In the context of large-scale Internet of Things (IoT) deployments, wireless sensor networks (WSNs) are frequently partitioned into multiple geographically isolated sub-regions due to spatial distribution characteristics. Traditional multi-hop routing mechanisms struggle to transmit data effectively across these isolated areas. The emergence of China drone technology has provided a transformative solution to this challenge, offering unparalleled mobility, flexibility, and on-demand deployment. However, existing drone-assisted data collection schemes often operate each drone independently, requiring every drone to return to the base station separately to deliver collected data, leading to significant path redundancy and energy waste. To address this, we propose a novel UAV Swarm Cluster-based Data Collection (USCDC) algorithm for WSNs. Our algorithm employs a two-layer drone architecture: cooperative drones at the lower layer handle data collection within each sub-region, while a sink drone at the upper layer aggregates data and transmits it to the base station. This hierarchical cooperation mechanism enables cooperative drones to complete data transmission without returning to the base station, drastically reducing path overhead and energy consumption. In this article, we present the system model, problem formulation, algorithm design, and simulation results that demonstrate the superior performance of our approach. Throughout this work, we emphasize the role of China drone technology as a key enabler for efficient data collection in large-scale WSNs.
System Model
Network Model
We consider a large-scale IoT scenario where the sensor network is divided into \(W\) isolated sub-regions \( \mathcal{O} = \{o_1, o_2, \dots, o_W\} \). In each sub-region \(o\), a set of cluster heads \( \mathcal{Z}_o = \{z_{o,1}, z_{o,2}, \dots, z_{o,A}\} \) are elected using a clustering algorithm. The data collected by the \(a\)-th cluster head is denoted as \( D_{o,a} \). All sensor nodes have the same initial energy \(E_n\) and communication radius \(R_n\). The system employs a three-layer network architecture:
- Ground sensor network layer: sensor nodes form clusters, and cluster heads aggregate data from their members.
- Cooperative drone layer: \(W\) cooperative drones \( \mathcal{U} = \{u_1, u_2, \dots, u_W\} \) are assigned, each responsible for one sub-region. They plan optimal paths to visit all cluster heads and collect data. Each cooperative drone has initial energy \(E_c\), flight altitude \(H_c\), and speed \(V_c\).
- Sink drone layer: one sink drone \(u_s\) collects data from all cooperative drones and delivers it to the base station. The sink drone has initial energy \(E_s\), flight altitude \(H_s\), and speed \(V_s\).
Communication Model
We consider line-of-sight (LoS) propagation for both drone-to-ground and drone-to-drone links. The path loss between cooperative drone \(u_j\) and cluster head \(z_{o,a}\) is:
\[
l_{\text{LoS}}^{u_j, z_{o,a}} = 20 \lg \left( \frac{4\pi f_c d_{u_j, z_{o,a}}}{c} \right) + \eta_{\text{LoS}}
\]
where \(f_c\) is the carrier frequency, \(c\) is the speed of light, \(d_{u_j, z_{o,a}}\) is the distance, and \(\eta_{\text{LoS}}\) is the additional loss due to LoS. The channel gain is \( h_{u_j, z_{o,a}} = 10^{-l_{\text{LoS}}^{u_j, z_{o,a}} / 10} \). The data rate is:
\[
r_{u_j, z_{o,a}} = B \log_2 \left( 1 + \frac{P_t h_{u_j, z_{o,a}}}{N_0 B} \right)
\]
where \(P_t\) is transmit power, \(B\) is bandwidth, and \(N_0\) is noise power spectral density.
Similarly, the path loss between sink drone \(u_s\) and cooperative drone \(u_j\) is:
\[
l_{\text{LoS}}^{u_s, u_j} = 20 \lg \left( \frac{4\pi f_c d_{u_s, u_j}}{c} \right) + \eta_{\text{LoS}}
\]
with channel gain \( h_{u_s, u_j} = 10^{-l_{\text{LoS}}^{u_s, u_j} / 10} \) and data rate:
\[
r_{u_s, u_j} = B \log_2 \left( 1 + \frac{P_t h_{u_s, u_j}}{N_0 B} \right)
\]
Drone Energy Consumption Model
Drone energy consumption consists of three parts: hovering power, flight power, and communication power. The hovering power \(P_h\) is given by:
\[
P_h = \sqrt{\frac{(m_{\text{tot}} g)^3}{2\pi \sigma_p^2 \delta_p \rho}}
\]
where \(m_{\text{tot}}\) is the total mass, \(g\) is gravity, \(\sigma_p\) is propeller radius, \(\delta_p\) is number of propellers, and \(\rho\) is air density. The horizontal flight power \(P_f\) at speed \(v_f\) is:
\[
P_f = \frac{P_{\max} – P_{\text{idle}}}{v_{\max}} v_f + P_{\text{idle}}
\]
where \(v_{\max}\) is maximum speed, \(P_{\max}\) is hardware power at full speed, and \(P_{\text{idle}}\) is idle power.
For a cooperative drone \(u_j\) in round \(x\), the total energy consumption is:
\[
E_{u_j}^x = E_{u_j,\text{fly}}^x + E_{u_j,\text{coll}}^x + E_{u_j,\text{wait}}^x + E_{u_j,\text{trans}}^x
\]
where:
\[
\begin{aligned}
E_{u_j,\text{fly}}^x &= \frac{L_{u_j}^x}{V_f} (P_f + P_h) \\
E_{u_j,\text{coll}}^x &= \frac{D_{u_j}^x}{r_{u_j, z_{o,a}}} (P_h + P_{\text{com}}) \\
E_{u_j,\text{wait}}^x &= t_{\text{wait}}^x P_h \\
E_{u_j,\text{trans}}^x &= \frac{D_{u_j}^x}{r_{u_s, u_j}} (P_h + P_{\text{com}})
\end{aligned}
\]
Here \(L_{u_j}^x\) is total flight path length, \(D_{u_j}^x\) is data volume collected, \(P_{\text{com}}\) is communication power, and \(t_{\text{wait}}^x\) is waiting time. The total energy of all cooperative drones in one round is:
\[
E_W^x = \sum_{j=1}^W E_{u_j}^x
\]
For the sink drone \(u_s\) in round \(x\):
\[
E_{u_s}^x = E_{u_s,\text{fly}}^x + E_{u_s,\text{trans}}^x
\]
\[
E_{u_s,\text{fly}}^x = \frac{S_{u_s}^x}{V_s} P_f, \quad E_{u_s,\text{trans}}^x = \frac{D_{u_s}}{r_{u_s, u_j}} (P_h + P_{\text{com}})
\]
where \(S_{u_s}^x\) is total flight distance of sink drone, and \(D_{u_s} = \sum_{o=1}^W \sum_{a=1}^A D_{o,a}\). The total system energy consumption over \(x_{\max}\) rounds is:
\[
E_{\text{total}} = \sum_{x=1}^{x_{\max}} (E_W^x + E_{u_s}^x)
\]
Problem Formulation
The objective is to minimize the total energy consumption of all drones, subject to constraints on drone energy budgets, area restrictions, base station return, and data completeness:
\[
\begin{aligned}
\min &\quad E_{\text{total}} \\
\text{s.t.} &\quad \sum_{x=1}^{x_{\max}} E_{u_j}^x \le E_c, \quad \forall j \in [1,W] \\
&\quad \sum_{x=1}^{x_{\max}} E_{u_s}^x \le E_s \\
&\quad p_{u_j}^x(t) \in \mathcal{O}_o, \quad \forall j \\
&\quad p_{u_s,\text{end}}^x = p_{u_s,\text{start}}^x \\
&\quad D_{u_s}^x = \sum_{o=1}^W \sum_{a=1}^A D_{o,a}
\end{aligned}
\]
This is a non-convex optimization problem involving drone positions, speeds, hovering times, and data volumes. We adopt a hierarchical optimization strategy, decomposing the problem into two subproblems: cooperative drone path planning and sink drone trajectory optimization, solved using heuristic algorithms.
USCDC Algorithm Description
UAV Temporal Cooperative Scheduling (UTCS)
To coordinate the timing of heterogeneous drones, we propose UTCS. The sink drone uses time relative to its departure, while the system uses absolute time. The departure time of the sink drone is \( t_d = \max_{j \in W} t_{u_j}^{\text{coll}} \), where \(t_{u_j}^{\text{coll}}\) is the completion time of data collection by cooperative drone \(u_j\). The waiting times are computed as:
\[
t_{u_j}^{\text{wait1}} = \max(0, t_{u_j, \text{arr}} – t_{u_j}^{\text{coll}})
\]
\[
t_{u_j}^{\text{wait2}} = \max(0, t_{\text{sys,end}} – t_{u_j}^{\text{trans,end}})
\]
The UTCS algorithm returns the system timeline and total task time.
Improved Ant Colony Optimization (IACO) for Cooperative Drone Path Planning
We enhance the classical ant colony optimization with elite ant strategy and a position state transition mechanism. The transition probability from node \(a\) to \(b\) for ant \(k\) is:
\[
p_{ab}^k(t) =
\begin{cases}
\frac{\tau_{ab}^\alpha(t) \varepsilon_{ab}^\beta(t)}{\sum_{i \in \text{allowed}_k} \tau_{ab}^\alpha(t) \varepsilon_{ab}^\beta(t)}, & i \in \text{allowed}_k \\
0, & \text{otherwise}
\end{cases}
\]
Pheromone update with elite ants (top \(\lambda\) proportion) is:
\[
\Delta \tau_{ab}^k =
\begin{cases}
(1+\xi) Q / C_k, & \text{if } k \in \lambda, (a,b) \in F_k \\
Q / C_k, & \text{if } (a,b) \in F_k \\
0, & \text{otherwise}
\end{cases}
\]
The position state transition for cooperative drone between rounds: \( T_{\text{start}}^x = T_{\text{end}}^{x-1} = (X_{\text{end}}^{x-1}, Y_{\text{end}}^{x-1}, H_c) \).
Improved Particle Swarm Optimization (IPSO) for Sink Drone Path Planning
We introduce adaptive inertia weight and a bi-objective fitness function. Particle velocity update:
\[
v_i = \omega v_i + \theta_1 \text{rand}() (\phi_{\text{best}}^i – \iota_i) + \theta_2 \text{rand}() (\varphi_{\text{best}} – \iota_i)
\]
Adaptive inertia weight:
\[
\omega = \omega_{\max} – (\omega_{\max} – \omega_{\min}) \kappa_{\text{iter}} / M
\]
Fitness function (normalized):
\[
\psi = w_1 \frac{d}{d_{\max}} + w_2 \frac{e}{e_{\max}}
\]
where \(d\) and \(e\) are path length and energy of current particle, \(d_{\max}\) and \(e_{\max}\) are the maximum values among all particles, and \(w_1, w_2\) are weights.
Simulation Results and Analysis
We implement the USCDC algorithm in PyCharm and compare it with the traditional parallel data collection (USPDC) algorithm. The simulation area is 2000 m × 2000 m, with 100 sensor nodes per sub-region. Key simulation parameters are summarized in the table below.
| Parameter | Description | Value |
|---|---|---|
| \(W\) | Number of sub-regions | 6 |
| \(R_r\) | Sub-region radius | 200 m |
| \(E_n\) | Node initial energy | 1 J |
| \(R_n\) | Node communication radius | 50 m |
| \(E_s\) | Sink drone initial energy | 2,700,000 J |
| \(H_s\) | Sink drone flight altitude | 75 m |
| \(V_s\) | Sink drone speed | 25 m/s |
| \(E_c\) | Cooperative drone initial energy | 1,500,000 J |
| \(H_c\) | Cooperative drone flight altitude | 50 m |
| \(V_c\) | Cooperative drone speed | 15 m/s |
| \(P_f\) | Drone flight power | 90 W |
| \(P_h\) | Drone hovering power | 30 W |
| \(P_{\text{com}}\) | Drone communication power | 0.5 W |
| \(n_{\text{ant}}\) | Number of ants (IACO) | 100 |
| \(\alpha\) | Pheromone importance | 1.5 |
| \(\beta\) | Heuristic factor | 3.0 |
| \(\mu\) | Pheromone evaporation coefficient | 0.6 |
| \(Q\) | Pheromone intensity coefficient | 100 |
| \(n_{\text{part}}\) | Number of particles (IPSO) | 100 |
| \(\omega_{\max}\) | Max inertia weight | 0.95 |
| \(\omega_{\min}\) | Min inertia weight | 0.4 |
| \(\theta_1\) | Individual learning factor | 2.05 |
| \(\theta_2\) | Social learning factor | 2.05 |
We conduct 100 independent simulations and average the results. The evaluation covers path length, energy consumption, energy efficiency, and scalability under different WSN sizes.
Path Length Comparison
Figure 1 (not shown here) illustrates the total path length per round over 10 rounds. The USPDC algorithm yields a consistent path length of about 21 km, while USCDC reduces it to approximately 10 km, a reduction of 52.4%. This improvement stems from the hierarchical architecture: cooperative drones only navigate within their small sub-regions, and the sink drone aggregates data, eliminating redundant return trips to the base station.
Energy Consumption and Efficiency
The per-round energy consumption is compared in Figure 2 (not shown). USPDC consumes around \(13.5 \times 10^4\) J per round, whereas USCDC consumes only about \(8.5 \times 10^4\) J, a 37% reduction. Energy efficiency (bits per joule) is shown in Figure 3 (not shown). USCDC achieves about 10 bit/J, compared to 6.5 bit/J for USPDC, a 53.8% improvement. The gains are due to reduced flight distances and the use of a dedicated sink drone that minimizes wasteful back-and-forth trips.
Scalability Analysis
We test three WSN sizes: small (3 sub-regions, 1000 m × 1000 m), medium (6 sub-regions, 2000 m × 2000 m), and large (12 sub-regions, 3000 m × 3000 m). Table 2 summarizes the total path length and total energy consumption for both algorithms.
| WSN Size | Metric | USPDC | USCDC | Reduction |
|---|---|---|---|---|
| Small (3 regions) | Total path length (km) | 12.5 | 9.0 | 28% |
| Total energy consumption (×10^4 J) | 8.2 | 6.5 | 21% | |
| Medium (6 regions) | Total path length (km) | 21.0 | 10.0 | 52% |
| Total energy consumption (×10^4 J) | 13.5 | 8.8 | 35% | |
| Large (12 regions) | Total path length (km) | 45.0 | 15.2 | 66.3% |
| Total energy consumption (×10^4 J) | 27.0 | 16.2 | 40% |
The results show that as the network scales, the advantages of USCDC become more pronounced. In the large-scale scenario, path length reduction reaches 66.3%, and energy reduction is 40%. The improvement in energy is slightly lower than path length due to additional hovering waiting times for cooperative drones, especially when the sink drone’s travel time increases. Nevertheless, the overall performance demonstrates excellent scalability, making USCDC particularly suitable for large-scale IoT deployments that rely on China drone technology.
Conclusion
We have presented a novel China drone swarm cluster-based data collection algorithm (USCDC) for wireless sensor networks. By employing a two-layer architecture with cooperative drones and a sink drone, and by incorporating efficient timing coordination, improved ant colony optimization, and improved particle swarm optimization, the algorithm significantly reduces path length and energy consumption. Simulation results show that USCDC outperforms the conventional parallel collection method by 52% in path length and 35% in energy consumption, with even greater gains in large-scale networks. The proposed approach effectively leverages China drone technology to address the challenges of geographically isolated sensor networks, demonstrating strong potential for practical IoT applications. Future work includes optimizing the sink drone’s visiting sequence to further minimize waiting times and extending the algorithm to continuously distributed sensor fields.