A Genetic Algorithm for Packing CAN FD Frame with Real-Time Constraints

Shan DING  Gang ZENG  Ryo KURACHI  Ruifeng HUANG  

Publication
IEICE TRANSACTIONS on Information and Systems   Vol.E100-D   No.10   pp.2505-2514
Publication Date: 2017/10/01
Online ISSN: 1745-1361
Type of Manuscript: PAPER
Category: Software System
Keyword: 
in-vehicle network,  Controller Area Network,  frame packing,  network bandwidth,  schedulability,  

Full Text: PDF(1.3MB)
>>Buy this Article


Summary: 
As a next-generation CAN (Controller Area Network), CAN FD (CAN with flexible data rate) has attracted much attention recently. However, how to use the improved bus bandwidth efficiently in CAN FD is still an issue. Contrasting with existing methods using greedy approximate algorithms, this paper proposes a genetic algorithm for CAN FD frame packing. It tries to minimize the bandwidth utilization by considering the different periods of signals when packing them in the same frame. Moreover, it also checks the schedulability of packed frames to guarantee the real-time constraints of each frame and proposed a merging algorithm to improve the schedulability for signal set with high bus load. Experimental results validate that the proposed algorithm can achieve significantly less bandwidth utilization and improved schedulability than existing methods for a given set of signals.