Hongjuan Li,Hui Kang,Geng Sun,Jiahui Li,Jiacheng Wang,Xue Wang,Dusit Niyato,Victor C. M. Leung
Abstract:Unmanned aerial vehicles (UAVs) have gained considerable attention as a platform for establishing aerial wireless networks and communications. However, the line-of-sight dominance in air-to-ground communications often leads to significant interference with terrestrial networks, reducing communication efficiency among terrestrial terminals. This paper explores a novel uplink interference mitigation approach based on the collaborative beamforming (CB) method in multi-UAV network systems. Specifically, the UAV swarm forms a UAV-enabled virtual antenna array (VAA) to achieve the transmissions of gathered data to multiple base stations (BSs) for data backup and distributed processing. However, there is a trade-off between the effectiveness of CB-based interference mitigation and the energy conservation of UAVs. Thus, by jointly optimizing the excitation current weights and hover position of UAVs as well as the sequence of data transmission to various BSs, we formulate an uplink interference mitigation multi-objective optimization problem (MOOP) to decrease interference affection, enhance transmission efficiency, and improve energy efficiency, simultaneously. In response to the computational demands of the formulated problem, we introduce an evolutionary computation method, namely chaotic non-dominated sorting genetic algorithm II (CNSGA-II) with multiple improved operators. The proposed CNSGA-II efficiently addresses the formulated MOOP, outperforming several other comparative algorithms, as evidenced by the outcomes of the simulations. Moreover, the proposed CB-based uplink interference mitigation approach can significantly reduce the interference caused by UAVs to non-receiving BSs.
What problem does this paper attempt to address?
The main problem that this paper attempts to solve is the significant interference problem caused by unmanned aerial vehicles (UAVs) to the ground network during data collection and transmission. Specifically, although the dominance of line - of - sight (LoS) communication between UAVs and ground stations improves communication efficiency, it also brings serious interference problems, affecting the communication efficiency between ground terminals. To meet this challenge, the paper proposes an uplink interference suppression method based on cooperative beamforming (CB). Multiple UAVs work together to form a virtual antenna array (VAA) to reduce interference to non - receiving base stations while improving data transmission efficiency and energy utilization efficiency.
### Specific description of the problem
1. **Interference caused by line - of - sight communication**: Since the line - of - sight communication path between UAVs and ground stations is short and the signal strength is high, this will cause serious interference to the ground network, especially when multiple UAVs transmit data to multiple base stations simultaneously, this interference is more obvious.
2. **Trade - off between energy consumption and transmission efficiency**: UAVs need to frequently adjust their positions to optimize the beamforming effect when performing tasks, which will increase energy consumption and thus shorten the service life of UAVs. Therefore, how to ensure effective energy utilization and high - efficiency data transmission while reducing interference is a key issue.
3. **Multi - objective optimization problem**: The paper formalizes the above problems into a multi - objective optimization problem (MOOP), aiming to simultaneously reduce interference, improve transmission efficiency and reduce energy consumption. Specifically, these goals are achieved by jointly optimizing the excitation current weights, hovering positions and data transmission sequences of UAVs.
### Overview of the solution
The solution proposed in the paper is an uplink interference suppression method based on cooperative beamforming (CB), which specifically includes the following aspects:
- **Construction of virtual antenna array (VAA)**: Multiple UAVs work together to form a virtual antenna array to enhance the directivity and gain of the signal, thereby effectively reducing interference to non - receiving base stations.
- **Modeling of multi - objective optimization problem (MOOP)**: Combine the three goals of interference suppression, transmission efficiency and energy consumption to construct a complex multi - objective optimization problem and prove that its computational complexity is NP - hard.
- **Chaos non - dominated sorting genetic algorithm II (CNSGA - II)**: An improved evolutionary computing method, chaos non - dominated sorting genetic algorithm II (CNSGA - II), is introduced to solve the above multi - objective optimization problem. This algorithm combines chaos theory, improves the quality of the initial population, and achieves a good balance between exploration and exploitation.
### Results and contributions
Through extensive simulation analysis, the paper verifies the effectiveness of the proposed method, indicating that this method can significantly reduce interference to the ground network and is superior in performance to several other comparison algorithms. In addition, this method can also flexibly adjust the positions and transmission parameters of UAVs in different scenarios to meet diverse application requirements.
In summary, this paper aims to solve the interference problem caused by UAVs to the ground network during data collection and transmission, and through the introduction of cooperative beamforming technology, proposes an effective solution to achieve the multiple goals of interference suppression, transmission efficiency improvement and energy consumption reduction.