WO2012057607A1

WO2012057607A1 - Decision engine to achieve efficient route optimization in mobile networks

Info

Publication number: WO2012057607A1
Application number: PCT/MY2011/000120
Authority: WO
Inventors: Mohd Ariff Abdullah; Muhammad Faheem Mohd Ezani; Shariq Haseeb
Original assignee: Mimos Berhad
Priority date: 2010-10-29
Filing date: 2011-06-22
Publication date: 2012-05-03
Also published as: MY155123A

Abstract

A method to allow a more efficient routing in Network Mobility by means of a mobile router deciding on a more efficient route to be taken as it roams between networks comprising the steps of starting a delay counter when the mobile router performed a handoff from one network to another network, deciding whether performing a Route Optimization is necessary based on the delay counter, stripping of a parent subnet from a new Care-of-Address, comparing the parent subnet with home subnet, deciding whether the mobile router is performing a micromobility or macromobility, deciding whether the mobile router is roaming in its parent network or in/between foreign networks and selecting an appropriate Route Optimization method based on type of mobility performed.

Description

DECISION ENGINE TO ACHIEVE EFFICIENT ROUTE OPTIMIZATION IN

MOBILE NETWORKS

TECHNICAL FIELD The present invention relates a method to select the best Route Optimization (RO) solution in Network Mobility (NEMO) by looking at the state of the mobility performed by the Mobile Router (MR) . By examining the period of time, MR is connected to a certain network, and a parent subnet of that network, an appropriate RO method will be chosen by the MR. The options are not only limited to which RO solution to pick, but also whether RO is necessary in that situation.

BACKGROUND ART

Network Mobility (NEMO) Basic Support (IETF RFC 3963) defines a protocol that provides IPv6 mobility support for entire moving networks, where all data packets go through the IPv6-in-IPv6 tunnel established between the Mobile Router (MR) and the Home Agent (HA) . This can have severe consequences on the communication performances, as it causes data packets to follow a path that can be very far from optimal and requires a double IPv6 header for every packet exchanged with a Correspondent Node (CN) in the Internet. Compared with a communication that uses the ideal packet routing and the normal IPv6 header size, these factors result in an increased delay and a reduced throughput, plus indirect consequences like increased packet fragmentation and overall less efficient usage of resources .

Patent US 7453864 discloses a mobile ad hoc network which includes a plurality of wireless mobile nodes and a plurality of wireless communication links connecting the nodes together. The method for operating the network includes discovering and using routes in the network, predicting route failure in the network, and performing route maintenance in the network based upon the predicted route failure. However, the prior art is distinct over the present invention wherein the prior art predicts the failure in a running route optimization technique and select a new method before the failure occurs. However, the present invention introduces a decision engine to decide which route optimization technique which would give best result every time mobile router performs a handoff. In another example of prior art, patent number US 6895398 discloses a decision engine and method of its application. In this system, the decision engine includes a Bayesian network. The method includes retrieving data from a client database and forming a focus database; applying a set of initial rules to the focus database to form at least two nodes; applying a first learning process to determine a set of states to be applied within each node; applying a third learning process to determine a set of probabilities; and applying a set of fourth learning process to update a structure of the at least two nodes, the set of arcs, the set of states within each node, and the set of probabilities for the states. However, the present invention selects the best route optimization solution in network mobility by looking at the state of the mobility performed by the mobile router wherein an appropriate route optimization method will be chosen by the mobile router in a network.

Therefore, there exists a need for a decision engine to achieve efficient route optimization in mobile networks that will automatically select which Route Optimization (RO) solution should be used at a specific situation. DISCLOSURE OF THE INVENTION

The present invention aims to provide a method to select the best Route Optimization (RO) solution in Network Mobility (NEMO) by looking at the state of the mobility performed by the Mobile Router (MR) .

In a preferred embodiment of the present invention, a method to allow a more efficient routing in Network Mobility by means of a mobile router deciding on a more efficient route to be taken as it roams between networks comprising the steps of starting a delay counter when the mobile router performed a handoff from one network to another network, deciding whether performing a Route Optimization is necessary based on the delay counter, stripping of a parent subnet from a new Care-of-Address , comparing the parent subnet with home subnet, deciding whether the mobile router is performing a micromobility or macromobility, deciding whether the mobile router is roaming in its parent network or in/between foreign networks and selecting an appropriate Route Optimization method based on type of mobility performed. In another preferred embodiment of the present invention, the mobile router has a record of its home parent subnet that is used as a benchmark to determine whether the Care- of-Address belongs to the home parent network or foreign network.

In another preferred embodiment of the present invention, the mobile router has a predefined delay timer that is used to determine whether mobile router has stayed in a same network long enough to make Route Optimization method necessary .

In another preferred embodiment of the present invention, the mobile router may have more than one Route Optimization installed.

In another preferred embodiment of the present invention, the Route Optimization method may be running in the background, staying as idle processes or remaining inactive for some period of time.

In another preferred embodiment of the present invention, the mobile router can trigger any of the Route Optimization method at any time. The present invention consists of features and a combination of parts hereinafter fully described and illustrated in the accompanying drawings, it being understood that various changes in the details may be made without departing from the scope of the invention or sacrificing any of the advantages of the present invention. BRIEF DESCRIPTION OF THE ACCOMPANYING DRAWINGS

To further clarify various aspects of some embodiments of the present invention, a more particular description of the invention will be rendered by references to specific embodiments thereof, which are illustrated, in the appended drawings. It is appreciated that these drawings depict only typical embodiments of the invention and are therefore not to be considered limiting of its scope. The invention will be described and explained with additional specificity and detail through the accompanying drawings in which:

FIG. la illustrates a mobile network FIG. lb illustrates mobile network moves to a foreign network

FIG. lc illustrates route optimization in a network mobility

FIG. 2a illustrates MR roaming between sub networks FIG. 2b illustrates roaming between parent networks

FIG. 2c illustrates roaming inside a home parent network FIG. 2d illustrates roaming outside home network FIG.3 illustrates a stationary vehicle and a moving vehicle FIG. 4 shows a basic flowchart for a RO decision engine FIG. 5 shows detailed operation of RO engine

FIG. 6 shows example of decision engine implementation FIG. 7 shows complementary RO protocol FIG.8 shows dynamic RO protocol

DETAILED DESCRIPTION OF THE ACCOMPANYING DRAWINGS Now referring to FIG. la, the figure shows a Mobile Network (MoN) (100), represented by a Mobile Router (MR) (13) and two Mobile Network Nodes (MNN) (12), in which one of the MNN (12) is having a communication session with a Corresponding Node (CN) (16), residing in a remote network. In this figure the MoN (100) is connected to its Home Network (HoN) (10), which is shown by MR (13) and Home Agent (HA) (11) are connected by the same network link, which is the home link. According to FIG. lb, MoN (100) has changed its attachment point from HoN (10) to the Foreign Network (FoN) (15), which in this situation; MR (13) has been assigned a new address by the FoN (15) . In conventional networks, the Home Address (HoA) of MNN (12) which is being used to communicate with CN (16) as shown in FIG. la will be rendered invalid. This is due to the fact that CN (16) will send the communication packets to MNN's (12) HoA, which is located in the HoN (10) . Since MR (13) and MNNs (12) are no longer in HoN (10), the packets will be dropped.

Luckily in NEMO protocol, Home Agent (HA) (11) will intercept the packets and tunnel it to MR (13) . MR (13) will then forward the packets to MNN (12) . In this way, MNN ' s (12) HoA will be in valid regardless where the MoN (100) physically available. Since all the communication packets between MNN (12) and CN (16) must be tunneled between HA (11) and MR (13), the packets' route is, in most cases, not the best path possible.

According to FIG. lc, the packets between CN (16) and MNN (12) do not go through MR (13) -HA (11) tunnel, instead they are sent through an RO tunnel (17) . Although this might seems like a perfect solution, this situation is, at best, only theoretical in nature. There is no method or protocol that can achieve this ideal solution yet. Some methods and algorithm work best only in certain condition, while some others work only in certain type of network.

The present invention aims at giving a solution where RO (17) can be efficiently performed regardless of the mobility situation or network condition. However, this invention is not proposing a method to achieve RO (17) per se, but instead, a decision engine to determine which RO (17) method will be most efficient at a certain time, and in a certain condition.

There are three major factors that influence the efficiency of a routing path in a mobile network.

1. The hierarchical level of roaming performed by the mobile network 2. The parent network that the foreign subnetwork belongs to

3. The amount of time the mobile network spends in a particular network

Figure 2a shows an example when MR (13) performs handoff between two attachment points that belong to the same parent network (15) . This is an example of micromobility, where MR (13) retains the first 32 bits of its CoA. For example :

Previous CoA: 1111 : 2222 : 3333 : 444 :AAAA: BBBB: CCCC: DDDD

Current CoA: 1111 : 2222 : 3333 : 5555 :AAAA: BBBB: CCCC: DDDD

Next CoA: 1111 : 2222 : 6666 : 7777 : AAAA: BBBB : CCCC : DDDD 32-bit subnet = 1111 : 2222 : : /32

48-bit subnet 1 = 1111 : 2222 : 3333 :: /48

48-bit subnet 2 = 1111 : 2222 : 6666 :: /48

64-bit subnet la = 1111 : 2222 : 3333 : 444 :: /64

64-bit subnet lb = 1111 : 2222 : 3333 : 5555 :: /64

64-bit subnet 2a = 1111 : 2222 : 6666 : 7777 :: /64

64-bit Node Address = :: AAAA: BBBB : CCCC : DDDD

As we can see, the 32-bit subnet of 1111:2222: is always there no matter which attachment point MR (13) is connected to. Moreover, if MR (13) performs handoff between two points that belong to the same 48-bit subnet (22), then it will retain the first 48 bits (22) of its CoA.

In most cases, which are also the best case, attachment points are only wireless access points that do not have their own subnet. The address subnet is provided by a router that sits behind these access points. In this case, MR (13) will retain the first 64 bits (23) of its CoA. Since the node address is unlikely to change except by special router requirement, this kind of handoff does not change the CoA of MR at all. The handoff will only be performed at the link layer, involving MAC address registration to the access point.

Considering that CoA does not change that much when mobility is performed inside the same parent network, the need for a new RO tunnel (17) from scratch is not crucial since the changes in the path between MNN (12) and CN (16) will be less than five hops. And if the CoA does not change at all, the old RO tunnel (17) can still used after MR (13) has performed its layer 2 handoff.

Figure 2b shows an opposite situation of Figure 2a. In this figure, MR (13) performs handoff between two foreign networks, Network 1 (51) and Network 2 (52) . Since neither Network 1 (51) nor Network 2 (52) sits under each other, both of them are advertising two sets of completely different subnets. In this case, the old RO tunnel (17) is not usable anymore. MR (13) has to either sets up a new RO tunnel (17) or operates with only MR (13) -HA (11) bidirectional tunnel. In Figure 2c, oN (100) is roaming between subnets inside its home parent network, denoted by the presence of a HA (11) in the network link. In short, MoN (100), HoA and the subnets' routers under the parent network are belonging to one organization.

This is an example where more control is available over the network. In this case, the task of performing RO (17) can be distributed between the routers inside the network. Every router in the network can co-operate to achieve the optimized routing path for MNNs ' (12) communication packets .

In an RO method that utilizes dynamic routing algorithm, the intermediate routers will automatically update the forwarding interface for the Mobile Subnet (MoS) every time MR (13) moves from one subnet to another.

Different from the scenario in Figure 2c, Figure 2d shows a situation when MoN (100) is roaming inside a network that is not its home parent network. With the routers inside this network are belong to some other entity, MR (13) does not have the same type of control over the routers as in Figure 6. In this case, MR (13) has to bear all the responsibility of performing RO between MNN (12) and CN (16) .

Although in theory dynamic RO can be applied here, but that concept remains only as a technical possibility. The idea of having dynamic routing set up for MoS that belongs to somebody else is not a practical solution. Therefore, MR has to opt for other alternatives, such as complementary RO protocol .

Another factor that affects the efficiency of RO is the amount of time MR (13) spends inside a network. Figure 3 shows two vehicle, namely Vehicle A (111) and Vehicle B (110), which we assume mounted with MRs (13) and MNNs (12) inside. Taking other factors aside, there is still a big difference between a stationary and a moving vehicle. Vehicle A (111) stays inside the coverage of Network Tower A (114) for a long time while Vehicle B (110) changes attachment points between several network towers in short amount of time.

Initiating and implementing RO requires some packet exchanges and translation mechanism inside MR (13). Several hundred milliseconds will be lost during this time. That amount of time is acceptable to ensure a higher throughput during the communication sessions. However, for MR (13) that roams between networks in a high velocity, these hundreds of milliseconds are not a plus factor. Therefore, for a MoN inside a moving vehicle, RO is not favorable.

Taking all the above factors into consideration, it is necessary to have a decision engine that will automatically select which RO scheme best fit the situation.

Figure 4 shows a basic flowchart for the process of deciding which method of RO will be used when a condition is met. Firstly, when MR performs a handoff, it will check whether RO is necessary. If MR decides that RO is necessary, it will check whether it is roaming inside the parent HoN . If it is inside HoN, it will select a comptible RO method, which is Method A. If MR is not inside HoN, it will then check whether it had just performed a micromobility handoff. If the handoff is hierarchically micro, MR will select Method B for RO. If this condition is not satisfied, MR will conclude that it is roaming between two foreign parent network, which will be compatible with Method C. Figure 5 shows an expanded version of Figure 9 where more details have been put to show the inner working of the process .

The flow of the decision engine starts by the MR periodically checking change s in the subnet. Once it detects a change, MR will assign the previous 32-bit subnet into a buffer and then wait for a pre-determined amount of time, to see if a new change will occur. If new change in the subnet is detected, MR will assume that MoN is in a constant movement, therefore RO is not necessary.

If there is no further change in the subnet, MR will then check whether the current 32-bit subnet is its 32-bit Home Subnet (HoS) . If the two subnets are found to be of the same subnet, MR will use a MoS to perform a dynamic RO protocol, co-operating with the intermediate routers. If current 32-bit subnet and HoS are not the same, MR will employ a complementary RO protocol. MR will finally check whether the previous 32-bit subnet and current 32-bit subnet is the same subnet, which will show whether MR has performed a micromobility or a macromobility. This will determined how MR will perform the complementary RO protocol .

Figure 6 shows a simple example of an implementation of this invention. This flowchart depicts a module that is triggered by an event of handoff where the first thing the module would do is to get a new CoS . After a new CoS has been acquired, the module starts a 60-seconds countdown timer. If before the 60-seconds is over, MR had performed yet another handoff, the process will start all over again. If another handoff did not occur, the module will take the four Most Significant Byte (MSB) from CoS and assign it to the new 32-bit subnet. This subnet will be compared with the current subnet, which is in the cache. Similar subnet will divert to a protocol called DROP, and different subnet to a protocol called CROP which will change the current subnet to the new subnet afterwards. This module will be idle until it was triggered by the next handoff. Examples of complementary and dynamic RO protocols are shown in Figure 7 & Figure 8.

In the foregoing specification, specific embodiments of the present invention have been described. However, one of ordinary skill in the art will appreciate that various modifications and changes can be made without departing from the scope of the present invention as set forth in the various embodiments discussed above and the claims that follow. Accordingly, the specification and figures are to be regarded in an illustrative rather than a restrictive sense, and all such modifications are intended to be included within the scope of present invention. The benefits, advantages, solutions to problems, and any element (s) that may cause any benefit, advantage, or solution to occur or become more pronounced are not to be construed as a critical, required, or essential features or elements as described herein.

Claims

1. A method to allow a more efficient routing in Network Mobility by means of a mobile router deciding on a more efficient route to be taken as it roams between networks comprising the steps of:

starting a delay counter when the mobile router performed a handoff from one network to another network;

deciding whether performing a Route Optimization is necessary based on the delay counter;

stripping of a parent subnet from a new Care-of- Address ;

comparing the parent subnet with home subnet;

deciding whether the mobile router is performing a micromobility or macromobility;

deciding whether the mobile router is roaming in its parent network or in/between foreign networks; and selecting an appropriate Route Optimization method based on type of mobility performed.

2. The method according to claim 1, wherein the mobile router has a record of its home parent subnet that is used as a benchmark to determine whether the Care-of- Address belongs to the home parent network or foreign network .

A method according to claim 1, wherein the mobile router has a predefined delay timer that is used to determine whether mobile router has stayed in a same network long enough to make Route Optimization method necessary . A method according to claim 1, wherein the mobile router may have more than one Route Optimization installed .

A method according to claim 4, wherein the Route Optimization method may be running in the background, staying as idle processes or remaining inactive for some period of time.

A method according to claim 4, wherein the mobile router can trigger any of the Route Optimization method at any time.