A Construction Method of Spectrum Map

Abstract

The invention provides a method for constructing a spectrum map, which includes the following steps: first step: using multiple spectrum sensors randomly deployed in the area to collect and obtain observations about power; second step: presetting the number of center points and the initial attenuation index value; use an improved algorithm to select the initial value of the center point position; the third step: use the least squares fitting method combined with iterative alternating minimization method to obtain the estimated value of weight, center point position and attenuation index; Step four: Obtain the estimated value of the power at any position in the area through a specific expression, so as to obtain the spectrum map of the power in the area. The effect of applying the method of the present invention is: the method of the present invention does not need to use relevant information about the transmitter and the propagation environment, so that the application scenario is more universal; the number of observation points required is less, so the number of sensors and deployment requirements are higher. Low cost and more economical; the method of the invention does not need parameter adjustment, which is more conducive to realization.

Description

technical field [0001] The present invention relates to the technical field of communications, and in particular, to a method for constructing a frequency spectrum map. Background technique [0002] The spectrum map is to superimpose the spectrum information on the map, which usually indicates the distribution of a certain radio parameter (such as received signal power, channel gain, interference power, etc.) in the area of ​​interest. Spectrum map information is more conducive to improving frequency utilization due to the consideration of spatial multiplexing of idle spectrum. [0003] The most commonly used spectral map construction methods are various deterministic and geostatistical spatial interpolation methods, such as inverse distance method, natural neighbor interpolation method, improved Shepard method and Kriging interpolation method, etc. Among the above interpolation methods, although the kriging interpolation method is the most complex and requires more observa...

AI Technical Summary

Problems solved by technology

[0005] The above methods have the following defects: (1) methods based on compressed sensing, methods based on dictionary learning, and methods based on matrix completion require a large number of observations, thus requiring the dense deployment of a large number of sensors in the region of interest, which is costly , and sometimes cannot be realized in practical applications; (2) The method based on matrix completion and the method based on radial basis function only consider the problem of spectrum map construction from a mathematical point of view, without considering its physical meaning, without using (3) The parameters in the method based on dictionary learning and the method based on radial basis function need to be carefully selected, and this parameter tuning process will bring a lot of extra work (4) The method based on transmitter position estimation requires relevant information about the transmitter and the propagation environment, which cannot be obtained in practical applications

Images