A method of predicting mobile phone signal strength based on mobile phone surfing logs

Abstract

The invention provides a method for predicting mobile phone signal intensity according to a mobile phone httplog (httplog data). The method comprises the following steps: finding a corresponding user triple in MME data according to the IMSI and time field in the httplog data, extracting the corresponding signal intensity data from the MR data through the user triple, combining the signal intensity data with the httplog data as the training data, performing the data statistic, the feature extraction, the model construction and the model training; predicting by using the trained model, thereby predicting the mobile phone signal intensity according to the httplog data.

Description

technical field [0001] The present invention belongs to the field of information technology. It specifically relates to a method for predicting the signal strength of a mobile phone according to a mobile phone surfing log. Background technique [0002] With the development of the mobile Internet, more and more people choose to use mobile phones to surf the Internet. For mobile operators, through the statistics and analysis of a large number of online logs, they can grasp the user's online status, and can speculate on the user's access behavior preferences, and then Carry out accurate advertising and marketing. It can also combine user behavior and base station conditions to optimize the network and improve user experience. In the network optimization process, the signal strength of the user's mobile phone is extremely important data, but it is only stored in the MR data. MR data means that information is sent every 480ms on the business channel (470ms on the signaling cha...

AI Technical Summary

Problems solved by technology

MR data means that information is sent every 480ms on the business channel (470ms on the signaling channel), that is to say, any mobile phone in the power-on state sends data to the base station every 480ms, so a large amount of MR data will be generated every day, and its storage cost Very large, it is difficult for operators to store all MR data for statistics and analysis

The existing technology can only obtain the signal strength through MR data, and there is no other technology to measure the signal strength of the mobile phone. Therefore, it is particularly important to predict the signal strength of the mobile phone through the log data of the mobile phone, which can save storage costs and combine the MR data with the log data. Data combined for network optimization

Images