GB2466469A - Sharing a common wireless communication channel - Google Patents

Abstract

Method, system, node and signal for coordinating communications from two or more nodes along a common channel, where each node is preferably associated with a respective Basic Service Set (BSS). In the method and system the nodes, upon detecting the existence of unacceptable co-channel interference or the future threat of unacceptable co-channel interference along said common channel, communicate with one another to determine two or more time segregated medium access periods P1, P2, for the common channel, where each time segregated medium access period P1, P2 is dedicated to the signals of a respective node, for example, dedicated to a respective node and its associated basic service set. One or more of the nodes may be an Access Point (AP). The invention is presented as a 'survival MAC protocol' that is an extension to the 802.11 MAC protocol. The invention enables two or more BSSs, APs or WLANs to co-exist on the same channel and can be used when a new Basic Service Set (BSS) is being created in an area where all channels are occupied by other BSSs or to maintain connectivity or Quality of Service (QoS) when there is high inter-channel or co-channel interference and no interference-free channels available for dynamic channel re-allocation.

Description

I

Method and System for utilization of a wireless network The present invention relates to wireless communication and more particularly to a method of and system for sharing a wireless communication channel.

The most basic topology of an IEEE 802.11 network is the BSS (Basic Service Set), in which the wireless nodes are able to communicate either directly (ad hoc) or withlthrough an Access Point AP (infrastructure). An example of a BSS is illustrated in Figure 1, the BSS consisting of three wireless nodes 2, 4, 6 and an AP8, all located within an area defined by the circumference of a circle 12. An Ethernet 10 is also depicted in Figure 1. The Ethernet 10 is not a part of the BSS but communication between the nodes 2, 4, 6 and the Ethernet 10 is established through the AP8.

Current art allows two co-existing BSSIAPs to share the same channel using the standard 802.11 MAC, e.g. by means of contending for access (based on the CSMA!CA protocol).

Such teelmiques are described in IEEE 802.11-2007, Standard for Information technology-Telecommunications and information exchange between systems-Local and metropolitan area networks-Specific requirements -Part 11: Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) Specifications, 2007, the whole contents of which are herein incorporated by way of reference; and M. Gast, 802.11 Wireless Networks: The Definitive Guide, Second Edition: O'Reilly Media, Inc., 2005, the whole contents of which are incorporated herein by way of reference.

This approach generally works well in many cases where various devices contend for access and their traffic loads vary in time. One problem, however, with this approach is that nodes are vulnerable to co-channel interference (e.g. due to noise from far transmitting nodes, hidden/exposed node problems, collisions, etc). There are certain scenarios when this problem is exacerbated, for example dependent on the relative positions of APs and wireless nodes within different, adjoining BSSs preventing certain wireless nodes, from communicating with one another.

The IEEE 802.11 specification referred to above defines the Distribution System (DS) as a generic architectural component used to interconnect BSSs (e.g. for range extension). The DS and BSSs allow IEEE 802.11 to create a wireless network of arbitrary size and complexity which is known as the Extended Service Set (ESS) network (as seen in Figure 2). The ESS shown in simplified diagrammatic form in Figure 2 illustrates two Basic Service Sets BSS I, BSS2, each provided with a respective AP (STA2, STA3), for interconnecting the Basic Service Sets BSSI, BSS2 by way of the Distribution System (DS). Nodes STA1 and STA4 are shown located in BSS1 and BSS2 respectively.

The technology that is mainly used for inter-AP channel sharing is the 802.11 Wireless Distribution System (WDS), which is based on Layer-2 bridging (e.g. by means of a four address format) and uses the standard 802.11 MAC protocol for medium access.

This technology is described in D. Gupta, J. LeBrun, P. Mohapatra, and C.N. Chuah, C.N., "WDS-based layer 2 routing for wireless mesh networks, Proceedings of the 1st international workshop on Wireless network test beds, experimental evaluation and characterization, pp. 99-100, 2006 the whole contents of which are incorporated herein by reference; and in Darwin Engwer, IEEE 802.1 1-0510710r0, Wireless LANs: "WDS" Clarifications, 2005-07-19, the whole contents of which are incorporated herein by reference. In the latter reference the WDS provides a framework in which APs form various network topologies and may either share a channel in overlapping areas or operate on different channels. The WDS uses standard 802.11 MAC protocols which include various techniques both at the PHY or MAC layer. These techniques, which will be described in more detail hereinafter, have been found inadequate in adverse network conditions.

In a WDS interconnected BSSs may either share a channel (through the standard carrier sensing mechanism) or operate on different channels. In either case dynamic optimisations may maximise network capacity and reduce interference. Such optimisations may be leveraged by an implementation of the Inter-Access Point Protocol (IAPP), which provides necessary capabilities to achieve multi-vendor AP interoperability within a DS.

An IAPP is described in IEEE, Recommended Practice for Multi-Vendor Access Point Interoperability via an Inter-Access Point Protocol Across Distribution Systems Supporting IEEE 802.11 Operation, IEEE Draft 802.lf, January 2003, the whole contents of which is incorporated herein by way of reference. The IAPP can be used by APs to communicate with each other and implement network optimisations.

As such, in the subject invention an IAPP can (optionally) be used by an AP for the specific purpose of negotiating survival parameter as will be described later.

Also, the IEEE 802.11k draft standard, the whole contents of which is enclosed herewith by reference, may be used for inter-AP communications. 802.11k specifies mechanisms for managing and communicating "measurement requests/reports" for some types of radio information. This exchanged information can then be available to higher network layers for further optimisations.

In terms of achieving the objective, which is the avoidance or reduction of inter-channel and co-channel interference, some technologies employ dynamic channel allocation and interference avoidance and cancellation techniques. A typical dynamic channel allocation algorithm is described in M. W. R. Silva, and J. F. Rezende, "A dynamic channel allocation mechanism for IEEE 802.11 networks", In 4th International Telecommunications Symposium (ITS), 2006, the whole contents of which is incorporated herein by way of reference. Dynamic channel (re)-allocation is, however, only applicable when there is an interference-free channel. This may not be possible in densely populated areas.

In terms of survivability mechanisms, relevant prior art includes using multi-hop routing / layer-2 routing / bridging and adding redundant APs (e.g. extending a BSS to an ESS). These techniques are described in D. Chen, S. Garg, and K. S. Trivedi, "Network survivability performance evaluation: A quantitative approach with applications in wireless ad-hoc networks," Proceedings of the 5thi ACM international workshop on Modelling analysis and simulation of wireless and mobile systems, pp. 61- 68, 2002; and in D. Chen, S. Garg, C. Kintala, and K. S. Trivedi, "Dependability enhancement for IEEE 802.11 wireless LAN with redundancy techniques," Dependable Systems and Networks, 2003. Proceedings. 2003 International Conference on, pp. 521- 528, 2003, the whole contents of both of these technical articles being incorporated herein by reference. However all these survivability techniques are different to the one proposed here, and they are used to address completely different problems (such as shadow regions and range extension).

Conventional technology typically makes use of a predefined number of channels, 14 channels are designated in the 2.4 GHz for the 802.11 b/g. However not all of them are necessarily available for use. Countries apply their own regulations to allowable channels (as well as allowed users and maximum power levels). For example, in Japan all 14 channels can be used, in USA 11 channels and in Europe 13 channels (excluding France and Spain that only 4 and 2 channels can be used, respectively).

What is more, adjacent BSS/APs (for the 802.llb/g amended IEEE Standard must be separated by five channels to prevent inter-channel interference. As such, only three channels can be used for APs with overlapping coverage (The 802.1 la -1999 operating in 5.4 GHz offers 12 non-overlapping channels). Figure 3 shows a typical frequency planning scheme for 802.1 lb/g networks, channels 2, 7 and 12 being chosen and each channel adjoining only a different channel. Thus channel 2 only adjoins channels 7 and

12 for example.

The discussed channel scarcity resorts in either having no channel available for use for a new AP, or having increased interference. There are two main sources of interference: co-channel and interchannel interference. Interchannel interference occurs when where are (in 802.llb/g) overlapping channels. Co-channel interference is a result of overlapping cells that share the spectrum on the same channel. According to a study by M.W.R. Silva and J.F. Rezende, in a paper entitled "A dynamic channel allocation mechanism for lEE 802.11 networks", in 411) International Telecommunications Symposium (ITS), 2006, the 802.11 transmissions generate three reception regions where receiving stations behave differently: * The reception zone, nearest to the transmitter, where receiving stations can notice and decode packets.

* The carrier sense zone (the intermediate one), where receiving stations can only notice packets transmissions due to energy levels but they cannot be decoded.

* The interference zone, where receiving stations cannot notice packet transmissions but their energy contribution is added to the noise floor as interference.

The three reception regions can be seen depicted in Figure 4: Stations (1), (2) and (3) share the same channel. When station (1) transmits, stations (2) aid (3) must wait the end of this transmission before starting another one. Station (4), which is outside the reception and carrier sense zones, will suffer co-channel interference generated as a result of station's (1) transmissions (which are received by station (4) as noise).

Also, co-channel interference will result in collisions in Station (3) when Station (1) accesses the medium to transmit to (3) and (at the same time) channel (4) also accesses the medium to transmit (it doesn't matter where to). This is another type of co-channel interference and is known as the hiddenlexposed node problem.

The IEEE 802.11 MAC uses two techniques to combat interference: physical carrier sensing and virtual carrier sensing.

Physical carrier sensing has limitations. For example, interference may happen at receivers, while physical carrier sensing at transmitters senses a free medium (e.g., in a hidden terminal situation). Hence, physical carrier sensing cannot help much, unless a very large carrier sensing range is adopted, which is limited by factors such as power and antenna sensitivity.

Virtual carrier sensing also has its own limitations. The virtual carrier sensing is implemented with the Network Allocation Vector (NAy). Part of the NAV mechanism is the RTS/CTS handshake, which is used to reduce the risk of collisions. However, the effectiveness of the RTS/CTS handshake is based on the assumption that hidden nodes are within transmission range of receivers. This assumption may not hold in overlapping cells or in ad hoc networks. One reason for that is that is the fact that the power needed for interrupting a packet reception is much lower than that of delivering a packet successfully.

The need to support applications with demanding QoS requirements has been recognised and addressed to some degree by IEEE 802.lle amendment, which has enhanced the MAC protocol with the Hybrid Coordination Function (HCF). The HCF offers a sophisticated way of prioritising traffic as well as reserving the medium (with HCCA and EDCA protocols). Still, HCF is based on CSMAICA is not exempt of the discussed interference problems. To make this clearer, we need to make a distinction between collisions due to inter-BSS traffic and co-channel interference in overlapping BSSs: Contention-based traffic (i.e. with the use of the EDCA protocol) collisions are due to the (physical and virtual) carrier sensing limitations discussed above. Allocation-based traffic (i.e. with the use of the HCCA protocol) collisions should (ideally) not take place when there is only one BSS, since the Hybrid Controller (HC) (e.g. the AP) fully controls which node accesses the network and when (one exception to this is that collisions may happen during the HC polling phase). When there are overlapping BSSs, however, HCCA is still subject to co-channel interference, as discussed.

The present invention strives to minimize the problems associated in the prior art systems.

According to one aspect of the invention there is provided a method of coordinating communications from two or more nodes along a common channel, each node being associated with a respective Basic Service Set (BSS), the method comprising detecting the existence of unacceptable co-channel interference or the future threat of unacceptable co-channel interference along said common channel and, when said existence or future threat of unacceptable co-channel interference is detected, changing the operating mode of communication of the nodes with the coninion channel, wherein the nodes communicate with one another to determine two or more time segregated medium access periods for the common channel, each of said time segregated medium access periods dedicated to the signals from a respective node.

In one embodiment at least one of the nodes is an Access Point (AP). Each BSS has a respective access point (AP), the method comprising detecting the existence of unacceptable co-channel interference or the future threat of unacceptable co-channel interference along said common channel, and when said existence or future threat of unacceptable co-channel interference is detected changing the operating mode of communication of the APs with the common channel, wherein the access points (APs) communicate with one another to determine two or more time segregated medium access areas dedicated to the different APs and their respective BSS.

In another embodiment the method includes the step of detecting whether there are interference-free channels available when the existence of unacceptable co-channel interference or the future threat of unacceptable co-channel interference along said common channel is detected, and, if an interference-free channel is detected, reallocating one or more nodes to the interference free channel or channels.

In a further embodiment if the threat of future unacceptable co-channel interference is detected one of the APs determines whether it has received a survival MAC request from a peer AP. If a survival MAC request has been received from a peer AP, the APs negotiate survival MAC parameters. After said survival MAC parameters have been negotiated, the APs switch to a survival MAC mode of operation. After switching to the survival MAC mode the APs determine whether the BSS nodes are compatible with the survival MAC mode; and if compatible with the survival MAC mode, the nodes update survival MAC parameters.

Alternatively, if the BSS nodes are not compatible with the survival MAC mode a compatibility mode is established; the compatibility mode is established by the APs scheduling survival NAV updates to the nodes of their respective BSS.

In yet a further embodiment if a survival MAC request has not been received from a peer AP, the first AP sends a request for switching to the survival MAC mode to the peer. If the request is accepted by the peer AP, the APs negotiate initial survival MAC parameters. If the request has not been accepted the AP waits until the future threat of co-channel interference is detected and if detected determines whether a survival MAC request has been received from a peer AP.

According to a further aspect of the invention there is provided a system for coordinating communication from two or more nodes along a common channel, the system comprising a detector for detecting the existence of unacceptable co-channel interference or the future threat of unacceptable co-channel interference along said common channel and means for changing the operating mode of communication with the common channel when said unacceptable co-channel interference or threat of unacceptable co-channel interference has been detected, whereby the nodes communicate with one another to determine two or more time segregated medium access periods for the common channel, each of said time segregated medium access periods dedicated to the signals from a respective node.

In various embodiments of the system, the system is adapted to carry out the embodiments of the method referred to above.

In a further aspect of the invention there is provided a signal for use in a wireless network comprising two or more data packets for transmission along a common channel, each data packet having a respective time segregated access period dedicated to a respective node, the time segregated access periods being determined prior to transmission by communication between the nodes.

According to a yet further aspect of the invention there is provided a node for use in wireless communication in which two or more nodes communicate along a common channel, wherein the node is operable to change its operating mode of communication along said common channel and can communicate with one or more other nodes to determine two or more time segregated medium access periods for the common channel, each of said time segregated medium access periods dedicated to signals from a respective node.

In various embodiments the node is adapted to be operable for carrying out the embodiments of the method referred to above.

More generally, the proposed MAC protocol (called a survival MAC) is an extension to the 802.11 MAC protocol, which allows two 802.11 Basic Service Sets (BSS) (or two Access Points (APs)) to segregate themselves in time on a common channel in a non-interfering manner. The survival MAC has two main uses: * Create a new BSS in an area where there is channel scarcity (all available channels are occupied by other BSSs) * Maintain connectivity (or else application QoS) when there is high inter-channel interference or co-channel interference, and there is no interference-free channel for dynamic channel re-allocation.

One of the novel ideas is a mechanism that enables two different WLAN networks (i.e. two BSSs or two APs) to co-exist on the same channel (i.e. on a common, single medium) by limiting each one's access to a portion of the medium time.

The survival MAC is a fault-response technique proposed here as an alternative / complementary solution to the problems described above. It may guarantee connectivity under extreme interference conditions at the expense of bandwidth.

The present invention will be described further by way of example with reference to, and as illustrated in, the accompanying drawings:-Fig. 1 illustrates a Basic Service Set (BSS) topology of an IEEE 802.11 network; Fig. 2 is an Extended Service Set (ESS); Fig 3 illustrates a typical frequency planning scheme for 802.1 lb/g networks; Fig. 4 is an example of how CSMAICA activity can avoid interference; Fig. 5 is an illustration of two neighbouring/overlapping BSSs; Fig. 6 is a survival MAC protocol architecture in accordance with an embodiment of the invention; Fig. 7 illustrates IEEE 802.1, MAC protocols for two BSSs; Fig. 8 illustrates a survival MAC protocol for two BSSs in accordance with an embodiment of the invention; Fig. 9 illustrates BSS segregation with a survival MAC protocol in accordance with an embodiment of the invention; Fig. 10 illustrates NAV updates for backwards compatibility with IEEE 802.11; and Fig. ii is a flowchart for a survival MAC process in accordance with an embodiment of the invention.

In the following description, specific implementations of the invention are described. It will be appreciated that these are provided by way of example only, and are not intended to provide restriction or limitation on the scope of the invention which is defined in the appended claims.

The embodiment described below considers the case in which there is no available interference-free WLAN channel (causing connectivity intermittence or no connectivity at all) due to interchannel or co-channel interference in overlapping or adjacent WLAN cells (Figure 5).

Figure 5 depicts two overlapping BSSs, namely BSSA and BSSB, each defined within respective circular areas 20, 22. The overlapping area is shown as zone 24. BSSA comprises three nodes 26, 27, 28 and an AP 30, whilst BSSB comprises three wireless nodes 32, 34, 36 and an AP 38. It can be seen that the nodes 26, 28, 34, 36 and both APs 30, 38 are located within the overlapping zone 24 of BSSA and BSSB.

Interference can be caused by a number of causes and there are a number of measures that current art offers to combat these problems (as discussed above).

The proposed survival MAC is another mechanism that may be applied in cases where current teclmiques fail to guarantee dependable connectivity and application Q0S requirements. For example: * Power control may not solve the problem if the APs of the neighbouring / overlapping BSS's are very close to each other (see Figure 5) and they cannot be moved (e.g. neighbours may not want to negotiate on an optimised AP installation location) * Dynamic channel selection may not be applicable, if there are no other available / non-interfering channels (e.g. in a highly dense area, a large number of BSS/APs could co-exist, occupying all available channels -see also Figure 3, for 802.1 lb/g channel assignment, as previously discussed) The fundamental idea is to organise the way two co-existing WLANs share the same mediunilchannel by limiting their access to well defined regions (in contrast with the more chaotic manner that is offered by the current CSMAICA-based MAC protocols).

With the survival MAC enhancements to the 802.11 MAC, the two BSS/APs can be logically segregated in time following appropriate negotiation of sharing co-existence parameters.

The survival technique may guarantee connectivity under extreme interference conditions for multiple co-existing WLANs, while maintaining each network's independence (and security). On the downside, this survival mechanism will significantly reduce each channel's resources, which could even lead to the service cessation of the least critical applications, in favour of the survival of applications with the most critical QoS requirements.

The survival MAC according to the invention is an enhancement to the standard MAC protocol that decides whether to initiate the survival MAC protocol enhancements, only use the "standard" MAC protocol, or make no change to the current state. This is illustrated in Figure 6.

Firstly will be described the structure of the survival MAC protocol (see Figure 6), and later the survival MAC operations and decision-making procedure will be discussed.

According to the standard IEEE 802.11 MAC protocol, a beacon (BCN) is periodically transmitted every Target Beacon Transmission Time (TBTT), subject to potential drifts that may occur. After the BCN, access to the medium is organised by the HCF protocol.

The HCF protocol is described in IEEE 802. lle-2005, Amendment to IEEE Standard for Information technology -Telecommunications and information exchange between systems -Local and metropolitan area networks -Specific requirements Part 11: Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) specifications: Medium Access Control (MAC) Quality of Service Enhancements, the whole contents of which is incorporated herein by way of reference. In Figure 7 there can be seen the MAC protocols of two different BSSs operating on two different channels (Chi and Ch2) and each one of them running its own HCF MAC protocol.

The structure of the survival MAC protocol in accordance with an embodiment of the invention, can be seen in Figure 8. The two BSSs are now both operating on the same channel and they are segregated in time. This segregation is organised along the following lines: The BCN for each channel will still be transmitted periodically every TBTT as specified in the IEEE 802.11 (note here that both BCNs need to have the same TBTT). The relative time P1 (see figure 8) determines when the BCN2 transmission will occur after the BCN1 transmission. Equally P2 determines when the BCN1 transmission occurs after BCN2. In overall TTT = P. +-P and P1 and P2 determine the portion of the medium that each BSS will have and will need to be negotiated, for the purposes of the co-existence.

Details about this negotiation will be described later below.

It should be noted that P1 and P2 may not be tightly maintained (for the same reasons TBTT drifts). For example, it is common for small delays to occur due to ongoing transmissions (e.g. a packet transmission is half way through when the time the next BCN comes and, thus, the AP needs to wait for the medium to go free before obtaining it, after a PIFS). Delays may also occur due to clock drifts. In both cases the absolute beacon transmission times may be shifted. When such drifts occur, the BCNs will simply maintain their TBTT, P1 and P2 timings as per normal (resulting in all future timings being shifted, which is of no particular importance). As far as clock drifts are concerned, both API and AP2 will need to keep on synchronising their clocks every time they receive the peer AP's BCN.

After the broadcast of BCN1, only the nodes associated with the first (logical) channel (Chi) (i.e. BSS1D1) are allowed to access the medium (with the IEEE 802.11 HCF protocol), up until the broadcast of the BCN2, after which only the nodes associated with BSSID2 are allowed to access the medium, and so on (see Figure 9).

In order to prevent devices of Clii from accessing the HCF-Ch2 period and devices of Ch2 from accessing the HCF-Chl period (which should normally happen according with the IEEE 802.11 standard) we propose the following mechanism, which maintains backwards compatibility: Our minimum requirement is only the APs of Chi and Ch2 to have the survival MAC protocol stack (Figure 6) (e.g. all other devices only have the standard 802.11 MAC protocol stack). In the survival state, before the broadcast of BCN2, the AP1 should always get access to the medium (e.g. with a PIFS, which will give it priority over all other nodes) and it should broadcast a control packet that all Chi nodes will read and will update their NAVs as indicated by the following formula: ckoff.

On an implementation node, that (special) control packet (that we call survival NA V update message) could, for example, be a DATA packet (with an empty payload), in the MAC header of which the destination MAC address is all l's (indicating broadcast) and the DurationllD field is set to ?cwkoff1 (for updating the NAy).

Thus, all Chi nodes will then defer up until BCN1 is again broadcasted, leaving the backoff duration for Ch2 access (BCN2 and HCF-Ch2). The AP2 will (similarly) defer all Ch2 traffic just before the broadcast of BCN1 for: ckoff.

This method of using NAV updates that maintain backwards compatibility can be seen in Figure 10.

Note that if all the wireless nodes in Chi and Ch2 are equipped with the survival MAC protocol, then the NAV update control frame is not necessary. All Chi nodes will read the BCN2. The BCN2 will contain a BSSID2, a DurationllD2 value (for Ch2 nodes) and the TTT value in the Beacon Interval subfield of its Frame Body field. All the Chi nodes will then calculate P. = 1, -T., where is the time BCN1 has been sent (which time the nodes will have kept in memory) and T: is the time BCN2 is transmitted. The Chi nodes will update their NAVs for: = TTT -P, = -One alternative / supplementary implementation is that BCN2 may be enhanced with an additional Duration/ID1 field for Chi nodes that will specify that nodes associated with the BSSID1 should defer for P. Similarly, the BCN1 will be used for updating the NAVs of Ch2 nodes (nodes associated with the BSSID2). In overall, all nodes will know that they cannot access the cross-channel MAC straight after receiving the cross-channel BCN.

If a node misses the survival NAV update message then it may attempt to access the peer ESS's portion of medium time (as it would normally do with the standard MAC protocol). In that case the following two alternative approaches may be used: No action can be taken to prevent from such an unfortunate event. That specific node may temporarily disrupt traffic in the peer BSS, which would happen anyway if the survival MAC was not in force. Alternatively, the AP2 (or some other node in Ch2) may periodically attempt to update the NAV1 (apart from only updating NAy2, as per standard). Alternative implementations of such mechanisms should be apparent to the skilled person.

In one embodiment, survival MAC parameter updates can be re-negotiated between AP1 and AP2 if their applicationltraffic requirements deem that necessary. For example, consider the situation where there are two BSSs in survival mode where a 50/50 split (i.e.?. = ?:) of the medium has been agreed. Suppose now that one BSS becomes very heavily loaded and the other isn't. In that case the 50/50 split is inefficient and the survival parameters need to be re-negotiated and updated. Re-negotiation may be done through an IAPP the same way that the initial negotiation may take place. New L and: can easily be updated and applied with either the survival NAV update messages or alternative implementations, as discussed.

In another embodiment, inter-BSS APs may not be in each other's transmission range.

In that case, the APs may communicate (and stay synchronised) via an intermediate node, on an ad hoc basis.

The two BSSs can still maintain their data security in the survival mode, exactly as they do when operating on separate channels as per IEEE 802.11 standard. Encryption with separate session keys, access control mechanisms and authentication are maintained.

The security of willingly sharing the MAC medium is the same with the security of having an outsider sniffing the whole channel. Channel time sharing does not mean or require Chi nodes to associate with the Ch2 and vice versa. However, it should be mentioned that this can easily be enabled, if, and only if, the two network administrators wish to do so (e.g. very friendly neighbours wish to be able to receive/send traffic to each other).

The decision whether going into a survival state or not (see Figure 6), is taken by the survival MAC decision-making algorithm. The basic reasoning for this decision-making has been described above, and the basic decision making procedure is shown in the flowchart of Figure 11.

We will next describe the basic components of this procedure.

Threat for survivability detection Whether it is decided to enter a state of survival MAC, or not (see Figure 6), depends on the detection of the associated threat and analysis of the current situation. This has been described earlier and is illustrated in the upper part of the flowchart in Figure 11.

Channel re-allocation If there is any available (inter-channel and co-channel) interference-free channel, then standard channel re-allocation may solve potential "threats" as discussed. Survival MAC is an alternative technique used when channel (re)-allocation is not possible (or does not solve the problem). It should be considered as a last resort to fall back on, as survival MAC limits BSSs to only use portions of their channel time, as described above.

Send / receive survival MAC request A switch to the survival MAC mode may be requested by any of the two (or more) peer survival MAC-enabled nodes (APs), from the overlapping BSSs.

It is anticipated that the peer APs in the overlapping BSSs will cooperate and will accept to share the channel (by restricting themselves to portions of it) as this will reduce interference for both of them.

Survival MAC parameter negotiation The peer survival MAC nodes (APs) will have to agree on the amount of access time that will be allocated for each one of their own BSSs. This negotiation will depend on the QoS application requirements that each AP needs to satisfy for its own network. For example the main two attributes that could be negotiated are P and 2.

One simple way of negotiating those two values is the following: Suppose that the traffic loads of BSSID1 and BSSID2 are and L2, respectively. In order to keep the portions of medium access analogous to the respective traffic loads, the following simple formulas can be used: : -TT On a note here, the peer APs also need to negotiate the TBTT in case they originally used or wish to use different ones. The one used could, for example, be the average of the two proposed ones.

On another note, the Survival MAC parameter: and P can be renegotiated and updated when necessary (for example, this can be done when the traffic requirements and L2 are significantly changed).

Survival MAC backwards compatibility All the nodes (in all BSSs) in survival MAC mode need to adhere to the survival MAC access rules. If all the nodes in a BSS are survival MAC-enabled, then the new access rules can easily communicate by the APs (and the nodes will comply accordingly). If some nodes are not survival MAC-enabled, then backwards compatibility can be maintained with appropriately scheduled NAV updates, as described in Section 9. Note that the two BSSs sharing the same channel do not have to use the same mechanism for making the rest of the nodes survival MAC compliant.

Leaving Survival MAC mode The Survival MAC will dynamically return to the normal 802.11 MAC mode (see Figure 6) when the conditions that invoke the switch to the Survival MAC mode do not hold (preferably for a certain amount of time that ensures that the conditions are stable enough). More specifically, the conditions are the following: * One AP (or node within a BSS) scans an (interference) free channel. If that condition holds for a certain amount of time (satisfying our requirement for stability) then the BSS can be re-allocated to the new channel.

* When traffic patterns and QoS requirements constantly change requiring frequent re-negotiation of the Survival MAC parameters, the co-existing APs/BSSs may find it preferable exit the Survival MAC mode and contend for access in the shared channel in the standard way.

The procedure to exit the Survival MAC mode is the following: An AP sends a message to the peer AP informing it that it will exit the Survival MAC mode (and either keep on contending on the same channel or re-allocate its BSS in another channel). The peer AP does not need to reply, unless it has reasons to object (in that case the process for Survival mode parameter re-negotiation may be initiated). If there is no objection, both APs will either simply terminate the survival NAV updates or announce the new mode of operation to the rest of the nodes (the choice depends on the implementation, as discussed above in the paragraph regarding Survival MAC backwards compatibility).

The two main advantages of the proposed survival MAC are: 1. It can offer to a new AP an interference-free portion of an already used channel (by another AP) to operate on, when there is no other available interference-free channel 2. It can combat co-channel interference in adjacent 802.11 channels, in WDS, in wireless mess 802.11 networks and multi-hop 802.11 networks, in cases that is otherwise unavoidable (e.g. in cases where the standard 802.11 MAC fails due to the hidden!exposed problem) Several Examples of how a survival MAC can be employed in different scenarios are described below.

Example 1:

A scenario that demonstrates how the survival MAC may outperform the standard MAC is the following (see Figure 5): * BSS-A and BSS-B operate on the same channel * There is no other available interference-free channel, so BSS-A and BSS-B need to share the medium * The APs are close to each other and cannot by physically relocated (bad neighbours) * The handheld on the left of the BSS-B's AP can listen to the AP but cannot listen to the handheld on the right of the BSS-A's AP * The handheld on the right of the BSS-A's AP can listen to the AP but cannot listen to the handheld on the left of the BSS-B's AP.

Suppose that only the two mentioned handhelds try to access the internet (through their associated APs). In this case these two devices will never listen to each other transmitting and the receivers (the APs) will always receive corrupted signals.

With the survival MAC, the APs will first negotiate the survival MAC parameters (using some control mechanism such as IAPP or 802.11k) and then they will enforce their BSSs to operate only on portions of the medium, where both devices will never interfere with each other.

Example 2:

A customer installs and turns on a new AP. The AP scans all available channels and finds no free channel (remember that in 802.llb.g there can be only three co-existing channels without interchannel interference). With the survival MAC, the AP can communicate with the closest (strongest signal) AP, or (alternatively in a load balancing manner) with the AP that has the least traffic load and negotiate a survival MAC mode where both APs can share the same channel without interfering with each other (The peer AP needs to accept this offer or sharing or else it may lose more as a result on excessive co-channel or inter-channel interference).

Example 3:

In an 802.11 ad hoc network, the overlapping APs need to both operate in order to offer range extension for some remote devices (e.g. by they could use WDS in order to create an ESS). In a traffic / topology scenario similar to the one described in Examples 1 and 2, the survival MAC may offer an advantage over the standard MAC.

Whilst the embodiments have been described above with respect to variants of the IEEE 802.11 standard, they are equally applicable to other wireless standards with suitable modifications as would be understood by those skilled in the art. With suitable modifications the embodiments may also be implemented in non-wireless networks.

The skilled person will recognize that the above described system and methods may also be embodied as processor control code, for example on a carrier medium such as a disk, CD-or DVD-ROM, programmed memory such as read only memory (Firmware) or on a data carrier such as an optical or electrical signal carrier.

Claims (36)

1. CLAIMS: 1. A method of coordinating communications from two or more nodes along a common channel, each node being associated with a respective Basic Service Set (BSS), the method comprising detecting the existence of unacceptable co-channel interference or the future threat of unacceptable co-channel interference along said common channel and, when said existence or future threat of unacceptable co-channel interference is detected, changing the operating mode of conmntnication of the nodes with the common channel, wherein the nodes communicate with one another to determine two or more time segregated medium access periods for the common channel, each of said time segregated medium access periods dedicated to the signals from a respective node.
2. 2. A method as claimed in claim 1 wherein each BSS comprises one or more nodes, said one or more of the nodes of each BSS operable to use a MAC mechanism (survival MAC) protocol defined by process steps comprised in claim 1.
3. 3. A method as claimed in claim 1 or claim 2 wherein one or more of the nodes is an Access Point (AP).
4. 4. A method as claimed in claim 3 wherein each BSS has a respective access point (AP), the method comprising detecting the existence of unacceptable co-channel interference or the future threat of unacceptable co-channel interference along said common channel, and when said existence or future threat of unacceptable co-channel interference is detected changing the operating mode of communication of the APs with the common channel, wherein the access points (APs) communicate with one another to determine two or more time segregated medium access areas dedicated to the different APs and their respective BSS.
5. 5. A method as claimed in any of claims 1 to 4, comprising the step of detecting whether there are interference-free channels available when the existence of unacceptable co-channel interference or the future threat of unacceptable co-channel interference along said common channel is detected, and, if an interference-free channel is detected, reallocating one or more nodes to the interference free channel or channels.
6. 6. A method as claimed in claim 3 or claim 4 wherein if the threat of future unacceptable co-channel interference is detected one of the APs determines whether it has received a survival MAC request from a peer AP.
7. 7. A method as claimed in claim 6 wherein if a survival MAC request has been received from a peer AP, the APs negotiate survival MAC parameters.
8. 8. A method as claimed in claim 7 wherein, after said survival MAC parameters have been negotiated, the APs switch to a survival MAC mode of operation.
9. 9. A method as claimed in claim 8 wherein after switching to the survival MAC mode the APs determine whether the BSS nodes are compatible with the survival MAC mode.
10. 10. A method as claimed in claim 9 wherein if the BSS nodes are compatible with the survival MAC mode, the nodes update survival MAC parameters.
11. 11. A method as claimed in claim 9 wherein if the BSS nodes are not compatible with the survival MAC mode a compatibility mode is established.
12. 12. A method as claimed in claim 11 wherein the compatibility mode is established by the Al's scheduling survival NAV updates to the nodes of their respective BSS.
13. 13. A method as claimed in claim 6 wherein if a survival MAC request has not been received from a peer AP, the first AP sends a request for switching to the survival MAC mode to the peer BSS.
14. 14. A method as claimed in claim 13 wherein if the request is accepted by the peer AP, the APs negotiate initial survival MAC parameters.
15. 15. A method as claimed in claim 13 wherein if the request has not been accepted the AP waits until the future threat of co-channel interference is detected and if detected determines whether a survival MAC request has been received from a peer AP.
16. 16. A method according to any one of claims 1 to 15 operating according to IEEE 802.11 standard.
17. 17. A system for coordinating communication from two or more nodes along a common channel, the system comprising a detector for detecting the existence of unacceptable co-channel interference or the future threat of unacceptable co-channel interference along said common channel and means for changing the operating mode of communication with the common channel when said unacceptable co-channel interference or threat of unacceptable co-channel interference has been detected, whereby the nodes communicate with one another to determine two or more time segregated medium access periods for the common channel, each of said time segregated medium access periods dedicated to the signals from a respective node.
18. 18. A system as claimed in claim 17 wherein, the system comprises two or more Basic Service Sets (BSSs), one or more of the nodes being associated with a respective BSS, said one or more nodes of each BSS operable to use a MAC mechanism (survival MAC) protocol defined by process steps comprised in claim 17.
19. 19. A system as claimed in claim 17 or claim 18 wherein one or more of the nodes is an Access Point (AP).
20. 20. A system as claimed in claim 19 wherein each node is associated with a respective BSS having a respective access point (AP), the detector detecting the existence of unacceptable co-channel interference or the future threat of unacceptable co-channel interference along said common channel, and when said existence or future threat of unacceptable co-channel interference is detected changing the operating mode of communication of the APs with the common channel, the access points (APs) having processing means and communication means whereby they can communicate with one another to determine two or more time segregated medium access areas dedicated to the different APs and their respective BSS.
21. 21. A system as claimed in any of claims 17 to 20, wherein the detector can detect whether there are interference-free channels available when the existence of unacceptable co-channel interference or the future threat of unacceptable co-channel interference along said common channel is detected, and, if an interference-free channel is detected, allocation means is provided for reallocating one or more nodes to the interference free channel or channel or channels.
22. 22. A system as claimed in claim 19 or claim 20 wherein if the threat of future unacceptable co-channel interference is detected one of the APs is adapted to determine whether it has received a survival MAC request from a peer AP.
23. 23. A system as claimed in claim 22 wherein if a survival MAC request has been received from a peer AP, the APs are adapted to negotiate survival MAC parameters.
24. 24. A system as claimed in claim 23 wherein, after said survival MAC parameters have been negotiated, the APs have switching means to switch to a survival MAC mode of operation.
25. 25. A system as claimed in claim 24 wherein after switching to the survival MAC mode the APs are adapted to determine whether the BSS nodes are compatible with the survival MAC mode.
26. 26. A system as claimed in claim 25 wherein if the BSS nodes are compatible with the survival MAC mode, the nodes are adapted to update survival MAC parameters.
27. 27. A system as claimed in claim 25 wherein if the BSS nodes are not compatible with the survival MAC mode a compatibility mode is established.
28. 28. A system as claimed in claim 27 wherein the compatibility mode is established by the APs scheduling survival NAV updates to the nodes of their respective BSS.
29. 29. A system as claimed in claim 22 wherein if a survival MAC request has not been received from a peer AP, the first AP is adapted to send a request for switching to the survival MAC mode to the peer BSS.
30. 30. A system as claimed in claim 29 wherein if the request is accepted by the peer AlP, the APs negotiate initial survival MAC parameters.
31. 31. A system as claimed in claim 29 wherein if the request has not been accepted the AlP waits until the future threat of co-channel interference is detected and if detected determines whether a survival MAC request has been received from a peer AP.
32. 32. A processor code product comprising processor code arranged to implement when run on a processor a method according to any one of claims 1 to 16.
33. 33. A computer readable medium storing a program that causes a computer to perform any one of the methods according to any one of claims ito 16.
34. 34. A node for use in wireless communication in which two or more nodes communicate along a common channel, wherein the node is operable to change its operating mode of communication along said common channel and can communicate with one or more other nodes to determine two or more time segregated medium access periods for the common channel, each of said time segregated medium access periods dedicated to signals from a respective node.
35. 35. A node as claimed in claim 34 wherein the node is an Access Point.
36. 36. A signal for use in a wireless network comprising two or more data packets for transmission along a common channel, each data packet having a respective time segregated access period dedicated to a respective node, the time segregated access periods being determined prior to transmission by communication between the nodes.