Deep Spectrum Cartography: Completing Radio Map Tensors Using Learned Neural Models
Deep Spectrum Cartography: Completing Radio Map Tensors Using Learned Neural Models
复制标题
DOI：
10.1109/tsp.2022.3145190
复制
发表时间：
2021-05
影响因子：
5.4
通讯作者：
S. Shrestha;Xiao Fu;Min-Fong Hong
中科院分区：
文献类型：
--
作者：
S. Shrestha;Xiao Fu;Min-Fong Hong
频谱制图（SC）技术从有限的测量中构建多域（如频率、空间和时间）射频（RF）地图，这可以被视为一个不适定张量补全问题。基于模型的制图技术通常依赖于手工制作的先验（例如，稀疏性、平滑性和低秩结构）来完成任务。这样的先验可能不足以捕捉复杂无线环境的本质，尤其是在发生严重阴影的情况下。为了规避这些挑战，SC考虑了离线训练的无线电地图深度神经模型，因为深度神经网络（dnn）能够从数据中“学习”复杂的底层结构。然而，这种基于深度学习（DL）的SC方法在离线模型学习（训练）和完成（泛化）方面都遇到了严重的挑战，可能是因为生成无线电地图的潜在状态空间太大了。在这项工作中，提出了一种基于发射器无线电地图分解的方法，在这种方法下，只有单个发射器的无线电地图被dnn建模。这样，学习和泛化的挑战都可以大大减轻。利用学习到的深度神经网络，提出了一种快速非负矩阵分解的两阶段SC方法和性能增强的迭代优化算法。理论方面-如无线电张量的可恢复性，样本复杂性和噪声鲁棒性-在提出的框架下进行了表征，这些理论性质在基于dl的无线电张量补全的背景下是难以捉摸的。利用室内和强阴影环境的合成数据和真实数据进行实验，验证了所提方法的有效性。
The spectrum cartography (SC) technique constructs multi-domain (e.g., frequency, space, and time) radio frequency (RF) maps from limited measurements, which can be viewed as an ill-posed tensor completion problem. Model-based cartography techniques often rely on handcrafted priors (e.g., sparsity, smoothness and low-rank structures) for the completion task. Such priors may be inadequate to capture the essence of complex wireless environments—especially when severe shadowing happens. To circumvent such challenges, offline-trained deep neural models of radio maps were considered for SC, as deep neural networks (DNNs) are able to “learn” intricate underlying structures from data. However, such deep learning (DL)-based SC approaches encounter serious challenges in both off-line model learning (training) and completion (generalization), possibly because the latent state space for generating the radio maps is prohibitively large. In this work, an emitter radio map disaggregation-based approach is proposed, under which only individual emitters’ radio maps are modeled by DNNs. This way, the learning and generalization challenges can both be substantially alleviated. Using the learned DNNs, a fast nonnegative matrix factorization-based two-stage SC method and a performance-enhanced iterative optimization algorithm are proposed. Theoretical aspects—such as recoverability of the radio tensor, sample complexity, and noise robustness—under the proposed framework are characterized, and such theoretical properties have been elusive in the context of DL-based radio tensor completion. Experiments using synthetic and real-data from indoor and heavily shadowed environments are employed to showcase the effectiveness of the proposed methods.