Hierarchical Addressing for Long Thin ZigBee Networks

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing ZigBee network routing methods, such as table routing and tree routing, are not suitable for long thin networks due to high transmission costs and limited address space, leading to network jams and inefficiencies.

Innovation Solution

A network address assignment method that assigns each node a unique address comprising a cluster ID and a node ID, allowing for efficient packet routing within and between clusters, with cluster IDs arranged in a specific sequence and node IDs assigned using conventional algorithms, enabling improved routing performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If table routing method is used to determine next node for packet transmission, then routing flexibility is improved, but transmission cost increases due to broadcasting for routing information

Engineering Contradiction:
Improverouting flexibilityVSAvoidtransmission cost
Core Design Contradiction:
Adaptability or versatilityVSLoss of energy

Solution Approach 1:

The patent pre-assigns hierarchical addresses (cluster ID + node ID) to all nodes during network initialization, so that routing decisions can be made locally without broadcasting. This preliminary address assignment eliminates the need for runtime routing information discovery, reducing transmission cost while maintaining routing flexibility through the hierarchical structure.

Inventive Principle:
Principle #10Preliminary action

2Ease of operation

If tree routing method is used with 16-bit network addresses, then routing direction can be determined by address comparison, but network depth is limited to maximum 15 levels

Engineering Contradiction:
Improverouting direction determinationVSAvoidnetwork depth
Core Design Contradiction:
Ease of operationVSLength of moving object

Solution Approach 1:

The patent segments the 16-bit address into two parts: a cluster ID (identifying the cluster) and a node ID (identifying the node within the cluster). This segmentation allows the network to support arbitrary depths by organizing nodes into clusters, where each cluster can have multiple levels of hierarchy. The cluster ID enables routing to the correct cluster, while the node ID handles routing within the cluster, effectively removing the 15-level depth limitation.

Inventive Principle:
Principle #1Segmentation

3Ease of manufacture

If conventional 16-bit network addresses are used, then address assignment is simple, but only 2^16 addresses are available which is insufficient for large networks

Engineering Contradiction:
Improveaddress assignment simplicityVSAvoidnumber of available addresses
Core Design Contradiction:
Ease of manufactureVSQuantity of substance

Solution Approach 1:

The patent segments the address space into cluster IDs and node IDs, where the cluster ID identifies the cluster and the node ID identifies the specific node within the cluster. This segmentation effectively multiplies the address space: instead of having only 2^16 unique addresses, the network can now support multiple clusters, each with multiple nodes, significantly increasing the total number of addressable nodes while keeping the individual address structure simple and manageable.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentUS7869384B2Network address assignment method and routing method for a long thin zigbee network
Publication Date: 2011.01.11 INSTITUTE FOR INFORMATION INDUSTRY
  • US7869384B2 patent drawing
  • US7869384B2 patent drawing
  • US7869384B2 patent drawing

AI summary

A network address assignment method and a routing method for a long thin ZigBee network are provided. The routing method includes the following steps. A network address is assigned to each node of the ZigBee network, wherein each network address includes a cluster ID and a node ID. The cluster ID is used for identifying a plurality of clusters of the ZigBee network. The Node ID is used for identifying a plurality of nodes of each cluster. When a packet is transmitted, check whether the current node holding the packet and the destination node are in the same cluster. If they are in the same cluster, the packet is routed within the cluster according to the node ID of the destination node and a predetermined algorithm. If they are not in the same cluster, the packet is routed among the clusters according to the cluster ID of the destination node.