The Fourth International Conference on Mobile Ubiquitous Computing, Systems, Services and Technologies - 2010
Improved Spatial and Temporal Mobility Metrics for Mobile Ad Hoc Networks Elmano Ramalho Cavalcanti*; Marco Aurélio Spohn* * Systems and Computing Department Federal University of Campina Grande, Brazil Email: {elmano,maspohn}@dsc.ufcg.edu.br 1
Outline • • • • • •
Introduction and Motivation Background New Mobility Metrics Simulation Results and Observation Conclusions and Future Work
2
Introduction and Motivation • Mobility models describes the movement patterns of mobile nodes, having an impact on several areas: – Protocol performance, topology and network connectivity, data replication, security…
• Mobility metrics are used to measure the features of mobility models.
3
Introduction and Motivation • The IMPORTANT framework (Bai et al. 2003) presented two spatial and temporal mobility metrics (DSD, DTD). • Cited by > 600 papers (google scholar)
• Several works were based on these metrics for a wide range of purposes…
However… • We show that these metrics have crucial drawbacks, and propose improved mobility metrics. 4
Outline • • • • • •
Introduction and Motivation Background New Mobility Metrics Simulation Results and Observation Conclusions and Future Work
5
Mobility Models in MANETs Mobility Models
Random Models
Temporal Dependency Models
Spatial Dependency Models
Geographic Restriction Models
e.g. Random Waypoint
e.g. GaussMarkov
e.g. Reference Point Group
e.g. Manhattan Model
Categories of mobility models in MANETs [Bai et al. 03]
6
Spatial and Temporal Metrics IMPORTANT framework [Bai et al. 03]
DSD - Degree of Spatial Dependence
DTD - Degree of Temporal Dependence
7
Limitations on Previous Metrics a. DSD: i. does not consider spatial dependence between nodes when they are not moving
Spatial dependence degree decay over pause time…
a. DTD i. does not properly identify temporal models from others. ii. is biased by node speed iii. has low correlation with the memory parameter in temporal models (e.g., Gauss-Markov, SMS)
8
Outline • • • • • •
Introduction and Motivation Background Contributions Simulation Results and Observation Conclusions and Future Work
9
New Spatial Mobility Metrics Improved Degree of Spatial Dependence
Degree of Node Proximity 10
New Temporal Mobility Metric Improved Degree of Temporal Dependence
11
Outline • • • • • •
Introduction and Motivation Background New Mobility Metrics Simulation Results and Observation Conclusions and Future Work
12
Simulation Feature Randomness Group-based Temporal Gripd-based
RWP RPGM GM MAN X
X
X
X
X X X
T: 900 s Pause time: 0, 100, 200, ... , 900 s N: 100 nodes (RPGM) nodes per group: 10, 25, 50 R: 100, 150, 200 m (GM): memory level: .0, .2, ... ,.8, .99 X,Y: 1000 m Speed: 10, 20, 30 m/s 13
Outline • • • • • •
Introduction and Motivation Background New Mobility Metrics Simulation Results and Observation Conclusions and Future Work
14
Results and Observations
DESCRIPTIVE STATISTICS FOR THE MOBILITY METRICS
15
Results and Observations Correlation Matrix (Mobility Metrics x Input Parameters)
16
Results and Observations Temporal Metrics – GM scenario I
(S =10 m/s, R = 150 m) 17
Results and Observations Temporal Metrics – GM scenario II
(S =20 m/s, R = 150 m) 18
Results and Observations Temporal Metrics – GM scenario III
(S =30 m/s, R = 150 m) 19
Results and Observations Correlation Matrix (Mobility Metrics x Input Parameters)
20
Results and Observations Spatial Metrics – RPGM scenario I
10 groups of 10 nodes 21
Results and Observations Spatial Metrics – RPGM scenario II
4 groups of 25 nodes 22
Results and Observations Spatial Metrics – RPGM scenario III
2 groups of 50 nodes 23
Results and Observations Degree of Node Proximity (DNP) RPGM
RWP MAN
24
Results and Observations
Histogram for Degree of Node Proximity 25
Outline • • • • • •
Introduction and motivation Background New Mobility Metrics Simulation Results and Observation Conclusions and Future Work
26
Conclusions and Future Work • Introduction of the concept of pause state movement dependence • New mobility metrics (IDSD, IDTD, DNP) that overcome the drawbacks found on current spatial and temporal metrics. • Future: we’re investigating the use of the proposed mobility metrics for designing mobility-aware adaptive routing protocols for MANETs. 27
Thanks!
28
Random Waypoint
29
RPGM
30
Manhattan
31
