A predictive rapid mobility extensions for Mobile-IP
Download
Report
Transcript A predictive rapid mobility extensions for Mobile-IP
An adaptive networking
protocol for rapidly mobile
environments
Ph.D. Research Proposal
Edwin A. Hernandez
April 22th, 2002
Motivation
Mobile-IP is the most widely used mobility
solution in IPv4 and IPv6 networks.
However, the performance for vehicles
moving at high-speeds is questionable.
Predictable trajectory and mobility, network
originated handoff, and distributed
registration can improve the performance of
Mobile-IP without the use of costly micromobility protocols.
Table of Contents
• Introduction
• Related Research
• Performance of Micro- and Macro-Mobility
protocols
• RAMON: A proposal for network
emulation.
• Predictable mobility in wireless networks
• Extensions for a predictable Mobile-IP
Introduction – Concepts in
Mobile Networks
• 1.
Forwarding
Agent when a MH is
foreign network
• 2. Location
Directory (LD),
location information
• 3. Address
Translation Agent
(ATA)
Home
network
Stationary
host
Internet routing
Foreign Network
g ( fowardingaddress) home address
f (homeaddress) forwardingaddress
Introduction – Packet Forwarding
Model
Location Directory
Home
network
LD
Address Translation Agent
Source
Foreign Network
Forwarding Agent
f
g
Cache
Internet routing
MH
f: Home Address
g: Forwarding Address
Forwarding Address
Home Address
Network-layer mobility is solved by registering in a centralized
database of location, LD, which also solves problems of
authentication, accounting, and authorization of mobile users in the
network. However, network delays, time for authentication, and
handoff render the packet-forwarding model unusable for fast moving
hosts
Related Research
• Solutions to the problem of mobility
– Macro-Mobility protocols
• Mobile-IP
• Hierarchical Mobile-IP (Hierarchical Foreign
Agents)
– Micro-Mobility protocols
• Cellular-IP
• HAWAII
[Perk95], [Perk96], Solo[98]
Mobile-IP
HA
MH
CH
Router
Router
•Mobile-IP follows the LD/FA model
•Encapsulation is required when
packets are forwarded
•Mobile node acquires care-of-address
thru DHCP.
Src
X
Router
Dest
MH
Proto
?
FA
payload
Src
HA
Home
Agent
Encapsulated datagram
Dest
FA
Proto Src
4 or 55
X
Dest
MH
MH
Triangular routing in Mobile-IP
Foreign
Agent
Mobile Host
Src
X
• Registration overhead of 1 sec.
Dest
MH
Proto
?
payload
Proto
?
payload
Handoff in Mobile-IP
MH
FA
HA
Send Solicitacions
Send Advertisements
Update Agent List
Handoff necessary
Registration Request
Forwarding Request
Registration Reply
Forwarding Reply
New COA
• Handoff overhead >= Registration Overhead
• Handoff Impact = confuses TCP
[Perk96a], [Cast98]
Reducing Registration Overhead
Hierarchical Foreign Agents
Early reaction of the
research community
Internet
FA1
HA
Registration
FA2
FA3
Registration
FA4
FA5
MH
FA6
MH
FA7
MH
An FA includes in its Agent
Advertisement message a
vector of care-of-addresses,
which are the IP address of all
the ancestors in the tree as well
as its own. By the time the
MH arrives to a new cell, it
makes an advance registration
to the HA, the FA, and the
ancestors of the FA.
• Reduces Registration Overhead
• Requires many wired-nodes/costly
During handoff of HFA
During Handoff, the MH compares the new
vector of care-of-addresses with the old one.
Again, it chooses the lowest level address of the
FA that appears in both-vectors and sends a
Regional Registration Request, which is
processed by the FA. There is no need to notify
any higher-level FA about this handoff since
those FA already point to the proper location to
where to tunnel the packets for the MH
Micro-mobility Protocols
• 2-tier solution
• Micro-mobility model
used by Cellular-IP and
HAWAII [Camb00,
Ramj00]
• Intra-domain handoff is
handled by a signaling
protocol while the interdomain handoff is taken
care by the Mobile-IP
protocol
CH
Network
Home Domain
Router
HA
FA
root
root
home domain
MH
Foreign Domain
Router
foreign domain
Movement between domains, HA is notified
Movement within domain,
HA not involved
Movement within domain,
HA not involved
Cellular-IP
Correspondant
host
Home
Agent
Mobile IP
Internetworking
R
IP routing
IP tunneling
Cellular IP routing
BS4
BS2
b
BS1
a
Mobile X
BS3
Mobile X
routing
handoff
Layer-2/3 routing and handoff management, use of Signal strength and
telephony-like signaling for paging and handoff management.
HAWAII: Handoff-Aware Wireless
Access Internet Infrastructure
(0): 1.1.1.1->B
(3): 1.1.1.1->C
A
B
3
(0):*, 1.1.1.1 ->B
(3):A,C, 1.1.1.1 -> B
B, 1.1.1.1->C
(6) : *, 1.1.1.1->C
Router 0
C
A
4
4
(0): 1.1.1.1->A
(4): 1.1.1.1->B
Router 2
1
A
Router 1
B
A
C
B
C
(0):*, 1.1.1.1 ->C
(4):A,B, 1.1.1.1 -> C
C, 1.1.1.1->A
(5) : C, 1.1.1.1->CA
Router 1
Router 0
B
C
3
A
B
(0):*, Default ->A
(2):*, 1.1.1.1 -> B
A
C
6
(0): 1.1.1.1 ->C
(2): 1.1.1.1 ->A
B
C
7
Router 2
5
2
A
(0): 1.1.1.1->B
(1): 1.1.1.1->A
Old BS
5
A
B
B
(0): Default ->A
(4): 1.1.1.1->B
New BS
(0):*, 1.1.1.1 ->B
(5):*, 1.1.1.1 -> A
Old BS
A
A
B
B
2
New BS
6
1
Mobile Host
Mobile Host
IP: 1.1.1.1
(a)
MSF (Multiple Stream
Forwarding)
IP: 1.1.1.1
(b) SSF (Single Steam
Forwarding)
(0):*, Default ->A
(1):*, 1.1.1.1 -> B
A closer look to Micro-mobility
• Signaling protocols based on telephony standards.
• Avoid Mobile-IP for handoff
• Costly implementation for a wide-spread area, e.g. train
track, tied to speed. Requires the modification of
intermediate routers and network infrastructure.
• The packet loss can be described by rThoff, where r is
the rate and Thoff represents the amount of time to reach the
cross-over router from the MH.
• Our research re-examined the performance of Macro- and
Micro-mobility protocols in a simulation environment.
Network Simulator (ns)
plots
network simulator
(ns)
tcl
protocols
ns
objects
4G
trace file
results
trace analysis
awk,
xgraph
stats
802.
11a
New Technologies (to develop)
networks
•The ns network simulator – Berkeley [Fall00]
• tcl/c++ object oriented, 20 Mbytes of code, wired and
wireless network protocols
PERFORMANCE OF MACRO- AND MICROMOBILITY PROTOCOLS IN A RAPID
MOBILE ENVIRONMENT
• Two simulation scenarios were used:
– Mobile-IP original Berkeley/CMU implementation
– Columbia University micro-mobility suite
• Results for Macro-mobility protocols were
published in LCN 2001.
– E. Hernandez and A. Helal, "Examining Mobile-IP
Performance in Rapidly Mobile Environments: The
Case of a Commuter Train," LCN 2001, Tampa, FL,
Nov 14-16, 2001
– E. Hernandez and A. Helal “RAMON: a network
emulation testbed”, submitted to Wireless
Communications Journal, Wiley & Son’s.
ns simulation scenarios
C-host
0.2.0
W(0)
0.0.0
W(1)
0.1.0
10ms
20ms
30ms
100ms
Mobile Host 1.0.1
(a) Real track
BS(1)
1.0.0
BS(2)
2.0.0
BS(3)
3.0.0
(b) ns scenario
BS(10)
10.0.0
Performance of Mobile-IP for
TCP transmissions (FTP)
100%
90%
80%
50
40
250 m
500 m
30
750 m
20
1000 m
Time transmitting
Throughput (Kbytes/s)
60
70%
60%
50%
40%
30%
250 m
500 m
750 m
1000 m
20%
10%
0%
10
0
0
20
40
60
80
100
0
20
40
60
80
Speed (m/s)
Speed (m/s)
(a) Average throughput
(b) Percentage of usable time (not in handoff)
ns-2 simulation. [Hern01]
Performance of Mobile-IP for
UDP transmissions
50
40
250 m
500 m
30
750 m
20
1000 m
Time transmitting
Throughput (Kbytes/s)
60
100%
90%
80%
70%
60%
50%
40%
30%
250 m
500 m
750 m
1000 m
20%
10%
0%
10
0
0
20
40
60
80
100
0
20
40
60
80
100
Speed (m/s)
Speed (m/s)
(a) Average throughput
(b) Percentage of usable time (not in handoff )
ns-2 simulation. [Hern01]
Performance of TCP/FTP transmissions
macro/micro-mobility
12
100%
Cellular-IP
8
Mobile-IP
6
HFA
4
Hawaii
80%
%Usable
Kbytes/sec
10
Cellular-IP
Mobile-IP
60%
HFA
40%
Hawaii
20%
2
0%
0
0
20
40
60
80
100
Speed (m/s)
(a) Average Throughput
0
20
40
60
80
100
Speed (m/s)
(b) Percentage of usable time
ns-2 Columbia micro-mobility suite
Performance of UDP transmissions with
macro/micro mobility
60
100%
Cellular-IP
40
Mobile-IP
30
HFA
20
Hawaii
80%
% Usable
Kbytes/sec
50
Cellular-IP
Mobile-IP
60%
HFA
40%
Hawaii
20%
10
0
0%
0
20
40
60
80
100
Speed (m/s)
(a) Average Throughput
0
20
40
60
80
Speed (m/s)
(b) Percentage of usable time
ns-2 Columbia micro-mobility suite
100
Problems with the simulations
• Columbia uses the NOAH (non-adhoc agent)
developed by Widmer [Wid00] as an extension for
ns
• The NOAH agent has a simplified version of
propagation model.
• The NOAH agent has a “improved” handoff
mechanism and assumes GPS information
– NOAH->getX() and NOAH->getY( ) methods
– Mobile-IP with NOAH outperforms its predecessor.
• It’s hard coded the bandwidth at 2Mb/s and
difficult to change in the simulator.
• Simulator code is more than 20 Mbytes, why not
implement it directly on a testbed?
RAMON : A network emulation
approach
• Criticism of network simulation approaches
[Paw02]
• Attenuators used to emulate velocity and handoff
• Real implementation and code-extensions made to
real mobility agents
• Network emulation language to facilitate,
academic and network-engineering work.
• ns scripts can be parsed and emulated with minor
modifications.
• Applications can be tested in rapid mobility
conditions
RAMON: The architecture
Path loss attenuation and data
rates with 802.11b access points
distance (m)
101
201
301
401
12
501
distance (m)
0
601
-20
PL(d) 100mW
-30
Data rate
8
6
-60
4
-70
-80
100
200
300
400
500
600
700
-20
-40
-50
0
10
PL(d) 1mW
Path Loss (dBm)
PL (dBm)
-10 1
Data Rate (Mb/s)
0
-40
-60
PL(d) (5mW)
PL(d) (30mW)
PL(d) 100 mW
-80
2
-90
-100
-100
0
-120
(a) Path loss and data rate for Cisco AP-350
(b) Path loss equations at different transmission power
levels (n=2.5)
• It’s necessary two provide actual bandwidth to
accurately estimate and reflect the effects of speed and
handoff on network cards
800
Attenuation Control with the
parallel port
74LS374
Attenuator 0
7
LD
74LS374
7
D0
D1
D2
D3
D4
D5
D6
Attenuator 1
LD
74LS374
7
Attenuator 2
D7
LD
74HC4051
SEL
Autofeed
S0
0
LDi
IN
1
Line 14 LPT1
S1
74LS374
2
SEL
74HC4051
AB
2
2
LD
Pseudo Code
WriteLPT1(0xxx xxABb); // Select Attenuator <AB> address
WriteLPT1(1xxx xxxxb); // Write data to the attenuator
Emulation of speed
Path Loss Equation:
PLrecv Ptx P(d o ) 10n log(
d
)
do
Scenario
Attenuator 0
Attenuator 1
Attenuator 2
No connectivity
-127 dB
-127 dB
-127dB
One cell
0 dB <set < -80
dB
-127 dB
-127dB
Two overlapped
cells
0 dB < set < -80
dB
0 dB < set < 80 dB
-127 dB
Three
overlapped cells
0 dB < set < -80
dB
0 dB < set < 80 dB
0 dB < set < 80 dB
RAMON emulation language
ns script
Emulation script
Description
$BS X_
$BS Y_
$BS name X=
$BS name Y=
Sets the coordinates of the Base-station
set BS [$ns node IP]
$BS name IP=
Sets an IP Address for the base-station
set power 0.289
$BS name power=xxx
The power level in mW in the access-point
Set HA… /FA…
$HA name IP
$FA name IP
Sets the HA/FA at an IP address
set mobile-ip 1
$protocol=”MIP“
The protocol being used
set wiredNode [$ns node $IP]
$WiredNode name IP1 IP2 IP3
Creates a Wired Node with three interfaces.
$ns duplex-link $node1 $node2 $bw
$latency DropTail
$Link IP1 IP2 bw latency
Creates a Link between two interfaces using certain
bandwidth and latency values
$ns at $time [$MH etdest x y speed]
$MH time x y speed
Sets the destination position and speed of mobile
host. Acceleration = 0.
$ns at $time start
-
Starts after it’s called
$ns at $time end
$end time
End of the emulation
$set opt(prop)
Propagation/TwoRayGround
$Propagation=”TwoRayGround”
|”PathLoss”|any other.
Sets the propagation model being used.
N/A
$granularity X
Updates attenuation and speed every X ms
Convert an ns script into
emulation code
Platform commands
$route – add …
ns.tcl
Create wired nodes
Routing/Emulation tables
Load Mobility agent,
Configure agents
Load mobility
Patter of node
Goal : Process a modified version of an
ns script and generate the emulation environment
$cnistnet –a ….
$fa @ IP…
$ifconfig –eth0:1..
$start attenuators
$load pattern
Go emulation!
Sample Emulation Script
$WiredNode node1 192.168.1.1 192.168.2.1 192.168.3.1
$WiredNode node2 192.168.2.2 192.168.4.1 192.168.5.1
$Link 192.168.2.2 192.168.2.1 10Mb 20ms
$Link 192.168.1.1 128.227.127.11 10Mb 1ms
….
$BS node7 X=250 Y=250 power=20dBm IP=192.168.7.1
$BS node8 X=750 Y=250 power=20dBm IP=192.168.8.1
$BS node9 X=1250 Y=250 power=20dBm IP=192.168.9.1
$BS node10 X=1750 Y=250 power=20dBm IP=192.168.10.1
$BS node11 X=2250 Y=250 power=20dBm IP=192.168.11.1
…
$MH 0 1000 250 20m/s
$start 10s
$end-time 1500s
$Propagation=”PathLoss”
$Protocol “MIP”
Emulation Code
•
•
•
•
•
•
•
•
•
•
•
Emulation(MH, granularity)
initializeResources( )
DetermineRoutes(route[][], time_end[], trajectory(MH));
while timer() > end_simulation
do
if timer>=timer_end[k]
then
k++
createRoute(route[k][1..3], time_end[k]);
expireRoute(route[k-1][1..3])
emulateMovement(granularity, MH )
return
NistNET emulator for wired networks
• Wired network emulation required for academic and
network engineering of rapidly mobile networks with may
service providers and heterogeneous networks.
Example
Architecture to emulate
CH
Link(i,j) {
Bandwidth,
latency,
error
}
2
4
7
MH
1
3
5
8
6
9
10
11
Emulation process
t=0
t=2(d+ )/v
default gw 1
7-4-8
8-4-2-5-9
7-4-2-5-9
default gw 1
7-4-8
7-4-2-5-9
8-4-2-5-9
7
8
9
7
9
MH @ v m/s
MH @ v m/s
(a)
10
8
(b)
t=4(d+ )/v
t=6(d+ )/v
default gw 1
8-4-2-5-9
9-5-2-1-3-6-10
8-5-2-1-3-6-10
default gw 1
9-5-2-1-3-6-10
9-5-2-1-3-6-11
8
9
MH @ v m/s
(c)
10
11
MH @ v m/s
(d)
9
Implementation of RAMON
antennas
Agents
Access
Points
Programmable Attenuators
Controller for attenuator
Attenuators
Foundations for a predictive mobile
environment
Mobile-IP fails when:
Tdwell Thandoff Tforward Tregistration
In the mobile environment,
We can define a mobile host as:
MHk IPH , IPC , P, SNR, G
0
1
2
3
4
0
1
2
IP@home, IP@care, P received,
SNR signal to noise ratio, and G where
3
4
G x, y, v, a
5
is a vector of position, speed,
and acceleration, while a Base Station,
6
7
BS ij G, MH care , MH
prev
, MH next
MHk
BSi,j
5
6
7
Simple representation of geographical
information
0
0
0
0
0
P rxy ( MH )
1
0
0
1
0
0
0
0
1
0
0
1
0
0
0
0
1
0
1
0
0
0
0
0
1
0
1
0
0
0
0
0
1
1
1
0
0
0
0
0
1
1
0
0
0
0
0
0
1
1
0
0
0
0
0
0
1
0
0
0
1
2
3
4
5
6
7
0
1
2
BSi,j
3
r1
r2
4
5
6
7
Rough mapping of an 8x8 grid of geographical information
to a matrix of probability of location.
Geographical information thru:
• GPS – not suitable for most PDA’s with low-power
• Indirect measures Signal Strength, Inter-Access Point
communication, etc
Uncertain position given P
received
rssi = -40 dB @ t = 1
BSi,j
rssi = -20 dB @ t = 2
rssi = -20 dB @ t = 3
rssi = -40 dB @ t = 4
7
2
3
r2
r1
6
4
5
1
0
By using Inter Foreign Agent or Inter Base-station messages
BS.db GIS.db PLossModel() = Location <x,y>
Predictable Mobility
With the Location information obtained :
• Trajectory predictors using mobility models and/or neural
networks
• Comparison with other predictable models
• Reviewing mechanisms.
– Kalman Filters
– Neural Networks
– Fuzzy Logic
• Matlab simulation of the mechanism selected isn in
progresss.
• Implementation of the protocol in RAMON
Extensions for Mobile-IP
query/update
query/update
Home Agent
(initial
implementation)
Location Registry
Database
Multicast
Extension
query/update
• Inter-agent protocols
IAP
FA
Context Information
IAP
FA
IAP
Extensios for predictable
mobility
• Network initiated handoff
• Rely in mobile-IP for slow
moving vehicles.
IAP
FA
IAP
FA
• Extensions for a
hierarchical-predictive
registration.
Predictable Home Agent
• Inter-layer
Communication (State
Manager) / Chinta’s work
Research timeline
Task
Expected Completion Date
Current Status
RAMON testbed
February – April 31st, 2002
Performance of micro- and
macro- mobility protocols
on testbed
Review of predictable
algorithms and simulation
for mobile protocol
Session and Context
Transfer for TCP
Integration of mobile-ip and
Predictable extensions
Windows XP/Pocket PC
implementation
Performance Testing of
extensions for mobile-IP
Dissertation Preparation
Dissertation Defense
April 20th to April 31st,
2002
- Some hardware needs to
be purchased.
- Set up parser for
simulation code into testbed
scripts.
- Ongoing investigation of
Mobile-IP platforms and
network emulation.
- pending – testbed
required.
February – May 2002
- Already started matlab
simulations .
February – May 2002
May – June 2002
- Madhav Chinta’s thesis.
TCP level implementation.
- ongoing
June 2002
- ongoing
June 15st to June 31th, 2002
- pending
Spring – Summer 2002
Summer 2002
- ongoing
TBD
completed
List of References
[Camb00] A. T. Campbell, Gomez, J., Kim, S., Turanyi, Z., Wan, C-Y. and A, Valko
"Design, Implementation and Evaluation of Cellular IP", IEEE Personal
Communications, Special Issue on IP-based Mobile Telecommunications Networks, Vol.
7, No. 4, pp. 42-49, August 2000.
[Cast98] C. Castelluccia. “A Hierarchical Mobile Ipv6 proposal”, Technical Report
INRIA, France, November 1998
[Fall00] K. Fall, K. Varadhan, editors. NS notes and documentation. The VINT project,
LBL, February 2000. http://www.isi.edu/nsnam/ns/
[Hern01] E. Hernandez and A. Helal, "Examining Mobile-IP Performance in Rapidly
Mobile Environments: The Case of a Commuter Train," Accepted to LCN 2001 in
Tampa, FL, Nov 14-16, 2001
[Perk95] C. E. Perkins, K. Luo “Using DHCP with computers that move”, Wireless
Networks 1(1995) 341-353.
[Perk96a] C. Perkins, IP mobility support, RFC 2002, IBM, October 1996
Perk96b] C. Perkins, Mobile-IP local registration with hierarchical foreign agents, Internet
Draft, Internet Engineering Task Force (February 1996
[Ramj00] R. Ramjee, T. La Porta, S. Thuel, K. Varadhan, L Salgarelli, IP micro-mobility
support using HAWAII , Internet draft submission , Jul 2000.
[Solo98] J. Solomon. Mobile-IP. Prentice Hall, New Jersey, 1998