Time-Frequency Resource Optimization Method for UAV-Assisted IAB Networks in Complex Building Emergency Scenarios
-
摘要: 无人机辅助的接入回传一体化(IAB)网络凭借其全空间灵活部署和集成接入回传特性，有望克服复杂建筑应急场景网络覆盖范围受限与回传链路容量受限等瓶颈问题。然而，应急场景频谱资源稀缺性和数据流高并发性造成网络干扰严重，且复杂信道环境与无人机节点移动性导致干扰关系呈现时变非平稳特性，造成资源分配决策与网络拓扑失配。为此，该文提出了一种基于非平稳多臂老虎机的时频资源联合优化方法，首先建立分段复合信道模型以表征复杂建筑应急场景电磁传播特性，然后利用图论将时变非平稳干扰关系建模为非平稳定向冲突图，进而将时频资源联合优化问题转化为非平稳定向二重图着色问题，最后采用非平稳多玩家多臂老虎机模型来求解，并在此基础上引入指数遗忘策略以适应时变非平稳干扰关系。仿真结果表明，所提算法在保证决策时效性的同时显著降低了接入回传一体化网络的平均冲突率，能够适应复杂建筑应急场景中的时变非平稳环境。Abstract:
Objective Disaster accidents in complex, high-rise, and ultra-high-rise buildings in China occur frequently and are exhibiting a continuous upward trend, necessitating the urgent development of highly adaptable emergency rescue command systems. The intricate structural characteristics of complex buildings inevitably cause severe wireless signal attenuation, leading to communication blind spots at indoor-outdoor transitions and floor boundaries. To address these communication bottlenecks, the construction of "dual-use" emergency communication networks—designed for both routine daily operations and disaster response—has become a vital strategy for enhancing urban safety resilience. Unmanned Aerial Vehicles (UAVs), relying on their lightweight design, high payload capacity, and high mobility, provide a highly reliable carrier for deploying these networks. However, UAV-assisted networks in complex building emergency scenarios face severe limitations in network coverage and backhaul link capacity. Integrated Access and Backhaul (IAB) technology offers a promising architectural solution by enabling wireless access and backhaul over shared spectrum resources. Despite its potential, the deployment of UAV-assisted IAB networks in emergency scenarios remains challenging. Localized traffic surges caused by concurrent user access, coupled with partial infrastructure damage, result in severe network interference. Furthermore, the dense deployment of nodes, the high mobility of UAVs, and the intricate channel environment cause these interference relationships to exhibit pronounced time-varying and non-stationary characteristics. Existing resource allocation methods primarily operate under the assumption of a stationary environment, which makes it difficult to match resource allocation decisions with dynamic network topologies. Therefore, this study aims to propose a joint time-frequency resource optimization method to reduce network conflicts under time-varying and non-stationary interference relationships. Methods To characterize the electromagnetic propagation characteristics in disaster-stricken complex buildings, a segmented composite channel model is established. This model accounts for line-of-sight transmissions between indoor or outdoor IAB nodes, as well as penetration transmissions between indoor and outdoor environments. To characterize sudden channel variations caused by complex environmental factors, a Rician small-scale fading model is introduced, wherein the Rician factor is defined to follow a piecewise constant probability distribution. Subsequently, graph theory is utilized to map the dynamically evolving network topology into a non-stationary directed conflict graph. In this framework, communication links are designated as vertices, while time-varying interference relationships are mapped as edges. Two types of conflicts are defined in the model: link-scheduling constraints and resource mutual-exclusion constraints. Consequently, the joint time-frequency resource allocation problem is transformed into a non-stationary directed dual-graph coloring problem. To facilitate optimization, a minimum-conflict model based on soft constraints is constructed, which incorporates link-scheduling constraints and resource mutual-exclusion constraints as soft penalty terms within the objective function. To solve this problem, an Exponential Forgetting Upper Confidence Bound (EF-UCB) algorithm is proposed. Operating within a multi-player Multi-Armed Bandit (MAB) framework, the algorithm treats communication links as independent players and the combined time-frequency resource blocks as selectable actions. Crucially, an exponential forgetting mechanism is embedded into the historical information updating process. By applying exponentially decaying weights to past observations through a forgetting factor, the algorithm reduces the influence of outdated historical information, thereby enabling adaptive online distributed decision-making in non-stationary interference environments. Results and Discussions The proposed EF-UCB algorithm is systematically evaluated through Monte Carlo simulations in a three-dimensional complex building emergency scenario and benchmarked against Greedy, Color Diffusion Algorithm (CDA), Q-Learning (QL), and ε-greedy-based MAB algorithms. Simulation results demonstrate that as the number of time slots increases, the link-scheduling constraint conflict rate, the resource mutual-exclusion constraint conflict rate, and the total network conflict rate all exhibit a continuous downward trend ( Fig. 3 ). A similar reduction in the resource mutual-exclusion constraint conflict rate and total network conflict rate is observed when the number of subchannels increases, whereas the link-scheduling constraint conflict rate changes relatively slightly (Fig. 4 ). Across the evaluated resource availability configurations, the proposed EF-UCB algorithm maintains the lowest average network conflict rate among the compared algorithms (Fig. 3 ,Fig. 4 ). In terms of computational efficiency and decision latency, the EF-UCB algorithm completes its online decision-making process at the millisecond level and requires less average decision time than the Greedy and CDA approaches (Fig. 3 ,Fig. 4 ). As the network scale expands with an increasing number of UAVs, the average decision time of the EF-UCB algorithm remains relatively low and increases only slightly, indicating good scalability in dense emergency deployments (Fig. 6 ). Furthermore, compared with the other learning-based methods, the network conflict rate of EF-UCB decreases more rapidly with the number of operating system frames (Fig. 5 ). When the number of environmental mutation points increases, the EF-UCB algorithm exhibits the smallest increase in the total network conflict rate among the compared algorithms (Fig. 7 ). The exponential forgetting mechanism reduces the influence of outdated historical information, enabling EF-UCB to maintain a relatively flat growth trend in the total conflict rate and demonstrating stronger adaptability to time-varying and non-stationary interference conditions (Fig. 7 ).Conclusions This study addresses the joint time-frequency resource optimization problem for UAV-assisted IAB networks operating under time-varying and non-stationary interference in complex building emergency scenarios. By integrating a segmented composite channel model with a non-stationary directed dual-graph coloring framework, the physical propagation characteristics and dynamic interference relationships of complex building emergency environments are characterized. The proposed EF-UCB algorithm, combined with an exponential forgetting strategy, balances exploration with adaptation to non-stationary environments. Simulation results show that the proposed method effectively reduces the average total network conflict rate while maintaining relatively low computational complexity and decision latency. Overall, this approach supports timely online resource allocation and improves the performance of UAV-assisted emergency communication networks under non-stationary interference conditions. -
Key words:
- Complex building /
- Emergency communication /
- IAB /
- Non-stationary environment /
- Resource scheduling
-
表 1 基于指数遗忘置信上界区间的时频资源联合调度算法
1：输入：时隙数M，子信道数K 2：输出：时频资源分配矩阵$ \bf{SlotAllocation} $、$ \bf{ChannelAllocation} $ 3：初始化：每个玩家的累积选择次数$ \bf{count}{\bf{s}}^{t}({p}_{i},{a}_{i}) $、累积即时奖励$ \bf{win}{\bf{s}}^{t}({p}_{i},{a}_{i}) $、遗忘因子λ 4：$ fort=1\colon T $ 5： 根据t系统帧内IAB节点位置和环境状态更新非平稳定向冲突图$ \bf{DIG}(t) $ 6： $ forl\in \boldsymbol{L}(t) $ 7： 基于UCB决策方法选择动作$ {\boldsymbol{c}}^{t}({p}_{i}) $ 8： $ endfor $ 9： 更新时频资源分配矩阵$ \bf{SlotAllocation} $、$ \bf{ChannelAllocation} $ 10： 计算即时奖励： 11： $ forl\in \boldsymbol{L}(t) $ 12： 根据定向干扰图$ \bf{DIG}(t) $计算决策后链路冲突状态$ \boldsymbol{C}_{l}^{t} $ 13： 结合公式(8)计算即时奖励 14： $ endfor $ 15： 更新历史统计信息： 16： 联合公式(9)、(10)、(11)以指数遗忘的方式更新历史统计信息 17： 输出当前系统帧时频资源分配矩阵、网络链路冲突状态$ {\boldsymbol{C}}^{t} $ 18：$ endfor $ 
下载: 导出CSV
表 2 仿真实验参数表
实验参数 设置值 系统帧T (个) 5000 时隙数M (个) 1～5 子信道数K (个) 2,4,6,8,10 信噪比阈值η (dB) 15 信干噪比阈值δ (dB) 10 莱斯因子κ (dB) 0～15 遗忘因子$ \lambda $ 0.98 实验频段(MHz) 3450～3550 室内无人机数目$ N_{\text{uav}}^{\text{In}} $ (个) 7,10,13,17,20 室外无人机数目$ N_{\text{uav}}^{\text{Out}} $(个) 13,20,27,33,40 突变点数s (个) 2～10
下载: 导出CSV
-
[1] KIM T Y, PARK J, KANG J, et al. Multiagent deep reinforcement learning for joint movement and user association of UAV-BS emergency indoor user service[J]. IEEE Internet of Things Journal, 2026, 13(1): 1402–1422. doi: 10.1109/JIOT.2025.3629374. [2] 王威, 朱峻佚, 王国盛, 等. 城市地下空间“平急两用”规划建设思路与路径研究[J]. 中国工程科学, 2025, 27(4): 82–94. doi: 10.15302/J-SSCAE-2025.04.001.WANG Wei, ZHU Junyi, WANG Guosheng, et al. Urban underground space planning idea and path based on a “dual-use” concept[J]. Strategic Study of CAE, 2025, 27(4): 82–94. doi: 10.15302/J-SSCAE-2025.04.001. [3] 陈新颖, 盛敏, 李博, 等. 面向6G的无人机通信综述[J]. 电子与信息学报, 2022, 44(3): 781–789. doi: 10.11999/JEIT210789.CHEN Xinying, SHENG Min, LI Bo, et al. Survey on unmanned aerial vehicle communications for 6G[J]. Journal of Electronics & Information Technology, 2022, 44(3): 781–789. doi: 10.11999/JEIT210789. [4] 胡杨林, 张天魁, 李博, 等. 无人机使能的通信感知一体化组网与技术研究综述[J]. 电子与信息学报, 2025, 47(4): 859–875. doi: 10.11999/JEIT241116.HU Yanglin, ZHANG Tiankui, LI Bo, et al. A survey on UAV-enabled integrated sensing and communication networking and technologies[J]. Journal of Electronics & Information Technology, 2025, 47(4): 859–875. doi: 10.11999/JEIT241116. [5] ZHENG Xiaoya, SUN Geng, LI Jiahui, et al. UAV swarm-enabled collaborative post-disaster communications in low altitude economy via a two-stage optimization approach[J]. IEEE Transactions on Mobile Computing, 2025, 24(11): 11833–11851. doi: 10.1109/TMC.2025.3583510. [6] 金飞鸿, 张静, 谢亚琴. 用户需求差异化场景下信息年龄优先的多无人机部署及资源分配方法[J]. 电子与信息学报, 2026, 48(1): 253–263. doi: 10.11999/JEIT251062.JIN Feihong, ZHANG Jing, and XIE Yaqin. AoI-prioritized multi-UAV deployment and resource allocation method in scenarios with differentiated user requirements[J]. Journal of Electronics & Information Technology, 2026, 48(1): 253–263. doi: 10.11999/JEIT251062. [7] 初航, 董志浩, 曹杰, 等. 面向无人机辅助电力巡检的短包通信资源优化[J]. 电子与信息学报, 2026, 48(1): 78–85. doi: 10.11999/JEIT250852.CHU Hang, DONG Zhihao, CAO Jie, et al. Optimization of short packet communication resources for UAV assisted power inspection[J]. Journal of Electronics & Information Technology, 2026, 48(1): 78–85. doi: 10.11999/JEIT250852. [8] 闫莉, 岳涛, 方旭明. 铁路应急场景下无人机通信感知一体化无线网络资源智能分配算法[J]. 电子与信息学报, 2024, 46(9): 3510–3519. doi: 10.11999/JEIT240254.YAN Li, YUE Tao, and FANG Xuming. Intelligent wireless resource allocation algorithm for unmanned aerial vehicle integrated communication and sensing networks in railway emergency scenarios[J]. Journal of Electronics & Information Technology, 2024, 46(9): 3510–3519. doi: 10.11999/JEIT240254. [9] DENG Jikang, ZHOU Hui, and ALOUINI M S. Two-timescale optimization framework for IAB-enabled heterogeneous UAV networks[J]. IEEE Internet of Things Journal, 2026, 13(2): 3536–3551. doi: 10.1109/JIOT.2025.3634466. [10] 焦慧颖, 张莉, 李红阳. 无线接入回传一体化关键技术及标准化进展[J]. 移动通信, 2022, 46(2): 13–18. doi: 10.3969/j.issn.1006-1010.2022.02.003.JIAO Huiying, ZHANG Li, and LI Hongyang. Key technologies and standardization of NR IAB[J]. Mobile Communications, 2022, 46(2): 13–18. doi: 10.3969/j.issn.1006-1010.2022.02.003. [11] WEN Chaokai, TSAI L S, SHOJAEIFARD A, et al. Shaping a smarter electromagnetic landscape: IAB, NCR, and RIS in 5G standard and future 6G[J]. IEEE Communications Standards Magazine, 2024, 8(1): 72–78. doi: 10.1109/MCOMSTD.0008.2300036. [12] CHUKHNO O, CHUKHNO N, PIZZI S, et al. Mission-critical connectivity enhanced by IAB in beyond 5G: Interplay of sidelink, directional unicasting, and multicasting[J]. IEEE Open Journal of the Communications Society, 2023, 4: 1826–1838. doi: 10.1109/OJCOMS.2023.3304022. [13] SHENG Min, ZHANG Yaqian, LIU Junyu, et al. Enabling integrated access and backhaul in dynamic aerial-terrestrial networks for coverage enhancement[J]. IEEE Transactions on Wireless Communications, 2024, 23(8): 9072–9084. doi: 10.1109/TWC.2024.3358296. [14] ZHANG Hongyi, QI Zhiqiang, LI Jingya, et al. 5G network on wings: A deep reinforcement learning approach to the UAV-based integrated access and backhaul[J]. IEEE Transactions on Machine Learning in Communications and Networking, 2024, 2: 1109–1126. doi: 10.1109/TMLCN.2024.3442771. [15] ZHANG Hongyi, LI Jingya, QI Zhiqiang, et al. Autonomous navigation and configuration of integrated access backhauling for UAV base station using reinforcement learning[C]. 2022 IEEE Future Networks World Forum (FNWF), Montreal, Canada, 2022: 184–189. doi: 10.1109/FNWF55208.2022.00040. [16] ZHANG Yongqiang, KISHK M A, and ALOUINI M S. Energy-efficient optimization in aerial IAB networks for emergency communications[J]. IEEE Transactions on Aerospace and Electronic Systems, 2025, 61(2): 4614–4626. doi: 10.1109/TAES.2024.3513278. [17] ZHANG Jiuzhou, HUO Wei, CHEN Xiaomeng, et al. Remote state estimation with discounted multi-armed bandits for non-stationary channel selection[J]. IEEE Control Systems Letters, 2025, 9: 787–792. doi: 10.1109/LCSYS.2025.3578934. [18] YE Shuai, WANG Tianyu, and WANG Shaowei. Dynamic spectrum access in non-stationary environments: A Thompson sampling based method[C]. GLOBECOM 2022–2022 IEEE Global Communications Conference, Rio de Janeiro, Brazil, 2022: 3989–3994. doi: 10.1109/GLOBECOM48099.2022.10000684. [19] ZHANG Yongqiang, KISHK M A, and ALOUINI M S. A survey on integrated access and backhaul networks[J]. Frontiers in Communications and Networks, 2021, 2: 647284. doi: 10.3389/frcmn.2021.647284. [20] SHAKHATREH H, KHREISHAH A, and JI Bo. Providing wireless coverage to high-rise buildings using UAVs[C]. 2017 IEEE International Conference on Communications (ICC), Paris, France, 2017: 1–6. doi: 10.1109/ICC.2017.7997403. [21] YUAN Long, QIN Lu, LIN Xuemin, et al. Effective and efficient dynamic graph coloring[J]. Proceedings of the VLDB Endowment, 2017, 11(3): 338–351. doi: 10.14778/3157794.3157802. [22] 孙彦景, 李林, 王博文, 等. 灾后无人机自组网高动态多信道TDMA调度算法[J]. 西安电子科技大学学报, 2024, 51(2): 56–67. doi: 10.19665/j.issn1001-2400.20230414.SUN Yanjing, LI Lin, WANG Bowen, et al. Highly dynamic multi-channel TDMA scheduling algorithm for the UAV ad hoc network in post-disaster[J]. Journal of Xidian University, 2024, 51(2): 56–67. doi: 10.19665/j.issn1001-2400.20230414. [23] MOHAMED E M, ALNAKHLI M, HASHIMA S, et al. Distribution of multi mmWave UAV mounted RIS using budget constraint multi-player MAB[J]. Electronics, 2023, 12(1): 12. doi: 10.3390/electronics12010012. -
下载:
图(7) / 表(2)
计量
- 文章访问数: 29
- HTML全文浏览量: 4
- PDF下载量: 6
- 被引次数: 0
