Feature quantity measuring device, radio wave environment measuring device, feature quantity measuring method, radio wave environment measuring method, and program
In order to acquire a feature containing only a fluctuation component of a radio wave signal from received data, a feature quantity measuring device calculates the feature, based on data in a predetermined time span of time series data from a received communication signal, and calculates a degree of similarity between the feature and supervised data representing a preset feature when there is a fading variation. In addition, the feature quantity measuring device determines, based on a comparison result indicating a magnitude relationship between the degree of similarity and a predetermined threshold value, whether a characteristic of the fading variation is contained within the time span and, when the characteristic is not contained, outputs the feature for the time span.
1. A feature quantity measuring device, comprising:
an acquiring circuit configured to acquire time series data from a received signal;
a feature quantity calculating circuit configured to calculate a feature, based on data in a predetermined time span of the time series data;
a degree of similarity calculating circuit configured to calculate a degree of similarity between the feature based on the data in the time span and supervised data representing a preset feature when there is a long-period variation;
a long-period variation detecting circuit configured to output a comparison result indicating a magnitude relationship between the degree of similarity and a predetermined threshold value; and
a control circuit configured to determine, based on the comparison result, whether a characteristic of the long-period variation is contained within the time span and, when the characteristic is not contained, output the feature for the time span,
wherein, when determining that the long-period variation is contained within the predetermined time span, the control circuit instructs the feature quantity calculating circuit of recalculation of a feature for a new time span reduced from the time span and the recalculation is repeated until the control circuit determines that the time span does not contain the characteristic of the long-period variation within the time span.
2. The feature quantity measuring device according to claim 1 ,
wherein the feature quantity calculating circuit calculates a plurality of corresponding features, based on data in a plurality of time spans having different durations of the time series data,
wherein the degree of similarity calculating circuit calculates respective degrees of similarity between the plurality of features and the supervised data,
wherein the long-period variation detecting circuit outputs comparison results respectively indicating a magnitude relationship between each of degrees of similarity between the plurality of features and the supervised data, and a predetermined threshold value; and
wherein the control circuit determines, based on the respective comparison results, whether the characteristic of the long-period variation is contained within the time span and, when the characteristic is not contained, outputs the feature for the time span.
3. The feature quantity measuring device according to claim 1 ,
wherein the supervised data are data representing features based on theoretical values or actual measured values under a plurality of different long-period variation environments.
4. A radio wave environment measuring device, comprising:
the feature quantity measuring devices according to claim 1 disposed at spatially different positional coordinates; and
a database configured to acquire and store the feature, which is calculated by the feature quantity measuring devices, for the time span that does not contain the characteristic of the long-period variation.
5. The feature quantity measuring device according to claim 2 ,
wherein the supervised data are data representing features based on theoretical values or actual measured values under a plurality of different long-period variation environments.
6. A radio wave environment measuring device, comprising:
the feature quantity measuring devices according to claim 2 disposed at spatially different positional coordinates; and
a database configured to acquire and store the feature, which is calculated by the feature quantity measuring devices, for the time span that does not contain the characteristic of the long-period variation.
7. A radio wave environment measuring device, comprising:
the feature quantity measuring devices according to claim 3 disposed at spatially different positional coordinates; and
a database configured to acquire and store the feature, which is calculated by the feature quantity measuring devices, for the time span that does not contain the characteristic of the long-period variation.
8. A feature measuring method, comprising:
acquiring time series data from a received signal;
calculating a feature, based on data in a predetermined time span of the time series data;
calculating a degree of similarity between the feature based on the data in the time span and supervised data representing a preset feature when there is a long-period variation;
outputting a comparison result indicating a magnitude relationship between the degree of similarity and a predetermined threshold value;
determining, based on the comparison result, whether a characteristic of the long-period variation is contained within the time span;
when the characteristic is not contained within the time span, outputting the feature for the time span;
when determining that the long-period variation is contained within the predetermined time span, recalculating a feature for a new time span reduced from the time span and repeating calculating the degree of similarity and outputting the comparison result until the time span does not contain the characteristic of the long-period variation within the time span.
9. A non-transitory storage medium storing instructions that are executable to perform processing comprising:
acquiring time series data from a received signal;
calculating a feature, based on data in a predetermined time span of the time series data;
calculating a degree of similarity between the feature based on the data in the time span and supervised data representing a preset feature when there is a long-period variation;
outputting a comparison result indicating a magnitude relationship between the degree of similarity and a predetermined threshold value;
determining, based on the comparison result, whether a characteristic of the long-period variation is contained within the time span;
when the characteristic is not contained within the time span, outputting the feature for the time span;
when determining that the long-period variation is contained within the predetermined time span, recalculating a feature for a new time span reduced from the time span and repeating calculating the degree of similarity and outputting the comparison result until the time span does not contain the characteristic of the long-period variation within the time span.