Wireless testing calibration method based on K-means algorithm

Abstract

The present invention relates to the wireless test field, especially to a wireless testing calibration method based on a K-means algorithm. The method comprises the following steps: the step 1, performing statistics of the difference value between channels through analysis of lots of past data of the same circuit board; the step 2, employing the K-means algorithm to perform cluster division of the channel to be tested, randomly giving K cluster centers, distributing the channel gain values to be classified to each cluster according to the nearest neighbor rule, and obtaining a cluster center channel; and the step 3, obtaining the debugging initial values of the residual channels in other channel clusters through calculation of the difference values with the cluster center channel of the located channel, and completing the calibration. The wireless testing calibration method based on the K-means algorithm greatly improves the first-pass yield of the calibration and reduces the number of the debugging times so as to greatly shorten the test time and improve the production efficiency.

Description

technical field [0001] The invention relates to the field of wireless testing, in particular to a K-means algorithm-based calibration method for wireless testing. Background technique [0002] In the Internet era, wireless networks play an increasingly important role in people's lives. Therefore, more and more products are equipped with wireless functions. In order to ensure the normal use of wireless functions, in the production process, it is necessary to To calibrate and test wireless chips, efficient calibration algorithms are needed to meet the needs of mass production. [0003] In the production process, each circuit board is often wirelessly calibrated according to different hardware designs, or even different circuit boards of the same hardware design, and the calibration parameters that need to be written are often different. However, due to the large number of channels and test items that need to be calibrated, it often takes a long time and requires high retestin...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a wireless test calibration method based on the K-means algorithm, which solves the problem that in the prior art, there are many channels and test items that need to be calibrated for the wireless network status test, resulting in low accuracy and low error rate. high problem

Images