GB2411549A

GB2411549A - Route discovery with quality of service check in ad hoc network

Info

Publication number: GB2411549A
Application number: GB0404445A
Authority: GB
Inventors: Zhong Fan
Original assignee: Toshiba Research Europe Ltd
Current assignee: Toshiba Europe Ltd
Priority date: 2004-02-27
Filing date: 2004-02-27
Publication date: 2005-08-31
Anticipated expiration: 2024-02-27
Also published as: GB0404445D0; GB2411549B

Abstract

The invention provides a method of admission control at an intermediate node between a source node 20 and a destination node 90 in an ad hoc network, e.g. a wireless ad hoc network, preferably a mobile ad hoc network (MANET). The source node 20 sends a route request message including quality of service requirement information to discover at least one route to a destination node 90 via one or more intermediate nodes 30, 40, 50, 60, 70. The destination node 90 then generates for at least one discovered route between the source node 20 and the destination node 90, a route reply message which includes the quality of service requirement information. The route reply message is sent to the source node 20 via the intermediate nodes defined in the discovered route (e.g. 50, 60). On receipt of a route reply message, each intermediate node determines whether it can meet the requirements specified by the quality of service requirement information and, if so, transmits the route reply message to the next node in the route; the node does not forward the route reply message if it determines that it cannot meet the requirement. Thus the source node 20 only receives replies from discovered routes which have recently determined that they satisfy the quality of service requirement information, which is advantageous in highly mobile ad hoc networks where up-to-date information is essential. A similar determination based on the forward transmission of the route request message (see fig. 2) may be used to ensure the destination node 90 only replies along routes that meet the quality requirements. The quality of service requirement determination may be based on more than one quality metric, e.g. the maximum available bandwidth at a node may vary depending on a MAC delay (fig. 5).

Description

2411 549

METHOD OF ROUTING IN AN AD HOC NETWORK

This invention relates to apparatus and a method for routing within an ad hoc network.

In particular, the invention relates to a method of determining a route from a source node to a destination node in a mobile ad hoc network comprising one or more mobile intermediate nodes between the source node and the destination node.

An ad hoc network is a network in which there is no central controller and wherein communications devices in the network are arranged to be capable of communication with one or more of the other devices in the network. A mobile ad hoc network (MANET) is an ad hoc network in which a plurality of devices are capable of wireless communication with one another, and in which one or more of the devices are capable of being physically moved relative to each other. It will be appreciated that a mobile ad hoc network may include one or more relatively immobile devices, and may include such devices as will allow connection to a wider area network, such as the Internet.

A communications network is considered to possess a topology, which is a map or graph, in mathematical terms, of nodes (representing the communications devices) of the network and edges between the nodes (representing the communications links established between the devices). In an ad hoc network, and in particular in a mobile network there is no predetermined topology. For example, in use of the network, a device may move away from another device with which it has established wireless communication, to the extent that the communications range of one or other of the devices constituting the nodes of the network is exceeded, and thus the communications link is lost.

In an ad hoc network, and in particular in a MANET, comprising a plurality of mobile communications devices, each device may be in communication with only some of the other devices of the network. This may be because of limitations on the range of wireless communication of the device in question. For example, if a mobile ad hoc network is established using Bluetooth communications technology (Bluetooth is a trade mark of Bluetooth SIC Inc.), the range, within which communication between two devices can be reliably established, is in the region of ten metres.

In these circumstances at least one communications device in the network needs to be operable to route data not intended for it but intended for other devices with which it may be directly or indirectly in communication. In other words, such a device operates as a router. A router, in such a system, will operate to receive a packet of data from a first device with which it has established communication, and will pass this packet of data to a second device, with which it has wireless communication. This second device may be the intended recipient, or may be a further router in the network, operable to pass the packet of data further towards its intended recipient device.

This routing operability is useful where, on the one hand, first and second devices of a network are not capable of direct wireless communication (whether for reasons of range, or otherwise) or, on the other hand, the quality of service available through a wireless communication link between the first and second devices is inferior to the quality of service available through using the indirect path via the intermediate device, in communication with the first and second devices and acting as a router.

Where a network is considered which comprises only three nodes, namely first and second devices not in communication with each other, and an intermediate device in communication with both of the first and second devices, it is straightforward to establish how packets of information should be transferred from the first device to the second device, as no alternative path exists. However, in larger networks, comprising a larger plurality of devices and thus a more complex topology, there can be several alternative paths between a first device and a second device, routed via a plurality of intermediate devices.

As there is no central control of the network, and the organisation and control of the communication of data between devices in the network is distributed amongst the devices, there is a requirement also for distributed control of topology and the manner in which data is passed from a first device to a second device where communication can only be effected via one or more intermediate devices.

The structure of a communications protocol is commonly determined in the context of an open system interconnection (OSI) model, which defines a networking framework for implementing protocols in terms of seven layers. In a node, communication of a packet of data on a communications channel to another node, is established by passing packet of data through seven layers of processing structure within the node. Then, at the other node, a received packet of data is passed through a corresponding seven layers, in reverse, to enable handling of the data at the other node.

The structure of the OSI model is well documented. Layer I of the OSI model is a physical layer, which conveys data physically through the network. Thus, in a wireless network, it is implemented by means of a wireless transceiver. Directly above this, passing data for transmission and receiving data from the transceiver, is a data link layer, which encodes and decodes packets of data into bits. The data link layer is further divided into two sub layers, namely a media access control (MAC) layer and a logical link control (LLC) layer. The MAC sub layer, in use, controls the manner in which a node gains access to data in the network, and gains permission to transmit data.

The LLC sub layer controls frame synchronization, flow control and error checking of data. As such, the MAC sub-layer records information relating to the quality of service that is achievable in communications channels established by the node.

The operation of the other layers of the OSI model has no impact on the understanding of the present invention.

Routing protocols are used to discover routes between two nodes in an ad hoc network.

Routing protocols can be broadly classified into two categories: ondemand (reactive) routing or proactive routing. On-demand protocols collect routing information only when necessary, and do not maintain unused routes, whereas pro-active protocols maintain optimal routes to all destinations at all times. When compared to proactive routing protocols, on-demand routing protocols are more efficient by minimizing control overhead and power consumption, since routes are only established when required. Two examples of on-demand protocols that are currently under active development in the IETF MANET working group are the AODV (ad hoc on-demand distance-vector) protocol disclosed in "Ad hoc on-demand distance vector (AODV) routing", by C. Perkins et al., Internet Draft, 2002, http://people.nokia.net/charliep/txt/aodvid/aodvid. txt and DSR (dynamic source routing) protocol disclosed in "The dynamic source routing protocol for mobile ad hoc networks (DSR)" by D. Johnson et al.,, Internet Draft, 2002.

http://www.monarch.cs.rice.edu/internet-drafts/draft-ietf-manet-dsr-07. txt In an on-demand protocol such as AODV, when a source node requires to communicate with a destination node that it is not in direct communication with, the a request from the source node is broadcast to the destination nodes via the intermediate nodes. The request is teemed a Route Request (RREQ) message, as it indicates that the source node is searching for a route to the destination node. The RREQ message is passed from intermediate node to intermediate node until it reaches the destination node. At each intermediate node the RREQ message is updated to include information about the path taken from the source.

On receipt of the RREQ message, the destination node responds by unicasting a Route Reply (RREP) message to the source node via the intermediate nodes in the route.

Information obtained through RREQ and RREP messages is kept with other routing information in the route table of each node.

These conventional protocols therefore enable a route to be discovered from a source node to a destination node via one or more intermediate nodes.

These protocols may establish that there are several paths between a source node and a destination node. A "best effort" approach is then used to select a route from the source node to the destination node from the available routes. Typically, the number of hops is the main criteria in such a best effort approach. In other words, the route with the least amount of hops between the source node and the destination node is selected as the transmission route. However, such protocols do not address the quality of service of the routes, and the quality of service of the links in each node may be very different.

Therefore a best effort approach is not sufficient to detennine a route if the transmission of data between the source node and the destination node requires stringent quality of service guarantees.

For many applications of wireless ad hoc networks, the quality of service of the link between the nodes in a route is extremely important. For example, any application in a wireless ad hoc network that requires the real-time traffic of data necessitates stringent quality of service requirements. An example of this is UDP based video or voice communications, which are associated with very strict bandwidth requirements, as the links between the nodes must be able to support the real time flow of substantial amounts of data. Delay is another quality of service metric that has a large bearing on the performance of a link for the transfer of real time data such as video and voice data.

For example, a delay of over 150 ms is considered by most users to be unacceptable for voice transmissions.

Therefore, quality of service is very important in some applications such as real-time high-quality audio/video applications, such as those envisaged in home entertainment networks but also in many other applications. For example, a display device providing a service of displaying a real time streamed video signal. The device generating the video signal may seek out a display device over the local ad-hoc network. The above described routing protocols such as AODV would be able to determine the available routes between the device and the display, but they would not be able to determine whether any of the discovered routes have the available bandwidth to support the transfer of the real time streamed video signal over the links between the nodes in the route.

At the MAC layer, the dynamic nature of ad hoc networks makes it difficult to assign a central controller to maintain connection states and reservations. Therefore "best-effort" distributed MAC controllers (such as the IEEE 802.11 Distributed Coordination Function) are widely used in existing ad hoc networks. Recently a number of distributed control schemes have been proposed to support service differentiation at the MAC layer. An example of such a control scheme is "Differentiation mechanisms for IEEE 802.11", by Aad and C. Castelluccia, IEEE lnfocom'01, Anchorage, Alaska, April 2001, which can be found at http://www.ieee-infocom.ora/2001/paper/214.ps At the network layer, most of the ad hoc routing schemes proposed so far are also best- effort. In other words, they have no quality of service support. "Improving UDP and TCP performance in mobile ad hoc networks with INSIGNIA" by S.-B. Lee et al., IEEE Communications Magazine, June 2001, which can be found at http://comet.columbia.edu/insienia/commag2001.pdf, discloses a system often referred to as "INSIGNIA".

"INSIGNIA" discloses an in-band signalling system that supports adaptive reservation- based services in ad hoc networks. It discloses a general-purpose approach to delivering quality of service (mainly the signalling aspect), but does not address the issue of admission control.

To support quality of service requirements of real-time applications, various traffic control mechanisms such as rate control and admission control are needed.

"Distributed call admission control for ad hoc networks" by S. Valace and B. Li, IEEE VTC'02, 2002, which can be found at http://www.eecg.toronto. edu/bli/papers/vtcO2.pdf, discloses a distributed call admission controller. This controller uses the concept of service curve provisioning, and the drawback of this scheme is that it is difficult to accurately measure the service curve of the network. Furthermore, the admission criterion that is disclosed in this paper is the deterministic universal service curve which can be very conservative.

W003094025 discloses an on-demand routing protocol in which intermediate nodes between a source node and a destination node in an ad hoc network perform a degree of admission control. As for other on-demand protocols, a route request is sent from a source node to a destination node, and this route request contains a quality of service parameter. On receipt of a route request, a node calculates a quality of service tag value, which is a function of a quality of service parameter specific to that node. The node then detennines whether to admit traffic in response to the route request based upon the calculated quality of service tag value and the quality of service parameter in the route request.

Thus WO 03094025 discloses a method of admission control, as nodes are only allowed to admit traffic if a calculated quality of service tag value meets a quality of service parameter in a route request.

However, all the above systems suffer from the disadvantage that they do not adequately provide for the quality of service requirements of mobile ad hoc networks.

According to a first aspect of the invention there is provided a method of admission control at an intermediate node between a source node and a destination node in a wireless ad hoc network, wherein: the source node is arranged to send a route request message including quality of service requirement information for determining at least one route to a destination node via one or more intermediate nodes; and the destination node is arranged to generate a route reply message for at least one discovered route between the source node and the destination node, each route reply message including the quality of service requirement information, and to send each route reply to the source node via the one or more intermediate nodes in that route; the method comprising, at an intermediate node: receiving a route reply message; and determining whether the intermediate node can meet the requirements specified by the quality of service requirement information and, if so, transmitting a further route reply message to the next node in the route.

Such a method involves "piggy backing" quality of service requirement information onto the routing messages of a conventional on-demand routing protocol. In a conventional routing protocol such as AODV, routes are established from the source node to the destination node. However, there is no consideration of the quality of service requirements of the source in the route determination steps, and therefore there is no admission control. The present invention provides admission control at intermediate nodes on receipt of a route reply message, i.e. on the backward path from the destination node to the source node. Therefore, the source node will only receive route reply messages from routes in which the quality of service requirements of those routes have been checked as recently as possible.

In some embodiments an intermediate node can receive a route request message and determine whether the intermediate node can meet the requirements specified by the quality of service requirement information in the route request message. If so, the intermediate node can transmit a further route request message.

Such embodiments therefore provide admission control at intermediate nodes on receipt of route request messages as well and on receipt of route reply messages, i.e. the forward path as well as the backward path. This has the advantage of reducing control overhead, as routes that do not meet the quality of service requirements of the source node on the forward path are dropped by the intermediate and not considered further.

The method can comprise at an intermediate node, on receipt of a route reply message, measuring a quality of service metric for the intermediate node; generating quality of service data based on the measured quality of service metric; and including the quality of service data in the further route reply message.

Therefore the intermediate nodes in such embodiments update the route reply message to include data relating to a quality of service metric measured at that node. This data could be used later by the source node to determine which route to choose.

The quality of service requirement information can include one or more sets of data, each of which specifies a quality of service criteria used by the intermediate nodes in the determining steps.

The quality of service requirement information can comprise a first quality of service metric relating to a resource requirement that must be available at the intermediate node for the intermediate node to meet the requirements specified by the quality of service requirement information, and the method can further comprise at the intermediate node: determining whether the intermediate node meets the requirements specified by the quality of service requirement information by calculating a value of the first quality of service metric on the basis of establishing the amount of resources currently allocated and a variable threshold value; and wherein the variable threshold value is determined based upon measuring a second quality of service metric.

Such a method can be used to provide adaptive admission control that can be used to vary a threshold value of a first quality of service metric based on a measured value of a second quality of service metric. For example, such a method could be used to vary the estimate of a maximum bandwidth threshold of a link according to measured MAC delay conditions The network can be a mobile wireless ad-hoc network.

According to a second aspect of the invention, there is provided a device arranged to act as an intermediate node between a source node and a destination node in a wireless ad hoc network, in which: the source node is arranged to send a route request message including quality of service requirement information for determining at least one route to a destination node via one or more intermediate nodes; and the destination node is arranged to generate a route reply message for at least one discovered route between the source node and the destination node, each route reply message including the quality of service requirement information for that route, and to send each route reply to the source node via the one or more intermediate nodes in that route; the intermediate node comprising: a receiver arranged to receive a route reply message; and a processor arranged to determine whether the intermediate node can meet the requirements specified by the quality of service requirement information; and a transmitter arranged to transmit a further route reply message to the next node in the route if the processor determines that the intermediate node can meet the requirements specified by the quality of service requirement information.

In some embodiments, the receiver is further arranged receive a route request message; and the processor is further arranged to detennine whether the intermediate node can meet the requirements specified by the quality of service requirement information in the route request message; and the transmitter is further arranged to transmit a further route request.

According to a third aspect of the invention, there is provided a method of determining a route from a source node to a destination node in an ad hoc network comprising one or more intermediate nodes between the source node and the destination node, the method comprising, at the source node: transmitting a route request message, the route request message including quality of service requirement information, at the destination node: receiving a route request message for each discovered route between the source node and the destination node; generating a route reply message for at least one discovered route between the source node and the destination node, each route reply message including the quality of service requirement information; and transmitting each route reply message to the source node via the one or more intermediate nodes in that route, at an intermediate node: receiving a route reply message; and determining whether the intermediate node can meet the requirements specified by the quality of service requirement information and, if so, transmitting a further route request message to the next node in the route, and at the source node: receiving one or more route reply messages; and determining which route to use on the basis of the received route reply messages.

The method can further comprise, at an intermediate node: receiving a route request message; and determining whether the intermediate node can meet the requirements specified by the quality of service requirement information in the route request message and, if so, transmitting a further route request message to one of the other intermediate nodes or the destination node.

At an intermediate node on receipt of a route reply message, the method can comprise: measuring a quality of service metric for the intermediate node; generating quality of service data based on the measured quality of service metric; and including the quality of service data in the further route reply message; at the source node: determining which route to use on the basis of the node specific quality of service data in the received route reply messages.

According to a third aspect of the invention, there is provided a wireless ad hoc network comprising, a source node comprising: a transmitter arranged to transmit a route request message, the route request message including quality of service requirement information, a destination node comprising: a receiver arranged to receive a route request message for each discovered route via one or more intermediate nodes between the source node and the destination node; a processor arranged to generate a route reply message for at least one discovered route between the source node and the destination node, each route reply message including the quality of service requirement information; and a transmitter arranged to transmit at least one route reply message to the source node via the one or more intermediate nodes in that route, and at least one intermediate node comprising: a receiver arranged to receive a route request message; and a processor arranged to determine whether the intermediate node can meet the requirements specified by the quality of service requirement information in the route request; and a transmitter arranged to transmit the a further route reply message to the next node in the route if the processor determines that the intermediate node can meet the requirements specified by the quality of service requirement information.

In some embodiments, the receiver of the intermediate node is further arranged receive a route request message; and the processor of the intermediate node is further arranged to determining whether the intermediate node can meet the requirements specified by the quality of service requirement information in the route request; and the transmitter of the intermediate node is further arranged to transmit a further route request.

The network can be a mobile wireless ad-hoc network.

Several embodiment of the invention will now be described by way of example only, with reference to the accompanying drawings, in which: Figure 1 illustrates a wireless mobile ad hoc network in accordance with a first embodiment; Figure 2 shows a flow chart of a first part of the route discovery process according to the first embodiment; Figure 3 shows a flow chart of a second part of the route discovery process according to the first embodiment; Figure 4 illustrates a wireless mobile ad hoc network in accordance with a second embodiment; Figure 5 shows a flow chart of the process of adaptive admission control; Figure 6 is graph of packet delay against number of sources; and Figure 7 is graph of throughput efficiency against number of sources.

Figures 6 and 7 show graphs that were obtained from an analytical model of 802.11 MAC. These graphs show that both throughput (available bandwidth) efficiency and packet delay deteriorate considerably as the number of traffic sources increase.

Therefore, admission control is necessary to maintain quality of service performance within an acceptable region.

Figure I shows a wireless ad hoc network of four nodes: a source node 2, node A 6, node B 10, and a destination node 14.

The nodes could be any devices capable of communicating wirelessly to form an ad hoc network. It will be appreciated by those skilled in the art that the nodes could comprise: laptop computers, personal digital assistants (PDAs), mobile phones, or any other such devices. Although the nodes communicate wirelessly in this embodiment, one or more of the nodes could be capable of communicating with each other over a wired link.

The source node 2 wishes to communicate over the wireless ad hoc network with the destination node 14. However, the source node 2 is not in wireless communications range with the destination node 14.

In this embodiment, any on-demand routing protocol such as AODV or DSR can be used as the basis for the routing protocol. However, quality of service requirement information is included in the routing messages, so that intermediate nodes between the source and destination nodes can perform admission control.

A routing mechanism used in the first embodiment will be now be explained with reference to Figures 2 and 3.

At step S 1 the source node commences the process of trying to establish a route with the destination node 14, and transmits a Route Request message (RREQ).

In a conventional on-demand routing protocol such as AODV or DSR, the RREQ messages comprise a number of fields, and it will be appreciated that the exact format of the fields will depend on the particular routing protocol. However, in contrast to a conventional on-demand routing protocol, the RREQ message used in embodiments of the invention contain extra fields, i.e. extensions, that relate to quality of service requirements. Therefore, it could be considered that embodiments "piggy back" quality of service requirement information onto the routing messages of a conventional on- demand routing protocol.

In this embodiment the quality of service requirements could relate to the minimum amount of bandwidth that must be made available along an acceptable path from the source node 2 to the destination node 14 via the intermediate nodes in the route. Thus the RREQ message comprises a requested bandwidth extension. However, the quality of service requirements could comprise one or more different quality of service metrics, including: available power, available bandwidth at the node, error rate, MAC delay, buffer occupancy, or packet loss.

At step S2, node A 6 receives the RREQ message over communications link 4.

When any node receives a RREQ message containing a quality of service requirement extension, it determines whether it can meet that quality of service requirement at step S3. For example, if the quality of service requirement in the RREQ message specifies a minimum amount of bandwidth that must be made available along an acceptable path from the source node 2 to the destination node 14, then the node 6 subtracts the current used bandwidth (i.e. the bandwidth that its current wireless communications is taking up) from the total bandwidth capacity at that node. This would require the node measuring a value of its current used bandwidth, whichcould be performed periodically and stored in an appropriate data store.

If node A 6 cannot meet the quality of service requirements specified in the RREQ message, then the node proceeds to S4 and does not forward the RREQ message on to another node, and does not process it further.

However, if node A 6 determines that it can meet the quality of service requirements, then node A 6 transmits the RREQ message at step S5. It will be appreciated that node A 6 may perform other processing as required by the particular on-demand routing protocol used as the basis of the routing protocol before transmitting the RREQ message.

If at step S6, the node that receives the RREQ message is the destination node 14, then the protocol proceeds to step S7. However, if the node cannot transmit the RREQ message directly to the destination node 14, then the RREQ message would be received by the next intermediate node.

Therefore, node A 6 transmits the RREQ message to node B 10 over link 8. Node B 10 then determines whether it can meet the quality of service requirements comprised in the RREQ message (step S3). If node B can meet the quality of service requirements, it then transmits the RREQ message, which would be received by the destination node 14 overlink 12.

The RREQ message is then processed by the destination node 14. It will be appreciated that, as for a conventional on-demand routing protocol, the RREQ message would have been updated on its path from the source node 2 to the destination node 14 to include information relating to the path of nodes from the source to the destination. Therefore, at step S7 the RREQ message may contain information that indicates that it came from the source node to the destination node 14 via node A 6 (over link 4) and node B (over link 8) and finally to the destination node 14 (over link 12) . This information enables the destination to unicast a reply back to the source.

Only one path from the source node 2 to the destination node l 4 is present in Figure 1.

Therefore, the destination node 14 would receive only one RREQ message. However, it will be appreciated that in a practical implementation of the invention, many nodes could be present in the wireless network, and there could be many routes between the source node 2 and the destination node 14. In this situation, the destination would receive a RREQ message from the source node 2 via every path whose intermediate nodes can each meet the quality of service requirements in the RREQ messages. In this situation, the RREQ message corresponding to each route from the source node would contain information relating to the sequence of nodes from the source node. Therefore, if the source node 2 broadcasts a single RREQ message, as many RREQ messages will be received by the destination node l 4 as there are available routes with nodes that meet the quality of service requirements. Furthermore, each RREQ message received from the destination node 14 would be different, as they would specify a different route from the source node 2 to the destination node 14.

At step S8 of Figure 3, the destination node 14 generates at Route Reply (RREP) message. As for a RREQ message, in a conventional on-demand routing protocol RREP messages comprise a number of fields. However, in this embodiment the RREP messages contain extra fields, i.e. extensions, that relate to quality of service requirements. These would include the same set of requirements that were included in the RREP message. However, additional quality of service metrics or fields could be included, as will be explained in more detail below.

The RREP message is then unicast back to the source node 2 via the intermediate nodes B and A. At step S9, node B 10 receives the RREP message over link 12 and determines whether it can still meet the quality of service requirements. if the RREP message includes additional quality of service metrics that were not present in the RREQ, then node B 10 also determines whether it can support the additional quality of service metrics.

If node B 10 can no longer meet the quality of service requirements, then node B 10 proceeds to step S 10, and does not forward the RREP message to node B 10.

Furthermore, node B 10 could send an error message to the source node 2, as required by the conventional routing protocol used as a base.

If node B 10 can still meet the quality of service requirements, then node B 10 transmits the RREP message to node A 6 over link 8 and steps S I I and S 12 are performed. Node A 6 then performs the step of determining whether it can still meet the quality of service requirements. If it can still meet the requirements, it forwards the RREP message to the source node 2 over link 4.

Therefore, as a result of the above protocol, the source node 2 has established that there is a route from the source node 2 to the destination node 14 via a path from node A 6 to node B. Furthermore, the source node 2 can be assured that the route supports its quality of service requirements. The source node 2 can then determine which of the routes to use on the basis of the received RREP messages (step S 13).

The quality of service requirement information included in extensions to the RREQ and the RREP messages are used by the intermediate nodes A and B to perform admission control. Only nodes that can meet the quality of service requirements pass on the routing messages, and therefore only routes that meet the quality of service specified by the source node 2 are set up.

On this basis, the above protocol combines route discovery with admission control in a way that is particularly suited to the requirements of a wireless mobile ad hoc network.

In a conventional routing protocol such as AODV, routes are established from the source node to the destination node. However, there is no consideration of the quality of service requirements of the source in the route determination steps, and therefore there is no admission control. Therefore, the source receives RREP messages via all the available routes. A best-effort approach is then used to determine which route to use, which for example could be done on the basis on the route with the fewest hops.

However, in the protocol described in relation to figures 2 and 3, the source node 2 is only informed about those routes that meet the quality of service criteria of the source node 2. One of the advantages of this is that, when determining which route to choose at S 13, the source node 2 only needs to consider those routes that can meet its quality of service requirements.

Routing requests such as RREP and RREQ messages are control packets, and therefore contribute to the control overhead of a wireless network. Control overhead is defined as the ratio of the number of control packets processed to the total number of packets.

Reducing the number of routing messages has the effect of reducing the number of control packets, and hence lowering the control overhead.

As a result of the intermediate nodes dropping RREQ messages if they do not meet the specified quality of service requirements, routes that have no chance of meeting the quality of service requirements of the source node 2 are not considered in the route finding process, and therefore the control overhead of the protocol is reduced.

Therefore dropping routing messages on the forward path from a source node to a destination node has the effect of reducing control overhead. However, this alone has been found not to be the optimum method for use in many mobile ad hoc networks. This is because, in many ad hoc networks, the nodes may be highly mobile. The mobility of the nodes could clearly affect the communications links between them. For example, the distance of two nodes from each other will affect many of the quality of service metrics associated with the link between those nodes. Furthermore, many quality of service metrics (such as delay) are greatly affected by the amount of network traffic. For example, MAC delay is affected to a large extent by congestion hotspots in an ad hoc network.

Therefore, the quality of service metric of each node could be highly variable within short time spans, and could depend on a number of dynamic parameters. On this basis, it is very important that the source receives as up to date information about the quality of service of the discovered routes as possible.

If the intermediate nodes only check whether they meet the quality of service requirements on the forward path from the source node 2 to the destination node 14, then there is a significant risk that one or more routes of the discovered routes will not be able to handle the quality of service requirements of the source node 2 by the time the source node 2 receives the RREP message for those routes.

If a source node receives a RREP message via a route that is no longer suitable for its quality of service requirements, and then determines to use this route, significant degradations in the performance of the network could arise. This is because if the source node begins transmitting real-time packets under the assumption that resources are available to meet the needs of this flow of data, excessive delays in the real-time traffic will arise.

For example, the source node and the destination node could comprise mobile phones, and the real time traffic could be voice transmissions. If a route is chosen under the false assumption that it can meet the bandwidth requirements for this voice transmission, then this may cause unacceptable packet delays. Furthermore, the other traffic passing through the intermediate nodes may also be subject to large delays due to the intermediate nodes running out of available bandwidth.

On this basis, this embodiment double checks the quality of service requirements, by the intermediate nodes dropping RREP messages if they cannot meet the quality of service requirements on the backwards path to the source node 2, as well as dropping RREQ messages on the forward path to the destination node 14.

The advantage of the intermediate nodes double-checking the quality of service requirements is to avoid false admission due to node mobility and other network dynamics. For example, in the forward phase of route discovery the resources along the path may available. However after a short while, some of the resources along the path may become unavailable due to node movement or due to an increase in traffic.

By having the intermediate nodes establish whether they still meet the quality of service requirements of the source node on the backwards path, then the problem of node mobility and sudden traffic increases can be alleviated to some extent. This is because, the source node 2 only receives RREP messages via routes that it has been established meet its quality of service requirements as recently as possible.

Furthermore, in other embodiments of the invention, the intermediate nodes only establish whether they meet the quality of service requirements of the source node on the backward path from the destination node 14 to the source node 2, i.e. in response to RREP messages only. In such situations, the source node 2 transmits a RREQ message, containing quality of service requirement information. However, on the forward path of the route discovery process (i.e. from the source node 2 to the destination node 14) the intermediate nodes would not consider their own quality of service metrics and will transmit the RREQ message on to the next intermediate node or the destination node 14 without determining whether they meet the quality of service requirements of the source node 2. On this basis, considering Figure 2, the intermediate nodes A and B would not carry out steps S3, S4 or S5.

Therefore, at step S7 the destination node 14 will have received a RREQ message from all the available paths from the source node 2 to the destination node 14. As the quality of service requirements of the source node 2 have not yet been considered up to this point, then some of these routes may not be able to support the quality of service requirements of the source node 2. However, the processing load on each node is reduced, as it simply needs to pass on the RREQ message on with the minimum of processing to update the route information.

However, the quality of service requirements are then considered in the backward path from the destination node 14 to the source node 2, and the method shown in Figure 3 is then carried out. Therefore, the destination node 14 sends out a RREP message, and the intermediate nodes in a route drop the RREP message for that route if they cannot meet the specified quality of service requirements.

The advantage of the intermediate nodes only checking if they meet the quality of service requirements on the backward path, as opposed to only on the forward path, is that the source node 2 will only receive RREP messages from routes in which the quality of service requirements of those routes have been checked as recently as possible.

Furthermore, methods that involve the intermediate nodes only checking whether they meet the quality of service requirements on the backward path have the advantage over methods in which the quality of service requirements are checked on both the forward and backward path, in that they ensure that routes are only dropped just before the source node 2 transmits data over the route.

This is advantageous in some circumstances because, in a method in which the intermediate nodes check whether they meet the quality of service requirements on both the forward and backward path, it is possible that some routes are rejected on the forward path that could become useable by the time that the RREP message flows along the backwards path.

For example, a route could be rejected on the forward path as a result of a poor link due a large distance between two particular intermediate nodes in that route. However, the nodes in the network could move such that this route becomes the optimum route by the time that the destination node would have received a RREP message via this route had the route not been dropped on the forward path.

In embodiments in which the intermediate nodes only check whether they meet the quality of service requirements on the backward path, the destination node receives a RREP message from every available route *om the source node 2 to the destination node 14, regardless of whether the intermediate nodes in that route meet the quality of service requirements. Therefore in the above example, after the nodes have moved positions (consequently improving the poor link between the two particular intermediate nodes), the RREP message will not be dropped and the source node 2 will receive a RREP message for this route.

Therefore, embodiments that only check whether the intermediate nodes meet the quality of service requirements on the backward path have the advantage of ensuring that the information received at the source node 2 most accurately represent the current status of the network. This can be a very important consideration in mobile ad hoc networks. However, because such embodiments involve no admission control on the forward path, there is an increase in the control overhead. This is due to the increased number of RREQ packets resulting from the forward path to the destination node, as a RREP message will be received message for every available route. However, this is Offset by the reduction in the amount of processing needed at the intermediate nodes, as they only need to establish whether they meet the quality of service requirements on the backward path.

Figure 4 shows a wireless network according to a second embodiment.

The wireless ad hoc network shown in Figure 4 comprises eight nodes: a source node 20, node E 30, node F 40, node G 50, node H 60, node X 70, node Y 80, and a destination node 90.

In this embodiment, there are shown three paths from the source node 20 to the designation node 90. The paths are: i) source node 20 to node E 30 via link 23, node E 30 to node F 40 via link 34, and node F 40 to the destination node 90 via link 49; ii) source node 20 to node G 50 via link 25, node G 50 to node H 60 via link 56, and node H 60 to the destination node 90 via link 69; and iii) source node 20 to node X 70 via link 27, node X 70 to node Y 80 via link 78, and node Y 80 to the destination node 90 via link 89; However, as for the network shown in Figure 1, it will be appreciated that in a practical implementation of the invention, many nodes could be present in the wireless network, and there could be many routes between the source node 20 and the destination node 90.

Table I shows a table of the available bandwidth and the delay over the links between the nodes shown in Figure 3.

Table 1.

Link Available Bandwidth / Mbps MAC delay / ms | source node to node E 2 2 node E to node F 1 6 node F to destination node 2 2 source node to node G 2 2 node G to node H 3 2 node H to destination node 3 3 source node to node X 2 2 node X to node Y 2 1 node Y to destination node 1.5 2 In order to commence the route discovery process, the source node 20 transmits a RREQ message that includes quality of service requirement information included as one or more extensions to a conventional RREQ message. As for the previously described embodiments, this embodiment uses a conventional on-demand routing protocol such as ADOV as the framework for the protocol, but piggy backs quality of service information on to the routing messages to enable the intermediate nodes to perform distributed admission control.

In this embodiment, the quality of service requirement information includes a minimum amount of bandwidth that must be made available along an acceptable path from the source node 20 to the destination node 90. In this embodiment, the minimum amount of bandwidth specified in the quality of service requirement information is 1.5 Mbps.

Nodes E,G and X receive the RREQ message from the source node 20. These nodes will then measure their available bandwidth. The available bandwidth at each node in the three possible routes from the source node 20 to the destination node 90 is shown in

Table 1.

Node E30will drop the RREQ message as it does not have enough available bandwidth to node F 40. However, nodes G and X can support the quality of service requirement in the RREQ message, and therefore continue processing the RREQ message as specified in the best-effort ADOV on- demand routing protocol on which this embodiment is based.

Node G 50 will transmit its RREQ message to node H 60, at which point node H 60 measure its available bandwidth to establish whether it can support the quality of service bandwidth requirement. As shown in Table 1, node H 60 can support the quality of service requirement, and so node H 60 transmits the RREQ message, which will be received by the destination node 90.

Similarly, node X 70 will transmit its RREQ message to node Y 80. As shown in Table I, node Y 80 can support the quality of service bandwidth requirement, and so node Y will transmit its RREQ message, which will be received by the destination node 90 At this point the destination node will have received two RREQ messages, one having come from the source via path ii) and the other via path iii). The two RREQ messages will contain information about the routes from the source node 20 from which they came.

The destination node 90 then generates a RREP message for both of the RREQ messages that it received. In this embodiment the destination node 90 generates a RREP message for every discovered route. However, in other embodiments, the destination node 80 could generate a RREP message for only a portion of the discovered routes, which could be a predetermined fraction, maximum number or otherwise according to some other criteria.

A requested bandwidth field included in the RREP packets, as one or more extensions to a conventional RREP message. In this embodiment, a delay field is also included in the RREP message, whose initial value is zero.

The two RREQ messages are then unicast back to the source node 20 via their respective routes ii) and iii). To minimise the possibility of false admission due to network dynamics, upon reception of a RREP message, each intermediate node establishes whether it can still meet the bandwidth requirements specified by the source node 20. If not, the RREP message is dropped.

As the RREP message propagates along the reverse path, each intermediate node forwarding the RREP message adds its own estimated MAC delay time for the next hop to the delay field. Therefore, the intermediate nodes update the RREP message, and the updated RREP message will comprise a cumulative indication of MAC delay for all the hops in the route up to that point.

MAC delay (in a RTS-CTS-DATA-ACK cycle) is a very useful metric to identify congestion hotspots and measure link interference in an ad hoc network. The MAC delay of a packet represents the time it takes to send the packet between a transmitting node and a receiving node including the total deferred time (including possible collision resolution) as well as the time to fully acknowledge the packet. It is measured by a transmitting node subtracting the time that a packet is passed to the MAC layer from the time an ACK packet is received from the receiving node. It is therefore a relatively easy quality of service metric for a node to measure when sending packets.

MAC delay is a specific concept in 802.1 1 -type wireless networks. It has been shown that, among various measurements such as packet loss, buffer occupancy, and MAC delay, MAC delay is the most useful metric to identify congestion hotspots in an ad hoc network. To discourage new routes from using badly congested or interfered links, MAC delay can be used as a routing metric, and is particularly a useful in determining packet congestion.

In this embodiment, it is assumed that the intermediate nodes in the network have been relatively immobile during the time between forwarding on the RREP message and receiving a RREQ message. Therefore, it is assumed that the values shown in Table 1 reflect the measure bandwidth on both the forward and the backwards paths.

Therefore, the source node 20 receives two RREP messages, one via path ii) and the other via path iii). As a result of the admission control performed by the intermediate nodes, the route chosen by the source will meet the bandwidth quality of service requirements.

Furthermore, each RREP message comprises a cumulative indication of MAC delay for all the hops in that route. The source node 20 can use the indication of MAC delay in the RREP messages to determine the minimum delay route to transmit real-time data to the destination node 90. In this embodiment, the source would determine that route iii) is the minimum delay route that satisfies its quality of service requirements. This route would not necessarily be the same as the minimum hop route.

However, the source node could use other ways of determining which of the discovered routes to choose.

It will be appreciated that, in on-demand protocols, as a RREQ message travels from a source node to a destination node, a reverse path from the destination node back to the source node is established. For timeslotbased MAC approaches (e.g. TDMA), free time slots can be reserved hop-byhop along the path as the RREP message travel back to the source node from the destination node. This is similar to connection-oriented resource reservation in wired networks. However, for 802.1 1 -like MACs, bandwidth reservation is not straightforward. In addition to admission control, some kind of traffic regulation is needed. After a flow is admitted and transmission starts, the traffic needs to be monitored and regulated as necessary.

In addition, in a network where both TCP-based best-effort traffic and real-time quality of service dependent traffic are present, effective rate control mechanisms have to be employed to ensure that greedy TCP traffic will not occupy excessive capacity and downgrade the performance of quality of service dependent traffic.

In order for the intermediate nodes in this embodiment to perform admission control, on receiving a RREQ message or a RREP message a node is required to measure its available bandwidth (also referred to as permissible throughput). This is performed by a node periodically measuring the bandwidth of its outgoing wireless links using a packet window. Such an approach is described in "End-to-end versus explicit feedback measurement in 802.11 networks" by M. Kazantzidis and M. Gerla,, IEEE ISCC, 2002, which can be found at: http://www.cs.ucla. edu/NRL/wireless/uploads/kazantz_7iscc.pdf This paper discloses that: Available bandwidth = ( l -u) * Throughput In the above equation u is the link utilization, which can be measured as the fraction of idle time in the duration of a packet window, and Throughput is measured in the following way: Throughput = S I (TACK reception-T'ransmsson) Where: S is the packet size in bits, TACK recep,o,, iS the timestamp of ACK reception and Transmisso,' is the timestamp of packet transmission.

In order to filter out the noise introduced by the measured raw throughput from packets of different sizes, Throughput is normalized to a pre-defined packet size to represent the bandwidth of a wireless link. Furthermore, a weighted moving average is used to smooth out the high variance of per packet measurements. Simulation results using the network simulator NS-2 have shown that this measurement technique is accurate enough for the purpose of call acceptance control. Simulations were performed using a topology of 20 mobile nodes in a 670m by 670m area, I I Mbps channel capacity, 10 constant bit rate traffic connections each with 200 kbps bandwidth requirement.

Link bandwidth measurement has been attempted in the past for networks with wireless links using a variety of approaches: link or end-to-end, passive or active. However, it has been found that the method above is able to provide bandwidth estimates accurate enough for the purpose of admission control, easy to implement (a network layer implementation is available) and is cost effective.

When an intermediate node measures its available bandwidth, it takes into account the maximum available bandwidth for the next hop. However, both theoretical analyses and experiments have shown that the maximum usable bandwidth in a wireless network is often far less than the nominal channel capacity. The maximum usable bandwidth will be only a percentage of the nominal channel capacity, and the specific percentage depends on several factors, for example: topology, number of nodes, traffic type, and interference. Therefore, in order to be conservative and avoid violating the quality of service requirements of the source, the effective maximum available bandwidth at a node is often chosen as a value far less than the nominal channel capacity in order to be conservative.

For example, in a network of 11 Mbps channel capacity, the admission threshold rate may be set as low as 0.7 Mbps. Therefore, a node will not accept any more bandwidth than 0.7 Mbps, which is clearly considerable lower than the maximum channel capacity.

Simulations using the network simulator NS-2 have shown that under favourable conditions a node in a network of 11 Mbps channel capacity may have a useable throughout of 2 Mbps. Therefore, using a static threshold of only 0.7 Mbps will, in effect, waste available bandwidth under favourable conditions in order to adequately provision for unfavourable conditions.

Therefore, in a further embodiment of the invention there is provided a means of adapting the admission threshold according to specific channel conditions, such that a higher threshold can be used when conditions are favourable than when the conditions are unfavourable.

Figure 5 shows a flow chart of a process of adaptive admission control based on available bandwidth.

In this process of adaptive admission control, MAC delay is used as an indication of the channel condition. Simulations have shown that the MACdelay measurements continuously fluctuate throughout the time, and that MAC delay is a very reliable indication on link contention. If there is a high degree of link contention, then the effective maximum bandwidth capacity of a link will be reduced compared to normal levels. For example, simulations have shown that for an 11 Mbps network, 2 Mbps represents the best possible value of the maximum bandwidth capacity of the link.

However, under high contention conditions, 0.7 Mbps can represent the effective maximum bandwidth capacity of the link.

Adaptive admission control can be used to vary the estimate of the maximum bandwidth threshold of the link according to measured delay conditions. At the start of the process the node uses a maximum bandwidth threshold of 2 Mbps, and figure S shows a flow chart for the process of the node monitoring the delay conditions and where necessary to switch it to a lower maximum bandwidth threshold of 0.7 Mbps.

At step S20 a node measures the MAC delay of a link. This could be performed continuously, or periodically. For example, MAC delay could measured every second, or every 100 ms.

At step S2 I the node determines whether the measured MAC delay is above a predetermined threshold D. In this example, the predetermined threshold D is 20 ms.

If the measured MAC delay is less than D, the process goes back to step S20, and MAC delay is measured again after a suitable time period.

If the measured MAC delay is greater than D a count value is set to 1, and the process moves to step S22 at which point the MAC delay is measured again.

If at step S23 the new measured MAC delay is less than D, the process goes back to step S20, and the node continues to use the upper maximum bandwidth threshold of 2 Mbps.

However, if the new measured MAC delay is greater than D, then the count value in incremented at step S24 and the process proceeds to step S25.

At step 25 the node establishes whether the count value has reached a predetermined threshold N. which in this case is 4. If the count has not reached the predetermined threshold N. then the process reverts back to step S22 and MAC delay is measured again. However, if the count has reached the predetermined threshold N. then the node switches to using the lower maximum bandwidth threshold of 0.7 Mbps.

Therefore, if the measured MAC delay measurements exceed a predetermined threshold D consecutively N times, a lower maximum bandwidth threshold is used. Thus when the wireless medium is busier, MAC delay will be higher, and the admission control policy at intermediate nodes becomes stricter and the intermediate node will accept fewer connections.

When D and N are configured as large values, admission control can be too loose and intermediate nodes can accept too many connections. This can render the protocol to be less effective against moderate congestion. In contrast, when N and D are configured with small values, admission control can be too conservative and the network utilization can be too low. Therefore, the appropriate choice of these parameters is important for admission control to function properly. In the above example, D is set to 20 ms and N to 4. However, the values of D and N could vary according to the network topology or various other factors. Furthermore, the choice of the value of N could depend on the time period used to perform the MAC delay measurements.

The node could revert to the higher maximum bandwidth threshold after a predetermined time, for example after 5 seconds. Alternatively, the process described above in relation to Figure 5 could be reversed, and the node could revert to higher maximum bandwidth threshold if the MAC delay is lower than a predetermined threshold for a predetermined number of measurements.

The above method of adaptive admission control could be used by a node to determine which of several maximum bandwidth thresholds to use.

Furthermore, the above method of adaptive admission control is applicable to other quality of service metrics. For example a node could determine whether to use a higher or lower maximum bandwidth threshold on the basis any quality of service metric that provides an indication of contention, such as buffer occupancy or packet drop in a predetermined time interval.

It will be appreciated that the system and method described above are capable of being implemented on commonplace hardware components, including on a general purpose computer including suitable transmitter/receiver functionality, configured by a suitable software product. The software product can be supplied as stored on a storage medium, such as an optically or magnetically readable storage medium, or by means of a signal transmitting a file executable by commonplace hardware to cause the hardware to become configured in accordance with an embodiment of the invention.

Though the specific embodiments described herein show particular implementations of the invention and demonstrate its advantages, other embodiments will be apparent from the description and the scope of protection sought herein is defined by the claims appended hereto with reference to the description and the drawings; no statement in this description should be interpreted as a specific limitation on the generality expressed in the claims.

Claims

CLAIMS: 1. A method of admission control at an intermediate node between a

source node and a destination node in a wireless ad hoc network, wherein: the source node is arranged to send a route request message including quality of service requirement information for determining at least one route to a destination node via one or more intermediate nodes; and the destination node is arranged to generate a route reply message for at least one discovered route between the source node and the destination node, each route reply message including the quality of service requirement information, and to send each route reply to the source node via the one or more intermediate nodes in that route; the method comprising, at an intermediate node: receiving a route reply message; and determining whether the intermediate node can meet the requirements specified by the quality of service requirement information and, if so, transmitting a further route reply message to the next node in the route.
2. A method according to claim 1, further comprising, at an intermediate node: receiving a route request message; and determining whether the intermediate node can meet the requirements specified by the quality of service requirement information in the route request message and, if so, transmitting a further route request message.
3. A method according to claim I or 2, further comprising, at an intermediate node on receipt of a route reply message: measuring a quality of service metric for the intermediate node; generating quality of service data based on the measured node specific quality of service metric; and including the quality of service data in the further route reply message.
4. A method according to any of claims l to 3, wherein the quality of service requirement information includes one or more sets of data, each of which specifies a quality of service criteria used by the intermediate nodes in the determining steps.
5. A method according to any one of the preceding claims, wherein the quality of service requirement infonnation comprises a first quality of service metric relating to a resource requirement that must be available at the intermediate node for the intermediate node to meet the requirements specified by the quality of service requirement information; the method further comprising, at the intermediate node: determining whether the intermediate node meets the requirements specified by the quality of service requirement infonnation by calculating a value of the first quality of service metric on the basis of establishing the amount of resources currently allocated and a variable threshold value; and wherein the variable threshold value is determined based upon measuring a second quality of service metric.
6. A method according to any one of the preceding claims, wherein the network is a mobile wireless ad-hoc network.
7. A device arranged to act as an intermediate node between a source node and a destination node in a wireless ad hoc network, in which: the source node is arranged to send a route request message including quality of service requirement information for determining at least one route to a destination node via one or more intermediate nodes; and the destination node is arranged to generate a route reply message for at least one discovered route between the source node and the destination node, each route reply message including the quality of service requirement information for that route, and to send each route reply to the source node via the one or more intermediate nodes in that route; the intermediate node comprising: a receiver arranged to receive a route reply message; and a processor arranged to detennine whether the intermediate node can meet the requirements specified by the quality of service requirement information; and a transmitter arranged to transmit a further route reply message to the next node in the route if the processor determines that the intermediate node can meet the requirements specified by the quality of service requirement information.
8. A device according to claim 7, wherein: the receiver is further arranged receive a route request message; and the processor is further arranged to determine whether the intermediate node can meet the requirements specified by the quality of service requirement information in the route request message; and the transmitter is further arranged to transmit a further route request.
9. A method of determining a route from a source node to a destination node in an ad hoc network comprising one or more intermediate nodes between the source node and the destination node, the method comprising, at the source node: transmitting a route request message, the route request message including quality of service requirement information, at the destination node: receiving a route request message for each discovered route between the source node and the destination node; generating a route reply message for at least one discovered route between the source node and the destination node, each route reply message including the quality of service requirement information; and transmitting each route reply message to the source node via the one or more intermediate nodes in that route, at an intermediate node: receiving a route reply message; and determining whether the intermediate node can meet the requirements specified by the quality of service requirement information and, if so, transmitting a further route request message to the next node in the route, and at the source node: receiving one or more route reply messages; and determining which route to use on the basis of the received route reply messages.
10. A method according to claim 9, further comprising, at an intermediate node: receiving a route request message; and determining whether the intermediate node can meet the requirements specified by the quality of service requirement information in the route request message and, if so, transmitting a further route request message to one of the other intermediate nodes or the destination node.
A method according to claim 9 or 10, further comprising, at an intermediate node on receipt of a route reply message: measuring a quality of service metric for the intermediate node; generating quality of service data based on the measured quality of service metric; and including the quality of service data in the further route reply message; at the source node: determining which route to use on the basis of the node specific quality of service data in the received route reply messages.
12. A wireless ad hoc network comprising, a source node comprsmg: a transmitter arranged to transmit a route request message, the route request message including quality of service requirement information, a destination node comprising: a receiver arranged to receive a route request message for each discovered route via one or more intermediate nodes between the source node and the destination node; a processor arranged to generate a route reply message for at least one discovered route between the source node and the destination node, each route reply message including the quality of service requirement information; and a transmitter arranged to transmit at least one route reply message to the source node via the one or more intermediate nodes in that route, and at least one intermediate node comprising: a receiver arranged to receive a route request message; and a processor arranged to determine whether the intermediate node can meet the requirements specified by the quality of service requirement information in the route request; and a transmitter arranged to transmit the a further route reply message to the next node in the route if the processor determines that the intermediate node can meet the requirements specified by the quality of service requirement information.
13. A wireless ad hoc network according to claim 12, wherein: the receiver of the intermediate node is further arranged receive a route request message; and the processor of the intermediate node is further arranged to determining whether the intermediate node can meet the requirements specified by the quality of service requirement information in the route request; and the transmitter of the intermediate node is further arranged to transmit a further route request.
14. A wireless ad hoc network according to claim 12 or 13, wherein the wireless ad hoc network is a mobile wireless network.
15. A method of admission control at an intermediate node between a source node and a destination node in a wireless ad hoc network substantially as hereinbefore described with reference to the accompanying drawings.
16. A device arranged to act as an intermediate node between a source node and a destination node in a wireless ad hoc network substantially as hereinbefore described with reference to the accompanying drawings.
17. A method of determining a route from a source node to a destination node in an ad hoc network comprising one or more intermediate nodes between the source node and the destination node substantially as hereinbefore described with reference to the accompanying drawings.
18. A wireless ad hoc network substantially as hereinbefore described with reference to the accompanying drawings