TY - GEN
T1 - Design and Optimization of Switched FlexRay Networks
AU - Han, Gang
AU - Lu, Jia
AU - Li, Jinhua
AU - Hu, Songqi
AU - Zhang, Jinying
N1 - Publisher Copyright:
© 2017 IEEE.
PY - 2018/8/24
Y1 - 2018/8/24
N2 - FlexRay is a promising communication protocol for in-vehicle systems. With the growth of functionalities delivered by in-vehicle systems, FlexRay is accommodating more and more signals (or messages) which have exceeded the capacity of a single FlexRay bus. Switched FlexRay networks have been studied to solve the problem. However, previous research mostly focused on reducing the number of overall consumed time slots, and hardly paid attention to load balance of each branch (i.e., switch-isolated FlexRay subnetwork). This paper proposes SSFN (Scalable Switched FlexRay Network). SSFN is based on a message-buffered Crossbar structure, and thus allows the inter-branch messages transmitting at different slots of sender and receiver branches. A MILP formulation is proposed to look for the optimal solution in terms of load balance on each FlexRay branch, number of consumed slots and size of switch buffer. We conduct experiments on two case studies to evaluate the proposed formulation and the result shows that a solution with far improved load balance and approximate time slot consumption (compared to existing research) can be found in a few seconds.
AB - FlexRay is a promising communication protocol for in-vehicle systems. With the growth of functionalities delivered by in-vehicle systems, FlexRay is accommodating more and more signals (or messages) which have exceeded the capacity of a single FlexRay bus. Switched FlexRay networks have been studied to solve the problem. However, previous research mostly focused on reducing the number of overall consumed time slots, and hardly paid attention to load balance of each branch (i.e., switch-isolated FlexRay subnetwork). This paper proposes SSFN (Scalable Switched FlexRay Network). SSFN is based on a message-buffered Crossbar structure, and thus allows the inter-branch messages transmitting at different slots of sender and receiver branches. A MILP formulation is proposed to look for the optimal solution in terms of load balance on each FlexRay branch, number of consumed slots and size of switch buffer. We conduct experiments on two case studies to evaluate the proposed formulation and the result shows that a solution with far improved load balance and approximate time slot consumption (compared to existing research) can be found in a few seconds.
UR - http://www.scopus.com/inward/record.url?scp=85053843000&partnerID=8YFLogxK
U2 - 10.1109/CYBER.2017.8446099
DO - 10.1109/CYBER.2017.8446099
M3 - Conference contribution
AN - SCOPUS:85053843000
SN - 9781538604892
T3 - 2017 IEEE 7th Annual International Conference on CYBER Technology in Automation, Control, and Intelligent Systems, CYBER 2017
SP - 769
EP - 774
BT - 2017 IEEE 7th Annual International Conference on CYBER Technology in Automation, Control, and Intelligent Systems, CYBER 2017
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 7th IEEE Annual International Conference on CYBER Technology in Automation, Control, and Intelligent Systems, CYBER 2017
Y2 - 31 July 2017 through 4 August 2017
ER -