Handover Management for mmWave Networks With Proactive Performance Prediction Using Camera Images and Deep Reinforcement Learning

Abstract

For millimeter-wave networks, this paper presents a paradigm shift for leveraging time-consecutive camera images in handover decision problems. While making handover decisions, it is important to predict future long-term performance-e.g., the cumulative sum of time-varying data rates-proactively to avoid making myopic decisions. However, this study experimentally notices that a time-variation in the received powers is not necessarily informative for proactively predicting the rapid degradation of data rates caused by moving obstacles. To overcome this challenge, this study proposes a proactive framework wherein handover timings are optimized while obstacle-caused data rate degradations are predicted before the degradations occur. The key idea is to expand a state space to involve timeconsecutive camera images, which comprises informative features for predicting such data rate degradations. To overcome the difficulty in handling the large dimensionality of the expanded state space, we use a deep reinforcement learning for deciding the handover timings. The evaluations performed based on the experimentally obtained camera images and received powers demonstrate that the expanded state space facilitates (i) the prediction of obstacle-caused data rate degradations from 500 ms before the degradations occur and (ii) superior performance to a handover framework without the state space expansion.

A Survey of Machine Learning Applications to Handover Management in 5G and Beyond

“…The camera images also enable the prediction of obstacles that will affect the mm-wave links. The authors in [140] developed a DRL framework using camera images for optimizing the HO timing by predicting the future data rate of mm-wave links and ensuring that proactive HO is performed before data rate degradation occurs due to presence of obstacles.…”

Analysis and Optimization for Robust Millimeter-Wave Communications

“…As shown in [56,99], MC can be used to avoid communication interruptions during a handover or a blockage by having multiple connections that are simultaneously active. Other solutions involve methods that use context information including SNR, beam ID, radio maps, users' position, and even camera images [100,101,102,103]. Given such information, the network can predict possible blockages and select the optimal mmAP to perform a handover with, or reduce the duration of the initial access and beam alignment.…”

Communication-Efficient Multimodal Split Learning for mmWave Received Power Prediction

“…For handover or positioning, RF-modalities, e.g., received power or channel state information, were studied [18], [19]. For mmWave received power prediction, handover, prior studies in [4], [11], [20], [21] leverage visual data to detect sudden LoS-and-NLoS transitions due to moving obstacles [5], [6]. While these aforementioned works demonstrate the feasibility of wireless systems benefitting from RF or non-RF modality, these works focus on the usage of a single modality.…”

“…To complement such insufficient mmWave RF features, non-RF domain data can be utilized, such as location information [7], [8], motion sensory data [9] as well as visual data obtained from RGB-depth (RGB-D) cameras [4], [10], [11], which is the focus of our contribution. As shown in Fig.…”