Distributed M2M Clustering via Capability-Based Head Selection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Machine-to-machine (M2M) networks face challenges in minimizing collisions among heterogeneous devices with varying resources, leading to packet loss, reduced reliability, and increased energy consumption due to existing clustering methods that are not suitable for resource-constrained devices.

Innovation Solution

A distributed clustering method that classifies nodes based on their capabilities, allowing higher-capability nodes to be selected as cluster heads, and forms clusters as cliques to minimize collisions within and between clusters through coordinated time slot reservations and data collection management.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If existing cluster methods (LEACH, periodic discovery packets) are used in M2M networks, then devices can be organized into clusters, but collision reduction is not considered and heavy overhead is generated

Engineering Contradiction:

Improvepacket delivery reliabilityVSAvoidcommunication overhead

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent changes the parameter of cluster head selection from random or periodic discovery-based methods to capability-based selection. Nodes broadcast their capabilities (transmission power, processing ability, energy levels) and are selected as cluster heads based on these parameters. This reduces collisions by ensuring capable nodes handle cluster management, while reducing overhead by eliminating periodic discovery packets in favor of capability-based one-time selection.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

Nodes autonomously determine their own capability parameters and broadcast them to the network. The selection process is self-organizing without requiring centralized control or periodic coordination messages. Each node independently evaluates its own suitability for cluster head role based on its resources, reducing the need for control overhead.

Inventive Principle:

Principle #25Self-service

2Reliability

If deterministic cluster head selection (LEACH) is used, then cluster organization is achieved, but resource constraints of M2M devices are not considered

Engineering Contradiction:

Improvecluster stabilityVSAvoidheterogeneous device compatibility

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent applies local quality by allowing different types of devices (sensors, gateways, smartphones) to have different capability parameters. Each node type contributes differently to cluster formation based on its local resources. Gateway nodes with high power and processing capability are selected for different roles than resource-constrained sensor nodes, enabling heterogeneous device compatibility while maintaining cluster stability.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The selection criterion changes from fixed deterministic rules to dynamic capability-based parameters. Nodes with higher transmission power, better processing ability, and sufficient energy are preferentially selected as cluster heads. This adaptability allows the network to accommodate various device types while maintaining stable cluster organization.

Inventive Principle:

Principle #35Parameter changes

3Use of energy by moving object

If time-division multiple access (TDMA) is used within clusters, then devices can be in standby mode, but collision minimization among clusters is not addressed

Engineering Contradiction:

Improveenergy consumptionVSAvoidinter-cluster collision reduction

Core Design Contradiction:

Use of energy by moving objectVSReliability

Solution Approach 1:

The patent extends TDMA from intra-cluster time scheduling to inter-cluster spatial-temporal coordination. Cluster heads negotiate time slots not only within their own cluster but also with neighboring cluster heads. This multi-dimensional approach (combining time slots with spatial cluster boundaries) minimizes inter-cluster collisions while allowing member devices to remain in standby mode during non-transmission periods.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

Cluster heads act as intermediaries that coordinate time slot allocation between their own cluster members and neighboring clusters. Instead of direct member-to-member coordination which would generate overhead and collisions, cluster heads negotiate and manage the time division, reducing inter-cluster collisions while maintaining energy efficiency for member devices.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS9225606B2Method for clustering devices in machine-to-machine networks to minimize collisions

Publication Date: 2015.12.29 MITSUBISHI ELECTRIC RESEARCH LABORATORIES INC

AI summary

Nodes in a network are clustered by first determining, in each node, a cluster head capability (CHC). The CHC is broadcasting, directly or indirectly, until all nodes have received the CHCs. Each node nominates, one or more candidate cluster heads based on the CHCs, and then, in each node, at least one cluster head is selected from the candidate cluster head nodes based on maximal CHCs.