Ad-Hoc Networks: A Requirements Analysis 7th International and - - PowerPoint PPT Presentation

▶

Apr 10, 2023 310 likes •465 views

Technology for Pervasive Computing Global Peer-to-Peer Classification in Mobile Ad-Hoc Networks: A Requirements Analysis 7th International and Interdisciplinary Conference on Modeling and Using Context Dawud Gordon, Markus Scholz, Yong Ding,

SLIDE 1

KIT – University of the State of Baden-Wuerttemberg and National Research Center of the Helmholtz Association

www.kit.edu Technology for Pervasive Computing

Global Peer-to-Peer Classification in Mobile Ad-Hoc Networks: A Requirements Analysis

7th International and Interdisciplinary Conference on Modeling and Using Context

Dawud Gordon, Markus Scholz, Yong Ding, and Michael Beigl Karlsruhe Institute of Technology (KIT), TecO

SLIDE 2

Technology for Pervasive Computing

2 29.09.2011

Overview

Motivating Scenario: Recognition of social group activities using mobile P2P devices

Define how, what and why

What are we trying to recognize? How are we trying to do it? Why is in-network recognition needed?

Requirements Analysis

Survival Recovery Mapping ability

Observing Individuality

Results from requirements and scenario Why it’s necessary

Resources

Bounds for distribution Brute force method (upper) Connectionist method (lower)

Prof. Dr.-Ing. Michael Beigl

SLIDE 3

Technology for Pervasive Computing

3 29.09.2011

GAR using Mobile P2P Devices

Devices collaborate to recognize group activity using embedded sensors

Dawud Gordon

SLIDE 4

Technology for Pervasive Computing

4 29.09.2011

The activity recognition community

My background: human activity recognition based on mobile sensor measurements Focus here: distributed input / processing

Dawud Gordon

SLIDE 5

Technology for Pervasive Computing

5 29.09.2011

What and Why?

What are we trying to recognize?

Behavior of a group of social individuals

Why and when on P2P devices?

Sporadic access to infrastructure Expensive access (energy, bandwidth, etc.) No access (Autonomous)

Dawud Gordon

James Cridland

SLIDE 6

Technology for Pervasive Computing

6 29.09.2011

Requirement 1: Survival

Recognition must survive nodes leaving without loss of recognition capabilities

Dawud Gordon

SLIDE 7

Technology for Pervasive Computing

7 29.09.2011

Requirement 2: Recovery

Recognition must not lose ability as individuals come and go

Dawud Gordon

SLIDE 8

Technology for Pervasive Computing

8 29.09.2011

Requirement 3: Mapping Ability

Which “social” context is to be recognized is not defined Approach must be able to model mapping from sensor values to contexts

Dawud Gordon

SLIDE 9

Technology for Pervasive Computing

9 29.09.2011

Observing Individuality

Assuming nodes are heterogeneous leads to problems! Constant subject “throughput” means data from new subjects are constantly introduced to system Eventually all original (training) subjects will be replaced

Dawud Gordon

SLIDE 10

Technology for Pervasive Computing

10 29.09.2011

Related Work

Parallel Computing:

Global access to data Or, central merging/computation unit

Collaborative Methods:

Distributed voting Counts vote, not voter

Organic Computing

Multi-agent stigmergy approaches Produce a distributed stigmergic map

Dawud Gordon

SLIDE 11

Technology for Pervasive Computing

11 29.09.2011

Possible solutions

Several different algorithmic approaches Brute force

redundant classifier Complete dissemination of all measurement data

Connectionist approach:

distribution of processing units across network Each node input, output and hidden processor

Self Organizing Maps: distribution of data representation across network

Dawud Gordon

SLIDE 12

Technology for Pervasive Computing

12 29.09.2011

Resource Consumption Analysis

Assumption: distributed algorithm meeting requirements N: number of nodes in the network P: total processing load (per classification phase) M: total memory required by algorithm

Dawud Gordon

Algorithm Messages Passed Processing/ Node Memory/ Node Brute Force (Worst Case) N(N-1) P M + SG ANN 2N P/N M/N + SL Best Case N P/N M/N + SL

SLIDE 13

Technology for Pervasive Computing

13 29.09.2011

Summary

In-network P2P classification is necessary:

No, Restricted, or Intermittent access

For functionality there are 3 requirements:

Survival, Recovery and Mapping

An upper and lower bound for resource consumption and distribution derived

Brute force approach Distributed reasoning approach

The importance of incorporating role elaborated on

Dawud Gordon

SLIDE 14

Technology for Pervasive Computing

14 29.09.2011

That’s All

Thank You!

Questions?

Dawud Gordon