Transcript Mobile P2P - Creating a mobile file

Mobile P2P - Creating a
mobile file-sharing
environment
Johnny Biström,
Ville Partanen
Agenda
Research questions
What is mobile P2P
Solution: Full mobile phone based P2P
Solution: Computer aided mobile P2P
Solution: The future of mobile P2P: JXTA
Threats to mobile P2P
Conclusion
Research questions
How can file sharing be realized in mobile
networks today?
How will the upcoming 3G-networks
support file sharing?
What are the threats for file sharing and
how can they be overcome?
Mobile P2P?
Transferring data from one mobile phone
to another
Mobile phone and network limit the
possibilities of mobile P2P
Low efficiency (CPU and memory)
Low bandwidth
Low power
Billing
Full mobile P2P in 2/2.5G
In 2/2.5 there are limitations that are
impossible to overcome:
Operators do not allow to see mobile phones
IP address
Operators control data traffic including ports
suitable for them
Network does not offer any way to sustain
active connection in all situations
Voice and data can not be transferred
simultaneously
A solution to 2/2.5 P2P:
MMS
MMS could be used as a way of sending
data from one mobile note to another.
However there are problems:
How to know who has the information you
need?
MMS size is limited
MMS costs more than GPRS data
A solution to 2/2.5 P2P:
MMS
We have to have a server
that keeps a record of
MSISDN number and the
data that can be found from
that number
Downloader asks the data
and the person who is
downloaded permits or
denies download
A better solution:
computer aided P2P
All the major limitations could be
overcome if the mobile phone would be
connected to a computer which has P2P
software
We would only need a software to
communicate between the computer and
mobile phone:
Short distance: IR, BT, PC suite etc.
Remotely: Over HTTP
Computer aided P2P: short
distance
Within short distance we would not have
true mobile P2P:
A better solution would be to control fixed
network peer remotely
Computer aided mobile
P2P: remotely
For example over http we could control
the fixed network peer by using a
program called mobile eMule
Computer aided mobile
P2P: eMule
1. login
2. search
3. download
to computer
4. download
to phone
eMule is a working solution but does
not currently implement full download
to mobile phone
JXTA – Tomorrow’s P2P
solution
Background
Software Architecture
Network Architecture
Protocols
Example Applications
JXME
JXTA - Background
started by Sun Microsystems in 2001
Open Source, royalty free licence
platform independent (mobile phone ->)
architecture and protocols
uses HTTP, TCP/IP and XML
builds virtual ad-hoc network on top of
physical network
JXTA – Software
Architecture
JXTA – Network
Architecture
JXTA
2
JXTA - Protocols
Core Specification Protocols
 Peer Resolver Protocol (PRP)
 Endpoint Routing Protocol (ERP)
Standard Service Protocols
 Rendezvous Protocol (RVP)
 Peer Discovery Protocol (PDP)
 Peer Information Protocol (PIP)
 Peer Binding Protocol (PBP)
JXTA – Example
Applications
JXTA Shell
command line interface
enables publishing, searching, messaging,
discovering, piping and more
MyJXTA
open soure example application
enables group chatting, secure chatting,
credential groups in addition to JXTA Shell
JXTA – MyJXTA2 interface
JXTA – JXME (JXTA for
J2ME)
Working Proxy based solution exists
JXTA – JXME (JXTA for
J2ME)
Proxyless solution under development
Threats to mobile P2P
In 3G true mobile P2P is possible due to
high bandwidth, efficient mobile phones
and simultaneous voice and data
capability
-> But will the operators allow P2P
software since is would lead to the loss of
revenues?
Viruses, spy- and adware
Digital rights management
Conclusions
2/2.5G is not ready for mobile P2P.
However with the aid of computer killer
applications could be developed
3G does not have technical restrictions so
the future of mobile P2P will be mainly in
the hands of the operators