Approximating Clustered Millimeter Wave Vehicular Channels by Sparse Subband Fitting

Approximating Clustered Millimeter Wave Vehicular Channels by Sparse Subband Fitting

Paper in proceedings (conference paper)

Understanding millimeter wave (mmWave) vehicular channels is crucial for the application of mmWave technologies in vehicle-to-anything settings. However, as of yet, few attempts of low-complexity approximation of such channels exist. Prior results have shown that such channels are often composed of clustered multipath components, and this work builds on those results. We present an approach to project a measured mmWave channel into subbands. Sufficiently narrow subbands do not resolve the cluster structures and are efficiently approximated as sparse channels. We thereby render sparse tapped delay line model fits possible. In this contribution, we optimize sparse fits in subbands, and then combine all fits to approximate the full band. We evaluate this approach using vehicular mmWave channel measurements, and demonstrate that subband fitting results in efficient leveraging of sparse structures of mmWave channel data.

Understanding millimeter wave (mmWave) vehicular channels is crucial for the application of mmWave technologies in vehicle-to-anything settings. However, as of yet, few attempts of low-complexity approximation of such channels exist. Prior results have shown that such channels are often composed of clustered multipath components, and this work builds on those results. We present an approach to project a measured mmWave channel into subbands. Sufficiently narrow subbands do not resolve the cluster structures and are efficiently approximated as sparse channels. We thereby render sparse tapped delay line model fits possible. In this contribution, we optimize sparse fits in subbands, and then combine all fits to approximate the full band. We evaluate this approach using vehicular mmWave channel measurements, and demonstrate that subband fitting results in efficient leveraging of sparse structures of mmWave channel data.

mmWave; Vehicular Channel Models; c-LASSO; Cluster

mmWave; Vehicular Channel Models; c-LASSO; Cluster

BLAZEK, T.; ZÖCHMANN, E.; MECKLENBRÄUKER, C.

2020

09.09.2018

978-1-5386-6009-6

Proceedings of 29th IEEE Annual International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC)

91

95

5