K-Nearest Neighbors Gaussian Process Regression for Urban Radio Map Reconstruction

Han, Min; Wang, Shaowei

doi:10.1109/lcomm.2022.3207210

Cited by 16 publications

(4 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Taking O E with (8), the rotation and scaling properties can be applied to the virtual obstacle map H, which helps reconstruct the geometry structure of the virtual environment. Using (13), it is also possible to downsample or oversample a feature map, as one can define the output size to be different to the input size, which can adjust the feature map in different scales flexibly for CNN layer.…”

Section: The Scattering Branchmentioning

confidence: 99%

“…Interpolation-based radio map construction techniques exploit real measurements taken at various locations of transmitters (TXs) and receivers (RXs) without explicitly exploiting the geometry structure of the environment. Some representative interpolation methods for radio map construction include k-nearest neighbor (KNN) interpolation [13], inverse distance weighted (IDW) interpolation [14], matrix completion [15], dictionary-based compressive sensing [16], and Kriging [17], etc. These methods are based on the spatial correlation of measurements, but they cannot differentiate the corresponding propagation conditions, such as LOS and non-line-of-sight (NLOS).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

UAV-aided Joint Radio Map and 3D Environment Reconstruction using Deep Learning Approaches

Zeng

Chen

2022

ICC 2022 - IEEE International Conference on Communications

View full text Add to dashboard Cite

Machine learning (ML) facilitates rapid channel modeling for 5G and beyond wireless communication systems. Many existing ML techniques utilize a city map to construct the radio map; however, an updated city map may not always be available. This paper proposes to employ the received signal strength (RSS) data to jointly construct the radio map and the virtual environment by exploiting the geometry structure of the environment. In contrast to many existing ML approaches that lack of an environment model, we develop a virtual obstacle model and characterize the geometry relation between the propagation paths and the virtual obstacles. A multiscreen knife-edge model is adopted to extract the key diffraction features, and these features are fed into a neural network (NN) for diffraction representation. To describe the scattering, as oppose to most existing methods that directly input an entire city map, our model focuses on the geometry structure from the local area surrounding the TX-RX pair and the spatial invariance of such local geometry structure is exploited. Numerical experiments demonstrate that, in addition to reconstructing a 3D virtual environment, the proposed model outperforms the state-of-theart methods in radio map construction with 10%−18% accuracy improvements. It can also reduce 20% data and 50% training epochs when transferred to a new environment.

show abstract

Section: The Scattering Branchmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

UAV-aided Joint Radio Map and 3D Environment Reconstruction using Deep Learning Approaches

Zeng

Chen

2022

ICC 2022 - IEEE International Conference on Communications

View full text Add to dashboard Cite

show abstract

“…To reduce the cost for establishing a radio map, some researchers have devoted to radio map interpolation by using, e.g., tailored raytracing acceleration [38]. These studies have corroborated the promising benefits of using radio map in indoor [39] and urban environments [40]. They also shed lights on the utilization of radio maps in remote areas.…”

mentioning

confidence: 95%

“…In light of the aforementioned studies and the unique characteristics of remote areas, we use a coarse-grained radio map in this work, which records the large-scale CSI only, rather than the RSS data [36], [37], [39], [40]. The large-scale CSI varies slowly and usually is easy to obtain in practice.…”

mentioning

confidence: 99%

Radio Map-Based Cognitive Satellite-UAV Networks Towards 6G On-Demand Coverage

Feng,

Lin,

Wang

et al. 2024

IEEE Trans. Cogn. Commun. Netw.

View full text Add to dashboard Cite

The sixth generation (6G) network is envisioned to cover remote areas, with the help of satellites and unmanned aerial vehicles (UAVs). Considering the vastness of remote areas and the sparsity of users therein, we investigate a cognitive satellite-UAV network, where satellites and UAVs coordinately share spectrum to provide low-rate and high-rate services in a complementary manner. Multiple UAVs form a virtual antenna array to serve unevenly distributed users via multipleinput-multiple-output (MIMO) non-orthogonal multiple access (NOMA). An on-demand coverage framework is proposed so as to dynamically focus the communication resources on target users. In the framework, a radio map recording the slowlyvarying large-scale channel state information (CSI) is utilized. Different from traditional pilot-based approaches, the large-scale CSI is obtained by a lookup in the radio map per the position information of users and UAVs, during the online optimization of the network. In this way, the system overhead could be largely reduced. To explore the potential gain of such a framework, we formulate a joint power allocation problem to maximize the minimum user rate, which is not only non-convex but also with implicit expressions. We recast the problem after uncovering its mathematical characteristics, and derive its locally-optimal solution in an iterative manner. Simulation results corroborate that the proposed framework can significantly improve the coverage performance at a low cost.Index Terms-Cognitive satellite-UAV network, large-scale channel state information (CSI), non-orthogonal multiple access (NOMA), on-demand coverage, unmanned aerial vehicle (UAV).

show abstract

An efficient position optimization method based on improved genetic algorithm and machine learning for sparse array

Yang,

Chen,

Zhou

et al. 2024

AEU - International Journal of Electronics and Communications

View full text Add to dashboard Cite

K-Nearest Neighbors Gaussian Process Regression for Urban Radio Map Reconstruction

Cited by 16 publications

References 14 publications

UAV-aided Joint Radio Map and 3D Environment Reconstruction using Deep Learning Approaches

UAV-aided Joint Radio Map and 3D Environment Reconstruction using Deep Learning Approaches

Radio Map-Based Cognitive Satellite-UAV Networks Towards 6G On-Demand Coverage

An efficient position optimization method based on improved genetic algorithm and machine learning for sparse array

Contact Info

Product

Resources

About