KR100616562B1 - A Radio Wave Model Optimization Method Using the Normal Distribution Characteristics of Radio Waves - Google Patents

Abstract

The present invention optimizes the sample measurement data by using the principle that the radio waves form a normal distribution, and then generates an accurate electromagnetic wave propagation prediction model to calculate accurate results. In addition to significantly improving the accuracy of the electromagnetic wave propagation model used for the purpose, users of the wireless network design software can enter the limiting variables (edge percentage of the normal distribution and the minimum number of removal data) according to the environment and conveniently to the desired conditions. The data can be refined, which greatly improves the designer's convenience over traditional manual data removal.

Normal distribution, histogram, optimized propagation model, cumulative probability distribution function, probability distribution function

Description

Method for optimizing an electro-wave model of the mobile communication system using a regular distribution characteristic of the electro-wave}

1 is an explanatory diagram for explaining a radio wave model optimization apparatus of a conventional mobile communication system.

2 is a flowchart for explaining a conventional propagation model optimization method.

3 is an explanatory diagram for explaining a radio wave model optimizing apparatus for realizing the method of the present invention;

4 is a flowchart of the present invention.

Figure 5 (a), (b) is a rare analysis and radio wave distribution diagram before the present invention is applied.

Figure 6 (a), (b) is a rare hole analysis and radio wave distribution diagram to which the method of the present invention is applied.

<Detailed Description of Codes>

1: test base station 2: measurement equipment part

3: data processing section 4: attenuation determination module

5: correction constant determination module 6: prediction model optimization processing unit

The present invention relates to a method for optimizing a propagation model of a mobile communication system using a normal distribution characteristic of a radio wave, and in particular, by optimizing sample measurement data using a principle that a radio wave forms a normal distribution, and then generating an accurate electromagnetic wave propagation prediction model for accurate results. The present invention relates to a method for optimizing a propagation model of a mobile communication system using a normal distribution characteristic of a radio wave to calculate a.

In general, mobile communication systems have been rapidly developed as the industrial society has spread rapidly. Especially in the early 1980s, in North America, the service area was divided into hexagonal cells of several km to several ten kilometers in diameter, and then in the geographically separated cells. The mobile communication system has grown rapidly by using a cellular mobile communication system that uses the same frequency channel repeatedly and implements a radio channel switching function between cells as the subscriber moves. However, the mobile communication system as described above needs to design a wireless network before constructing the corresponding communication equipment. In this case, the wireless network design measures the strength of the radio signal transmitted from the base station at various places and measures the measurement. The electromagnetic wave propagation model is optimized and reflected by using the signal strength of the radio waves.

Then, referring to FIG. 1, the apparatus for optimizing the electromagnetic wave propagation model of the conventional mobile communication system as described above, the data processor 70 extracting only necessary data from the collected data by measuring signal strength of a specific place. The attenuation determination module 71 determines the degree of attenuation of the received signal strength according to the distance with the data extracted by the processing unit 70, and the correction constant is determined using the result value calculated by the attenuation determination module 71. It consists of a correction constant determining module 72.

Meanwhile, referring to the operation of the conventional electromagnetic wave propagation model optimization apparatus as described above, first, in order to apply the electromagnetic wave propagation model optimization technique, data according to the received signal strength is collected and collected (S101). Then, the data processing unit 70 performs a process (S102, S103) of removing the data if the collected data is below or above a specific value by using the distance and the strength of the received signal, and continuously according to the subjectivity of the user. When the measured data is different from the surrounding data, the process of removing the data is performed (S104). After the data removal process (S104), the process proceeds to the prediction model setting step (S105), and the attenuation determination module 71 reduces the received signal strength according to the distance showing the minimum error deviation using the data on which the data removal process is performed. The accuracy is applied as the slope of the electromagnetic wave propagation model. In addition, after the predictive model setting step S105, the constant compensation step S106 is performed, and the correction constant determining module 72 corrects the predictive model value calculated through the regression analysis to a constant value. In addition, the correction constant determination model 72 removes data having a value greater than a value applied to a theoretical free space loss model or smaller than a noise floor when removing data. Optimize the propagation model.

However, the apparatus for optimizing the electromagnetic wave propagation model of the conventional mobile communication system as described above is determined by performing analysis using the least square method with the measured data extracted to obtain the degree of attenuation of the received signal strength according to the distance. There are several methods to remove unnecessary data but still have a declination gradient to one side, which adversely affects the reliability of the regression results of the whole data. The user has to remove the graphs of the signal strength according to the distance or the measurement order. However, this process has been very inconvenient. Furthermore, the data can be removed by subjective judgment. Degrading door Problems have also occurred.

Accordingly, the present invention has been invented to solve the above-mentioned general problems, and the characteristics of the normal distribution of the radio waves which can significantly improve the accuracy of the electromagnetic wave propagation model used to predict the received signal strength from the base station in the design of the wireless network. The purpose of the present invention is to provide a method for optimizing the propagation model of a mobile communication system.

Another object of the present invention is to input the limiting variables (edge percentage of the normal distribution and the minimum number of removal data) of the wireless network design S / W user can conveniently refine the data to the desired conditions accordingly The present invention provides a method of optimizing a propagation model of a mobile communication system using a normal distribution characteristic of a radio wave, which considerably improves a designer's convenience compared to a manual data removing method.

In order to achieve the above object, the present invention provides a histogram for generating a histogram using measured equations of regression analysis of the electromagnetic wave propagation model using the difference value (A) of the estimated distance attenuation equation and the measured distance data. And a normal distribution generating step of obtaining an average value using the standard deviation of the difference value A calculated after the histogram generating step and preparing a normal distribution thereof, and after the normal distribution preparing step A removal condition setting step of setting a removal condition, a removal range checking step of checking whether both edges of the CDF of the normal distribution currently created after the removal condition setting step are within a range of a data removal setting value (E or "Ref") ; During the removal range checking step, both edges of the CDF of the normal distribution currently created are not within the range of the data removal setting value (E). In this case, the PDF comparison confirmation step of checking whether the normal distribution PDF is smaller than the probability of the histogram calculated by the measurement data and the PDF comparison confirmation step confirm that the normal distribution PDF is smaller than the probability of the histogram calculated by the measurement data. And a step of setting the number of data to be removed (B) if necessary, and calculating the propagation optimization model by removing the set number (B) from the measurement data after the step of setting the number of data to be removed. The present invention provides a method for optimizing a propagation model of a mobile communication system using a normal distribution characteristic of radio waves.

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

The apparatus to which the present invention is applied includes a measuring equipment unit 2 for collecting received signal strength data measured from each test base station 1 through a drive test as shown in FIG. 3, and the measuring equipment unit 2. An attenuation decision for determining the degree of attenuation of the received signal strength according to the distance with the data processor 3 for extracting only the data needed from the received signal strength collected data measured by the data and the data extracted by the data processor 3; Module 4, a correction constant determination module 5 for determining a correction constant using the result value calculated by the attenuation determination module 4, and data calculated from the correction constant determination module 5 The predictive model optimization processing unit 6 includes repeating the data removal operation until the predetermined value ("Ref") becomes smaller than the predetermined value ("Ref") to calculate the propagation optimization model.

Next, the method applied to the apparatus of the present invention as described above will be described in detail.

The method of the present invention proceeds from the initial state S1 to the radio wave measurement step S2 as shown in FIG. 4 and collects the received signal strength data measured through the drive test. After the radio wave measurement step S2, the data processing step S3 is performed to extract data to be used for analysis from the collected data, and to remove the short-range data and the remote data unnecessary for analysis. Eliminate data that is too strong or too weak.

Here, during the data processing step (S3), also, the strength of the specific point is significantly removed because the data is too low or high, unlike the surroundings, which adversely affects the overall analysis. After the data processing step S3, the process proceeds to the attenuation correction step S4, and the first regression analysis is performed to obtain a slope of the degree of attenuation for each distance using the processed data.

That is, by measuring the data according to the received signal strength transmitted from the test base station 1 through the measurement equipment unit 2 is collected. Then, the data processing unit 3 performs a process of removing the data if the data collected by the measuring equipment unit 2 is less than or equal to a specific value Ref using the distance and the strength of the received signal, and the user If the continuous measurement data differs from the surrounding data according to the subject, the process of removing the data is performed. After the data removal process, the attenuation determination module 4 calculates the received signal strength attenuation according to the distance showing the minimum error deviation using the data on which the data removal process is performed by applying the slope of the electromagnetic wave propagation model. Next, input to the correction constant determination module (5). In addition, the correction constant determination module 5 corrects the predictive model value calculated through the regression analysis to a constant value.

On the other hand, after the attenuation correction step (S4) proceeds to the histogram preparation step (S5) and the histogram using the difference value (A) of the estimated distance attenuation equation and the distance-dependent measurement data for the regression analysis of the measured data Write. After the histogram preparation step S5, the process proceeds to the normal distribution preparation step S6 to obtain an average value using the standard deviation with respect to the difference A, and prepare a normal distribution thereof. After the normal distribution creation step S6, the data removal input setting step S7 is made to set a specific value E of the normal distribution to be used as a reference for data removal.

Here, the specific value used during the data removal input setting step (S7) can be used in two ways: to set the percentage value (%) of the edge of the normal distribution and to set the size of the standard deviation of A.

After the data removal input setting step S7, the data removal minimum reference value setting step S8 is performed to set the minimum reference value T from which the number of data can be removed.

On the other hand, after the data removal minimum reference value setting step (S8) to proceed to the removal range check step (S9), both edges of the CDF (cumulative probability distribution function: CUMULATIVE DISTRIBUTION FUNCTION) of the currently generated normal distribution are set to the data removal setting value (E or It is determined whether it is within the range of Ref) . In this case, when both edges of the CDF of the currently generated normal distribution are within the range of the data removal set value E, that is, within the normal range, the optimization model calculation step S12 is determined during the removal range checking step S9. Proceed to

However, as a result of the determination in the removal range checking step S9, if both edges of the CDF of the currently generated normal distribution are not within the range of the data removal setting value E or Ref, that is, in the abnormal range, the PDF (probability distribution function PROBABILITY DISTRIBUTION FUNCTION The process of comparing and confirming (S10) determines whether the normal distribution PDF is smaller than the probability of the histogram calculated by the measured data. In this case, when the normal distribution PDF is greater than the probability of the histogram calculated by the measurement data, the process proceeds to the optimization model calculation step S12. However, if the normal distribution PDF is smaller than the probability of the histogram calculated by the measurement data, as determined by the PDF comparison confirmation step (S10), the process proceeds to the removal data number setting step (S11) to determine the number of data B to be displayed. Set it.

That is, the number B in the removal data number setting step S11 is calculated as (measurement data probability)-(normal distribution PDF * total measurement data number).

On the other hand, after the removal data number setting step S11, the process proceeds to the optimization model calculation step S12, and the radio wave optimization model is calculated by removing the set number B from the measurement data.

That is, the prediction model optimization processor 6 repeats the data removal operation until the data calculated from the correction constant determination module 5 becomes smaller than the value determined by the user through the above process to calculate the optimization model of the radio wave. Done.

Therefore, according to the present invention, it is possible to more accurately design the wireless network by optimizing the electromagnetic wave propagation model used to predict the strength of the signal received from the base station in the wireless network design simulation S / W. In particular, the data are refined in consideration of the normal distribution characteristics of the radio waves, which are data at the edge of the normal distribution, that is, the biased data distributed far from the predictive decay line resulting from the regression analysis (FIGS. 5A and 5B). By avoiding unnecessary influence by the dashed line inside), an accurate electromagnetic wave propagation prediction model is generated.

In other words, if any measurement data is obtained as shown in the regression graph (a of FIG. 5) and the data distribution (b of FIG. 5), the data of the present invention is removed by applying the method of FIG. Regression graphs and data distributions as shown in b of FIG. 6 are obtained.

As shown in (a) and (b) of FIG. 6, the refined data has less skew to one side when compared with the previous data, which makes it possible to predict a more accurate electromagnetic wave propagation model.

In addition, the wireless network design S / W user can input data such as limit variables (normal edge ratio (S203), minimum number of removal data (S204)) according to the environment to conveniently refine the data to the desired condition. It is more convenient than the manual data removal method of the electromagnetic wave propagation model used in the network design software.

As described above, the present invention optimizes the sample measurement data by using the principle that the radio waves form a normal distribution, and then generates an accurate electromagnetic wave propagation prediction model to calculate accurate results. This has the advantage of significantly improving the accuracy of the electromagnetic wave propagation model used to predict signal strength.

In addition, according to the present invention, the user can easily refine the data to the desired condition by inputting the limiting variables (edge percentage of the normal distribution and the minimum number of removal data) according to the environment. Compared to manual data removal method, it greatly improves the convenience of designer.

Claims (4)

A histogram generating step of creating a histogram of the measured data after the regression analysis of the electromagnetic wave propagation model of the mobile communication system using the estimated distance attenuation equation and the difference value (A) of the measured data for each distance, and the histogram A normal distribution creation step of obtaining an average value using the standard deviation of the difference value (A) calculated by the creation step and creating a normal distribution thereof, and a removal condition for setting a removal condition of measured data after the normal distribution creation step A removal range checking step of checking whether both edge data of the CDF of the normal distribution currently created after the setting step and the removing condition setting step fall within a range of the data removal setting value E, and the normality created during the removal range checking step If both edges of the distribution's CDF data are not within the range of the data removal setting (E), then the normal distribution PDF is displayed. The PDF comparison confirmation step of checking whether the histogram calculated by the measurement data is smaller than the probability of the histogram calculated by the measurement data, and when the PDF comparison confirmation step confirms that the normal distribution PDF is smaller than the probability of the histogram calculated by the measurement data. A propagation characterized in that it comprises a removal data number setting step of setting the number B and an optimization model calculation step of calculating the propagation optimization model by removing the set number B from the measurement data after the removal data number setting step. A Method for Optimizing Propagation Model of Mobile Communication Systems Using Normal Distribution Characteristics The data removal input setting step of setting a specific value (E) of a normal distribution to be used as a reference value for data removal, and the minimum number of data numbers can be removed after the data removal input setting step. A method of optimizing a propagation model of a mobile communication system using a normal distribution characteristic of a radio wave, characterized in that it further comprises a step of setting a minimum reference value for data removal for setting a reference value (T). The number B of the removal data number setting step is calculated as (measured data probability)-(normally distributed PDF * total number of measured data). Method of optimizing propagation model of communication system. The method of claim 2, wherein the specific value used during the data removal input setting step is one of a method of setting a percentage value (%) of the edge of a normal distribution and a method of setting the size of the standard deviation of A. A propagation model optimization method of a mobile communication system using a normal distribution characteristic of a radio wave, characterized in that used.

Images