In this paper, we propose a deep learning-based beam tracking method for millimeter-wave (mmWave) communications. Beam tracking is employed for transmitting the known symbols using thesounding beamsand tracking time-varying channels to maintain a reliable communication link. When the pose of a user equipment (UE) device varies rapidly, the mmWave channels also tend to vary fast, which hinders seamless communication. Thus, models that can capture temporal behavior of mmWave channels caused by the motion of the device are required, to cope with this problem. Accordingly, we employ a deep neural network to analyze the temporal structure and patterns underlying in the time-varying channels and the signals acquired by inertial sensors. We propose a model based on long short term memory (LSTM) that predicts the distribution of the future channel behavior based on a sequence of input signals available at the UE. This channel distribution is used to 1) control the sounding beams adaptively for the future channel state and 2) update the channel estimate through themeasurement update stepunder a sequential Bayesian estimation framework. Our experimental results demonstrate that the proposed method achieves a significant performance gain over the conventional beam tracking methods under various mobility scenarios.

In this paper, we propose an efficient beam training technique for millimeter-wave (mmWave) communications. Beam training should be performed frequently when some mobile users are under high mobility to ensure the accurate acquisition of the channel state information. To reduce the resource overhead caused by frequent beam training, we introduce a dedicated beam training strategy which sends the training beams separately to a specific high mobility user (called a target user) without changing the periodicity of the conventional beam training. The dedicated beam training requires a small amount of resources because the training beams can be optimized for the target user. To satisfy the performance requirement with a low training overhead, we propose the optimal training beam selection strategy which finds the best beamforming vectors yielding the lowest channel estimation error based on the target user's probabilistic channel information. This dedicated beam training is combined with the greedy channel estimation algorithm that accounts for sparse characteristics and temporal dynamics of the target user's channel. Our numerical evaluation demonstrates that the proposed scheme can maintain good channel estimation performance with significantly less training overhead compared to the conventional beam training protocols.

In this paper, we propose a new greedy channel estimation technique which can reconstruct dynamic sparse signals for millimeter wave (mmWave) communication system. In presence of the mobile user under high mobility, the angle of arrival (AoA) and angle of departure (AoD) is changing with time. We assume that the AoA and AoD are varying by discrete-state Markov random process. We formulate the estimation problem for mmWave channel as joint estimation problem of two variables; 1)AoA and AoD indices, and 2)the amplitude vector in the angular domain. The proposed greedy algorithm effectively estimate the AoD and AoA indices accounting for sparse characteristics and dynamics of the channel. Our experimental evaluation demonstrates that the proposed algorithm can obtain good channel estimation performance with small amount of computational power.

In this paper, we propose a new recursive sparse channel recovery algorithm which can track time-varying support of angular domain channel response vector in mobility scenario for millimeter wave-band communications. We model the angle of departure (AoD) and the angle of arrival (AoA) using discrete state Markov random process and derive joint estimation of the time-varying support and amplitude of the angular domain channel vector. Using sequential Monte Carlo (SMC) method, the proposed channel estimation scheme tracks the support by drawing the samples from a posteriori distribution of the support indices while capturing the dynamics of time-varying amplitude using Kalman filter. Our simulation results show that the proposed algorithm yields significantly better tracking performance than the existing compressed sensing schemes.

In this paper, we propose a near-ultrasound chirp signal-based communication for the TV's 2nd screen services. While near-ultrasound (with under 20 kHz frequency) communication has been developed for various applications recently, none of the previous work provides a perfect solution for 2nd screen services between TVs and smart devices. This is due mainly to the following real world challenges. The embedded signal in TV contents should be successfully received in a typical TV-watching environment by (i) delivering information at least at 15 bps with significantly low volume to avoid human perception, (ii) despite the presence of ambient noise, e.g., a tick, a snap, or a knock. To fulfill (i), we design chirp quaternary orthogonal keying (QOK) symbols. Especially, we aim to minimize inter-symbol interference (ISI) effects by symbol design in order to completely eliminate guard intervals. To resolve (ii), we propose the novel J-shape detection algorithms for both frame synchronization and carrier sensing. The proposed modem achieves almost zero frame error rate on a smartphone 2.7 m away from the TV even with minimal receive sound pressure level of 35 dBSPL, i.e., the noise level in a very quiet room. Moreover, throughout experiments and log analysis of 2nd screen service deployed in a nation-wide TV broadcasting system, J-shape detection algorithms are proven to achieve highly resilient performance for both frame synchronization and carrier sensing compared to previous schemes.