IMPRESS: Indoor Mobility Prediction Framework for Pre-Emptive Indoor-Outdoor Handover for mmWave Networks


Turkmen A., Ansari S., Klaine P. V., Zhang L., Imran M. A.

IEEE Open Journal of the Communications Society, vol.2, pp.2714-2724, 2021 (ESCI) identifier

  • Publication Type: Article / Article
  • Volume: 2
  • Publication Date: 2021
  • Doi Number: 10.1109/ojcoms.2021.3133543
  • Journal Name: IEEE Open Journal of the Communications Society
  • Journal Indexes: Emerging Sources Citation Index (ESCI), Scopus, Compendex, INSPEC, Directory of Open Access Journals
  • Page Numbers: pp.2714-2724
  • Keywords: femtocells, heterogeneous network indoor-to-outdoor handover, indoor mobility, Markov Chain, mmWave 5G, online learning, pre-emptive handover, user trajectory
  • Bilecik Şeyh Edebali University Affiliated: No

Abstract

Millimeter-wave (mmWave) communication, the main success behind the fifth generation of mobile communication networks, will increase the ultra-dense small cell deployment under its limited coverage characteristics. Therefore, providing a seamless connection to its users, to whom transitioning between indoor and outdoor in a heterogeneous network environment particularly is a significant issue that needs to be addressed. In this paper, we present a two-fold contribution with a comprehensive study on mm-wave handovers. A user-based indoor mobility prediction via Markov chain with an initial transition matrix is proposed in the first step. Based on this acquired knowledge of the user's movement pattern in the indoor environment, we present a pre-emptive handover algorithm in the second step. This algorithm aims to keep the QoS high for indoor users when transitioning between indoor and outdoor in a heterogeneous network environment. The proposed algorithm shows a reduction in the handover signalling cost by more than 50%, outperforming conventional handover algorithms.