Mobility Models in Mobile

Ad Hoc Networks

Chun-Hung Chen
2005 Mar 8th
Dept. of Computer Science and Information Engineering
National Taipei University of Technology
What is Mobile Network? Why is it called
Ad Hoc?
Mobile Ad Hoc Networks (MANET) is a
network consisting of mobile nodes with
wireless connection (Mobile) and without
central control (Ad Hoc)
Self-Organized Network
Each mobile node communicates with other
nodes within its transmitting range
Transmission may be interrupted by nodal
movement or signal interference
It seems that something is watching

Sensor Network is recently a hot topic

Specified Application in MANET
Nodes are equipped some observation devices
(GPS, Temperature, Intruder Detection and so on)
Nodes may move spontaneously or passively
There are more restrictions in Sensor Networks
Power
Transmission Range
Limited Computation Power
Moving Speed (if does)
Realistic Mobility Model is better than
Proposed Mobility Models
Although MANET and Sensor Network are being researched for
a while, there do not exist many realistic implementation
To implement a real MANET or Sensor Network environment for
emulation is difficult
Signal Interference

Boundary

We will need to evaluate MANET by simulation

To have useful simulation results is to control factors in
simulation
Traffic Pattern

Bandwidth

Power Consumption

Signal Power (Transmission Range)

Mobility Pattern

And so on
How Many Mobility Model are inspected?

Entity Mobility Models

Random Walk Mobility Model
Random Waypoint Mobility Model
Random Direction Mobility Model
A Boundless Simulation Area Mobility Model
Gauss-Markov Mobility Model
A Probability Version of the Random Walk
Mobility Model
City Section Mobility Model
 Group Mobility Models
Exponential Correlated Random Mobility Model
Column Mobility Model
Nomadic Community Mobility Model
Pursue Mobility Model
Reference Point Group Mobility Model
Random Walk

Parameters:
Speed =
[Speed_MIN,Speed_MA
X]
Angle = [0,2]
Traversal Time (t) or
Traversal Distance (d)
Random Waypoint

Parameters
Speed =
[Speed_MIN,Speed_MA
X]
Destination =
[Random_X,Random_Y]
Pause Time >= 0
Random Direction

Parameters:
Speed =
[Speed_MIN,Speed_MA
X]
Direction = [0,] (Initial
[0,2])
Pause Time
Go straight with
selected direction until
bump to the boundary
then pause to select
another direction
Boundless Simulation Area

Parameters:
v(t+t) =
min{max[v(t)+v,0], Vmax}
(t+t) = (t) +
x(t+t) = x(t) + v(t) *
cos(t)
y(t+t) = y(t) + v(t) *
cos(t)
v = [-Amax*t , Amax*t]
= [-*t , *t]
Gauss-Markov

Parameters:
sn sn1 (1 )s (1 2 )sn1

d n d n1 (1 )d (1 2 )d n1

When approach the

edge, d value will
change to prevent the
node stocking at the
edge
Probability Version of Random Walk
City Section
Group Mobility

Exponential Correlated Random Mobility Model

Column Mobility Model

Given a reference grid, each node has a reference point on the
reference grid
Each node can move freely around its reference point (Ex:
Random Walk)
Nomadic Community Mobility Model
Similar to Column Mobility Model, but all nodes of a group share
a reference point
Pursue Mobility Model
new_pos = old_pos + acceleration(target-old_pos) +
random_vector
 Reference Point Group
Mobility Model
Group movements are
based on the path traveled
by a logical center for the
group
GM: Group Motion Vector
RM: Random Motion
Vector
Random motion of
individual node is
implemented by Random
Waypoint without Pause
Simulation Parameters
Simulation Package: ns-2
50 Mobile Nodes
100m Transmission Range
Simulation Time from 0s ~ 2010s
Results is collected from 1010s

Routing Protocol: Dynamic Source Routing (DSR)

Each mobility model has 10 different runs and show with 95%
confidence interval
Simulated Model
Random Walk

Random Waypoint

Random Direction
RPGM
100% Intergroup communication
50% Intergroup communication, 50% Intragroup communication
What results do different mobility models
bring?
Observation of Different Mobility Models

The performance of an ad hoc network protocol can

vary significantly with different mobility models
The performance of an ad hoc network protocol can
vary significantly when the same mobility model is
used with different parameters
Data traffic pattern will affect the simulation results
Selecting the most suitable mobility model fitting to
the proposed protocol
Considerations when Applying Each
Mobility Model
Random Walk
Small input parameter: Stable and static networks
Large input parameter: Similar to Random Waypoint
Random Waypoint
Scenarios such as conference or museum
Random Direction
Unrealistic model
Pause time before reaching the edge Similar to Random Walk
Boundless Simulation
Although moving without boundary, the radio propagation and neighboring
condition are unrealistic
Gauss-Markov
Most realistic mobility model
Probabilistic Random Walk
Choosing appropriate parameters to fit real world is difficult
City Section
For some protocol designed for city walk
 Exponential Correlated Random Mobility Model
Theoretically describe all other mobility models, but
selecting appropriate parameter is impossible
Column, Nomadic Community, Pursue Mobility
Models
Above three mobility models can be modeled by RPGM
with different parameters
RPGM Model
Generic method for group mobility
Assigned an entity mobility model to handle groups and
individual node
Random Waypoint Considered Harmful

Yoon has published a paper indicating that

network average speed of RWP will decay to
zero in simulation
Setting the minimum speed larger than zero
Besides speed decaying, the nodes with
RWP tend to cross the center of the area
(0, Ymax) (Xmax, Ymax)
( X max X 1 )(Ymax Y1 )
First : X max Ymax
( X 1 0 )(Ymax Y1 )
(X1 - 0) * (Ymax - Y1) (Xmax - X1) * (Ymax - Y1)
Second : X max Ymax
(X1, Y1)

( X 1 0 )(Y1 0 )
(X1 - 0) * (Y1 - 0) (Xmax - X1) * (Y1 - 0) Third : X max Ymax
( X max X 1 )(Y1 0 )
Fourth : X max Ymax
(0, 0) (Xmax, 0)
If you cant move quickly, please dont go
too far

V2
Dshorter

V1
V4 V3

Dshorter

V min<V 3,V 4<V lower<V 1,V 2<V max

If you are close to the wall, dont be afraid

Start Position
Target Position
Fixed Lower Speed, Varied Shorter
Distance
Fixed Lower Speed, Varied Shorter Distance
Random Waypoint
12 Lower Speed, Shorter Distance (2, 400)
Lower Speed, Shorter Distance (2, 300)
10 Lower Speed, Shorter Distance (2, 200)
Lower Speed, Shorter Distance (2, 100)
Average Speed (m/s)

8 Lower Speed, Shorter Distance (2, 50)

0
0 5000 10000 15000 20000 25000 30000 35000
Simulation Time (seconds)
Varied Lower Speed, Fixed Shorter
Distance
V aried Speed Fixed D istance(200m ) R andom W aypoint
Speed 1m /s
12
Speed 2m /s
Speed 3m /s
10 Speed 4m /s
Speed 5m /s
A verage Speed (m /s)

8 Speed 6m /s
Speed 7m /s

0
0 5000 10000 15000 20000 25000 30000 35000
Sim ulation Tim e (seconds)
Ripple Mobility Model with LSSD

R ipple M obility M odel(R adius:1000m ,Low er Speed:2m /s) W ithout LSSD

Shorter D istance 400m
12
Shorter D istance 300m
Shorter D istance 200m
10 Shorter D istance 100m
Shorter D istance 50m
A verage Speed (m /s)

0
0 5000 10000 15000 20000 25000 30000 35000
Sim ulation Tim e (seconds)
Ripple Mobility Model with LSSD and
Shorter Radius
R ipple M obility M odel(R adius:500m ,Low er Speed:2m /s) W ithout LSSD
12 Shorter D istance 400m
Shorter D istance 300m
Shorter D istance 200m
10 Shorter D istance 100m
Shorter D istance 50m

8
A verage Speed (m /s)

0
0 5000 10000 15000 20000 25000 30000 35000
Sim ulation Tim e (second)
How does the speed distribute during the
simulation
Random Waypoint
The Number of Speed Generation Lower Speed Shorter Distance (2, 50)
Lower Speed Shorter Distance (6, 50)
4500
Ripple Mobility Model(2, 50)
4000
3500
Number of Events

3000
2500
2000
1500
1000
500
0
0 2 4 6 8 10 12 14 16 18 20
Speed Distribution
Thats the Answer

The D uration of Speed R andom W aypoint

Low er Speed Shorter D istance (2, 50)
0.7
Low er Speed Shorter D istance (6, 50)

0.6 R ipple M obility M odel(2, 50)

0.5
D uration R atio

0.4

0.3

0.2

0.1

0
0 2 4 6 8 10 12 14 16 18 20
Speed D istribution
Spatial Distribution Comparisons

Cumulate the nodal appearance in the square of

100mx100m (Moving area is 1000mx1000m)
Conclusions

The defects in RWP are explored by some

research and solutions are also proposed
Decaying Average Speed
Unfair Moving Pattern
Most proposed solutions target to one of the
defects but not both
To generate different moving pattern within
the same speed range is a reasonable
requirement but it has not been brought up
Reference

Tracy Camp, Jeff Boleng, and Vanessa

Davies, A Survey of Mobility Models for Ad
Hoc Network Research, Wireless
Communication and Mobile Computing
(WCMC) 2002
Guolong Lin, Guevara Noubir, and Rajmohan
Raharaman, Mobility Models for Ad Hoc
Network Simulation, INFOCOM 2004
First Model

X1, X2,, Xn as non-overlapping time periods

During each time periods Xi, the node travels with
some random speed Vi for some random distance
Di
Xi=Di/Vi Xi+1=Di+1/Vi+1 Xi+1=Di+1/Vi+1
{Xi} is a renewal process
This model is applied on the homogeneous
metric space
 Analytical Results of First Model
E[ D] z 1
( z) dFV ( x)
t
lim
t
F V
E[ X ] 0 x
E[ D]
E[V ]

f ( z )dz
V
t E[ D] 1 0 E[ X ]
lim
t
f V ( z) E[ X ] z fV ( z)

2 2 Rmax Vmax
E[ D] Rmax E[ X ] ln( )
3 3(Vmax Vmin ) Vmin

Vmax Vmin
E[V ]
Vmax
ln( )
Vmin
 Residual distance t is represented the remaining
distance in time period Xi contains t
t0 t1

Xi = (t0, t1) Di ,
t0 2
Di
2

Residual waiting time t is represented the

remaining time in time period Xi contains t
z
1 1
lim Ft ( z ) [1 FD ( x)]dx lim f ( z )
t
[1 FD ( z )]
t E[ D] 0 t E[ D]
x
1
G( x) lim Pr{ t x} [1 Fx ( y)]dy
t E[ X ] 0
Second Model

The initial Speed and Distance are derived

from the stable state of First Model
The movement state of a node at time t is
characterized by Vt, t, t
For X0,
D0 is selected according to F(z)
V0 is selected according to FV(z)
The remaining periods are chosen as the
same as the first model
Analytical Observation

Second model has the following distribution

E[ D] x 1

Pr{Vt x}
S
y dFv ( y )
E[ X ] 0
x
1
Pr{ t x}
S
[1 FD ( y)]dy
E[ D] 0
If the following formula is satisfied,
2
fV ( y ) ky, y [Vmin ,Vmax ], k 2
Vmax Vmin
2

Speed distribution is uniformly distributed within [Vmin, Vmax]

Simulation Results in Boundless Area
(Rmax=1000m, (0,20))
Original:
r2
Distance D is picked from FD (r ) 2
Rmax
Speed V is picked
uniformly from [Vmin, Vmax]
Modified:
3 r3
D0 is picked from F (r )
d
0
(r 2 )
2 Rmax 3Rmax
Vo is picked uniformly
from [Vmin, Vmax]
2
r
Di is picked from F ( rD) 2
Rmax
Vmin
2 2

Vi is picked from F ( x) VX
max Vmin
V 2 2
Speed Distribution Comparison between
Modified and Original in Boundless Area
Simulation Results in Bounded Area
(1500mx1500m, (0,20))
Original:
Destination is picked
uniformly within the
simulation region
Speed is picked uniformly
from (Vmin, Vmax)
Modified
Destination is picked
uniformly within the
simulation region
First Speed is picked
uniformly from (Vmin, Vmax)
Following Speed is picked
according to F ( x) X V
2 2
min
V
V 2 2
Vmax min
Speed Distribution Comparison between
Modified and Original in Bounded Area
Extended Version of First and Second
Model
Pause Time Z is considered in the extended
model
Z is independent of D, V (hence X)
It generates the following time sequence
X1, Z1, X2, Z2, X3, Z3,, Xn, Zn,
The sequence is not renewal process
If pair of X and Z is considered, then {Yi} is
renewal process
Y1(X1+Z1), Y2(X2+Z2), Y3(X3+Z3),, Yn(Xn+Zn),
Discussions and Conclusions

Renewal theory is useful to analyze and solve the

problem of speed decay in the RWP model
Derive target speed steady state distribution from
the formula to simulate the different impact on
routing protocol
Do not need to discard the initial simulation data to
achieve steady state
How is the nodal distribution?
RWP in Bounded Area is not derived from analytical
work