Burstiness-aware bandwidth reservation for ultra-reliable and low-latency communications in tactile internet

Zhanwei Hou, Changyang She, Yonghui Li, Tony Q.S. Quek, Branka Vucetic

研究成果: Article同行評審

32 引文 斯高帕斯(Scopus)


The Tactile Internet that will enable humans to remotely control objects in real time by tactile sense has recently drawn significant attention from both academic and industrial communities. Ensuring ultra-reliable and low-latency communications with limited bandwidth is crucial for Tactile Internet. Recent studies found that the packet arrival processes in Tactile Internet are very bursty. This observation enables us to design a spectrally efficient resource management protocol to meet the stringent delay and reliability requirements while minimizing the bandwidth usage. In this paper, both model-based and data-driven unsupervised learning methods are applied in classifying the packet arrival process of each user into high or low traffic states, so that we can design efficient bandwidth reservation schemes accordingly. However, when the traffic-state classification is inaccurate, it is very challenging to satisfy the ultra-high reliability requirement. To tackle this problem, we formulate an optimization problem to minimize the reserved bandwidth subject to the delay and reliability requirements by taking into account the classification errors. Simulation results show that the proposed methods can save 40%-70% bandwidth compared with the conventional method that is not aware of burstiness, while guaranteeing the delay and reliability requirements. Our results are further validated by the practical packet arrival processes acquired from experiments using a real tactile hardware device.

頁(從 - 到)2401-2410
期刊IEEE Journal on Selected Areas in Communications
出版狀態Published - 2018 十一月

All Science Journal Classification (ASJC) codes

  • 電腦網路與通信
  • 電氣與電子工程


深入研究「Burstiness-aware bandwidth reservation for ultra-reliable and low-latency communications in tactile internet」主題。共同形成了獨特的指紋。