Nautilus6 - Incremental Testbeds

Our testbeds are designed to develop the necessary protocol suite through a number of steps: specification, implementation, validation, demonstration, integration with the overall system architecture, and operational validation and evaluation.

Because each protocol required in our architecture are at various stages of their development process, we cannot demonstrate everything at once right now. Rather, we will go threw this process at the pace of each protocol's development. We are therefore developing incremental testbeds to match the level of development of each mobility technology we develop.

At the very early stage of the development of a particular protocol, we will implement and test any given new protocol on our in-door testbed. Mature implementations will then be demonstrated on a light-weight demonstration testbed to validate and demonstrate the new mobility features. The third stage is an operational testbed where we will carry on validation and evaluation of the protocol with the overall system already in place. The forth stage will be a large-scale real-conditions testbed we will build to convince business operators once the IPv6 mobile technologies is mature for real deployment.

Each protocol will be moved further up from the in-door testbed to the real-conditions testbed according to our progress in each of the activities highlighted earlier. Not to say, the testbeds will have to be adapted for each new protocol that is brought in.

So far, only host mobility has reached the necessary maturity to move on to an operational testbed while other protocols such as network mobility are only at the earliest stage of development. Our Indoor Testbed is thus designed to test the implementations of NEMO Basic Support and FMIPv6 under several configurations. Our E-Bike Demonstration Testbed and E-Wheelchair Demonstration Testbed are designed to demonstrate and validate NEMO Basic Support and multihoming mechanisms, while the Zaurus Operational Testbed is designed to validate and evaluate the operation of Mobile IPv6.

In the process of developing new protocols, finding the problems in existing solutions is one of the most important processes. Even though developed protocols should be tested in real situations, many problems can be discovered and addressed using a small in-door testbed.

Our indoor testbed is designed to validate our protocol implementations during the course of their development. It is currently configured to test a wide range of configurations. For network mobility, possible configurations include "multihomed mobile networks" (mobile network connected to the Internet via several mobile routers), and "nested mobile networks" (ability to bring a mobile node or a mobile network into another mobile network without breaking open connections by performing a combination of Mobile IPv6 and NEMO) and a combination of both. It has recently been extended to test seamless mobility protocols.

Light-weight demonstration testbeds are necessary to validate the new protocols developed by Nautilus under more complex situations, to demonstrate the new protocols, to demonstrate their combination with other existing IPv6 features, and to evaluate their efficiency.

Since the implementation of NEMO Basic Support has reached its completion, we have started to develop the E-Bike testbed and we are investigating ways to achieve a E-Wheelchair testbed targeted to disabled persons.

The E-bike testbed will be used to demonstrate the benefit of network mobility for all types of configurations (i.e. nested mobile networks and multihomed mobile network). The testbed is currently under development. We are waiting for the completion of the implementation of the NEMO Basic Support protocol currently tested in the indoor testbed to start the actual demonstration.

The E-Wheelchair testbed is a new plan under investigation. The objective would be to demonstrate the benefit of network mobility protocols, particularly NEMO, for health-care applications. Various IPv6 devices could be embedded in a wheelchair in a PAN-fashion. We expect to work in collaboration with a team of people working in relation with hospitals and doctors.

The concept of Layer 3 mobility protocols was established long time ago. There were a great effort for researching and developing such technologies. Especially, Mobile IPv6 is the oldest of such technologies and we have almost completed its standardization procedure after a long lasting discussion at the IETF. There are many well tested implementations of Mobile IPv6 already. However it is hard to say that layer 3 mobility technologies are deployed widely. We think that one of the reasons why mobility protocols aren't deployed widely is that most people don't know their usefulness.

The most efficient way to know how a technology is useful is to use the technology in a real environment. In the past few WIDE meetings, some efforts to try to make attendees of the meeting use the mobility technology were made. However such efforts were not so effective as a method to convince the attendees, since there were many restrictions. For example, attendees must install additional software extension modules to their computers, which is not acceptable for some people who are doing serious daily works on their computers. Another reason is that the network services for mobility is terminated when the meeting finishes. The attendees can use such advanced network features only during the meetings. Thus, we have concluded that it is necessary to provide mobility service as a general service to users so that they can use mobility features in their own networks everyday, to accelerate the deployment of mobility service.

In general, the word 'Mobility' doesn't mean only host mobility such as Mobile IPv6. 'Network mobility' is another important technology in the Nautilus project. However, compared to the host mobility technology, it is difficult to show the usefulness of network mobility in user's regular network because network mobility is useful for like vehicles with wireless networks installed on wide area. Of course, there is good applications for network mobility such as PAN (Personal Area Network), however, considering the current technical restrictions, it is hard to operate PAN.

We thus decided to start our Zaurus Operational experiment by providing host mobility functions as a first step.

The objectives of this experiment are

The goals of the first stage are: