Self-Routed Packet Network for Wire-Speed Path Selection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current data networks, particularly those using OSI Layer 3 packet networks, face issues with slow routing processes, congestion, insecurity, unreliable paths, and complexity in managing network flows, which are exacerbated in dynamic environments like ad-hoc or mobile networks, and are not suitable for neural networks due to slow and imprecise optimization.

Innovation Solution

A Self-Routed Packet (SRP) network architecture that uses a distributed hardware-based design to enable fast, secure, and reliable connection establishment and management of Label Switched Paths (LSPs) through hunting packets, token-based access control, and pipelined checking of bandwidth and QoS constraints, allowing real-time optimization and self-correction.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If traditional IP routing is used in Layer 3 packet networks, then network flexibility and simplicity are maintained, but routing speed is slow and path selection is imprecise

Engineering Contradiction:

Improverouting speedVSAvoidrouting process complexity

Core Design Contradiction:

SpeedVSDevice complexity

Solution Approach 1:

The patent replaces traditional software-based IP routing mechanisms with a hardware-based neural network system. The neural network uses specialized hardware circuits to perform path selection and routing decisions at wire speed, substituting the mechanical/software routing process with a parallel hardware neural network that can simultaneously evaluate multiple paths and make optimal selections without the delays inherent in conventional routing protocols.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent changes the fundamental parameters of routing by using neural network algorithms instead of traditional routing protocols. The neural network processes routing decisions based on dynamic parameters such as traffic patterns, network conditions, and optimization criteria, allowing for adaptive and intelligent path selection that responds to real-time network changes rather than relying on static routing tables and protocols.

Inventive Principle:

Principle #35Parameter changes

2Reliability

If connectionless packet routing is used, then network simplicity is maintained, but congestion handling and time-sensitive traffic performance deteriorate

Engineering Contradiction:

Improvetime-sensitive traffic reliabilityVSAvoidnetwork management complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The neural network system performs self-optimization and self-management of network paths. It automatically learns from network conditions, adjusts routing decisions, and handles congestion dynamically without requiring external intervention or complex manual configuration. The system serves itself by continuously optimizing its own performance based on observed traffic patterns and network state.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent implements feedback mechanisms where the neural network continuously monitors network conditions, traffic flow, and path performance. This feedback information is fed back into the neural network to refine future routing decisions, enabling adaptive response to congestion and optimization of time-sensitive traffic. The feedback loop allows the system to learn from past performance and make increasingly optimal routing choices.

Inventive Principle:

Principle #23Feedback

3Reliability

If flow-based routing with specialized hardware is implemented, then quality of service improves, but system complexity and cost increase

Engineering Contradiction:

Improvequality of serviceVSAvoidspecialized hardware complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The neural network hardware is designed to perform multiple functions including path selection, congestion avoidance, quality of service management, and adaptive routing. Rather than requiring separate specialized hardware components for each function, the neural network architecture integrates these capabilities into a unified system that can handle diverse network management tasks through its learning and optimization algorithms.

Inventive Principle:

Principle #6Universality (Multi-functionality)

4Measurement precision

If traditional routing protocols are used in dynamic environments, then network adaptability is maintained, but routing accuracy and optimization speed deteriorate

Engineering Contradiction:

Improvepath optimization precisionVSAvoidrouting decision time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The neural network performs preliminary learning and analysis of network patterns, traffic characteristics, and optimal paths during idle periods or low-traffic times. By pre-computing and preparing routing decisions based on historical data and predicted patterns, the system can make rapid routing decisions when traffic demands arise, reducing real-time decision delays while maintaining high optimization precision.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS9542642B2Packet data neural network system and method

Publication Date: 2017.01.10 WOOD SAMUEL F

AI summary

This application discloses a neural network that also functions as a connection oriented packet data network using an MPLS-type label switching technology. The neural network uses its intelligence to build and manage label switched paths (LSPs) to transport user packets and solve complex mathematical problems. However, the methods taught here can be applied to other data networks including ad-hoc, mobile, and traditional packet networks, cell or frame-switched networks, time-slot networks and the like.