Predicting wireless communication quality a few seconds ahead using spatial informationWireless-telecommunication link-quality-prediction technology using spatial information

This technology predicts the link quality of wireless telecommunication a few seconds into the future by using spatial information such as image information collected from cameras and location information obtained from sensor information and by modeling of the relationship between the spatial information and the link quality using machine learning.

The use of high-frequency bands is necessary to cope with large amounts of mobile traffic. However, due to the linearity of radio waves, communication quality is easily affected by surrounding structures and moving objects. This technology predicts wireless communication quality more than one second ahead based on spatial information obtained from cameras and sensors. We believe that the quality of service can be maintained by controlling the transmission data rate, buffers, etc. in advance. We are working to establish a highly reliable communication system that can be used in high-frequency bands.

• The technology predicts the future behavior of peripheral objects from image and location information to determine the impact on wireless communication quality one second in advance.

• Transmission data rate and buffer control in video distribution services
• Terminal operation and behavior control using link quality as optimization parameter

NTT Network Innovation Laboratories – Frontier Communication Laboratory