Cell load forecasting method

Abstract

The invention relates to a method for predicting the region load, which comprises that: a, collecting the history data of region load and randomly generating weight and bias; b, using the weight and bias to calculate out the predicted value of region load, and the difference between predicted value and real output value; c, using iterate formula to calculate out the adjusting value of weight and bias, refreshing weight and bias; d, judging if the iterate times is over preset time, or judging if the difference between predicted value and output value is lower than preset value; if it is lower, using refreshed weight and bias to calculate the predicted value, ending, or else, entering into step b. The invention can improve the predict accuracy with better track property.

Description

technical field [0001] The invention relates to the field of mobile communication radio resource control, in particular to a method for realizing cell load prediction. Background technique [0002] Since the Wideband Code Division Multiple Access (WCDMA, Wideband Code Division Multiple Access) system is a self-interference system, any user in the system is the potential interference of other users. In order to facilitate system planning, it is necessary to know the forward and reverse According to various interference situations, then calculate the size of the cell and balance the front and reverse links of the system, and the actual cell size is determined by the propagation model of the area and the tolerable path loss of the front and reverse links, so in When performing radio resource control, the uplink and downlink load is an important evaluation factor. At present, the load control method is usually used to evaluate the uplink and downlink load of the cell. The load c...

AI Technical Summary

Problems solved by technology

However, when the filter coefficient is set too large, there will be more historical data to be measured, which will lead to a decrease in tracking characteristics, which in turn leads to low accuracy of cell load prediction; if the filter coefficient is set too small, the accuracy of the cell load prediction value will not only decrease When the filter coefficient is 1, the cell load prediction value is high, and it is relatively easy to be affected by abnormal values ​​when performing cell load prediction, which leads to low accuracy of cell load prediction

Based on the above reasons, the accuracy of the cell load prediction value in the above-mentioned prior art is not high, and setting too large a filter coefficient will cause load tracking delay, and setting a too small filter coefficient will make the load prediction easily affected by abnormal values

Images