IC Interconnection Network Using Redundant Check Codes for Fault Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Integrated circuits face challenges in providing reliable data transfer between nodes due to hardware faults, such as transient and permanent faults, which can lead to data corruption or misrouting, especially in applications requiring functional safety like automotive systems, where existing redundancy methods are costly in terms of circuit area and power consumption.
Innovation Solution
An interconnection network with a primary network for data transfer and a redundant network that transmits a check code with fewer bits than the primary payload, using error checking circuitry at the destination to verify payload integrity, reducing circuit area and power consumption by minimizing redundant data transmission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If redundant data transmission is used to detect hardware faults, then reliability is improved, but circuit area and power consumption increase
Solution Approach 1:
The patent extracts only the essential error detection information (check codes) from the full payload data and transmits separately through the redundant network. Instead of transmitting complete redundant payloads, the system transmits compact check codes (e.g., CRC codes) that can detect errors without requiring full data duplication, thereby reducing circuit area while maintaining fault detection capability
Solution Approach 2:
The patent creates a simplified copy of the payload in the form of check codes rather than complete data copies. The check codes are mathematical transformations of the original data that enable error detection with minimal overhead, achieving redundancy benefits without the area cost of full data duplication
2Reliability
If redundant data transmission is used to detect hardware faults, then reliability is improved, but power consumption increases
Solution Approach 1:
The patent extracts only the essential error detection information (check codes) from the full payload data and transmits separately through the redundant network. This extraction approach minimizes the volume of data requiring redundant transmission, thereby reducing power consumption while maintaining fault detection capability
Solution Approach 2:
The patent creates a simplified copy of the payload in the form of check codes rather than complete data copies. The check codes require significantly less power to transmit and process compared to full payload redundancy, achieving energy efficiency without sacrificing reliability
3Measurement precision
If full payload redundancy is transmitted, then fault detection accuracy is improved, but data transmission efficiency decreases
Solution Approach 1:
The patent extracts compact check codes from the full payload that retain sufficient information for accurate error detection. These check codes are transmitted separately and require minimal bandwidth compared to full payload redundancy, thereby maintaining detection accuracy while improving transmission efficiency
Solution Approach 2:
The patent transforms the payload data into check codes through mathematical operations (e.g., cyclic redundancy check algorithms). This parameter transformation reduces the data volume from full payload size to a small fixed-size code, enabling efficient transmission while preserving error detection capability
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
An interconnection network for providing data transfer between a plurality of nodes of an integrated circuit comprises a number of endpoints for exchanging data with respective nodes of the integrated circuit, a primary network (102) to route a primary payload (104) from a source endpoint (1) to a destination endpoint (2); and a redundant network (100) to route, to the destination endpoint, a redundant payload (110) comprising a first check code (112) calculated based on at least a portion of the primary payload, the first check code having fewer bits than said at least a portion of the primary payload. The destination endpoint comprises error checking circuitry (48) to perform an error checking operation to calculate a second check code (114) based on the primary payload received via the primary network, and verify integrity of the primary payload based on a comparison of the second check code (114) with the first check code (112) received via the redundant network.