Fast Beam Tracking for Millimeter-Wave Systems Under High Mobility

Zhang, Deyou; Li, Ang; Shirvanimoghaddam, Mahyar; Cheng, Peng; Li, Yonghui; Vucetic, Branka

doi:10.1109/icc.2019.8761896

Cited by 7 publications

(6 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The prior and posterior probabilities work hand-in-hand: the former admits changing channel conditions while the latter resists change, fostering beam stability over time. And as opposed to most other greedy algorithms [20]- [26], [28]- [32], [34] [35], the Markov process supports multiple hypotheses: it propagates all 𝐾 states over time instead of committing to only the most likely. This allows hypotheses that persist over time to emerge, in which the prior reinforces the posterior, rendering the algorithm robust.…”

Section: (4)mentioning

confidence: 99%

“…• number of hypotheses: Most papers commit to a single hypothesis per time [20]- [26], [28]- [32], [34][35], meaning that only information about the DD beam that is estimated per chain is propagated over time, whereas some papers propagate information about multiple candidate DD beams per chain, either through particle filtering [27][33] [36] or Bayesian inference [37][38], and so are more robust to harsh, uncertain, or rapidly changing channel conditions.…”

Section: Introductionmentioning

confidence: 99%

“…• tracking mechanism: Many papers assume an underlying motion modeltypically lineareither directly for the DD angle of the dominant paths or vis-à-vis R's motion, and solve a least-squares fit between the angles predicted from the motion and the angles within the codebook, either through Kalman filtering [21][23] [30][34], gradient descent [24][29] [31], or particle filtering [27][33] [36]. Others, rather, localize the scan within the codebook by means of a probability distributionuniform [22][28], Gaussian [24][32] [37], or exponential [26][38]centered at a previously estimated angle; the codebook angles are then conditioned by the probability when generating a new estimate, e.g. through Bayesian inference [37][38].…”

Section: Introductionmentioning

confidence: 99%

“…Most papers assume a sparse channel, i.e. with just a couple of dominant paths [22] [30] or even just the LoS path [20][24][25] [26]. The single paper that employs actual measurements only considers motion over a meter or so and hence does not capture large-scale variation over which the number of scatterers and their properties change significantly.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Markov Multi-Beamtracking on 60 GHz Mobile Channel Measurements

Blandino

Senic

Gentile

et al. 2022

IEEE Open J. Veh. Technol.

View full text Add to dashboard Cite

Beamforming training refers to the exhaustive scan over which the transmitter and receiver jointly steer their beams along a predefined set of double-directional angles to determine the beam pairs that coincide with the dominant propagation paths of the channel, for millimeter-wave spatial multiplexing at millimeter-wave. When mobile, training necessitates a high refresh rate to maintain connectivity and so, to reduce overhead, beamtracking algorithms exploit the spatial-temporal consistency of the channel to localize the scan around the beam pairs determined at a previous time. The algorithms' true performance, however, is still unknown since results reported to date are based on oversimplified channel models. In this paper, we propose a novel beamtracking algorithm formulated as a first-order Markov process that supports multiple beam pairs. The algorithm is evaluated through actual channel measurementsnot a channel modelrecorded with our high-precision 3D double-directional 60 GHz channel sounder. The measurement campaign, to our knowledge, is unprecedented: with 10,895 large-scale measurements, spaced 8.8 cm apart on average to emulate continuous motion, over which the mobile receiver traversed a total of 900.2 m. We demonstrate that four beam pairs can be sustained always and that eight pairs can be sustained 57% of the time.

show abstract

Section: (4)mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Markov Multi-Beamtracking on 60 GHz Mobile Channel Measurements

Blandino

Senic

Gentile

et al. 2022

IEEE Open J. Veh. Technol.

View full text Add to dashboard Cite

show abstract

“…The scattering in the mmWave spectrum has been shown to induce attenuation of over 20dB [13]. Hence, the line of sight beam is considered for the purpose of channel estimation and tracking [14]- [17]. This assumption is reasonable due to the recent measurements that showed sparsity of mmWave communications [13], [18], [19].…”

Section: B Channel Modelmentioning

confidence: 99%

Extended Kalman Filter Beam Tracking for Millimeter Wave Vehicular Communications

Shaham¹,

Kokshoorn²,

Ding³

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

Millimeter-wave (mmWave) communication is a promising technology to meet the ever-growing data traffic of vehicular communications. Unfortunately, more frequent channel estimations are required in this spectrum due to the narrow beams employed to compensate for the high path loss. Hence, the development of highly efficient beam tracking algorithms is essential to enable the technology, particularly for fast-changing environments in vehicular communications. In this paper, we propose an innovative scheme for beam tracking based on the Extended Kalman Filter (EKF), improving the mean square error performance by 49% in vehicular settings. We propose to use the position, velocity, and channel coefficient as state variables of the EKF algorithm and show that such an approach results in improved beam tracking with low computational complexity by taking the kinematic characteristics of the system into account. We also explicitly derive the closed-from expressions for the Jacobian matrix of the EKF algorithm.

show abstract