A prediction method of inter-frequency signal strength based on AI model

Abstract

The invention discloses a method for predicting the strength of different frequency signals based on an AI model. Including the following steps: step 1, data collection, step 2, sample preprocessing, step 3, obtaining a prediction model, step 4, model detection, step 5, prediction and use based on the model: step 6: update and maintenance of the model. The present invention The superior effect is that the inter-frequency signal strength is accurately predicted, and the inter-frequency measurement process is avoided in the scenario where inter-frequency handover is performed on the UE to select an inter-frequency target cell or multiple secondary carriers are configured during carrier aggregation. Obtain the signal strength information of inter-frequency adjacent cells without relying on the measurement time slot, and at the same time solve the inaccurate prediction of inter-frequency signal strength caused by too large grid granularity in the prior art and the prediction model caused by not regularly updating the grid The problem of inaccurate inter-frequency signal strength prediction caused by failure.

Description

technical field [0001] The invention belongs to the technical field of mobile communication, and in particular relates to an inter-frequency signal strength prediction method based on an AI (Artificial Intelligence, artificial intelligence) model. Background technique [0002] At present, with the gradual popularization of smart phones, the rapid development of mobile communication needs has been driven. The Long Term Evolution (LTE) wireless network, which is a mainstream communication technology, has an increasing business volume. Carriers can no longer meet the needs of society. Therefore, operators usually deploy more wireless frequency bands to provide greater communication capacity. For example, multiple carriers are used to cover the same area to improve the communication capacity of the system. Due to frequency differences and deployment differences in different frequency bands, the signal strengths of different frequency bands at the same location may vary greatly. ...

AI Technical Summary

Problems solved by technology

The disadvantage of the scheme is that when the area where users are concentrated is limited by the fixed step size (such as 5dB) and the number of adjacent cells, the grid division granularity is large, and the characteristics of different grids are very different, which leads to low prediction accuracy. Problem; if the fixed step size is too small, it will cause too many grids, resulting in too large storage space and too long search time; and the area with few users will not have reasonable statistical characteristics due to the small number of samples, which will lead to The raster is not available, or the prediction accuracy is not high after use

Images