UAV-Assisted Three-Dimensional Spectrum Mapping Driven by Spectrum Data and Channel Model

Abstract

As the number of civil aerial vehicles increase explosively, spectrum scarcity and security become an increasingly challenge in both the airspace and terrestrial space. To address this difficulty, this paper presents an unmanned aerial vehicle-assisted (UAV-assisted) spectrum mapping system and a spectrum data reconstruction algorithm driven by spectrum data and channel model are proposed. The reconstruction algorithm, which includes a model-driven spectrum data inference method and a spectrum data completion method with uniformity decision mechanism, can reconstruct limited and incomplete spectrum data to a three-dimensional (3D) spectrum map. As a result, spectrum scarcity and security can be achieved. Spectrum mapping is a symmetry-based digital twin technology. By employing an uniformity decision mechanism, the proposed completion method can effectively interpolate spatial data even when the collected data are unevenly distributed. The effectiveness of the proposed mapping scheme is evaluated by comparing its results with the ray-tracing simulated data of the campus scenario. Simulation results show that the proposed reconstruction algorithm outperforms the classical inverse distance weighted (IDW) interpolation method and the tensor completion method by about 12.5% and 92.3%, respectively, in terms of reconstruction accuracy when the collected spectrum data are regularly missing, unevenly distributed and limited.

Deep Learning for Reduced Sampling Spatial 3-D REM Reconstruction

“…Consequently, DL models with a large number of layers and filters, popular in the field of computer vision, have been employed for the interpolation problem. These include statistical PL modeling, orthogonal matching pursuit (OMP), long-short term memory (LSTM), convolutional neural networks (CNNs), CAEs, or a combination between CNNs and generative adversial networks (GANs) [6], [8], [12], [13], [17], [22], [27]. These methods employ convolutionalpooling layer structure, rather than fully-connected ones, to both reduce the computational complexity, and preserve the environmental characteristics, i.e.…”

“…By including pre-processing of information about the height of the buildings and locations of the TXs, the influence of 3D physical environment is taken into account. Alternatively, modified IDW and PL information is used to infer the unknown values of the RSS in 3D space [17]. The OMP method is utilized in [22] as the 3D data includes the RSS of the transmitters on their corresponding locations in the grid, and zeros on the other points.…”

“…Which domains will be considered depends on the specific scenario for which the algorithm is applied. Most often (as also in this work), the data used for REM reconstruction are obtained from either a small number of stationary sensors or a single mobile receiver and thus, the spatial domain is most important as usually the bandwidth remains the same, and the REM is updated over a period of time [8], [17]. Consequently, the dimension of space affects the model's complexity depending on the spatial grid resolution, and as described in Section V.D, the volume of data grows exponentially in the 3D case.…”

“…This is true especially when it comes to UAV-based sensors, as they are usually limited to operate in very specific areas used for drone flights [18], [20], [21]. 3) PL models -classical method for obtaining the channel gain between a TX and a RX, via statistical models fitted by RW measurements [17], [22], [23]. This method considers each kind of environment, i.e.…”

Limited Sampling Spatial Interpolation Evaluation for 3D Radio Environment Mapping

“…Alternatively, a similar method was applied to infer the path loss from 2D REMs and floor plan images for different indoor environments using DL for REM estimation [20][21][22]29]. The literature has placed a particular focus on the spatial interpolation of REMs under a constrained number of samples, due to some influential works such as [10,[12][13][14][15]18]. This section describes their main features and limitations, to outline the contributions of this paper.…”