Radio Environment Map Construction Based on Spatial Statistics and Bayesian Hierarchical Model
Leverage (statistics)
DOI:
10.1109/tccn.2021.3066566
Publication Date:
2021-03-17T19:27:01Z
AUTHORS (6)
ABSTRACT
Constructing an infrastructure for spectrum sensing and sharing on the cloud is a promising technology. In this paper, we propose analysis framework based data gathered by distributed sensors to construct radio environment map (REM). To best of our knowledge, first attempt leverage power Bayesian Markov chain Monte Carlo (MCMC) in REM. Specifically, three-stage hierarchical model (BHM) established imitate generation process under spatially correlated shadow fading. Parameters BHM are estimated with MCMC algorithm from collected sensor network. Then, address space-dimension inference problem aim interpolate signal strength where there no node composition sampling. At each point area interest, posterior predictive distribution receive can be obtained kernel density smoother. We make spatial two location modes (square lattice located randomly sensors) scenarios (with without information about source), respectively. Simulation results demonstrate that although mode suitable parameter estimation, performance not better than square mode. Quantitative confirms effectiveness compelling.
SUPPLEMENTAL MATERIAL
Coming soon ....
REFERENCES (47)
CITATIONS (23)
EXTERNAL LINKS
PlumX Metrics
RECOMMENDATIONS
FAIR ASSESSMENT
Coming soon ....
JUPYTER LAB
Coming soon ....