GB2460701A - Method of implementing opportunistic relaying in a wireless network - Google Patents

Abstract

A roaming user equipment identifies candidate relay nodes based on information contained in monitored signals, which may include a network identifier and/or an identifier of a destination node. The information may be located in a particular portion of the signal, for example a portion of the signal indicative of the destination and recognised as such by the user equipment. In this way, the user equipment can better ensure the robustness of a relay path. The user equipment may also adopt a similar approach to determine when a current communication link to the destination node can be reconfigured from an indirect link to a direct link.

Description

I

The present invention relates to the provision of networking facilities between wireless communications apparatus and is particularly, but not exclusively, concerned with the management of networking in the context of the provision of such facilities by a central controller or access point.

The reader will appreciate that the invention can be considered in the context of WiFi technology but is not exclusively concerned therewith.

Figure 1 illustrates a simple network 10 comprising an access point 20 and a plurality of (in this example, two) wireless terminals 22. Each of the access point 20 and the wireless terminals 22 is equipped with the technology to enable establishment of wireless communication, such as in accordance with the 802.11 standard and series of amendments specified by the IEEE. In the illustrated example, the workable range of the access point 20 is signified by the broken line X-X. One of the wireless terminals 22 is within this range and can thus establish directed communication with the access point 20, although the other wireless terminal 22 is outside range and so some form of ad hoc routing is required.

Although the word "routing" is used herein, it will be understood that this word has specific meaning within the IEEE 802.11 C Standard. In particular, this is a layer 3 function (i.e. the IP layer), whereas "bridging" is a layer 2 (MAC layer) function. A repeater is typically defined in the art as a function with little or no assumed intelligence, where all transmission are echoed (i.e. repeated) to extend range. Although the term "relay" may be considered the broadest term, this can also be taken to be synonymous with "repeater". The present disclosure is not concerned with the manner in which routing is achieved, and a repeater, a bridge or any other level of technology could be used in conjunction with the invention.

Further, it should be noted that the diagram illustrated in Figure 1 is also simplified in that it indicates that there is a single cut-off range boundary for an access point. In fact, many wireless LAN technologies (including 802.11 C) include a number of different rates and encoding technologies. As range between two devices increases, adaptive selection of transmission rate and encoding technologies is well known and, although a particular transmission rate and encoding technology will have a specific range of operation, the access point will in fact have a number of cut-off points in operation.

Therefore, if a wireless terminal moves gradually away from an access point, rate adaptation will take place through all of the operation modes but, when the distance increases beyond the operational range of the most robust (and probably the slowest) mode of operation, then, in practice, the link between the access point and the wireless terminal will be dropped. Also, it must be recognised that the notion of a cut-off range boundary being a perfect circular arc with a transmitter at the centre is a simplification designed for purposes of illustration only. Not only is the boundary of a highly irregular nature, but as propagation conditions change (temperature, humidity, moving reflectors and obstructers) the boundary, whatever its shape, will vary and "breathe".

It is already well known in the field of the present invention to implement opportunistic relaying. Examples of this include ODMA in 3GPP UMTS. Further, opportunistic routing, intended to extend range of AP based networks is also well known. One instance of this is the ExOR protocol, put forward by Biswas et a! in "ExOR: "Opportunistic Multi-Hop Routing for Wireless Networks" (B.iswas S and Morris R, M.I.T. Computer Science and Artifical Intelligence Laboratory). According to these disclosures, and others, it is well established to use multi-hop ad hoc network connections for ad hoc purposes in wireless networks.

Indeed, some existing WiFi devices already incorporate an optional repeater mode.

This is usually provided as part of the wireless distribution system (WDS), as set out in M. Gast, 802.11 Wireless Networks -The Definitive Guide, First ed: O'Reilly & Associates Inc., 2002. Accordingly, it is already known to forward packets through adhoc connections made by a WiFi device with a repeater mode. Other WiFi devices also incorporate an optional "secondary" mode that permits passive repeaters, or even passive access points, to wait until the repeater (or, as the case may be, the access point) is deemed to be required, before assuming the active repeater (or active access point) role. This autonomous detection behaviour is reliant on the dormant repeater (or access point) being placed at the edge of the practical range of the access point and altruistically detecting that there are now orphaned nodes beyond the reaches of coverage.

The IEEE 802.11 standard and series of amendments has been used in the field of WiFi networking for some time. The latest draft amendment (802.lls) provides frame forwarding functionality and multi-hop/mesh link setup primitives. In the 802.11 s draft, it is put forward that stations could be mesh points (MPs) able to forward traffic onto other stations or MiPs. Each constituent hop is referred to as a mesh link, and the full end to end path (within the 802.11 mesh basic service set (BSS)) is known as the mesh path.

The direct link functionality provided in IEEE 802.11 e allows stations to communicate with each other directly and not via the access point, using direct link setup (DLS) messages. Figure 2 illustrates the use of direct link setup as extracted from that standard. Figure 3 illustrates a teardown process in accordance with the same standard amendments. As these will be familiar to the reader, no further description will be required at this time.

In the absence of 802.lls implementations, 802.11 DLS could be used to establish a point-to-point link between a roaming station and a relay station. However, additional functionality is then required in the relay station to handle messages from remote nodes so that they get forwarded securely to the access point. Additionally, the DLS mechanism is always overseen by the access point. Therefore, if the initialising node is no longer in range of the access point, this technology is unavailable for use.

The power saving mechanism of 802.11 also puts forward the concept of reserving windows of time, notably beacon periods. Infrastructure BSSs make use of the delivery traffic indication message (DTIM) and, equivalently, for independent BSS arrangements, the ad hoc traffic indication message (ATIM). In both of these cases, only beacons and traffic indication messages (DTIM or ATIM as the case may be) are permitted within the guarded interval. Therefore, extending the protocol to allow for a new type of message, to be transmitted by a node now out of contact of its access point, would require an amendment to the initial standard. Alternatively, the standard could be violated but this would only be permitted within a proprietary network.

Other fields of technology might be considered to offer solutions to the above problems.

In particular, Japanese patent application JP 2007 036421A refers to a distributed reservation protocol (DRP) such as used in WiMedia. This approach is further exemplified in the following references: G. R. Hiertz, Z. Yunpeng, J. Habetha, and Sirin, "Multiband OFDM Alliance -The next generation of Wireless Personal Area Networks," pp. 208-214, 2005; J. d. P. Pavon, Sal Shankar N, V. Gaddam, K. Challapali, and C.-T. Chou, "The MBOA-WiMedia specification for ultra wideband distributed networks," IEEE Communications Magazine, vol. 44, iss. 6, pp. 128-134, 2006; WiMedia-Alliance (Ecma Internation (Ecma)), "Standard ECMA-368 High Rate Ultra Widwband PHY and MAC Standard," 2005, Available: http:f/www.wimedia.org/enlresources/eis.asp?ithres.

However, this approach involves a completely different design of MAC structure, to support distributed reservation from the outset. As such, this is not applicable given the constraints to the development of IEEE 802.11 WLAN related systems.

The reader might consider use of other radio access technologies, such as Bluetooth or even UWBIWiMedia to extend the range of a wireless terminal and/or an access point.

This is considered in European patent application EP 0811286. However, that approach raises additional problems.

Firstly, all devices would have to support the secondary radio access technology (RAT) as well, with cost and complexity implications. Needless to say, this would also be detrimental to legacy devices.

Secondly, there is disparity in data rate between these various technologies. For instance, the data rates of 802.11 and Bluetooth are markedly different. This would give rise to a need for buffering at relay nodes, with consequent perfonnance issues.

Thirdly, not only would a relay node need to have the capability to buffer data in order to deal with these data rate issues, it would also have to perform relatively complex transcoding (or even decoding and recoding) of packets to achieve transfer from one format or address scheme to another. This again implies the need for complicated, sophisticated and potentially very expensive stations. All stations would need to have this functionality in order to guarantee that the potential relay station (which could be any node in the network) could support the relay request when required. Otherwise, the probability of a roaming node finding a relay that is both prepared to, and capable of, achieving relay role is greatly affected.

Another physical solution to the above problems would be to increase the number of access points in order to increase coverage. Then, if an access point failed, the overall set of access points could be managed to accommodate any resulting orphaned stations.

An example of such an approach is detailed in F. E. d. Deus, and 3. Kabara, "on Survivability of IEEE 802.11 WLAN," (Sensor Networks, Ubiquitous and Trustworthy Computing 2006, IEEE International Conference, pp. 462-471, 2006). However, this does not lead to a more effective wireless communications network and greatly increases the amount of physical infrastructure required in order to support wireless communications quality of service.

An aspect of the invention seeks to provide opportunistic relaying through candidate-relay-identification and virtual-carrier-sense-protected contact periods within IEEE 802.11 (or similar) protocols with minimal changes. An aspect of the invention is also exemplified in due course by a mechanism for tearing down this opportunistic relaying mode of operation and returning to normal operation.

The underlying concept of opportunistic relaying is for the roaming node to identify a candidate relay node in range and have an opportunity to then set up an ad hoc opportunistic multi-hop link back to the AP via a node that is still in range.

There are a number of known solutions to identifying candidate relay nodes for opportunistic relaying purposes.

One such solution is for the AP to deliver a list or "book" detailing all associated nodes of the AP to each station on association, thereby enabling a station to identify other stations of the same network who could act as repeaters. However, this requires updating the list or book as stations join and leave the association, making this solution neither particularly scalable nor easy to administer. For example, the approach described in European patent application EP 0811286 expects every station in an ad hoc network to maintain a list of connectivity information for all in-range nodes, and to actively probe the network to ensure that this information is always up-to-date and accurate. It will be apparent that such an approach places a significant overhead burden on stations in the network.

The problem of placing a continuous additional load on connected devices, i.e. in-range devices, is considered in European patent application EP 1641184. There, a so-called "promiscuous mode" is implemented on roaming terminals. The term "promiscuous mode", as used in the art of wireless networking, refers to receiving packets within wireless radio range of a device and processing the packets irrespective of whether they are intended for the device or not. In this mode, the device in question must be associated to an access point or ad hoc network.

However, in EP 1641184, a diagnostic server of an infrastructure network is relied upon to store and periodically update all the information about the clients. When a client is disconnected, the system responds by running diagnostics on the disconnected client and on a client connected to the wireless medium. A new ad hoc network is then established between the connected client and the disconnected client. This may require the connected client to straddle two networks, with the additional complexity that this entails, such as maintaining synchronisation with both networks. Further, no provisions are made for a roaming terminal to determine whether a potential relay node is on the network with which the roaming terminal was in association with.

Another known solution is for the roaming node to broadcast probe packets to all neighbouring nodes, with those nodes that can potentially serve as relay nodes responding accordingly, as explored in M. He, T. Todd, D. Zhao, and V. Kezys, "Ad Hoc Assisted Handoff for Real-time Voice in IEEE 802.11 Infrastructure WLANS," (Proceedings of IEEE Wireless Networking and Communications, WCNC 2004). The skilled reader will appreciate that such indiscriminate flooding of a wireless medium may impact on network availability and performance.

Other background infonnation may be found in E. M. Royer and T. Chai-Keong, "A Review of Current Routing Protocols for Ad Hoc Mobile Wireless Networks," IEEE Personal Communications, vol. 6, Iss. 2, pp. 46-5 5, 1999, and PCT patent application WO 2005/025110. However, the approaches described in these references suffer from a variety of drawbacks, including excess overhead and traffic. For example, in the system disclosed in WO 2005/025110, the determination of when to activate one or more relay resources in order to facilitate a relaying operation is made at a base site in response to a request therefor by a remote unit; in the approach described in Royer and Chai-Keong, every node of a network has to maintain a routing table in which all possible destinations within the network, as well as the number of hops to each destination, are recorded.

Aspects of the invention seek to mitigate or eliminate at least some of the above-mentioned problems or disadvantages.

Broadly speaking, the present invention recognises that opportunistic relaying in wireless networks can be improved by the discriminatory identification of candidate relay nodes based on information contained in signals monitored at a roaming node. By distinguishing nodes as candidates and non-candidates, a roaming node can attempt to contact identified candidate nodes (or a single candidate node, as the case may be) directly, rather than flooding all the nodes that are able to listen with relay requests, thereby reducing medium access overhead.

By running the identification process at the roaming node, opportunistic relaying becomes more distributable and power efficient by only being used when needed.

Further, such an approach is more scalable than maintaining a list of associated MAC addresses, which may become quite large, and also doesn't require updates, which incur additional signalling overhead proportional to the number of associated nodes and their mobility. Moreover, the identification of candidate relay nodes becomes manageable by implicitly limiting candidate relay node identification to those nodes that are within radio range of the roaming node.

The information which is searched for by the roaming node may comprise one or more identifiers, such as a network identifier and/or a node identifier, thereby allowing a specific node (or nodes) to be pinpointed. Also, an identifier may be correlated to a particular portion of a monitored signal, such as a destination address field. In this way, a roaming node can satisfy itself that a candidate relay node is in communication range of a destination node, thus substantially increasing the likelihood that a message intended for the destination node will reach it.

The described method and apparatus may be implemented in both infrastructure-based networks, where connectivity between an access point and an associated roaming node changes to an ad hoc configuration as a result of out-of-range problems due to mobility, and networks with no infrastructure, where connectivity between a source node and a destination node is maintained in a decentralised manner.

It is also recognised that existing network traffic may need to be protected. A classic problem is that of the "hidden node", where two nodes which are out of range of each other, and so are unable to sense each other's transmissions, attempt to transmit to a destination (or even two separate destinations) in range of both. This results in both transmissions mutually interfering and colliding. In the specific case of the scenario described herein, it will be apparent that a candidate relay node may be a common destination node of the access point and the roaming node when the latter two nodes are out of broadcast range of each other, such that their transmissions to the candidate relay node are at considerable risk of mutually interfering, e.g. if the roaming node transmits a relay request to the candidate relay node, it may unwittingly transmit at the same time as the access point, resulting in both its request and the regular traffic within the network being lost. Consequently, any relay requests to a candidate relay node from a roaming node may be transmitted during a quiet period of the network.

Finally, roaming nodes may be able to determine when a return to direct connectivity is feasible, so that a point-to-point communication can be restored and the relay channel torn down. In particular, the roaming node may periodically sample network traffic in order to identify packets sent directly from the destination node. This could be based on an analysis of each signal's source address field, for example. In order to reduce the amount sampling that the roaming station has to perform, the received signal strength of the relay station may be used as a trigger.

Aspects of the invention provide methods and apparatus for implementing opportunistic relaying in a wireless network. A roaming node identifies candidate relay nodes based on information contained in signals monitored by the roaming node, which may include a network identifier and/or an identifier of a destination node. The information may be correlated to a particular portion of the signal, for example a portion of the signal indicative of the destination of the signal. In this way, the roaming node can better ensure the robustness of a relay path. The roaming node may also adopt a similar approach to determine when a current communication link to the destination node can be reconfigured from an indirect link to a direct link.

Aspects of the invention may use BSS ID to spot relevant packets, and may identify useful nearby stations by noting the destination MAC address of packets sent by these stations as being the MAC address of a known AP. Aspects of the invention may also use the quiet period after beacons, which are protected by NAy, to transmit relay requests. Aspects of the invention may further comprise checking to see if the AP comes back into range again.

Accordingly, a first aspect of the invention provides a method of retaining connectivity to a network by a wireless communication device, the network comprising a plurality of nodes, the method comprising: monitoring for signals transmitted by neighbouring nodes; searching said signals for information content, said information content comprising an identifier of the network; and transmitting a request to at least one of said 1o' neighbouring nodes if said information content is found, each of the at least one of said neighbouring nodes corresponding to a signal in which said information content is found, said request comprising a request to serve as a relay between the wireless communication device and a destination node.

A second aspect of the invention provides a method of reconfiguring a communication link by a wireless communication device, from a current indirect link to a destination node over a relay node, to a direct link to the destination node, the method comprising: monitoring for signals transmitted by neighbouring nodes; searching said signals for information content, said information content comprising an identifier of the destination node; checking whether said identifier of the destination node is located in a portion of said signal that is indicative of an origin of said signal; and requesting establishment of a direct link to the destination node if said checking is true.

A third aspect of the invention provides a wireless communication device adapted to retain connectivity to a network comprising a plurality of nodes, the device comprising: means for monitoring for signals transmitted by neighbouring nodes; means for analysing said signals for information content, said information content comprising an identifier of the network; and means for transmitting a request to at least one of said neighbouring nodes if said information content is found, each of the at least one of said neighbouring nodes corresponding to a signal in which said information content is found, said request comprising a request to serve as a relay between the wireless communication device and a destination node.

A fourth aspect of the invention provides a wireless communications network comprising a plurality of nodes at least one of which is in accordance with the third aspect of the invention.

Aspects of the invention may comprise a computer program product comprising computer executable instructions operable to cause a computer to become configured to perform a method in accordance with any of the above identified aspects of the invention. The computer program product can be in the form of an optical disc or other computer readable storage medium, a mass storage device such as a flash memory, or a read only memory device such as ROM. The method may be embodied in an application specific device such as an ASIC, or in a suitably configured device such as a DSP or an FPGA. A computer program product could, alternatively, be in the form of a signal, such as a wireless signal or a physical network signal.

These and other aspects of the invention will now be further described, by way of example only, with reference to the accompanying figures, in which: Figure 1 is a schematic diagram of a simple network including an access point and two wireless terminals; Figure 2 is a schematic diagram of a direct link setup process in accordance with the IEEE 802.11 E standard; Figure 3 is a schematic diagram of a teardown process in accordance with the IEEE 802.11 E standard; Figure 4 is a schematic diagram of a network depicting opportunistic relaying in accordance with a specific embodiment of the invention; Figure 5 is a schematic diagram of a node of the network illustrated in figure 4; Figure 6 is a schematic diagram of the MAC frame format in accordance with the IEEE 802.11 standard; Figure 7 is a flow diagram of a process of establishing an opportunistic relay in accordance with a specific embodiment of the invention; Figure 8 is a flow diagram detailing a process of analysing captured packets of figure 7; Figure 9 is a schematic diagram of a management frame format in accordance with the IEEE 802.11 standard; Figure 10 is a schematic diagram of message sequences of normal operation, initiating opportunistic relaying, opportunistic relaying, and opportunistic relay teardown pro cesses in accordance with a specific embodiment of the invention; Figure 11 is a flow diagram of a process of tearing down an opportunistic relay in accordance with a specific embodiment of the invention; and Figure 12 is a schematic diagram of a state machine of a repeater as known in the art.

As illustrated in figure 4a, an exemplary network 400 in which the invention can be applied comprises, by way of example only, an infrastructure network in compliance with the IEEE 802.11 Standard. The network includes an access point (AP) 402 and four wireless terminals 402, 404, 406, 410. The AP 402 manages the communications of terminals 404, 406, 408, 410 associated to its network 400. Also depicted in figure 4 are two other terminals 420, 422 which are not associated to network 400.

In the context of this description, terminals 404, 406, 408, 410, 420, 422 and AP 402 are also referred to as nodes', which is a generic term referring to a point of communication in a network. Further, where reference is made to a wireless communication device having an association to a network, this generally means that, in operation, the device is connected to the network such that it may communicate (directly or indirectly) with one or more other nodes in the network, and vice versa. For purposes of this description, which is couched in the context of the 802.11 Standard, the terms "signals" and "packets" are used interchangeably, though it will be apparent that signals in accordance with implementations of the invention are not limited to packet-based signals.

In the illustrated example, the dashed line X-X signifies the workable range of Al 402.

It will be appreciated that the radio coverage need not be a perfect circular arc with a transmitter at the centre, as shown, but may be highly irregular, omnidirectional or directional, and may vary with coding/modulation scheme and with time. Three of the terminals 404, 406, 408 are within this range and can thus establish direct connections with the access point 402. However, terminal 410, having roamed outside the range of AP 402 (as indicated by the dotted arrow), will require some form of ad hoc routing in order to maintain connectivity to the network 400.

While this example assumes that terminal 410 intends to remain associated to network 400, it will become apparent in due course that the described apparatus and method may alternatively be implemented in such a way that terminal 410 associates to another network.

In the exemplary network 400, each of the access point 402 and the wireless terminals 404, 406, 408, 410 is equipped with the technology to enable establishment of wireless communication, such as in accordance with one of the IEEE 802.11 standard.

Exemplary tenninals include mobile phones, Personal Digital Assistants (PDA), general purpose computers, integrated devices combining one or more of the preceding devices, and the like. The AP may be a general-purpose computer, standard laptop, fixed terminal, or other electronic device configured to transmit, receive and process data according to an air interface method.

Figure 4b will be refened to in due course. However, turning first to figure 5, in an embodiment of the present invention, a node 500 comprises a processor 502 operable to execute machine code instructions stored in a working memory 504 and/or retrievable from a mass storage device 506. By means of a general-purpose bus 508, user operable input devices 510 are in communication with the processor 502. The user operable input devices 510 comprise any means by which an input action can be interpreted and converted into data signals, for example, DIP switches.

Audio/video output devices 512 are further connected to the general-purpose bus 508, for the output of information to a user. Audio/video output devices 512 include any device capable of presenting information to a user, for example, status LEDs.

A communications unit 514 is connected to the general-purpose bus 508, and further connected to an antenna or set of antennas 516. By means of the communications unit 514 and said antenna 516, the node 500 is capable of establishing wireless communication with other nodes. The communications unit 514 is operable to convert data passed thereto on the bus 508 to an RF signal carrier in accordance with a communications protocol previously established for use by a system in which the node 500 is appropriate for use, for example 802.11.

In the node 500 of figure 5, the working memory 504 stores applications 518 which, when executed by the processor 502, cause the establishment of an interface to enable communication of data to and from other nodes. The applications 518 thus establish general purpose or specific computer implemented utilities and facilities that are used in linking nodes.

A method of opportunistic relaying in accordance with the present invention will now be described in the context of a WLAN Basic Service Set (BSS) network, which is an area of coverage or "cell" established by an 802.11 wireless AP. Reference to such a network is made for purposes of discussion and explanation only, and it will be understood that other wireless networks may equally be suitable.

In 802.11, the BSS ID (network ID) is included as part of the MAC header. Such a header is depicted in the 802.11 MAC frame format of figure 6, taken from "IEEE Wireless LAN Edition -A compilation based on IEEE Std 802.1 1TM-1999 (R2003) and its amendments" 2003. The BSS ID field is "Address 3".

Also in 802.11, packets transmitted to or from a given node will include an identifier corresponding to that node in the MAC header, such as in the source address field ("Address 1") or the destination address field ("Address 2"). This could be the MAC address of the wireless AP, for example.

Suitable candidate relay nodes can therefore be identified by making use of information contained in signals monitored at the roaming node. For example, the network ID helps to distinguish nodes associated to different networks, while the source and destination fields enables identification of nodes that are within radio coverage of the AP. This is because any transmissions destined for the AP that an outlying roaming station can detect must be from a node that is in range of the AP (assuming channel reciprocity is in effect, i.e. if one of a pair of nodes can hear the transmissions of the other node of the pair, then the same holds true for the reverse transmissions). Similarly, if an outlying roaming station (suddenly) receives a transmission with a source address file dequal to the MAC address of the AP, then it must now be (back) in range of the AP again in order to have received that packet.

With reference to the flow diagram of figure 7, in a first step of the method (step S702), a roaming node determines that its connectivity to a network is dropped or in danger of being dropped. The manner in which this determination is performed is not relevant to the present invention, and can be of a conventional nature, for example based on signal strength measurements, time outs, and so forth. In particular, it will be understood that, although a drop in connectivity is described in terms of the coverage range of an AP, other events could contribute to the drop in connectivity. Such events include, but are not limited to, a loss in line of sight between the access point and the wireless terminal due to the appearance of an obstacle, which may be prevalent in urban environments for

example.

Having determined that routing is required (or more optimal), the roaming node then captures and analysis passing packets (step S704) in order to identifS' candidate relay nodes. Packet analysis will be described in more detail in due course. In the context of the present disclosure, the terms "capturing", "monitoring", "listening" and "sniffing" all refer generally to the activity of receiving and decoding, at a wireless communication device, wireless traffic from any network and any nodes within radio range of the device, and which may or may not be bound for the device. Generally, the roaming node does not acknowledge receipt of such monitored wireless traffic.

To protect ongoing or intended network traffic from potential collisions, the roaming node may optionally wait for a "quiet" period (step S706) before attempting to contact identified candidate relay nodes (step S708). Generally, the quiet period is a time during which the nodes of the network refrain from transmitting. Exemplary quiet periods will be discussed in more detail later.

A roaming node may contact one, some, or all identified candidate relay nodes. If more than one candidate agrees to serve as a relay (step S710, Yes'), the roaming node may need to select a candidate that it thinks is most suitable before communicating via that candidate (step S714). Selection could be based on signal strength, which may be indicative of candidate location, or the like. Otherwise, the roaming node may need to contact other identified candidates, or begin again the process of capturing and analysing passing packets (step S710, No'). The capture and analysis process may be performed once, periodically, or continuously. As this is a power-intensive process, the choice will generally depend on power reserves available to the terminal.

Packet analysis is now described in more detail with reference to figure 8.

For each captured packet, the roaming node first determines whether a network identifier corresponding to the target network is present (step S802). In general, the target network is the network to which the roaming node was most recently associated to, though this is need not always be the case. Packets that do not contain the target network identifier are discarded (step S802, No'). Otherwise (step S802, Yes'), the packet is analysed further for the presence of an identifier corresponding to the destination node (step S804). Packets that do not satisfy this criterion are also discarded (step S804, No').

As indicated above, the network identifier may comprise BSS ID of the network, determined from field "Address 3", while the destination node identifier may comprise the MAC address of the AP, determined from the source address field, "Address 1", or the destination address field, "Address 2" (figure 6).

Where the network comprises a wireless network in which an AP acts as a hub to provide connectivity for devices associated with the AP, and given that the roaming node may no longer be able to detect transmissions from the AP (this condition being a trigger for such a mode of operation), it may be fairly safe to assume that any transmissions detected by the roaming node and containing the identifier of that network are destined for the AP. As such, step S806 may be considered optional and may be implemented by the roaming node as a check that packets are indeed bound for the AP (i.e. the Destination Address field is equal to the AP's MAC address).

In general, a signal that satisfies step S806 can be considered as having been sent by a node that is currently in range of the AP and which is therefore a good candidate relay node. The skilled person will appreciate that there is, of course, a slight possibility that this signal originates from a node that has just gone out of range of the AP, and that that node will receive no acknowledgement from the AP. In such situations, a connection timeout will eventually take effect, ultimately causing the roaming node to select a different candidate relay node.

Thus, and referring again to figure 4a, as node 410 roams outside or to the edge of the coverage range of AP 402, it sniffs packets transmitted from nodes 404, 406, 408, 420, 422, and segregates packets according to whether an identifier corresponding to network 400 is present. Assuming that the roaming node 410 intends to remain connected to network 400, packets originating from nodes 420, 422 will be discarded.

Where network 400 comprises an infrastructure BSS, all communications are relayed through access point 402, such that each of nodes 404, 406, 408 may be identified as a candidate relay nodes based on the identifier of AP 402 included in their packets.

Alternatively, where network 400 comprises an ad hoc network having a plurality of nodes 402, 404, 406, 408 but no central hub or AP managing the network, roaming node 410 may analyse packets for the identifier of a particular destination node that it wishes to converse with.

In more detail, and referring now to figure 4b, which schematically depicts an exemplary message exchange occurring in the network 400 of figure 4a, roaming node 410 overhears an omnidirectional transmission 407 from node 406 to node 402, but not the acknowledgment 409 sent in reply. In transmission 407, the network identifier field is set to the identifier of network 400, while the destination address field is set to the identifier of node 402. It will be apparent that this transmission satisfies all of the conditions set out in order to identify a suitable candidate relay node; that is, steps S802 and S804 are answered in the affirmative, and step S806 is also confirmed. In the event that node 410 does hear the acknowledgement, then node roaming 410 has moved back in range of node 402 and direct communication may be possible again.

Exemplary packet fragments that a roaming node may encounter while trying to identify a candidate relay node are given in Table 1.

Table 1: Packet Fragments Fragment Destination SourceAddress BSS1D Packet Address Any (not AP; not Roaming node's A AP MAC Address roaming node's) network address B Unrecognised Unrecognised Unrecognised Any (not AP; not Roaming node's C AP MAC Address roaming node's) network address In a first exemplary packet, packet A, the destination address field is set to the MAC address of an AP, the BSS ID field is set to the network address corresponding to that of the roaming node, while the source address field is set to an identifier of a node other than that of the AP or the roaming node. Such a combination of fragments is typical of a signal transmitted by a potential candidate relay node, and may be seen by the roaming node when it is out of range of the AP.

Meanwhile, packet B is a typical part-packet corresponding to a node belonging to another network/BSS. Here, each field is set to an identifier that is unrecognisable to the roaming node. Of course, one or more of the identifiers could be known to the roaming node in advance, which may aid in handover decision making if no suitable candidate relay node is identified.

In packet C, as with packet A, the BSS ID field is set to the identifier of the network address corresponding to that of the roaming node. However, in this case, the destination address field is set to an identifier of a node other than that of the AP or the roaming node, while the source address field is set to the MAC address of an AP. Such a packet is unlikely to be seen unless the roaming node has roamed back into range of the AP, and may trigger the roaming node to attempt to contact the AP directly.

The above discussion has mainly considered some of the ways in which a roaming node may identif' candidate relay nodes in accordance with the present invention. We now consider some of the conditions under which a roaming node might attempt to contact an identified candidate relay node to request routing, and the process of tearing do an established relay route and returning to direct communication with the destination node.

Turning first to the broadcasting of route requests, the skilled person will recognise that a roaming node may be a hidden node from the perspective of the destination node. This is to say that, in its attempts to contact a candidate relay node, the roaming node could disrupt a network's communications inadvertently. For example, an ongoing communication between the destination node and the candidate relay node could be interrupted by the roaming node's opportunistic routing requests to the candidate relay node.

One solution is for the roaming node to wait until a time period within which the network is quiet (step S706 in figure 7). Such "quiet periods" or "quiet intervals" may comprise mandatory idle time intervals defined by a wireless protocol.

In IEEE 802.11, a number of different frame types are defmed for use in managing and controlling wireless links, as well as data communications. Figure 9 depicts the format of an 802.11 management frame, and in particular a beacon frame. In infrastructure network configurations, the beacon frame is a periodic frame sent by the A? to announce its presence and relay information, such as BSSID and other parameters regarding the access point, to nodes within radio coverage. The duration field is provided to allow a station to keep an accurate Network Allocation Vector (NAy), and accounts for a time period during which the virtual carrier sense mechanism holds off from transmitting.

The duration field is typically set to an artificially long value, to cover both the beacon transmission at the control data-rate, and the ensuing quiet period, i.e. the time interval to the next frame, known as the inter-frame space (IFS). During this quiet period, any station that has fallen out of range can try to make contact with nearby nodes. If the station receives no response, it can try in a next time window. This incurs an overhead (of not transmitting in that window) but is a straightforward and reliable solution.

The message sequence chart of figure 10 shows the messages exchanged between AP 1002, relay node 1004 and roaming node 1006 over time, taking into account the quiet periods of a network. This message sequence chart is for a generic protocol, though it will be apparent that could be populated with particular primitives from the IEEE 802.11 standard, as partially depicted in figures 2 and 3.

During normal operation, both nodes 1004, 1006 will hear the beacon 1008a of the AP, and data transfer 1010 between the AP 1002 and roaming node 1006 can be effected directly. In this case, there is no need for the roaming node to make use of the quiet period 1012a for opportunistic relaying purposes.

However, if node 1006 roams beyond the coverage range of the AP 1002, it will no longer bear transmissions from the AP and will initiate opportunistic relaying. In figure 10, an acknowledgement 1016 transmitted by node 1004 responsive to a transmission 1014 from the AP 1002 is captured and analysed by the roaming node 1006, which identifies node 1004 as the sender and its association to the network of AP 1002. The roaming node may also check that the acknowledgement is destined for the AP 1002, as described previously. Node 1004 is therefore identified as a candidate relay node.

The roaming node 1006 is likely to know when the next beacon transmission 1008b and quiet period 101 2b occur, since it has until recently been associated to that network. It will therefore wait until such a time, before performing relay set-up messaging 1018, which typically comprises a relay request and a relay request confirmation.

It will be apparent that an attempt to make contact during a quiet period is as much at risk of collision and loss as any other wireless transmissions. To that end, a random-back-off access parameter may be implemented (using the same philosophy as that of the fundamental medium access, the distributed beacon transmission, or the ATIMJDTIM access mechanisms of 802.11) so that neighbouring nodes do not all try to initiate communication at the same time. In addition, a time-out value in the transmitting station may be implemented that will trigger a retransmission if no response is heard. However, once the roaming node receives a reply from the prospective relay node 1004, it cancels the time-out that would have triggered a second relay set-up attempt.

During opportunistic relay, transmissions 1020a from the roaming node are relayed 1020b to the AP, and vice versa (i.e. transmissions 1020c and 1020d). Using current, known technologies, the relay station will have to store and forward beacons (and relevant multicasts) to the roaming node, though the development of future technologies in the area of simultaneous transmission and reception may obviate this.

Once the opportunistic relaying link has been set-up and is in operation, there is a chance that the roaming station 1006 will wander back into range of the AP 1002. To cater for this, the roaming station needs to continue to monitor transmissions from other stations. Figure 11 depicts an exemplary procedure of this type. In step Si! 02, the roaming station monitors transmissions from neighbouring nodes in order to identify passing packets having a Source Address (SA) equal to the Access Point's MAC address (step Si 104), which would indicate that the AP is back in range (step S 1104, Yes'). Thus, having served its purpose, the opportunistic relay mode may be torn down 1024, and a return to normal operation takes place (step S 1106).

Various modifications to the above-described embodiment will be apparent to the skilled reader.

For example, an alternative collision avoidance solution to random back-off includes predicting hidden AP transmissions from a priori knowledge of the BCN scheduling andlor traffic patterns obtained from when the roaming node was still in range of the AP. Alternatively still, as the signal strength from the AP drops (subject to hysteresis), the roaming node may pre-emptively send a message to the AP indicating an imminent drop in communication. The AP may then respond with a message indicating that a quiet period at a time I has been reserved for the roaming node, during which period the A? won't transmit in order to try and make contact with a relay node. A number of assumptions are made in this solution, including: that there is enough time for the roaming node to react to the sensed signal strength drop and send its message, as well as enough time for a response from the AP to reach it; that the roaming node is given channel access to send its message; that the message(s) do not get lost; and that not a lot of nodes fall off the network at the same time (e.g. due to a blast of interference or a sudden node migration), which would result in the AP getting swamped with messages.

In addition, although the quiet period has been described in the context of beacon frames, a quiet period can be achieved without incurring the overhead of a dead-period being set aside after every beacon on the off-chance that a node moves too far from the AP. An exemplary time interval of this kind is the Traffic Indication Message window in the IEEE 802.11 power saving mechanism (PSM), for use in operating modes with both infrastructure and without.

In IEEE 802.11 PSM, a node can save energy by going into a dormant state (doze mode). In doze mode, a node consumes much less energy compared to normal mode, but cannot send or receive packets. In IEEE 802.11 PSM, time is divided into beacon intervals, and every node in the network is synchronized by periodic beacon transmissions. In this way, every node will start and finish each beacon interval at about the same time. Once the synchronization beacon has been transmitted, power management is planned based on Ad hoc Traffic Indication Messages (ATIMs), which indicates pending unicast traffic. Announcements of broadcast and multicast traffic is performed by means of Delivery Traffic Indication Messages (DTIM).

Thus, an alternative solution is to use the AT[M/DTIM guarded periods, and signalling the presence of the incoming "mayday" message via the ATIIM traffic indication map.

This then warns stations in range that the transmission will occur, though collisions due to hidden nodes might still occur. Cleary, there is therefore a trade-off between reliability and efficiency.

In order to reduce the amount of promiscuous packet reception that the roaming station has to perform (which has a power impact), the received signal strength of the relay station may be used as a trigger. As a generalisation, if the signal strength from the relay station is getting weaker, then it would not be worth scanning for messages from the AP. This is because it is unlikely in such a scenario that the roaming node is actually getting closer to the AP. This assumption could be bolstered with a periodic sweep of all passing packets just as a back-up. Such an approach would still consume less power than continually scanning all passing messages for the duration of the opportunistic relaying mode.

In certain network configurations, nodes only receive and decode packets sent from the AP (that is, ignoring all packets having a Source Address set to anything other than that of the AP's identifier), in order to save power and to avoid duplicate packet issues. In the art, this is referred to as "infrastructure mode" or "managed mode". In such circumstances, unless a node goes into "master mode" and becomes an AP, or switches to a promiscuous mode of operation, relay requests transmitted from a roaming node Therefore, the AP may activate a node into a more promiscuous receiver mode when needed. In particular, the AP may initiate the relay function of another node in response to an observation of excessive retransmissions to that node or based on signal level degradation. The AP may either receive information about a candidate relay node from the outlying node before it loses contact, or determine the identity of the candidate relay node itself based on estimated direction-of-arrivals of the signals from the outlying node. The AP could then request that the identified candidate relay node contacts the outlying node, knowing that the outlying node has switched to a promiscuous mode of operation and is looking for candidate relay nodes. During relay operation, the MAC address of the roaming node could be included in transmissions from the AP. In other words, the remote node may receive packets having its own MAC address as well as packets having the MAC address of the roaming node, which it then forwards. This is the simplest and most transparent solution: that of a basic repeater.

The state machine for a passive repeater known in the prior art is depicted in figure 12.

Such repeaters continuously sample transmissions of the network (step S 1202). When a problem is detected (step Sl204, Yes'), for example when a relay request is received from a roaming node, the repeater enters a passive repeater mode (step S 1206). Here, the repeater is configured to merely forward data it receives without translating or changing the data in any way. Alternatively, the relaying node can decode and repackage the relayed packets with different addresses, acting as a proxy for the AP when talking to the remote node and vice versa. The node then has to operate as a layer 2 bridge (like the Network Address Translation mechanism at layer 3) and be more intelligent. IEEE 802.1 is compliance supports this functionality. In step S 1208, the repeater again continuously analysis traffic so that it can determine whether the problem is gone (step Si 210), which could be based on a relay tear down message from the roaming node.

Further variations, modifications and additional features will be apparent to the skilled person considering the above disclosure and no statement above is intended to limit the scope of protection sought for the invention, which is to be determined by reference to the appended claims, interpreted in the light of, but specifically limited to, the above description of specific embodiments. In particular, it will be apparent that although the present invention can be easily implemented in already established networks and MAC protocols in order to obtain additional reliability, the invention is not limited to the exemplary network-and protocol-types described herein.

Claims (23)

1. CLAIMS: 1. A method of retaining connectivity to a network by a wireless communication device, the network comprising a plurality of nodes, the method comprising: monitoring for signals transmitted by neighbouring nodes; searching said signals for information content, said information content comprising an identifier of the network; and transmitting a request to at least one of said neighbouring nodes if said information content is found, each of the at least one of said neighbouring nodes corresponding to a signal in which said information content is found, said request comprising a request to serve as a relay between the wireless communication device and a destination node.
2. 2. A method in accordance with claim 1 wherein the information content further comprises an identifier of the destination node.
3. 3. A method in accordance with claim 2 further comprising checking whether said identifier of the destination node is located in a portion of said signal that is indicative of a destination of said signal.
4. 4. A method in accordance with any one of the preceding claims wherein said transmitting a request is performed in a time interval during which said network is substantially quiet.
5. 5. A method in accordance with claim 4 wherein said time interval is reserved by the destination node, in response to a request therefor by the wireless communication device.
6. 6. A method in accordance with claim 4 wherein said time interval is defined by one of: a medium access contention mechanism, a distributed beacon mechanism, and an ATIM/DTIM mechanism.
7. 7. A method according to any of claims 1 to 6 further comprising receiving a reply from a said neighbouring node corresponding to a signal in which said information content is found, said reply comprising information indicating a willingness to serve as a relay between the wireless communication device and the destination node.
8. 8. A method according to claim 7 wherein the neighbouring node from which said reply is received has been selected by the destination node to serve as a relay node between the wireless communication device and the destination node.
9. 9. A method in accordance with claim 8 wherein the selection of said neighbouring node by the destination node is based on information indicative of a direction to or a location of said neighbouring node relative to the wireless communication device.
10. 10. A method of reconfiguring a communication link by a wireless communication device, from a current indirect link to a destination node over a relay node, to a direct link to the destination node, the method comprising: monitoring for signals transmitted by neighbouring nodes; searching said signals for information content, said information content comprising an identifier of the destination node; checking whether said identifier of the destination node is located in a portion of said signal that is indicative of an origin of said signal; and requesting establishment of a direct link to the destination node if said checking is true.
11. 11. A method according to claim 10 wherein said monitoring is performed in response to a relative increase in signal strength of the relay node.
12. 12. A method in accordance with any one of claims 1 to 9 further comprising implementing the method of claim 10 or 11.
13. 13. A method according to any one of claims 1 to 12 wherein the network is in accordance with an IEEE 802.11 standard.
14. 14. A method according to any one of claims 1 to 13 wherein the destination node comprises an access point managing communications of said network using a first protocol.
15. 15. A wireless communication device adapted to retain connectivity to a network comprising a plurality of nodes, the device comprising: means for monitoring for signals transmitted by neighbouring nodes; means for analysing said signals for information content, said information content comprising an identifier of the network; and means for transmitting a request to at least one of said neighbouring nodes if said information content is found, each of the at least one of said neighbouring nodes corresponding to a signal in which said information content is found, said request comprising a request to serve as a relay between the wireless communication device and a destination node.
16. 16. A wireless communication device in accordance with claim 15 wherein said information content further comprises an identifier of the destination node.
17. 17. A wireless communication device in accordance with claim 15 wherein said means for analysing is further operable to check whether said identifier of the destination node is located in a portion of said signal that is indicative of a destination of said signal.
18. 18. A wireless communication device in accordance with any one of claims 15 to 17 wherein the device is adapted to transmit said request in a time interval during which said network is substantially quiet.
19. 19. A wireless communication network comprising a plurality of nodes at least one of which is configured in accordance with any one of claims 15 to 18.
20. 20. A storage medium storing computer executable instructions which, when executed on general purpose computer controlled communications apparatus, cause the apparatus to become configured to perform the method of any of claims ito 14.
21. 21. A signal carrying computer receivable information, the information defining computer executable instructions which, when executed on general purpose computer controlled communications apparatus, cause the apparatus to become configured to perform the method of any of claims ito 14.
22. 22. A storage medium storing computer executable instructions which, when executed on general purpose computer controlled communications apparatus, cause the apparatus to become configured to a wireless communication device in accordance with any one of claims 15 to 19.
23. 23. A signal carrying computer receivable information, the information defining computer executable instructions which, when executed on general purpose computer controlled communications apparatus, cause the apparatus to become configured to a wireless communication device in accordance with any one of claims 15 to 19.AMENDMENTS TO THE CLAIMS HAVE BEEN FILED AS FOLLOWS1. A method of retaining connectivity to an access point of a network by a wireless communication device, the network comprising a plurality of nodes, the method comprising: monitoring for signals transmitted by neighbouring nodes; searching said signals for information content, said information content comprising an identifier of the network and an identifier of the access point; and transmitting a request to at least one of said neighbouring nodes if said information content is found, each of the at least one of said neighbouring nodes corresponding to a signal in which said information content is found, said request comprising a request to serve as a relay between the wireless communication device and the access point.2. A method in accordance with claim 1 further comprising checking whether said : identifier of the access point is located in a portion of said signal that is indicative of a *. destination of said signal. * **3. A method in accordance with any one of the preceding claims wherein said *** transmitting a request is performed in a time interval during which said network is substantially quiet. **** S. * * . S * *.4. A method in accordance with claim 3 wherein said time interval is reserved by the access point, in response to a request therefor by the wireless communication device.5. A method in accordance with claim 3 wherein said time interval is defined by one of: a medium access contention mechanism, a distributed beacon mechanism, and an ATIMIDTIM mechanism.6. A method according to any of claims 1 to 5 further comprising receiving a reply from a said neighbouring node corresponding to a signal in which said information content is found, said reply comprising information indicating a willingness to serve as a relay between the wireless communication device and the access point.7. A method according to claim 6 wherein the neighbouring node from which said reply is received has been selected by the access point to serve as a relay node between the wireless communication device and the access point.8. A method in accordance with claim 7 wherein the selection of said neighbouring node by the access point is based on information indicative of a direction to or a location of said neighbouring node relative to the wireless communication device.9. A method of reconfiguring a communication link by a wireless communication device, from a current indirect link to an access point of a network over a relay node of the network, to a direct link to the access point, the method comprising: monitoring for signals transmitted by neighbouring nodes; * : searching said signals for information content, said information content * comprising an identifier of the network and an identifier of the access point; * *, checking whether said identifier of the access point is located in a portion of said signal that is indicative of an origin of said signal; and S..requesting establishment of a direct link to the access point if said checking is t * . S... S. * * S** 10. A method according to claim 9 wherein said monitoring is performed in response to a relative increase in signal strength of the relay node.11. A method in accordance with any one of claims 1 to 8 further comprising implementing the method of claim 9 or 10.12. A method according to any one of claims 1 to 11 wherein the network is in accordance with an IEEE 802.11 standard.13. A method according to any one of claims 1 to 12 wherein the access point manages communications of said network using a first protocol.14. A wireless communication device adapted to retain connectivity to an access point of a network comprising a plurality of nodes, the device comprising: means for monitoring for signals transmitted by neighbouring nodes; means for analysing said signals for information content, said information content comprising an identifier of the network and an identifier of the access point; and means for transmitting a request to at least one of said neighbouring nodes if said information content is found, each of the at least one of said neighbouring nodes corresponding to a signal in which said information content is found, said request comprising a request to serve as a relay between the wireless communication device and the access point.15. A wireless communication device in accordance with claim 14 wherein said means for analysing is further operable to check whether said identifier of the access point is located in a portion of said signal that is indicative of a destination of said : signal.* S. S * S ** * 16. A wireless communication device in accordance with claim 14 or 15 wherein the device is adapted to transmit said request in a time interval during which said network is substantially quiet. S... * . *...* 17. A wireless communication network comprising a plurality of nodes at least one of which is configured in accordance with any one of claims 14 to 16.18. A storage medium storing computer executable instructions which, when executed on general purpose computer controlled communications apparatus, cause the apparatus to become configured to perform the method of any of claims 1 to 13.19. A signal carrying computer receivable information, the information defining computer executable instructions which, when executed on general purpose computer controlled communications apparatus, cause the apparatus to become configured to perform the method of any of claims 1 to 13.20. A storage medium storing computer executable instructions which, when executed on general purpose computer controlled communications apparatus, cause the apparatus to become configured to a wireless communication device in accordance with any one of claims 14 to 16.21. A signal carrying computer receivable information, the information defining computer executable instructions which, when executed on general purpose computer controlled communications apparatus, cause the apparatus to become configured to a wireless communication device in accordance with any one of claims 14 to 16. * * * ** S **** * S S... * *S * S S *55 S *S *S.. * S *5*I S. SS SS * .5