Novel Multipath Mitigation Technique for Antenna Arrays Based on Multi-Correlator Structures

Multipath propagation is to this day a major source of error in global navigation satellite systems (GNSSs). Objects in the nearer receiver environment can reflect the broadcast satellite signals, leading to multiple delayed signal replicas at the receiver. A conventional GNSS receiver may provide a reduced level of accuracy under such conditions. Numerous approaches have been proposed in the past to mitigate the effect of multipath propagation. However, distinguishing between the direct and reflected signal replicas remains in particular for short multipath delays a challenging task. Multi-antenna systems depict in this context a powerful solution. The additional spatial dimension allows to differentiate between signals impinging from different direction of arrivals (DOAs). Though, the spatial separation has to be sufficiently large for this to work effectively. Therefore, we propose in this work an algorithm that combines both, the temporal and the spatial domain to effectively mitigate multipaths. It is based on a multipath mitigation algorithm developed recently for the single-antenna case. It relies on a multi-correlator structure with an unstructured underlying signal model. That means, multipaths are not individually resolved, instead the channel impulse response (CIR) is estimated to consider for additional signal replicas. The extension to antenna arrays is achieved by linking the individual CIRs for each array element through the line-of-sight (LOS) steering vector. The functionality of this approach has been demonstrated with simulations. It turned out to be a valid competitor against state-of-the-art approaches relying on a deterministic beamformer (DET) with a favorable noise performance.