Network-on-Chip TMR Messaging for Branched Routing Reliability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network-on-chip (NoC) technologies face limitations in functional scope, particularly in terms of message transmission efficiency, redundancy management, and throughput optimization.
Innovation Solution
Implementing multi-source communication methods with combined and validated message transmission, branched routing, and virtual bus scheduling to enhance NoC functionality, including triple-modular redundancy (TMR) for error detection and virtual bus systems for efficient data transfer.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple source nodes transmit identical messages to a destination node through different routes, then message reliability is improved through redundancy checks, but device complexity increases due to the need for combining and validating messages from multiple sources
Solution Approach 1:
The message transmission process is segmented into distinct phases: transmission phase where multiple source nodes send identical messages through different routes, combination phase where intermediate nodes merge messages using bit-by-bit operations, and validation phase where destination nodes verify messages using check algorithms. This segmentation allows each phase to be optimized independently, managing complexity while maintaining reliability.
Solution Approach 2:
Intermediate nodes act as mediators that receive messages from multiple source nodes, combine them using bit-by-bit OR or AND operations, and forward the combined message to the destination. This intermediary function simplifies the destination node's task by pre-processing the redundant messages before they arrive, reducing the validation complexity at the final destination.
2Productivity
If messages are transmitted from multiple source nodes along different routes to reach the receive node together at a predetermined time, then throughput is improved by allowing branched routing, but loss of time increases due to the need for precise timing coordination
Solution Approach 1:
The system performs preliminary routing calculations and timing adjustments before message transmission begins. Static routing tables are pre-configured with timing information, and source nodes are assigned specific transmission times in advance. This preliminary preparation enables messages from multiple sources to arrive simultaneously at the receive node without requiring complex real-time coordination, thus improving throughput while minimizing time loss.
Solution Approach 2:
The NoC operates with periodic time slots and scheduled transmission windows. Messages are transmitted in coordinated bursts during predetermined time intervals, allowing the system to achieve high throughput through regular, predictable patterns. This periodic operation simplifies timing coordination compared to continuous random access, as all nodes synchronize to the same periodic schedule.
3Productivity
If the receive node combines multiple identical messages into a single combined message, then data transmission efficiency is optimized, but manufacturing precision requirements increase for ensuring identical message pairs from different source nodes
Solution Approach 1:
The system changes the parameter of message identity verification from requiring absolute bit-perfect duplication to allowing verification through check algorithms. Source nodes transmit messages with attached checksums or parity bits, and the receive node validates combined messages using these check data rather than requiring exact manufacturing precision of the message content itself. This parameter change from content-identity to structural-validity verification maintains transmission efficiency while reducing manufacturing precision requirements.
Solution Approach 2:
Instead of requiring multiple source nodes to manufacture identical message copies with high precision, the system uses one source node to generate the original message with check data, and other source nodes to transmit copies. The check data serves as a template for verification, allowing the system to tolerate minor manufacturing variations while ensuring message integrity through the validation process.
Data Source
AI summary
A method for multi-source communication for triple-modular redundancy (TMR), a method for branched communication, and a method for virtual buses are disclosed. A method includes a) transmitting by at least two different source nodes in each case at least two identical messages which contain at least flow control data, payload data and check data to at least one predetermined receive node where the messages reach the receive node together at a predetermined time, b) combining by the receive node the messages received by the receive node into a combined message containing flow control data, payload data and check data, or comparing by the receive node of messages received by the receive node, and c) further processing of the combined message by the receive node or further processing of at least one of the messages received by the receive node based on comparison from step b).