Automotive Network Gateway Co-Processors for Low-Latency Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing software-based network gatewaying approaches in automotive communication networks fail to guarantee real-time performance, particularly in terms of latency, jitter, bandwidth, and throughput, especially for next-generation autonomous-driving vehicles due to the coexistence of heterogeneous in-vehicle network technologies and protocols.
Innovation Solution
A hardware/software co-designed network device with a central processing unit and multiple co-processors for efficient data switching, routing, and gatewaying, including frame normalization, queuing, filtering, policing, and traffic shaping, capable of handling heterogeneous network technologies like LIN, CAN, and Ethernet, with embedded algorithms for reliability and cyber security.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If software-based approaches are used for network gatewaying, then device complexity is reduced and ease of manufacture is improved, but real-time performance (latency, jitter, bandwidth, throughput) deteriorates
Solution Approach 1:
The gateway device is segmented into multiple specialized co-processors, each handling specific network protocols (CAN, LIN, FlexRay, Ethernet). This segmentation allows parallel processing of different protocol types simultaneously, reducing overall latency while maintaining software-based flexibility for configuration and updates.
Solution Approach 2:
A central processing unit acts as an intermediary between the specialized co-processors and the external network interfaces. This mediator coordinates data flow between heterogeneous protocols, enabling efficient gatewaying functions while maintaining real-time performance through optimized intermediate processing.
2Adaptability or versatility
If software-based approaches are used for network gatewaying, then adaptability to different protocols is improved, but processing speed and real-time performance worsen
Solution Approach 1:
The system is divided into protocol-specific co-processors that provide hardware-accelerated processing for each protocol type. This segmentation enables high-speed processing while maintaining adaptability through software-configurable parameters and centralized control by the main processor.
Solution Approach 2:
Each co-processor is designed with universal interfaces and configurable parameters that allow it to handle multiple protocol variations. The centralized processing unit provides unified control and configuration, enabling the system to adapt to different network standards while maintaining high processing speeds through hardware acceleration.
3Adaptability or versatility
If multiple heterogeneous network technologies are supported, then versatility and adaptability are improved, but device complexity and processing overhead increase
Solution Approach 1:
The gateway device is segmented into dedicated co-processors for each network technology (CAN, LIN, FlexRay, Ethernet). This segmentation isolates protocol-specific complexity into separate modules, making the overall system more manageable while maintaining versatility through standardized interfaces between modules.
Solution Approach 2:
Multiple protocol handling capabilities are merged into a single integrated gateway device through the coordination of multiple co-processors under centralized control. This combining approach provides versatility across heterogeneous networks while managing complexity through modular architecture and unified configuration management.
Data Source
AI summary
A communication network comprises a central processing unit, data ingress ports and data egress ports configured to exchange data with a further network device of the communication network, and a plurality of co-processors that comprises frame normalization co-processors, ingress queuing co-processors, filtering and policing co-processors, intermediate queuing co-processors, a gatewaying co-processor, egress queuing co-processors, and a traffic shaping co-processor. The central processing unit configures and controls the data ingress ports, the data egress ports, and the plurality of co-processors to implement data processing paths in parallel or in a pipeline between the ingress ports and the egress ports.