Large-scale MIMO dynamic environment fingerprint positioning method based on domain adaptive network

Abstract

The invention provides a large-scale MIMO (Multiple Input Multiple Output) dynamic environment fingerprint positioning method based on a domain adaptive network, which comprises the following steps of: training a positioning model by combining a small number of unsupervised samples in a current environment on the basis of an initial environment position fingerprint database by adopting a domain adaptive learning mode; the working capability of the model in the current environment can be improved, and the trained model has relatively high robustness in the current environment; according to the scheme, the positioning precision improvement effect in the dynamic environment is good, the positioning model in the current environment can be trained and obtained through a small amount of label-free data in the current environment, and the problems that the positioning model loses efficacy and the resampling cost is high after the environment changes can be relieved to a certain extent.

Description

technical field [0001] The invention belongs to the technical field of wireless communication, and in particular relates to a large-scale MIMO (multiple input multiple output) dynamic environment fingerprint positioning method based on a domain adaptive network. Background technique [0002] With the rapid development of Internet of Things technology, the demand for Location Based Service (LBS) has increased significantly, so that precise positioning technology has received extensive attention. At present, the main positioning technology is divided into geometric method and fingerprint method. The geometric method usually uses multiple base stations to receive the Angle of Arrival (AOA), Time of Advent (TOA) and Received Signal Strength (RSS) of the Line of Sight (LOS) of the user signal. Measured values ​​for co-location. However, in complex environments such as tall buildings and indoors, the received signal contains a lot of Non-line of Sight (NLOS) interference or even...

AI Technical Summary

Problems solved by technology

However, there are still deficiencies: the fingerprint method has high requirements on the environment and cannot meet the positioning requirements in a dynamic environment.

When the multi-path environment in the positioning area changes (for example: parking of outdoor vehicles and movement of indoor homes, etc.), the wireless fingerprint of the same location changes during the online phase, resulting in the invalidation of the location fingerprint database collected in the original environment.

In the new environment, the mapping relationship between wireless fingerprints and locations has changed, so that the positioning model established by the initial location fingerprint database training is no longer applicable to the new environment, and the cost of re-collecting samples to update the location fingerprint database is too high

Images