The Internet is becoming wireless

New Service and Architecture Requirements

for the Future Internet: The Wireless, Mobile and Sensor

Network Perspective

based on the NSF WMPG

report, Rutgers, Oct 2005

CCW OCT 2005

Mario GerlaCSD, UCLA

The Internet is becoming wireless

Laptop sales exceeded desktop PC sales in July 2003 2B mobile phones in use by the end of 2005 > ~1B Internet users

>~0.5B networked PC’s …most new phones also have packet data capability

Overall, this means that by 2015, # wireless Internet terminals >> # wired! Laptops, cell-phones, PDA’s, iPoD’s ~ 10x PC’s/servers Embedded devices (sensors, actuators, RFID,…) ~ 10-100x PC’s & growing

This has important implications for network architecture, both wired and wireless: Wireless access networks must scale and handle new types of devices (sensors,

etc.) The Internet, which was designed in the 70’s for wired PC-PC/server connections,

needs to scale and evolve towards changing service needs

Wireless Internet Access Evolution

Internet

Infostation cache

WLANAccess Point

WLANHot-Spot

VOIP(dual-mode)

Low-tier clusters(e.g. low power 802.11 sensor)

Ad-hocnetwork

extension

Public Switched Network(PSTN)

BTS

VOIP

Broadband Media cluster(e.g. UWB or MIMO)

BTS

BSC

MSC

CustomMobileInfrastructure(e.g. GSM, 3G)

CDMA, GSMor 3G radio access network

Mobile/wireless service enhancements

Today Future?

Impact on Internet Architecture

Previous attempts at upgrading of IP spec have not had the expected result:

IPv6 standardized but not widely deployed... Little progress with end-to-end QoS in the Internet Mobile IP for first wave of wireless needs not implemented IP’s lowest common denominator (best effort datagram) also its greatest strength!

Earlier attempts at utopian new network architectures mostly ended in failure, in spite of technical merits

B-ISDN/ATM did not take off (...complexity, lack of organic growth model) Significant standards activity and community endorsement not sufficient to launch new network

architectures... Problems with 3G wireless

This doesn’t mean that new networks aren’t needed, but architectures needed to encourage bottom-up transformation without loss of investment in legacy system:

Evolutionary strategies preferable New approaches to protocol standards: hierarchies, modularity, open-source,.. Economic incentives for deployment

Internet Architecture: Caveats

Evolutionary approach Design a new wireless, ad-hoc and sensor “low-tier IP network profile to be “compatible”

with IP global network (e.g. IPv6, BGP routing, MPLS, etc.) Identify critical hierarchy and core IP extensions needed and pass requirement to IETF, etc. Evolve IP functionality via new RFC’s As wireless service needs proliferate, new low-tier IP may replace current IP intra-network

Internet Architecture: Strategies for Change

BorderRouterfor IPw Border

Routerfor IPw

BorderRouterfor IPv4

GLOBAL INTERNET

IP Wireless/SensorAccess Network (IPw)

IP Wireless/SensorAccess Network (IPw)

IP AccessNetwork

(e.g. IPv4)

New Interface Spec

New Protocol Spec

IPv6 extensions

Overlay approach Design new wireless, ad-hoc or sensor access net to work across global overlay network Specify and build new overlay networks optimized for wireless needs May include concept of an “IP knowledge plane” accessible by overlay If successful, IP is pushed down to a “layer 3-” service, while overlay is “3+” Permits significant flexibility in advanced service features, but tight optimization of packet overhead more

difficult due to IP encapsulation

Internet Architecture: Strategies for Change

BorderRouter

GLOBAL OVERLAY NETWORK

New Wireless/SensorAccess Network

IP AccessNetwork

New Design (non-IP)

new wireless-specific services

GLOBAL INTERNET

Overlay NetGateway

Overlay NetGateway

IP Tunnel

Overlay NetGateway

New Wireless/SensorAccess Network

new knowledge plane?

Revolutionary approach Specify a new “beyond IP” network optimized for mobile/wireless/sensor Build a prototype nationwide network and offer it for experimental use Use this network for emerging mobile data and real-time sensor actuator applications with

demanding performance and efficiency requirements Most radical, risks being marginalized by Internet evolution and legacy staying power

Internet Architecture: Strategies for Change

New Designs (beyond IP)optimized for

emerging needs including wireless-specific services

Next-Gen GLOBAL INTERNET

New Access Network

optimized forwireless, etc.

New Access Network

BorderGateway

IP AccessNetwork

The NSF WMPG (Wireless Mobile Planning Group) Workshop Aug 2-3, 2005

NSF Wireless Mobile Planning Group (WMPG) Workshop - Rutgers Aug 2-3, 2005

A group of about 30 researchers in the wireless area met at Rutgers (under the leadership of Ray Dipankar) to discuss: Unique requirements posed by wireless mobile users Potential impact on the Internet architecture Experimental facilities required to explore the new Internet architecture solutions

A report was issued in October:

“New Architectures and Disruptive Technologies for the Future Internet:Wireless, Mobile and Sensor Network Perspective”

www.winlab.rutgers.edu/WMPG

The “wireless” requirements

Identify new requirements placed by wireless users on the Internet “network layer”

These new requirements may trigger a “redesign” of the IP stack (or more generally the way we do networking)

We were not concerned with SOLUTIONS at this point Questions to be addressed:

What is the wireless scenario/application you are addressing? What is the problem to be solved? What are the new qualitative requirements on the network layer? What is the impact of these innovations on user performance?

The wireless scenarios

We identified three representative scenarios:

The individual mobile user, interacting only with Internet resources

The mobile “constellation”: the users equipped with several devices/interfaces, interacting with the Internet, with environment (instrumented user) and with each other (opportunistic ad hoc networking). This model applies to individuals while they walk, drive cars, fly planes, ride trains etc.

The “dynamic” pervasive sensor fabric”: this concept includes the traditional environment sensor fields as well as the mobile sensor fields (people, car sensor fabrics). This latter scenario is clearly connected with the instrumented constellation scenario

Wireless Service Requirements

Summary of Network Requirements and Architecture Challenges

1. Naming and addressing flexibility 2. Mobility support for dynamic migration of end-users and network devices 3. Location services that provide information on geographic position 4. Self-organization and discovery for distributed control of network topology 5. Security and privacy considerations for mobile nodes and open wireless

channels 6. Decentralized management for remote monitoring and control 7. Cross-layer support for optimization of protocol performance 8. Sensor network features such as aggregation, content routing and in-network

processing 9. Cognitive radio support 10. Economic incentives to encourage efficient sharing of resources

Wireless Requirements: Mobile Data Fast growth of (conventional) mobile data terminals with wireless access

link implies a need for new services on the Internet: Terminal mobility (authentication, roaming and dynamic handoff)…mobile IPv6 Multicasting …IP multicast Security …e.g. protection against AP spoofing Efficient transport layer protocols (..non TCP)

Major topic in research & standards during 90’s, but limited use..

INTERNETINTERNET

AccessPoint (AP)

Mobile dataterminal

High packetError rate

mobility

Radio multicasting

Roaming,handoff

Wireless Requirements: Mobile P2P P2P, 7DS, Infostations, etc. represent another emerging category

of mobile applications on the Internet Router mobility Network may be disconnected at times …delayed delivery? Caching and opportunistic data delivery …. In-network storage Content- and location- aware data delivery

Internet

Low-speed wide-areaaccess

Infostationcell

Mobile Infostation

Roadway Sensors

Mobile User

Data Cache

Ad-HocNetwork

OpportunisticHigh-Speed Link

(MB/s)

Infostation

OpportunisticHigh-Speed Link

(MB/s)

Wireless Requirements: Ad-Hoc Nets Ad-hoc nets with multiple radio hops to wired Internet useful for various

scenarios including mesh 802.11, sensor, etc. Discovery and self-organization capabilities Seamless addressing and routing across wireless-wired gateway Geographic routing options Support for end-to-end cross-layer protocol approaches where needed Privacy and security considerations

Relay Node

Access Point

Sensor

Wireless link withvarying speed and QoS

Local Interferenceand MAC Congestion

Dynamically changingNetwork topology

Best sensor-to-mobile path via wired network(needs unified routing)Wired Internet

Ad-HocNetwork

IP-Ad-hoc NetProtocol Conversion

Gateway

Wireless Requirements: Sensors Sensors and actuators with size/power constraints

Limited CPU processing & memory (?) Communication speed may be low Intermittent connectivity (power saving modes) Relatively unreliable components Very different application requirements

Important new paradigm, since # sensors potentially in the billions Protocols & system designs still at an early stage First sensor nets for simple measurement applications More complex “closed-loop” sensor/actuator in future

MIT DVS

UC Berkeley MOTE

Sensors in roadway interact with sensor/actuator in cars Opportunistic, attribute-based binding of sensors and cars Ad-hoc network with dynamically changing topology Closed-loop operation with tight real-time and reliability constraints

Sensor Applications: Highway Safety

The Experimental Facilities

Experimental Infrastructure for Future Wireless Network Research

Techniques for Flexible Experimental Wireless Networks Virtualization of Wireless MAC

Cognitive Radio

Wireless Network Monitoring and Measurement Measuring and characterizing mobility. Measuring heterogeneous networks overlapping in space. Measuring cellular and DTN networks. Cooperative sharing of measurements

Wireless Network Repository

Emulation and Simulation Testbeds for Wireless

Wireless Networking Platforms

Platform Software and End-to-End Architecture

Experimental Infrastructure for Future Wireless Network Research (cont)

Wireless Network Repository

Emulation and Simulation Testbeds for Wireless

Wireless Networking Platforms

Platform Software and End-to-End Architecture

Virtualized Multi-MAC

Experimental Wireless Networks

Integration of Existing Testbeds

Ad Hoc Mesh Networks

Summary of Recommendations

Recommendation 1: the Internet will undergo a fundamental transformation over the next 10-15 years; invest in research programs aimed at creating necessary technical foundations.

Recommendation 2: Increase research focus on central network architecture questions related to future mobile, wireless and sensor scenarios.

Recommendation 3: Invest in development of flexible wireless technologies and platforms necessary to implement programmable and evolvable experimental networks.

Recommendation 4: Fund development of large-scale experimental wireless networks for effective validation and competitive selection of new architecture and protocol concepts.

Recommendation 5: Encourage collaborative research that would result in end-to-end deployment and evaluation of future wireless/mobile and sensor networks and applications over the global Internet.

Examples of Research enabled by the new testbed platforms

Vehicle Grid Applications Car Torrent Ad Torrent Car to Car Games Vehicle Sensor Network

Co-operative Downloads: Car-torrent, Ad torrent

Vehicle-Vehicle Communication

Internet

Exchanging Pieces of File Later

Car2Car Games: Game Server Architecture + Car-networking Scenario

Vehicular Sensor Network (VSN)

Infostation

Car-Car multi-hop

1. Fixed Infrastructure2. Processing and storage

1. On-board “black box” 2. Processing and storage

Car to Infostation

Applications Monitoring road conditions for Navigation Safety or Traffic control Imaging for accident or crime site investigation

Thank you!