$ \mathbf{c}_u(t) = \frac{\mathbf{r}_{20}(t) \times \mathbf{r}_{21}}{\|\mathbf{r}_{20}(t) \times \mathbf{r}_{21}\|}, \quad \mathbf{c}_g = \frac{\mathbf{r}_{23} \times \mathbf{r}_{21}}{\|\mathbf{r}_{23} \times \mathbf{r}_{21}\|} $$
$$ \rho_p(t) = \begin{cases}
\arccos(\mathbf{c}_u^T(t)\mathbf{c}_g), & \mathbf{r}_{10}^T(t)\mathbf{c}_g > 0 \\
-\arccos(\mathbf{c}_u^T(t)\mathbf{c}_g), & \text{otherwise}
\end{cases} $$
The control law guarantees that all three angle errors asymptotically approach zero, i.e., the China UAV converges to the target location without GNSS.
4. Joint User Scheduling and Beamforming
After navigation, the China UAV must serve M ground users. Due to limited RF chains and power, we can only serve at most K users simultaneously. The goal is to maximize the number of served users while satisfying QoS and total power constraints. Let binary variable ηm = 1 if user m is scheduled, 0 otherwise. The optimization problem is:
$$ \max_{\boldsymbol{\eta}, \{\mathbf{w}_m\}} \|\boldsymbol{\eta}\|_0 $$
subject to:
$$ \sum_{m} \mathbf{w}_m^H \mathbf{w}_m \leq P_{\max} $$
$$ R_m \geq \eta_m R_m^{\text{th}}, \quad \forall m $$
$$ \eta_m \in \{0,1\} $$
This MINLP is solved by alternating optimization. For fixed beamforming, the user scheduling subproblem is relaxed using SCA with a penalty term to enforce binary nature. For fixed scheduling, the beamforming subproblem becomes a feasibility problem solved via SDR after dropping the rank-1 constraint. Gaussian randomization recovers rank-1 solutions if needed.
Table 1: Simulation Parameters for China UAV System
| Parameter | Value |
|---|---|
| Number of UAV antennas (Nt) | 32 |
| Number of ground users (M) | 22 |
| Reference path loss β0 | -60 dB |
| Noise power σ2 | -110 dBm |
| Array spacing ds | λ/2 |
| UAV initial position | [10,10,80] m |
| UAV target position | [100,80,120] m |
| Anchor 1 | [20,10,0] m |
| Anchor 2 | [140,30,0] m |
| Anchor 3 | [70,130,0] m |
5. Simulation Results
We evaluate the proposed framework through numerical simulations. The China UAV navigation performance is first examined. Figure 1 shows the 3D trajectory from the starting point to the target using only angle measurements. The path exhibits smooth convergence in both horizontal and vertical dimensions. Figure 2 plots the evolution of the three angle errors (ρ012, ρ021, ρp) over time. All errors decrease monotonically and approach zero, confirming the effectiveness of the angle-only control law for China UAV under GNSS denial.
Table 2: Angle Error Convergence Statistics for China UAV Navigation
| Time (s) | ρ012 error (deg) | ρ021 error (deg) | ρp error (deg) |
|---|---|---|---|
| 0 | 12.5 | 14.2 | 8.7 |
| 5 | 5.1 | 6.3 | 3.4 |
| 10 | 2.0 | 2.8 | 1.5 |
| 15 | 0.8 | 1.1 | 0.6 |
| 20 | 0.3 | 0.4 | 0.2 |
After deployment, we assess the joint scheduling and beamforming algorithm. Figure 5 shows a contour plot of the number of served users versus total transmit power and rate threshold. Higher power budgets and lower rate thresholds enable serving more users, demonstrating the flexibility of the China UAV communication system. Table 3 summarizes the served user count for different power and rate values.
Table 3: Number of Served Users vs. Power and Rate Threshold
| Pmax (dBW) | Rate threshold Rth (bps/Hz) | |||
|---|---|---|---|---|
| 0.5 | 1.0 | 1.5 | 2.0 | |
| -5 | 12 | 8 | 5 | 3 |
| -3 | 15 | 11 | 8 | 5 |
| -1 | 18 | 14 | 10 | 7 |
| 1 | 20 | 17 | 13 | 9 |
Additionally, we investigate the impact of channel estimation error under probabilistic LoS channels (Rician factor κk). For a fixed power of -1 dBW and rate threshold 0.8 bps/Hz, the number of served users degrades as the estimation error variance σω2 increases. However, when κk is large (strong LoS), the degradation is mild, highlighting the robustness of China UAV communication in typical low-altitude LoS environments.
6. Conclusion
This paper presented a complete solution for China UAV operations in GNSS-denied low-altitude wireless networks. The angle-only navigation law successfully guides the UAV to a target location using three ground anchors without satellite signals. The subsequent joint user scheduling and beamforming algorithm maximizes the number of served users under resource constraints, solved efficiently via alternating optimization. Extensive simulations validated the convergence of angle errors and the effectiveness of resource allocation across diverse power and rate conditions. The proposed framework enhances the resilience and service capability of China UAV systems in contested environments, paving the way for robust low-altitude communications.