CN109033389A - A kind of spectrum monitoring data processing platform (DPP) in knowledge based library - Google Patents

Abstract

The invention discloses a kind of spectrum monitoring data processing platform (DPP)s in knowledge based library, belong to spectrum monitoring field, for spectrum monitoring data processing.Data access layer can handle the real time data of Historical Monitoring data file or frequency spectrum monitoring system, after being accumulated to the data of access with one hour dimension, various monitoring data are converted to the data of unified format by data prediction layer, Business Logic is for statistical analysis to data, regular presumption, abnormal signal detection, and conclusion is stored in knowledge base；A kind of frequency point feature descriptive model is designed in knowledge base, describes a frequency point and its changing rule from multiple dimensions；Operator is managed knowledge base by application layer, retrieves, browses data in knowledge base, application layer also supports external system to pass through interface and access.The present invention realizes statistics unified to spectrum monitoring data, efficient, analysis, processing.

Description

A Spectrum Monitoring Data Processing Platform Based on Knowledge Base

technical field

The invention belongs to the field of spectrum monitoring, in particular to the field of spectrum monitoring data processing.

Background technique

my country's radio management agencies at all levels conduct daily monitoring of radio frequencies. With the increasingly complex electromagnetic environment, the continuous monitoring of the electromagnetic environment by monitoring stations has produced massive monitoring data. Spectrum monitoring is mainly divided into three steps: data collection, data analysis, and data accumulation. At present, it is difficult for spectrum monitoring data to timely, accurately and comprehensively reflect the usage status and changing trend of radio spectrum resources. The accumulated monitoring data is fragmented, and the data availability and developability are not high. The continuous historical change monitoring information during the frequency point monitoring cannot be fully and accurately displayed quantitatively. The existing monitoring software has a low degree of intelligence, and data processing and analysis methods are backward. Traditional abnormal signal alarms use fixed thresholds and fixed monitoring data items as alarm trigger conditions, which lack flexibility. In different electromagnetic environments, different systems, and different Under conditions such as frequency time periods and different frequency business scopes, it shows obvious defects such as poor adaptability, insufficient scalability, and high false alarm rate.

Contents of the invention

The purpose of the present invention is to provide a spectrum monitoring data processing platform based on a knowledge base to avoid the disadvantages in the background technology. The present invention describes a frequency point and its changing law from multiple dimensions by establishing a frequency point feature description model based on a knowledge base, and through statistical analysis, induction and sorting of historical monitoring data and real-time monitoring data, and policies, regulations, and data detection Algorithms, abnormal signal alarm rules and prediction rules, abnormal cases and other information and functions are unified to the spectrum monitoring data processing platform.

In order to solve the problems of the technologies described above, the present invention is achieved through the following technical solutions:

A knowledge base-based spectrum monitoring data processing platform adopts a layered design and includes the following functional modules of each layer:

The data access layer is used to access the monitoring data of the spectrum monitoring system through files or network communication, and output the monitoring data to the data preprocessing layer; the monitoring data includes broadband spectrum data, narrowband spectrum data, ITU parameter measurement data, Signal demodulation data, communication system identification data, broadband direction finding data and narrowband direction finding data;

The data preprocessing layer is used to convert the monitoring data into formatted data, and output the formatted monitoring data to the business logic layer;

The business logic layer is used for statistical analysis and signal early warning of the formatted monitoring data, and stored in the knowledge base;

The knowledge base is established by the relational database Oracle, which uses a series of two-dimensional tables to store the information output by the business logic layer;

The application layer is responsible for the interaction between the user and the platform. It is used to retrieve and view the statistical information and change rules of each frequency point data in the knowledge base; set the parameters of abnormal alarm rules and display abnormal alarm information; it is also used to provide external interfaces through which external systems can retrieve , Access knowledge base information, obtain abnormal alarm information.

Wherein, the business logic layer includes a data accumulation module, a data processing module, a data statistics module, a rule estimation module, a signal warning module and a data storage module:

The data accumulation module is used to cache various formatted monitoring data in a set time dimension, and output the cached monitoring data to the data processing module;

The data processing module is used to output the various formatted data to the data statistics module after unified resolution;

The data statistics module is used to calculate the maximum value, average value and amplitude-probability value of the background noise amplitude according to the wideband spectrum data and the narrowband spectrum data; calculate the azimuth-probability value and the maximum probability azimuth value according to the broadband direction finding data and the narrowband direction finding data; ITU parameter measurement data, signal demodulation data and communication system identification data calculate signal bandwidth-probability value, modulation style-probability value; obtain information domain information according to signal demodulation data; and output various calculated values and information domain information To the rule estimation module and data storage module;

The rule estimation module is used to estimate the time, day, week and month rules of each frequency point and dimension data in the frequency domain, modulation domain, azimuth domain and information domain according to the data output by the data statistics module, and output to the signal early warning module and data storage module;

The signal early warning module is used to compare the rules in the frequency domain, modulation domain, azimuth domain and information domain output by the rule estimation module with the knowledge base's compliance use and rule description information of the frequency point, and generate an alarm message if there is any change , outputting the alarm information to the data storage module;

The data storage module is used to store the received data into the knowledge base.

Wherein, the knowledge base includes a policy domain module, a frequency domain module, a modulation domain module, an azimuth domain module, an information domain module and an abnormal alarm record module;

The policy domain module is used to record the radio frequency allocation of the People's Republic of China, the radio frequency allocation data of ITU Regions 1, 2, and 3, and the frequency database and station database data issued by the competent department;

The frequency domain module is used to record the maximum value, average value and amplitude-probability information of the background noise amplitude within the set time, as well as the frequency domain change rule information by time, day, week, and month;

The modulation domain module is used to record the signal bandwidth-probability information and modulation style-probability information within the set time, as well as the modulation domain change rule information by time, day, week and month;

The azimuth field module is used to record the azimuth-probability information and maximum probability azimuth within the set time, as well as the change law information of the azimuth field by time, day, week, and month;

The information domain module is used to record the information of the information domain within the set time and the change rule information of the call sign by time, day, week and month;

The abnormal alarm recording module is used to record signal alarm information within a set time.

The beneficial effect of the present invention compared with background technology is:

(1) A spectrum monitoring data knowledge base is established to describe the characteristics and changing rules of a frequency point from multiple dimensions;

(2) Strong scalability, compatible with multiple types of radio monitoring data, and processing common heterogeneous data;

(3) Provides flexible configuration technology of customization and self-maintenance, which is convenient for users to configure, add and change data statistical analysis and abnormal signal alarm algorithm by themselves;

(4) Based on the abnormal signal intelligent alarm technology based on alarm rules and historical laws, the alarm level of abnormal signal alarms is determined according to the abnormal occurrence probability of each dimension.

Description of drawings

Fig. 1 is a schematic diagram of the composition of the data processing platform of the present invention;

Fig. 2 is a schematic diagram of a data analysis and processing flow in the business logic layer of the present invention;

Fig. 3 is a schematic diagram of the main components of the knowledge base of the present invention.

Detailed ways

The technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

The invention provides a frequency spectrum monitoring data processing platform based on a knowledge base, including the composition of the data processing platform, the processing process of data statistical analysis, the processing process of signal alarm, and the main components of the knowledge base.

Figure 1 shows the layered structure of the data processing platform, including data access layer, data preprocessing layer, business logic layer, knowledge base and application layer. The data access layer accesses the broadband spectrum data, narrowband spectrum data, ITU parameter measurement data, signal demodulation data, communication system identification data, broadband direction finding data and narrowband direction finding data monitored by the spectrum monitoring system through file or network communication The data preprocessing layer converts the received monitoring data into formatted data; the business logic layer completes the statistical analysis of the monitoring data and signal early warning; the knowledge base is established by the relational database Oracle, and a series of two-dimensional tables are used to store knowledge ; The application layer is responsible for the interaction between the user and the platform, and is composed of a knowledge base management and maintenance function module, a knowledge base data display module, a signal warning display module, and an external access interface.

Fig. 2 shows the flow chart of data analysis and processing in the described business logic layer, and the business logic layer includes data accumulation module, data processing module, data statistics module, rule deduction module, signal early warning module and data storage module: data accumulation module is paired with Various monitoring result data are cached in the time dimension of one hour, and data processing and statistics start on the hour. The data processing module processes the spectral data with a resolution of less than 1kHz through conversion calculation at a resolution of 1kHz; the direction finding result is processed at a resolution of 1 degree through conversion calculation; the analog demodulation data (audio file) is processed for language recognition and call sign recognition. The data statistics module calculates the maximum value and average value of the background noise amplitude from the spectrum data, and calculates the amplitude-probability value; the direction-finding data calculates the azimuth-probability value, the maximum probability azimuth value; calculates the signal bandwidth-probability value, modulation style-probability value. The law estimation module is to estimate the law of time, day, week, and month of each frequency point and each dimension data in a manual and automatic way. The data storage module is to store the analysis results of data statistics and law description information into the corresponding knowledge base.

Figure 3 shows the data model describing the characteristics of frequency points in the knowledge base, describing its characteristics from several aspects such as policy domain, frequency domain, azimuth domain, modulation domain, information domain and abnormal alarm records. The policy domain module records the compliance use of frequency points, including the radio frequency allocation data of the People's Republic of China, the radio frequency allocation data of ITU Regions 1, 2, and 3; the frequency database and station database data released by the competent authority. The frequency domain module records the maximum value and average value of the background noise amplitude per hour at the frequency point; the amplitude-probability information per hour; the frequency domain change information by time, day, week, and month. The modulation domain module records the signal bandwidth-probability information of the frequency point per hour; records the signal modulation pattern-probability information of the frequency point per hour; the modulation domain change rule information by time, day, week, and month. The azimuth domain module records the hourly azimuth-probability information of the frequency point; the hourly maximum probability azimuth; and the azimuth change rule information by hour, day, week, and month. The information domain module records data such as audio, call sign, fax, and communication system identification results of the frequency point per hour; the change rule information of call sign by hour, day, week, and month. The abnormal alarm recording module records the signal alarm information of the frequency point.

Combined with Figure 1, a brief description of the working process of the spectrum monitoring data processing platform is given.

When working, the spectrum monitoring data processing platform can process the real-time data and historical data files of the spectrum monitoring system. First, use a unified format to unify the data format of various spectrum monitoring systems, and the data of various monitoring results are accumulated by hour. After the data accumulation is completed, it is processed separately, the frequency data is converted, and processed at a uniform resolution of 1kHz; the direction finding result data is converted, and the frequency is uniformly processed at a resolution of 1kHz, and the azimuth is uniformly processed at a resolution of 1 degree ; ITU parameter measurement data, signal demodulation data, communication system identification data and data related to the center frequency are converted and processed uniformly at 1kHz resolution. The processed data files are statistically calculated separately to calculate the maximum value and average value of the background noise amplitude, amplitude-probability information, direction-degree-probability information, and maximum probability direction-degree of each frequency point within this hour. The corresponding background service estimates the change rules of each frequency point and dimension according to time, day, week, and month. The operator imports and edits data such as the frequency division table through the application layer software; can retrieve and view the statistical information and change rules of each frequency point data in the knowledge base; set abnormal alarm rule parameters and display abnormal alarm information; at the same time, the spectrum monitoring data platform provides External interface, the external system can retrieve and access knowledge base information through the interface, and obtain abnormal alarm information.

Claims (3)

1. a kind of frequency spectrum monitoring data processing platform based on knowledge base, is characterized in that, adopts layered design, comprises the following each layer functional modules: The data access layer is used to access the monitoring data of the spectrum monitoring system through files or network communication, and output the monitoring data to the data preprocessing layer; the monitoring data includes broadband spectrum data, narrowband spectrum data, ITU parameter measurement data, Signal demodulation data, communication system identification data, broadband direction finding data and narrowband direction finding data; The data preprocessing layer is used to convert the monitoring data into formatted data, and output the formatted monitoring data to the business logic layer; The business logic layer is used for statistical analysis and signal early warning of the formatted monitoring data, and stored in the knowledge base; The knowledge base is established by the relational database Oracle, which uses a series of two-dimensional tables to store the information output by the business logic layer; The application layer is responsible for the interaction between the user and the platform. It is used to retrieve and view the statistical information and change rules of each frequency point data in the knowledge base; set the parameters of abnormal alarm rules and display abnormal alarm information; it is also used to provide external interfaces through which external systems can retrieve , Access knowledge base information, obtain abnormal alarm information. 2. A kind of spectrum monitoring data processing platform based on knowledge base according to claim 1, it is characterized in that, described business logic layer comprises data accumulation module, data processing module, data statistics module, regularity deduction module, signal early warning Modules and data storage modules: The data accumulation module is used to cache various formatted monitoring data in a set time dimension, and output the cached monitoring data to the data processing module; The data processing module is used to output the various formatted data to the data statistics module after unified resolution; The data statistics module is used to calculate the maximum value, average value and amplitude-probability value of the background noise amplitude according to the wideband spectrum data and the narrowband spectrum data; calculate the azimuth-probability value and the maximum probability azimuth value according to the broadband direction finding data and the narrowband direction finding data; ITU parameter measurement data, signal demodulation data and communication system identification data calculate signal bandwidth-probability value, modulation style-probability value; obtain information domain information according to signal demodulation data; and output various calculated values and information domain information To the rule estimation module and data storage module; The rule estimation module is used to estimate the time, day, week and month rules of each frequency point and dimension data in the frequency domain, modulation domain, azimuth domain and information domain according to the data output by the data statistics module, and output to the signal early warning module and data storage module; The signal early warning module is used to compare the rules in the frequency domain, modulation domain, azimuth domain and information domain output by the rule estimation module with the knowledge base's compliance use and rule description information of the frequency point, and generate an alarm message if there is any change , outputting the alarm information to the data storage module; The data storage module is used to store the received data into the knowledge base. 3. A kind of spectrum monitoring data processing platform based on knowledge base according to claim 1, is characterized in that, described knowledge base comprises policy domain module, frequency domain module, modulation domain module, azimuth domain module, information domain module and abnormal alarm recording module; The policy domain module is used to record the radio frequency allocation of the People's Republic of China, the radio frequency allocation data of ITU Regions 1, 2, and 3, and the frequency database and station database data issued by the competent department; The frequency domain module is used to record the maximum value, average value and amplitude-probability information of the background noise amplitude within the set time, as well as the frequency domain change rule information by time, day, week, and month; The modulation domain module is used to record the signal bandwidth-probability information and modulation style-probability information within the set time, as well as the modulation domain change rule information by time, day, week and month; The azimuth field module is used to record the azimuth-probability information and maximum probability azimuth within the set time, as well as the change law information of the azimuth field by time, day, week, and month; The information domain module is used to record the information of the information domain within the set time and the change rule information of the call sign by time, day, week and month; The abnormal alarm recording module is used to record signal alarm information within a set time.