Research On Event Synchronization Measures Based On Functional Communication Network Analysis

With the continuous development of mobile communication technology,a large number of mobile communication spatio-temporal data emerge.The spatio-temporal data of mobile communication includes the voice traffic data and throughput traffic data generated in the geographical range covered by the communication base station.It is helpful to study the concurrency of high voice traffic and high throughput traffic in base station for rational planning and optimization of network resources.Use complex network theory to build functional communication network and to study event synchronization measures.A new method of event synchronization measures is proposed from a new aspect,which provides new ideas for more event synchronization research.For the synchronization measures based on the mobile communication base station data in a region,the following researches are mainly done in this paper:(1)Dynamic synchronization measures are studied based on dynamic functional communication network.The threshold method is used to identify the high-traffic events in the time series,and the events are proposed to get the high-traffic event sequence.The statistical relationships among high-traffic event sequences are identified by event coincidence analysis.The complex network theory is applied to construct static functional communication network,and the synchronous spatio-temporal distribution characteristics of high-traffic events are analyzed.The dynamic functional communication network was constructed with an hour as the time window to analyze the dynamic synchronous changes of high-traffic events.The results show that the hightraffic events of base stations have high spatial continuity and regular change,and there is an interactive relationship between the high-traffic events of different base stations.(2)The directional synchronization measure is proposed based on coupled communication function network.With the advent of the Internet era,the demand of users has gradually changed from voice traffic to throughput traffic.Oriented synchronization between high voice traffic and throughput traffic is helpful to predict the demand of throughput traffic.The event coincidence analysis method is used to identify the directional synchronization between high voice traffic and throughput traffic in the same geographical location and to quantify the extent of directional synchronization.The directional coupling between events is represented by the directional synchronization between events.Based on the event coincidence analysis method,a novel method for constructing directional networks is proposed to study the coupling relationship between two types of events in different geographical locations.It is found that in the area with large pedestrian flow,the coupling extent of high voice traffic and high throughput traffic in the same geographical location is high.There is a coupling correlation between high voice traffic and high throughput traffic in different geographical locations.(3)The optimization synchronization measure based on functional communication network is proposed.Aiming at the limitation of event synchronization method,a new optimization event synchronization method is proposed.Based on the base station mobile communication traffic data,the accuracy of the event synchronization method,the declustering event synchronization method and the optimization event synchronization method are verified.Three methods were used to construct functional communication networks.The differences of network topology characteristics of different methods were compared and the reasons for the differences were explored.The sensitivity of different methods to parameters is analyzed.From the aspects of application and theory,it shows that the optimization event synchronization method accurately identifies the event synchronization and quantifies the event synchronization extent.It is also found that the declustering event synchronization method is sensitive to the maximum dynamic delay,and it will destroy the network strength structure with no extra methods.