An Efficient and Secure Wireless Controller Area Network for Autonomous Vehicle

Abstract

<div class="section abstract"><div class="htmlview paragraph">Controller area network (CAN) buses, the most common intravehicle network (IVN) standard, have been used for over 30 years despite their simple architecture for connecting electronic control units (ECUs). Weight, maintenance costs, mobility promotion, and wired connection complexity increase with ECU count, especially for autonomous vehicles. This paper aims to enhance wired CAN with wireless features for autonomous vehicles (AVs). The proposed solutions include modifying the traditional ECU architecture to become wireless, implementing a hidden communication environment using a unique complementary code keying (CCK) modulation equation and presenting a strategy for dealing with jamming signals using two channels. The proposed wireless CAN (WCAN) is validated using OPNET analysis for performance and reliability. The results show that the bit error rate (BER) and packet loss of the receiver ECU are stable between different CCK modifications, indicating the robustness of the basic features of CCK modification. However, intercepting and decoding the signal by the eavesdropping ECU is challenging, with packet loss ranging from 63% to 100% across different CCK states. Anti-jamming results show that when packet loss reaches 2%, the passive channel is automatically activated, ensuring secure data transmission. The IEEE 802.11b network accommodated internal and external ECUs and maintained delay deadline requirements, but the LiDAR ECU requires high bandwidth to accommodate 13,500 point cloud data every 200 ms. The IEEE 802.11a standard was chosen, but it did not meet deadline requirements for delay, leading to the implementation of 100BaseT Fast Ethernet with a data rate of 100 Mbps and throughput of 500,000 bytes per second.</div></div>