Fast packet classification for V2X services in 5G networks

Vehicle-to-everything (V2X) is one of the key applications of upcoming 5G systems. The 5G systems are highly anticipated to adopt software defined network/network function virtualization technology in their core networks as it provides high flexibility and maintainability. However, if the flow tables in switches, called OpenFlow switches, overflow due to excessive policy rules, the networks suffer from large packet delay and frequent packet drops, and fail to support V2X services. To resolve this issue, we propose a new packet classification algorithm that runs fast even with the huge number of policy rules. There is the minimal concern for the overflow in our scheme since it is a pure software-based approach. Furthermore, simulation results show that our algorithm enjoys both the fast packet classification speed and the short policy update time. Thus, our scheme is very suitable for highly dynamic environments like 5G networks.

[1]  Paolo Zanier,et al.  Mobile Low Latency Services in 5G , 2015, 2015 IEEE 81st Vehicular Technology Conference (VTC Spring).

[2]  Baohua Yang,et al.  Practical Multituple Packet Classification Using Dynamic Discrete Bit Selection , 2014, IEEE Transactions on Computers.

[3]  Jonathan S. Turner,et al.  Scalable packet classification using distributed crossproducing of field labels , 2005, Proceedings IEEE 24th Annual Joint Conference of the IEEE Computer and Communications Societies..

[4]  Li Zhao,et al.  LTE-V: A TD-LTE-Based V2X Solution for Future Vehicular Network , 2016, IEEE Internet of Things Journal.

[5]  Martín Casado,et al.  The Design and Implementation of Open vSwitch , 2015, NSDI.

[6]  Pankaj Gupta,et al.  Packet Classification using Hierarchical Intelligent Cuttings , 1999 .

[7]  Venkatachary Srinivasan,et al.  Packet classification using tuple space search , 1999, SIGCOMM '99.

[8]  Xiaohong Guan,et al.  Taming the Flow Table Overflow in OpenFlow Switch , 2016, SIGCOMM.

[9]  T. N. Vijaykumar,et al.  EffiCuts: optimizing packet classification for memory and throughput , 2010, SIGCOMM '10.

[10]  Nick McKeown,et al.  Packet classification on multiple fields , 1999, SIGCOMM '99.

[11]  Ralf Philipsen,et al.  Public perception of V2X-technology - evaluation of general advantages, disadvantages and reasons for data sharing with connected vehicles , 2016, 2016 IEEE Intelligent Vehicles Symposium (IV).

[12]  Antonius P. J. Engbersen,et al.  Fast and scalable packet classification , 2003, IEEE J. Sel. Areas Commun..

[13]  Hyogon Kim,et al.  Scalable Packet Classification Through Rulebase Partitioning Using the Maximum Entropy Hashing , 2009, IEEE/ACM Transactions on Networking.

[14]  Thomas Y. C. Woo.  A modular approach to packet classification: algorithms and results , 2000, Proceedings IEEE INFOCOM 2000. Conference on Computer Communications. Nineteenth Annual Joint Conference of the IEEE Computer and Communications Societies (Cat. No.00CH37064).

[15]  Saewoong Bahk,et al.  FRFC: Fast Table Building Algorithm for Recursive Flow Classification , 2010, IEEE Communications Letters.

[16]  George Varghese,et al.  Packet classification using multidimensional cutting , 2003, SIGCOMM '03.

[17]  Jonathan S. Turner,et al.  Packet classification using extended TCAMs , 2003, 11th IEEE International Conference on Network Protocols, 2003. Proceedings..

[18]  Sartaj Sahni,et al.  Packet classification consuming small amount of memory , 2005, IEEE/ACM Transactions on Networking.

[19]  Taoka Hidekazu,et al.  Scenarios for 5G mobile and wireless communications: the vision of the METIS project , 2014, IEEE Communications Magazine.

[20]  George Varghese,et al.  Fast and scalable layer four switching , 1998, SIGCOMM '98.