Network Capacity Optimization for Cellular‐Assisted Vehicular Systems by Online Learning‐Based mmWave Beam Selection

Abstract

Directional communication is helpful to improve the performance of millimeter Wave (mmWave) links. However, the dynamic nature of vehicular scenarios raises the complexity of directional mmWave vehicular communications. Also, a mmWave link is susceptible to blockages. Therefore, a mmWave vehicular communication system requires high environmental adaptability and context-awareness. Due to inadequate context information and insufficient beam settings in the existing related algorithm, it is difficult to pick out the set of beams with more reasonable widths and directions, which hinders the further promotion of network capacity in vehicular networks. Therefore, we propose an improved fast machine learning (IFML) algorithm to overcome this shortcoming. In order to improve network capacity while suppressing the additional beam search overhead, a partitioned search method is designed in the IFML. Also, in order to be robust to occasional fluctuations and timely adapt to significant changes in communication environments, the IFML adopts a flexible beam performance update approach based on adjustable weight coefficient. The simulation results show that the IFML significantly outperforms the existing related algorithm in terms of aggregate received data after a certain number of online learning time periods.

Enhancing Energy Efficiency for Cellular-Assisted Vehicular Networks by Online Learning-Based mmWave Beam Selection

“…For Assumption 1, although the parameters are somewhat different from those described in [2,3], they are essentially the same.…”

“…The control function ( ) should be adequately chosen to ensure that the EFML scheme obtains an expected good performance with respect to its regret. The Theorem 1 in [2] provides an appropriate choice for the control function and it is also described in [3]. For the convenience of readers, we repeat it as follows.…”

“…The EFML scheme is most similar to the works in [2,3]. However, the work in [2] only considers the one-dimensional context vector.…”

“…Due to the many communication parameters involved in beam selection, pure mathematical modeling is very complex. So some researchers are trying to use machine learning to figure out how to solve this problem, where beam selection methods based on online learning [2,3] are used without prior training and very suitable for dynamic vehicular networks. In the existing online learning-based mmWave beam selection methods, based on the assumption that the performance of a specific beam is similar under similar contexts, by self-exploration, learning and adapting to the dynamic communication environment, each mmWave Small Base Station (mmSBS) can select right beams, where it independently learns from its previous decisions and the relationship to available vehicular contexts, and becomes conscious of the performance of each mmWave beam.…”

“…In the Fast Machine Learning (FML) scheme in [2], the assumption that there are no overlapping mmWave beams limits the optional beam width and azimuth range. In the Improved Fast Machine Learning (IFML) scheme in [3], the beams are allowed to overlap and the virtual beam concept is proposed. Unlike the number of Radio Frequency (RF) chains, which is limited by form-factor and manufacturing cost [4], a virtual beam is only a region that limits energy propagation and its shape is usually a cone (or sector), where the number of virtual beams can be infinite at each mmSBS.…”