Millimeter-wave vehicular networks incur enormous beam-training overhead to enable narrow-beam communications. This paper proposes a learning and adaptation framework in which the dynamics of the communication beams are learned and then exploited to design adaptive beam-tracking and training with low overhead: on a long-timescale, a deep recurrent variational autoencoder (DR-VAE) uses noisy beam-training feedback to learn a probabilistic model of beam dynamics and enable predictive beam-tracking; on a short-timescale, an adaptive beam-training procedure is formulated as a partially observable (PO-) Markov decision process (MDP) and optimized viapoint-based value iteration(PBVI) by leveraging beam-training feedback and a probabilistic prediction of the strongest beam pair provided by the DR-VAE. In turn, beam-training feedback is used to refine the DR-VAE via stochastic gradient ascent in a continuous process of learning and adaptation. The proposed DR-VAE learning framework learns accurate beam dynamics: it reduces the Kullback-Leibler divergence between the ground truth and the learned model of beam dynamics by ~95% over the Baum-Welch algorithm and a naive learning approach that neglects feedback errors. Numerical results on a line-of-sight scenario with multipath and 3D beamforming reveal that the proposed dual timescale approach yields near-optimal spectral efficiency, and improves it by 130% over a policy that scans exhaustively over the dominant beam pairs, and by 20% over a state-of-the-art POMDP policy. Finally, a low-complexity policy is proposed by reducing the POMDP to an error-robust MDP, and is shown to perform well in regimes with infrequent feedback errors.

This paper proposes a dual timescale learning and adaptation framework to learn a probabilistic model of beam dynamics and concurrently exploit this model to design adaptive beam-training with low overhead: on a long timescale, a deep recurrent variational autoencoder (DR-VAE) uses noisy beam-training observations to learn a probabilistic model of beam dynamics; on a short timescale, an adaptive beam-training procedure is formulated as a partially observable Markov decision process and optimized using point-based value iteration by leveraging beam-training feedback and probabilistic predictions of the strongest beam pair provided by the DR-VAE. In turn, beam-training observations are used to refine the DR-VAE via stochastic gradient ascent in a continuous process of learning and adaptation. It is shown that the proposed DR-VAE learning framework learns accurate beam dynamics and, as learning progresses, the training overhead decreases and the spectral efficiency increases. Moreover, the proposed dual timescale approach achieves near-optimal spectral efficiency, with a gain of 85% over a policy that scans exhaustively over the dominant beam pairs, and of 18% over a state-of-the-art POMDP policy.

Mobility may degrade the performance of next-generation vehicular networks operating at the millimeter-wave spectrum: frequent loss of alignment and blockages require repeated beam training and handover, thus incurring huge overhead. In this paper, an adaptive and joint design of beam training, data transmission and handover is proposed, that exploits the mobility process of mobile users and the dynamics of blockages to optimally trade-off throughput and power consumption. At each time slot, the serving base station decides to perform either beam training, data communication, or handover when blockage is detected. The problem is cast as a partially observable Markov decision process, and solved via an approximate dynamic programming algorithm based on PERSEUS [2]. Numerical results show that the PERSEUS-based policy performs near-optimally, and achieves a 55% gain in spectral efficiency compared to a baseline scheme with periodic beam training. Inspired by its structure, an adaptive heuristic policy is proposed with low computational complexity and small performance degradation.

Millimeter-wave (mm-wave) systems rely on narrow-beams to cope with the severe signal attenuation in the mm-wave frequency band. However, susceptibility to beam mis- alignment due to mobility or blockage requires the use of beam-alignment schemes, with huge cost in terms of overhead and use of system resources. In this paper, a beam-alignment scheme is proposed based on Bayesian multi-armed bandits, with the goal to maximize the alignment probability and the data-communication throughput. A Bayesian approach is proposed, by considering the state as a posterior distribution over angles of arrival (AoA) and of departure (AoD), given the history of feedback signaling and of beam pairs scanned by the base- station (BS) and the user-end (UE). A simplified sufficient statistic for optimal control is identified, in the form of preference of BS-UE beam pairs. By bounding a value function, the second-best preference policy is formulated, which strikes an optimal balance between exploration and exploitation by selecting the beam pair with the current second-best preference. Through Monte-Carlo simulation with analog beamforming, the superior performance of the second-best preference policy is demonstrated in comparison to existing schemes based on first-best preference, linear Thompson sampling, and upper confidence bounds, with up to 7%, 10% and 30% improvements in alignment probability, respectively.

Millimeter-wave will be a key technology in next-generation wireless networks thanks to abundant bandwidth availability. However, the use of large antenna arrays with beamforming demands precise beam alignment between the transmitter and the receiver and may entail huge overhead in mobile environments. This paper investigates the design of an optimal interactive beam alignment and data communication protocol, with the goal of minimizing power consumption under a minimum rate constraint. The base station selects beam alignment or data communication and the beam parameters, based on the feedback from the user end. Based on the sectored antenna model and uniform prior on the angles of departure and arrival (AoD/AoA), the optimality of a fixed-length beam-alignment phase followed by a data-communication phase is demonstrated. Moreover, a decoupled fractional beam-alignment method is shown to be optimal, which decouples the alignment of AoD and AoA over time, and iteratively scans a fraction of their region of uncertainty. A heuristic policy is proposed for non-uniform prior on AoD/AoA, with provable performance guarantees, and it is shown that the uniform prior is the worst-case scenario. The performance degradation due to detection errors is studied analytically and via simulation. The numerical results with analog beams depict up to 4dB, 7.5dB, and 14dB gains over a state-of-the-art bisection method and conventional and interactive exhaustive search policies, respectively, and demonstrate that the sectored model provides valuable insights for beam-alignment design.

Millimeter-wave (mm-wave) communications incur a high beam alignment cost in mobile scenarios such as vehicular networks. Therefore, an efficient beam alignment mechanism is required to mitigate the resulting overhead. In this paper, a one-dimensional mobility model is proposed where a mobile user (MU), such as a vehicle, moves along a straight road with time-varying and random speed, and communicates with base stations (BSs) located on the roadside over the mm-wave band. To compensate for location uncertainty, the BS widens its transmission beam and, when a critical beamwidth is achieved, it performs beam-sweeping to refine the MU position estimate, followed by data communication over a narrow beam. The average rate and average transmission power are computed in closed form and the optimal beamwidth for communication, number of sweeping beams, and transmission power allocation are derived so as to maximize the average rate under an average power constraint. Structural properties of the optimal design are proved, and a bisection algorithm to determine the optimal sweeping - communication parameters is designed. It is shown numerically that an adaptation of the IEEE 802.11ad standard to the proposed model exhibits up to 90\% degradation in spectral efficiency compared to the proposed scheme.

This paper presents a comparison between cooperative networks with full relay participation and limited participation (LP), in terms of the energy used by the entire network and the bit error rate (BER) of the received signal. The Universal Software Radio Peripheral (USRP) platform was used to perform the transmissions while signal processing, combination and time synchronization was performed on the GNU Radio Companion (GRC). The experiments were performed in an indoor office environment with cooperative transmission (CT) taking place over a single hop network, using binary phase shift keying (BPSK) as the modulation scheme. It has been demonstrated that relay selection provides better energy efficiency for a required quality of service (QoS).

This paper presents an experimental comparison between cooperative communication and single-input single-output (SISO) system, in terms of the corresponding ranges of signal reception in each case. Transmissions have been achieved using Universal Software Radio Peripheral (USRP) platform, with the signal processing and time synchronization occurring in the GNU Radio environment. The method has been implemented in a typical office environment, with cooperative transmission (CT) taking place over a multiple hop network, using the binary phase shift keying (BPSK) scheme. Presented herein are comparisons between SISO and cooperative networks with regards to ranges and bit error rate (BER) performance.

In this letter, a virtual multiple-input single-output (VMISO) network employing non-coherent frequency shift keying (FSK) is considered. In the VMISO network, spatially separated single-antenna nodes transmit the same information cooperatively to a single receiver where each received signal is affected by an independent carrier frequency offset (CFO). The existing works in this area assume a perfect carrier synchronization between the nodes. However, in this letter, the effects of CFOs on the performance degradation of this network are analyzed. For that purpose, the expression for the probability of symbol error has been derived. The results indicate that the performance is degraded due to CFOs, which is dependent upon the magnitude of CFOs, signal-to-noise ratio (SNR), number of transmitting nodes, and the modulation order of FSK. At high SNR, the CFOs affect the system severely and the performance margin is minimum.

In this paper, we propose a mathematical model for timing synchronization errors in a cascaded virtual multi-hop multiple-input multiple-output (MIMO) system that incorporates cooperation at each hop using decode-and-forward (DF) algorithm. Specifically, the DF relays are grouped to form clusters and one cluster transmits the same data to the next cluster over orthogonal fading channels. The proposed error model is used to study the statistics of the timing error from one cluster to the next and it has been shown that the variance of the timing errors gradually increases as the data traverse the multi-hop network. The effects of the timing errors on the bit error probability (BEP) performance of the system have been quantified. For that purpose, the closed-form expressions of the BEP for the network are derived for every hop with a particular number of nodes per hop. The results indicate that the BEP performance of the system degrades as the data propagates from hop to hop. We quantify the minimum required SNR and the optimal number of relays per cluster that guarantee successful traversal of data to a specified number of hop while keeping overall BEP below a defined threshold.

In this paper, we present a game theoretic framework for spectrum allocation in distributed cognitive radio networks containing both unicast and broadcast traffic. Our proposed scheme aims to minimize broadcast latency for broadcast traffic and minimize interference and access contention for both types of traffic. We develop a utility function that ensures that both objectives are met yielding a higher network throughput. Our proposed spectrum allocation game is also formulated as a potential game and is guaranteed to converge to a Nash equilibrium if the sequential best response dynamics is followed. A proof of concept of the proposed algorithm has been implemented on the Orbit radio testbed and the results verify the convergence of the potential game. Our simulation and experimental results also reveal that the choice of utility function improves the average network throughput for a mixed traffic profile.