Differences
This shows you the differences between two versions of the page.
— |
d2.3_rationale_for_ipv6_in_its [2014/03/19 15:19] (current) stevant created |
||
---|---|---|---|
Line 1: | Line 1: | ||
+ | == Proposed text == | ||
+ | |||
+ | === Emerging ITS market and solutions === | ||
+ | |||
+ | Cooperative ITS technologies are now moving from research to industrialization. Wireless communications for Vehicle-to-Vehicle (V2V) or Vehicle-to-Infrastructure (V2I) are standardized and implementations are emerging. On the other hand, services based on V2V and V2I are developing quickly, as stakeholders converge on relevant use cases and business cases. | ||
+ | But these first attempts of Cooperative ITS services focus on their intrinsic added value and inevitably miss the point about their integration with other systems. For example, a newly developed red-light management system may perfectly work with the public transport vehicles, but what about the emergency vehicles if such interactions are not plan at the beginning ? | ||
+ | It is a common flaw in the emergent Cooperative ITS services to restrict by design their perimeter, sometimes without possible extension. It may be understandable as such systems focus on high availability and predictability in critical conditions. Interoperability and extensibility may not fit in this picture. But what could happen in a near future is such vision remains persistent in a context where ITS services are multiplying ? | ||
+ | |||
+ | === What not to do : define another silos-based architecture === | ||
+ | |||
+ | The Internet is built on an open architecture, not on what we call “silo-based | ||
+ | architectures” that we can witness in the Telephony or Digital TV world: | ||
+ | networks based on tailor-made protocols, controlled by a few big entities that | ||
+ | dictate what people and corporation can do on “their” network, having | ||
+ | difficulties to cope with the evolution of services. These models have proven | ||
+ | to be locked markets, having high barriers to entry which hinder new | ||
+ | competitors. They are walled gardens: at first, they look nice and safe, but | ||
+ | the given solutions will soon depreciate and make the platform stagnate. | ||
+ | |||
+ | If ITS were to build a new silo-based architecture, it would rapidly ossify and | ||
+ | prohibit greater growth and innovation. As the transport industry values | ||
+ | security above all, the silo approach might sound appealing, but this is not a | ||
+ | good solution for a long-term infrastructure project: the silo architectures | ||
+ | base their security on the secrets of their methods (security by obscurity), which is known to be a | ||
+ | very bad security practice in the long run. On the contrary, open Internet | ||
+ | technologies rely on widely and deeply studied security measures, all based on | ||
+ | a solid and future-proof protocol. | ||
+ | |||
+ | === Openness is the key for a long term ITS architecture === | ||
+ | |||
+ | The good property that Cooperative ITS should develop to protect itself from stagnation, business interlocking and deprecation is openness. Openness to other systems (not necessary ITS), openness to technology and security evolution, openness to third parties are the keys for the durability of Cooperative ITS systems. | ||
+ | An inspiring example of a long term cooperating system is Internet. Internet is open by design as it was born from the will of networks to cooperate. This basic property allows Internet 40 years after its creation to continue evolving, to improve its robustness and to be the ground for new business development. | ||
+ | Hopefully the core technology of Internet, the Internet Protocol (IP) is integrated into the Cooperative ITS standards. It has a valuable importance as IP fulfills an important requirement for an open cooperative system: each part of the system speaks the same protocol, without the need of intermediate gateways or translation boxes. The last version of the IP protocol, IPv6, was especially design to that end. IPv6 should become the common ground on which open Cooperative ITS services will be developed. | ||
+ | |||
+ | === IPv6 strengths vs. weaknesses === | ||
+ | |||
+ | At the center of IPv6 strength is the end-to-end principle: the network is as | ||
+ | simple as possible and the end-point provide the intelligence. The end-points | ||
+ | are directly addressable through their IP address, without relying one some | ||
+ | particular application-specific gateway in the middle. This way, we don't tie | ||
+ | the network to some particular application, but allow for any future innovation | ||
+ | to still use the same network basis. | ||
+ | |||
+ | The security is thus also end-to-end: each network node can and has to ensure | ||
+ | that its communications are secure. Far from being complicated, this principle | ||
+ | rather puts each node in control of the security of its communication: there is | ||
+ | no way for a middle man to eavesdrop. The trust doesn't lie in the network, | ||
+ | thus enabling a very stable and simple network design. This is also what makes | ||
+ | it scale very well: the Internet, and particularly IPv6, is the only | ||
+ | technology to support network nodes in the range of billions. | ||
+ | |||
+ | Having a stable core is also a guaranty that IPv6 will stay robust: it is | ||
+ | proven to work well and we can keep the network functions as-is for decades to | ||
+ | come. ITS can base their communications on it with the assurance that it will | ||
+ | still functions in the long run; it is the applications built on top of it that | ||
+ | will bring the services and innovations needed. | ||
+ | |||
+ | === IPv6 will fill the gap between ITS and IoT and IoServices === | ||
+ | |||
+ | IPv6 has proven to be a remarkably consistent link between the vast majority of telecommunication technologies. It gathers advantages from the historic Internet protocol (openness, security, flexibility, robustness, …) and comes with its own (addressing size, restoration of end-to-end, long-term perspective, …). | ||
+ | |||
+ | These features make IPv6 become the central technology for various fields (Core Internet, LTE in mobile wireless networks, 6LowPAN in sensor networks, …), already replacing in some cases the IPv4 protocol due to its resources shortage. | ||
+ | |||
+ | We are seeing an evolution which is placing IPv6 as a common denominator for all these domains, each time adapted to the particularities and constraints of the considered use-case. We call it "Evolution at the fringe". | ||
+ | |||
+ | [Schema "Evolution at the fringe"] | ||
+ | |||
+ | This evolution considers an unmodified IPv6 protocol running the Internet backbone (Core Internet), and an adapted IPv6 at the periphery (Fringe). | ||
+ | |||
+ | The best and most advanced example of a fringe protocol is 6LowPAN, which is an adaptation of the protocol for sensor networks. The protocol has been simplified to fit within the computing capabilities on sensors, compressed to adapt to the wireless link, extended to provide features mainly applicable to sensor networks. Despite all these modifications, the protocol itself remains compatible with the base protocol through the usage of a transparent gateway. This means that a sensor could communicate with any host on the Internet. | ||
+ | |||
+ | This evolution, applied to ITS, enables lots of interactions and bridges between domains: | ||
+ | * ITS networks with Home networks: Smart Home applied to Smart Vehicles ; | ||
+ | * ITS networks with Sensor networks: In-vehicle Sensor Networks ; | ||
+ | * ITS networks with Content networks: Infotainment. | ||
+ | By extension, ITS networks automatically becomes compatible with future use-cases or services (Smart cities, …). | ||
+ | |||
+ | Openness of IPv6 and Internet protocols in general is a driver for innovation. It has been for twenty years the winning recipe of the Internet. Its application to ITS provides the opportunity to benefit from all this environment, it being aimed at ITS or not. | ||
+ | |||
+ | At first, it makes existing Internet services for ITS integrate in a better way with the vehicular environment, and it expands it with new features. It also permits to services developed for another domain to be transposed to the ITS domain. Finally, it makes a lot easier for newcomers with new ideas to innovate on ITS, without constraining entry cost since they only have to comply to open Internet standards. | ||
+ | |||
+ | By its openness and its flexibility, IPv6 allows to fill the gap between ITS, Internet of Things, and Internet of Services. This is a key piece for allowing Internet services to fuel the mobility market. | ||
+ | |||
+ | === Standard bodies are cooperating on this convergence === | ||
+ | |||
+ | TODO: ET | ||