TY - GEN
T1 - Wireless network virtualization by leveraging blockchain technology and machine learning
AU - Adhikari, Ashish
AU - Rawat, Danda B.
AU - Song, Min
N1 - Publisher Copyright:
c 2019 Association for Computing Machinery.
PY - 2019/5/15
Y1 - 2019/5/15
N2 - Wireless Virtualization (WiVi) is emerging as a new paradigm to provide high speed communications and meet Quality-of-Service (QoS) requirements of users while reducing the deployment cost of wireless infrastructure for future wireless networks. In WiVi, Wireless Infrastructure Providers (WIPs) sublease their RF channels through slicing to Mobile Virtual Network Operators (MVNOs) based on their Service Level Agreements (SLAs) and the MVNOs independently provide wireless services to their end users. This paper investigates the wireless network virtualization by leveraging both Blockchain technology and machine learning to optimally allocate wireless resources. To eliminate double spending (aka over-committing) of WIPs’ wireless resources such as RF channels, Blockchain - a distributed ledger - technology is used where a reputation is used to penalize WIPs with past double spending habit. The proposed reputation based approach helps to minimize extra delay caused by double spending attempts and Blockchain operations. To optimally predict the QoS requirements of MVNOs for their users, linear regression - a machine learning approach - is used that helps to minimize the latency introduced due to (multiple wrong) negotiations for SLAs. The performance evaluation of the proposed approach is carried out by using numerical results obtained from simulations. Results have shown that the joint Blockchain and machine learning based approach outperforms the other approaches.
AB - Wireless Virtualization (WiVi) is emerging as a new paradigm to provide high speed communications and meet Quality-of-Service (QoS) requirements of users while reducing the deployment cost of wireless infrastructure for future wireless networks. In WiVi, Wireless Infrastructure Providers (WIPs) sublease their RF channels through slicing to Mobile Virtual Network Operators (MVNOs) based on their Service Level Agreements (SLAs) and the MVNOs independently provide wireless services to their end users. This paper investigates the wireless network virtualization by leveraging both Blockchain technology and machine learning to optimally allocate wireless resources. To eliminate double spending (aka over-committing) of WIPs’ wireless resources such as RF channels, Blockchain - a distributed ledger - technology is used where a reputation is used to penalize WIPs with past double spending habit. The proposed reputation based approach helps to minimize extra delay caused by double spending attempts and Blockchain operations. To optimally predict the QoS requirements of MVNOs for their users, linear regression - a machine learning approach - is used that helps to minimize the latency introduced due to (multiple wrong) negotiations for SLAs. The performance evaluation of the proposed approach is carried out by using numerical results obtained from simulations. Results have shown that the joint Blockchain and machine learning based approach outperforms the other approaches.
UR - http://www.scopus.com/inward/record.url?scp=85066612326&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85066612326&partnerID=8YFLogxK
U2 - 10.1145/3324921.3328790
DO - 10.1145/3324921.3328790
M3 - Conference contribution
AN - SCOPUS:85066612326
T3 - WiseML 2019 - Proceedings of the 2019 ACM Workshop on Wireless Security and Machine Learning
SP - 61
EP - 66
BT - WiseML 2019 - Proceedings of the 2019 ACM Workshop on Wireless Security and Machine Learning
T2 - 2019 ACM Workshop on Wireless Security and Machine Learning, WiseML 2019
Y2 - 15 May 2019 through 17 May 2019
ER -