Ambient Networks Die Weiterentwicklung mobiler Netztechnik nach UMTS
Ambient Networks Die Weiterentwicklung mobiler Netztechnik nach UMTS Andreas Schieder Ericsson Research This presentation has been produced in the context of the Ambient Networks Project. The Ambient Networks Project is part of the European Community's Sixth Framework Program for research and is as such funded by the European Commission. All information in this presentation is provided "as is" and no guarantee or warranty is given that the information is fit for any particular purpose. The user thereof uses the information at its sole risk and liability. For the avoidance of all doubts, the European Commission has no liability in respect of this presentation, which is merely representing the authors view. FFV 2006 - Ambient Networks 31 March 2006 1
Outline v. AN architecture and basic concepts v. Network Composition v. Multi-Radio Access v. Media Delivery v. Dissemination and outlook to phase 2 FFV 2006 - Ambient Networks 31 March 2006 2
AN architecture and basic concepts FFV 2006 - Ambient Networks 31 March 2006 3
WWI organisation in Phase I WWI Collaboration WWI Steering Board WWI Coordination Team User requirements System architecture Operability Resilience Quality of Service Mobi. Life Ambient Networks WINNER E 2 R Security and trust Validation Reconfigurability Current WWI Cross Issues Source: WWI Current WWI Integrated Projects Chaired by Ambient Networks FFV 2006 - Ambient Networks 31 March 2006 4
Ambient Networks project phases Work - Areas I: Concepts and Architecture II: Key Technical Problems III: Business Interfaces and Commercial Viability IV: Prototyping and Validation Phase 1 2004 -2005 Establishing the Ambient Networks Concept and its Feasibility Concepts & Architecture Design Feasibility, Functions of AN Control Space Business Feasibility Usability/Networks Test Cases Phase 2 2006 -2007 Technology Development Phase 3 2008 -2009 System Synthesis Architecture, Standards Architecture: Scalability, Evolvability Detailed Specification, Technical Development, Performance Optimisation for Performance and Deployment Business Interfaces Market Dissemination Usability/Networks Prototypes FFV 2006 - Ambient Networks 31 March 2006 Integration across WWI, Validation, Market Dissemination 5
Requirements posed on the AN Architecture 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. Heterogeneous Networks Mobility Composition Security and Privacy Backward Compatibility and Migration Network Robustness and Fault Tolerance Quality of Service Multi-Domain Support Accountability Context Awareness Extensibility of the Network Services Provided Application Innovation and Usability FFV 2006 - Ambient Networks 31 March 2006 6
An Ambient Network and its links to the environment Reference Points Ambient Service Interface Ambient Control Space Ambient Network Interface Context Information Routing Group Information Ambient Connectivity Multi-Radio Resource Management Ambient Network Interface Overlay Ambient Support Resource Layer Interface Agreement Naming Establishment Traffic Engineering FFV 2006 - Ambient Networks 31 March 2006 7
ACS Architecture FFV 2006 - Ambient Networks 31 March 2006 8
The Ambient Connectivity Abstraction B C A o A ear A nn pp bs e lic tr r tr ct ac at ac iv io tio ity ns n n Ambient Network A Ambient Network B Application Bearer Endpoint “AN-node” Application Ambient Network Interface Bearer Endpoint Intermediary Bearer “AN-link” Flow Endpoint Flow Transit Flow tiv ity Flow Endpoint Session C on ne c Node Path FFV 2006 - Ambient Networks 31 March 2006 Link 9
Internetworking Architecture DNS retrieves Node ID of peer (B), and Node ID of peer’s core router (NR 3) Locator domains have independent address spaces Nodes have HIP-like Node IDs, self-generated public keys A LD 1 (IPv 4 Core) DNS LD 2 At the core, we do not want a global DHT of all nodes. Instead, peer’s core NID router (NR 3) is passed in the packets. A small DHT can look up the locator of NR 3. DHT NR 2 NR 3 LD 3 DNS Domains are connected by “NID routers” that route based on destination NIDs FFV 2006 - Ambient Networks End-to-end security is based on NIDs. B 31 March 2006 10
Traits of the Internetworking Concept v Integrates local mobility, end-to-end mobility, and network mobility v Even makes network-based multi-homing possible (a) (b) (c) B A A v Initial handshake (~ HIP) provides an always-on security A model; subsequent packets are protected v The handshake also has basic Do. S protection v Additionally, nodes can manage their NIDs and NID routers in a Hi 3 -like manner to provide network-based Do. S protection v NIDs can be changed on the fly for privacy reasons, and NID routers provide location privacy FFV 2006 - Ambient Networks 31 March 2006 11
Network Composition FFV 2006 - Ambient Networks 31 March 2006 12
Motivation for Composition v The central concept of Ambient Networks is Composition v Composition is… Ø a uniform, dynamic procedure for network interworking on the control plane v Uniform procedure Ø independent of network type and technology v Dynamic procedure Ø minimize human intervention v Control Plane Interworking regarding Ø routing, addressing, mobility, Qo. S, security, charging, . . FFV 2006 - Ambient Networks 31 March 2006 13
Composition Use Cases v Types of Composition • Involved networks merge into one common network • E. g. creation of a PAN Ø Control Delegation or Control Sharing • One AN delegates certain control functions to the other AN • 3 GPP-WLAN interworking: WLAN delegates authentication, authorization and charging to 3 GPP network Ø Network Interworking • Cooperation according to the Composition Agreement but no control delegation • E. g. dynamic roaming agreements FFV 2006 - Ambient Networks 31 March 2006 14 Increasing control plane cooperation Ø Network Integration
Composition - Standardization v The Ambient Networks Project established a Study Item “Network Composition” in 3 GPP SA 1 v TR 22. 980 „Network composition feasibility study; (Release 7)” v Content Ø Ø Ø Purpose and benefits of composition Use cases Requirement Composition Process New functionality in 3 GPP networks Relation to other functionality in evolving 3 GPP architecture • AIPN, … FFV 2006 - Ambient Networks 31 March 2006 15
Multi-Radio Access FFV 2006 - Ambient Networks 31 March 2006 16
Research Concepts Resource sharing between cooperating operators Resource sharing with cooperating local access provider Access Provider Multi-radio multi-hop Access Provider 3 G Radio WLAN Local Access Provider 4 G Radio 3 G Radio 4 G Radio Relaying Terminal Load management and(instant)mapping of data flows between accesses 4 G Radio Wi. MAX Relaying Network WLAN 4 G Radio Simultaneous/seamless multi-operator access Terminal FFV 2006 - Ambient Networks 31 March 2006 17
Dynamic Access Selection Algorithms v Taxonomy of Algorithms Type of Access Selection 1 ms Time Scale of Access Selection 100 ms 1 s 10 ms 1 min 1 h 1 d Link Quality Based Access Selection Fast channel variations (scattering, diffraction, fast fading) Slow channel variations (shadowing, slow fading, cell change) Stability of multi-hop routes Fast load variations (interference, data bursts, user handover, . . . ) Load Based Access Selection Slow load variations (new users / sessions, handover) Fast Access Selection Slow Access Selection FFV 2006 - Ambient Networks 31 March 2006 18
Access Selection Results “Slow” access selection “Fast” access selection Order of seconds Order of milliseconds Selection per flow Selection per IP-Packet or MAC-PDU Selection based on Averaged Link Quality (“Slow”) Instantaneous link quality (“Fast”) Complexity Moderate Complexity Timescale Network/Cell load, … Complex with high monitoring load Drastic changes to legacy link layers Capacity Gains 15%-20% additional capacity 15%-60% throughput and spectral gains by considering load efficiency gain (compared to signal strength only) Signalling delay decreases performance, In hotspot scenario up to 50% especially if delay > timescale of channel variations User Gains 15%-80% average bit-rate gain by 15%-60% user throughput gain with considering load single RAT at a time FFV 2006 - Ambient Networks 31 March 2006 19
Media Delivery FFV 2006 - Ambient Networks 31 March 2006 20
Media Delivery using SSONs v Establishment of Service-Specific Overlay Networks (SSONs) using Bearers Ø SSONs are independent virtual networks for every media delivery service (or group of services) Ø SSONs allow dynamically tailoring the network to the specific needs of a service v Why do we need SSONs? Ø Means to include network-side media processors into end-to-end media delivery path Ø Control application of media functions (where, how, which costs, etc. ) v What do SSONs achieve? Ø SSONs decouple the configuration of virtual networks from the underlying network Ø SSONs enable flexible configuration of virtual networks in terms of • • Topology Addressing Routing Qo. S FFV 2006 - Ambient Networks 31 March 2006 21
Ambient Networks Dissemination and outlook to phase 2 FFV 2006 - Ambient Networks 31 March 2006 24
Standardisation v >30 contributions in total v IETF submissions to hip and nsis Ø several hip submissions for security extensions Ø Nsis extensions v IEEE 802. 21 submissions Ø Triggering framework submitted v 3 GPP SI „composition“ Ø Approved at TSG SA Meeting #30, 05 - 07 December 2005, supported by Siemens, Ericsson, Nokia, Vodafone, NTT Do. Co. Mo, NEC FFV 2006 - Ambient Networks 31 March 2006 25
Structure of Work in Phase 2 FFV 2006 - Ambient Networks 31 March 2006 26
Thank you for your attention! Questions? www. ambient-networks. org FFV 2006 - Ambient Networks 31 March 2006 27
- Slides: 25