Browse Prior Art Database

ADAPTIVE OBJECTIVE FUNCTION FOR LOW-POWER AND LOSSY NETWORKS

IP.com Disclosure Number: IPCOM000252452D
Publication Date: 2018-Jan-12
Document File: 3 page(s) / 138K

Publishing Venue

The IP.com Prior Art Database

Related People

Li Zhao: AUTHOR [+4]

Abstract

An adaptive object function (OF) mechanism is presented herein. This mechanism may address the complex deployment issue when a user sets up a network. Coefficients of the OF may be derived from environmental information that may be captured by different sensors, such as those that are external to the Routing Protocol for Low-Power and Lossy Networks (LLNs) node. The computation of the optimized coefficients for a particular node may be offloaded to learning machines. As opposed to identical static Routing Protocol for LLNs (RPL) OFs for all nodes, each RPL node benefits from a tailor-made variation of a generic OF that adapts its operation to the particular Radio Frequency (RF) and physical environment of the node.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 53% of the total text.

Copyright 2018 Cisco Systems, Inc. 1

ADAPTIVE OBJECTIVE FUNCTION FOR LOW-POWER AND LOSSY NETWORKS

AUTHORS: Li Zhao

Pascal Thubert Chuanwei Li Huimin She Charlie Chen

CISCO SYSTEMS, INC.

ABSTRACT

An adaptive object function (OF) mechanism is presented herein. This mechanism

may address the complex deployment issue when a user sets up a network. Coefficients of

the OF may be derived from environmental information that may be captured by different

sensors, such as those that are external to the Routing Protocol for Low-Power and Lossy

Networks (LLNs) node. The computation of the optimized coefficients for a particular node

may be offloaded to learning machines. As opposed to identical static Routing Protocol for

LLNs (RPL) OFs for all nodes, each RPL node benefits from a tailor-made variation of a

generic OF that adapts its operation to the particular Radio Frequency (RF) and physical

environment of the node.

DETAILED DESCRIPTION

An OF defines how RPL nodes select and optimize routes within an RPL instance.

An OF defines how nodes translate one or more metrics/constraints into a value called rank,

which approximates the node’s distance from a Destination-Oriented Directed Acyclic

Graph (DODAG) root. The OF is identified by an Objective Code Point (OCP) within the

DODAG Information Object (DIO) configuration option, which is decided by the root node.

The OF is a single function with fixed parameters in current implementations.

In the context of network runtime of a Distributed Automation (DA) network,

higher throughput and lower packet loss rate is the goal. However, the situation for each

node is different in deployment. Some nodes have high background noise, some nodes are

behind a high building, some nodes have a weak Received Signal Strength Indication

(RSSI) but a high throughput, some nodes need to communicate with external nodes, some

Copyright 2018 Cisco Systems, Inc. 2

nodes only communicate with neighboring nodes, etc. A fixed OF cannot effectively

handle such variations.

As a general concept, an RPL OF is a piece of logic that operates on metrics to

select a set of parent nodes and a rank in a DODAG for each node. A typical OF is very

simple and derives from one metric, such as hop count or expected transmission count

(ETX). Other possible metrics for an OF include RSSI, the parent node’s packet loss rate,

link latency, link throughput, hop count, Global Positioning System (GPS) location, etc.

Even when one metric is chosen for the OF, some parameters may require configuring. For

example, in a high background noise environment, the square of ETX is more suitable than

the classic ETX to calculate the rank, because the square of ETX takes the packet loss ratio

into higher account. But in a low background noise environment, linear packet loss rate is

suitable to calculate the rank of ETX. Depending on the level of noise, the most efficient

metric would be a combination of ETX and ETX squa...