In this paper we consider a multipath channel subject to variations in time due to the presence of a Doppler drift which imposes a phase rotation on the original channel impulse response. Accordingly, a decision-directed state-space approach based on the extended Kalman filter (EKF) is proposed to track the channel variation. This approach is particularly efficient since we are able to reduce the number of training symbols used for channel tracking to be as few as two times the number of transmitters. These training symbols are required to initialize the EKF. Furthermore, using a decision-directed approach we are able to provide the EKF with state-observations without additional training symbols. Our approach is specially designed to work with hybrid analog/digital multiple-input multiple-output (MIMO) architectures.
A State-space Approach for Efficient Channel Tracking in Hybrid Analog/Digital MIMO Architectures
2021-04-01
2980756 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch