GB2535176A

GB2535176A - Wireless communications networks

Info

Publication number: GB2535176A
Application number: GB1502237.9A
Authority: GB
Inventors: Mcconnell Ray; Ryan Stuart
Original assignee: Bluwireless Technology Ltd
Current assignee: Bluwireless Technology Ltd
Priority date: 2015-02-11
Filing date: 2015-02-11
Publication date: 2016-08-17
Also published as: WO2016128711A1; GB201502237D0

Abstract

A wireless mesh communications network comprises a plurality of network nodes 12 and a controller. Each node has at least one radio frequency communications means 16 operable to transmit and receive data items over a radio frequency wireless communications link 18, such that the plurality of network nodes defines a plurality of radio frequency wireless communications links between adjacent nodes of the network. The controller (50, figure 7) is operable to determine a transmission schedule relating to timing of transmission of a data packet over a plurality of wireless communications links for a predetermined group of the plurality of network nodes, and to transmit information relating to such a transmission schedule to the network nodes in that group. The controller is operable to determine the transmission schedule such that data transfer latency for the data item concerned across the network is minimized. One of the plurality of network nodes is operable to modify a transmission schedule received from the controller for a communications link used by that network node in dependence upon conditions local to that at least one network node, wherein the network node concerned is operable to determine such local conditions with reference to delay measurements.

Description

The present invention relates to wireless communications networks, and, in particular, to wireless mesh communications networks.

BACKGROUND OF THE INVENTION

Communications networks are used to transfer data between a source and a destination.

Typical modern communications networks include a mixture of wired and wireless communication link s over which these data are communicated. Wired communications networks, whether a local area network (LAN) or a wide area network (WAN), make use of bidirectional communications links (so-called "full-duplex" communications) in which communications take place in both directions along the link simultaneously.

Wireless networks make use of radio frequency communications links in order to transfer the data between two stations. Unlike wired networks, however, a wireless link typically operates in one direction at any one time (so-called "half-duplex" communications). That is, data are communicated across a wireless link firstly in one direction, and then in the opposite direction. It is to be noted that there are examples of full-duplex wireless links where different frequency bands are used for send and receive. However, such full-duplex wireless communications are more expensive since they require concurrent send and receive circuitry. Half duplex wireless systems are lower cost and, hence, more widely deployed Half-duplex wireless links, therefore, require a technique to prevent interference on that link between competing transmissions. One known technique is use in consumer WiFi uses a CSMA protocol (Carrier Sense, Multiple Access) where a station first listens to see if a channel is idle or not. If it is idle the station will transmit. If that transmission results in detection of a collision (for example, when two stations attempt to access the channel at the same time and the data are corrupted), then both stations will "back-off" and wait for a random period and attempt transmission again. For each unsuccessful attempt that random wait is increased. This can lead to poor throughput in congested areas.

A scheduling methodology can be used in order to control the use of the link. For example, a scheduled access protocol, such as TDMA -Time Division Multiple Access, avoids the need for the random back-off periods as the station is guaranteed that the channel will be clear during its assigned transmission period. Typically, control is performed for a single wireless link (or "hop") independently of any other link. For example, Figure 1 of the accompanying drawings illustrates schematically a very simple wireless network 1 comprising first to fourth stations 2, 4, 6, 8. In this simple example, the first station 2 is able to communicate over respective wireless communications links with the second station 4, with the third station 6 and with the fourth station 8. Figure 1 may represent a short range wireless network, such as a WiFi network, with the first station 2 representing the wireless network hub. The stations 2, 4, 6 and 8 then form a basic service set (BSS).

In this example, the first station 2 is designated as the control station (sometimes referred to as a PCP, personal basic service set control point) for the single wireless hops from the first station 2. The first station 2 is operable to control the schedule of transmissions on the wireless links connected with the first station 2. The first station 2 uses so called "beacon" transmissions to the other stations being controlled.

An example schedule is shown in Figure 2 of the accompanying drawings. Each possible transmission is given a predetermined timeslot in which the transmission is to take place. In the example of Figure 2, the first station 2 transmits to the second station 4 during time period A, to the third station during time period C, and to the fourth station during time period D. The second, third and fourth stations 4, 6, 8 communicate with the first station 2 during the respective time periods B, E and F. It will be appreciated that this scheduling is adjusted as appropriate for any particular network.

Such a scheduling technique works well for single hop communications. However, in a multi-hop wireless network, such scheduling can lead to increased latency for data transfer across the network since each hop is individually scheduled.

It is, therefore, desirable to provide a multi-hop wireless communications network that reduces latency of data transfer

SUMMARY OF THE INVENTION

According to one aspect of the present invention, there is provided a wireless mesh communications network comprising a plurality of network nodes, each node having at least one radio frequency communications means operable to transmit and receive data items over a radio frequency wireless communications link, the plurality of network nodes thereby defining a plurality of radio frequency wireless communications links between adjacent nodes of the network, and a controller operable to determine a transmission schedule relating to timing of transmission of a data item over a plurality of wireless communications links for a predetermined group of the plurality of network nodes, and to transmit information relating to such a transmission schedule to the network nodes in that group.

In one example, the controller is operable to determine the transmission schedule such that data transfer latency across the network is minimized.

In one example, at least one of the plurality of network nodes is operable to modify a transmission schedule received from the controller in dependence upon conditions local to that at least one network node. The conditions may be determined by reference to delay measurements.

According to another aspect of the present invention, there is provided a method of transmitting a data item across a plurality of wireless communications links in a wireless mesh communications network which includes a plurality of network nodes, each node having at least one radio frequency communications means operable to transmit and receive data items over a radio frequency wireless communications link, the plurality of network nodes thereby defining a plurality of radio frequency wireless communications links between adjacent nodes of the network, the method comprising at a central controller, determining a transmission schedule relating to timing of transmission of a data item over a plurality of wireless communications links for a predetermined group of the plurality of network nodes, and transmitting information relating to such a transmission schedule to the network nodes in the group, and at the group of network nodes, transmitting a data item over a plurality of wireless communications links according to the transmission schedule.

In one example, the transmission schedule is determined in order to reduce latency of data transfer across the network.

One example further comprises, at one of the network nodes, modifying the transmission schedule received from the controller for that network node in dependence upon conditions local to the network node concerned. The conditions may be determined by reference to delay measurements.

According to a further aspect of the present invention, there is provided a method of scheduling transmission of a data item across a plurality of wireless communications links of a wireless mesh communications network which includes a plurality of network nodes, each node having at least one radio frequency communications means operable to transmit and receive data items over a radio frequency wireless communications link, the plurality of network nodes thereby defining a plurality of radio frequency wireless communications links between adjacent nodes of the network, the method comprising at a central controller, determining a transmission schedule relating to timing of transmission of a data item over a plurality wireless communications links for a predetermined group of the plurality of network nodes, and transmitting information relating to such a transmission schedule to the network nodes in the group.

It will be appreciated that reference to a data item refers to a data packet of any appropriate size an constitution, or to any other appropriate data format.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 is a schematic diagram of a simple wireless communications network; Figure 2 illustrates a scheduling technique for the network of Figure 1; Figure 3 is a schematic diagram of an exemplary wireless mesh communications network; Figure 4 is a schematic diagram illustrating a data path in the network of Figure 3; Figure 5 illustrates a scheduling technique for the data path of Figure 4; Figure 6 illustrates a data packet structure; Figure 7 illustrates a network including a controller embodying an aspect of the present invention; Figure 8 illustrates a first scheduling technique for the network of Figure 3; Figure 9 illustrates a second scheduling technique for the network of Figure 3. 25 DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Figure 3 is a schematic diagram of a simplified mesh network 10. The mesh network 10 is an example network, and an actual mesh network may be of any appropriate layout and configuration.

The network 10 comprises a plurality of nodes 12n interconnected by communications links 18. Each node 12n includes a switch 14n and a plurality of communications interfaces 16.

The communications interfaces 16 operate to communicate with a corresponding interface of an adjacent node in the network. The switch 14n operates to direct data received by one of the interfaces 16 to another of the interfaces 16 for transmission from the node concerned. The switch 14n and the communications interfaces 16 may be provided by discrete components or devices, or may be provided by a single integrated component or device.

In the example of Figure 3, a first node 121 is connected, via respective links 1812 and 1814, to second and fourth nodes 124 and 124. A first interface 1612 of the first node 121 transfers data with a first interface 1621 of the second node 122, and a second interface 1614 of the first node 121 transfers data with a first interface 1641 of the fourth node 124. The switch 141 of the first node 121 is operable to direct data received at one of the interfaces 16 of the first node 121 to another of the interfaces of the first node 121. It will be readily appreciated that each of the nodes operates in a similar manner, and that detailed explanation of the operation of each node is not included here for the sake of clarity. For communications between a node n and a node m, an interface 16nm communicates with an interface 16,in over a link 18..

For wireless mesh networks, the communications links 18 are provided by radio frequency links, and each of the interfaces 16 is a radio frequency interface, or "air interface". As discussed above, such wireless links operate in one direction at any given time, and so it is necessary to provide a scheduling technique for each link. Previously-considered techniques result in undesirable latency for data flow across such a multi-hop mesh network.

In order to describe this in more detail, reference will now be made to Figure 4, which is a simplified schematic diagram of part of a mesh network.

Figure 4 illustrates a data path 20 across a mesh network from a source 21 to a sink 22. The source 21 transmits data to a first interface 271 of a first node 241 via a first link 291. A switch 261 of the first node 241 transfers the received data to a second interface 281 of the first node 241. The second interface 281 of the first node 241 transmits the data to a first interface 272 of a second node 242 via a second wireless link 292. A switch 262 of the second node 242 transfers the received data to a second interface 282 of the second node 242. The second interface 282 of the second node 242 transmits the data to a first interface 273 of a third node 243 via a third wireless link 293. A switch 263 of the third node 243 transfers the received data to a second interface 283 of the third node 243. The second interface 283 of the third node 243 transmits the data to the sink 22 via a fourth wireless link 294.

In this manner, data are transferred across a multi-hop wireless network from the source 21 to the sink 22. Data are typically transferred across such a network in the form of discrete data packets having a defined structure and size.

Figure 5 illustrates the progression of a data packet 40 across the network from the source 21 to the sink 22 of the data path 20 illustrated in Figure 4. Figure 5 shows the results of a previously-considered scheduling technique, in which each wireless link is managed locally and independently of each other wireless link in the network. The Figure shows the respective schedules for the second, third and fourth links 292, 293, and 294 of the data path 20. In this schedule, on the second link 292, the first node 241 transmits and receives data to and from the second node 242 in alternate time periods. On the third link 293, the second node 242 transmits and receives data to and from the third node 242 in alternate time periods, with the transmission to the third node 243 coinciding with transmission to the first node 241, such that the second node 242 is transmitting to both nodes simultaneously. On the fourth link 294, the third node 243 transmits and receives data to and from the sink 22 in alternate time periods, with the transmission to the second node 242 coinciding with transmission to the sink 22.

A data packet 40 received from the source 21 is transmitted from the first node 241 to the second node 242 in a first time period AA. The data packet 40 is then held at the second node 242 until it is able to be transmitted from the second node 242 to the third node 243 during a subsequent time period BB. The data packet 40 is then held at the third node 243 until it is able to be transmitted from the third node 243 to the sink 22 during a subsequent time period CC. Each time period AA, BB, CC is separated by a guard time period 30 in order to ensure sufficient spacing of transmissions.

In the case when the time taken to transmit the data packet from a node to the next node in the data path, the time during which the data packet is held at a node is effectively wasted time, since there could be sufficient time in a transmission time period to transmit the data packet at least one further hop. Such delays results in undesirable latency of data transfer through the data path 20.

Figure 6 illustrates an example structure of the data packet 40 suitable for transmission across a wireless mesh network. The data packet 40 includes a payload portion 42 which includes the actual data being transmitted, and a header portion 44 which defines the source and destination network addresses for the payload portion 42 (this header portion 44 is also termed the MAC address header). In addition, in order to be transmitted across a wireless mesh network, the header portion 44 preferably also includes information relating to the specific route across the wireless mesh network. In addition, the data packet 40 includes a radio training portion including a channel estimate field (CEF) 46 and a short training field (STF) 48, which include data to enable the receiver to lock to the correct incoming signal for the data packet 40. The details of the channel estimate field 46 and short training field 48 are well known, and defined, for example, in the IEEE802.11ad wireless communications standards published by the Institute of Electrical and Electronics Engineers (IEEE). This standard relates to wireless communications using bands around 60GHz. The elements of the data packet 40 are transmitted in the following order in time: short training field 48, channel estimate field 46, header portion 44 and then payload 42.

In such a data packet structure, the header information added simply to enable transmission of the data packet across the wireless mesh network can be considerably larger than the payload and payload header 42 and 44. Accordingly, it is possible to aggregate multiple payloads and associated payload headers into a single data packet for transmission across the wireless network with a single forwarding header 46 and training header 48 pair. However, aggregated data packets of this sort take longer to transmit across the network. It is, therefore, desirable to balance the time taken to process the training and forwarding headers 46 and 48 with the time taken to process an aggregated data packet.

Figure 7 illustrates the data path 20 of Figure 4 in combination with a controller operating in accordance with an aspect of the present invention. The controller 50 is operable to control centrally the transmission schedules for the nodes 24i, 242, 243 in order that a data packet passing along the data path 20 experiences reduced latency.

The controller 50 receives information concerning the performance characteristics of links in the network, and concerning the data transfer requirements for the network. The controller 50 issues instructions over a control channel to each node so that data transfer is scheduled across the network as a whole. The control channel may be a radio independent of the main data transmission link, or may be an "in-band" communication using the same wireless link as the data transfer. In those networks in which a particular node is designated as a local controller for single hop links connected to the node concerned, the controller provides the scheduling information to each local controller, which then uses existing signalling techniques to disseminate the scheduling information to the other nodes. The controller 50 may provide each node or controlling node with information specific to that node, or scheduling information for a group of nodes may be broadcast to all of the nodes in the group. The group may include all of the nodes in the network, or may be a predetermined subset of those nodes.

Figure 8 illustrates a scheduling technique embodying one aspect of the present invention that makes use of the central controller 50 of Figure 6. In the scheme illustrated in Figure 7, all three nodes 241, 242, 243 are scheduled to transmit data packets in the direction of the sink 22 at the same time. A time period between time ti and time t2 is defined for transmission of data packets from the nodes 241, 242, 243 towards the sink 22. Accordingly, the data packet 40 can be transmitted from the first node 241 to the second node 242, from the second node 242 to the third node 243 and from the third node 243 to the sink 22 during a single time period. The data packet does not need to be held at any of the nodes 241, 242, 243. In this way, the delay (latency) of transmission of the data packet 40 across the data path 20 can be reduced.

The scheme illustrated in Figure 8 assumes that the data packet 40 is instantaneously available to be transmitted from a node as soon as it is received by that node. The scheme also assumes that the data packet 40 takes no time to travel across each wireless link. In a practical example, however, there will be time taken to route the data packet through the switch from the incoming interface to the outgoing interface, and time taken for the data packet to travel across the wireless link. For these reasons, an alternative scheme embodying an aspect of the present invention is illustrated in Figure 9.

In the alternative scheme of Figure 9, the time periods for transmission from the nodes 241, 242, 243 are offset with respect to one another. In this example, the time period for transmission of the data packet 40 from the first node 241 to the second node 242 over link 292 starts at time ti and lasts a predetermined period until time t2. The time period for transmission from the second node 242 to the third node 243 over link 293 is offset from the time ti by an amount Oti determined by the controller 50. Similarly, the time period for transmission from the third node 243 to the sink 22 over link 294 is offset from the time ti by an amount 6t2, greater that the amount oti. In such a way, the controller 50 is able to adjust the transmission periods from each of the nodes such that the data packet 40 is able to follow the data path 20 without impediment.

The controller 50 operates to adjust the time offsets a and length of aggregated data packets at each node of the network in dependence upon prevailing network conditions. Such a technique is particularly relevant to a software defined network in which the controller 50 has access to a software definition of the network. The definition includes physical location and connection information, and information regarding the link characteristics that can be expected. Such information is "long term" and changes slowly over time, since a network topology tends not to change quickly. The software definition of the network also allows the controller access to expected data transfer requirements. These data transfer requirements relate to the services that can be expected to require servicing at a particular node, and depend upon the connections being made at that node.

Such long term information enables a baseline set up to be communicated from the controller 50 to the nodes of the network, so that in base line conditions latency is reduced. However, wireless links will experience changes in link conditions, primarily due to local environmental factors. Short term factors may include changing weather conditions, such as rain or snow, which can adversely affect the wireless communications link performance. For example, communications using the 60GHz band are susceptible to attenuation due to rain and snow. Another cause of short term effects, particularly in an urban environment, is road traffic. Although nodes will be located above most road traffic, it is possible that larger vehicles will cause interruption to a wireless link for a short period. The nature of these short term issues means that it is not feasible to report the effects back to the controller 50 for adjustment of the scheduling scheme for the network. Reduction in link quality can result in data packet transfer failure, leading to retransmission attempts being needed. Such retransmissions take time, and so can affect the overall schedule.

Accordingly, local nodes may be provided with the ability to make local changes to the baseline schedule provided by the controller 50, within predefined bounds. Individual local nodes are then operable to communicate with adjacent nodes about actual prevailing link conditions. This information may also be communicated to the controller 50 for longer term updating of the baseline conditions.

In one example, each of a pair of nodes periodically exchanges messages (in the form of a delay measurement data packet) with each other to measure the current average delay on the wireless communications link therebetween. Such information can then be used to determine local changes to the transmission schedule, and can be shared with the controller 50. For example, as illustrated in Figure 9, the link 293 has a delay offset of 6ti which is set according to the expectation of the controller 50. Upon measurement of the link (through nodes 242 and 243 exchanging delay measurement data packets) an updated offset Ot,' can be estimated and shared through a message with node 241 to update the scheduling of transmissions on link 292.

The delays are likely to be asymmetrical between for transmission and reception at any given node.

Claims

CLAIMS: 1. A wireless mesh communications network comprising: a plurality of network nodes, each node having at least one radio frequency communications means operable to transmit and receive data items over a radio frequency wireless communications link, the plurality of network nodes thereby defining a plurality of radio frequency wireless communications links between adjacent nodes of the network; and
2.
3.
4.
5.a controller operable to determine a transmission schedule relating to timing of transmission of a data item over a plurality of wireless communications links for a predetermined group of the plurality of network nodes, and to transmit information relating to such a transmission schedule to the network nodes in that group.

A network as claimed in claim 1, wherein the controller is operable to determine the transmission schedule such that data transfer latency for the data item concerned across the network is minimized.

A network as claimed claim 1, wherein one of the plurality of network nodes is operable to modify a transmission schedule received from the controller for a communications link used by in dependence upon conditions local to that at least one network node.

A network as claimed in claim 3, wherein the network node concerned is operable to determine such local conditions with reference to delay measurements.

A method of transmitting a data item across a plurality of wireless communications links of a wireless mesh communications network which includes a plurality of network nodes, each node having at least one radio frequency communications means operable to transmit and receive data items over a radio frequency wireless communications link, the plurality of network nodes thereby defining a plurality of radio frequency wireless communications links between adjacent nodes of the network, the method comprising: at a central controller, determining a transmission schedule relating to timing of transmission of a data item over a plurality wireless communications links for a predetermined group of the plurality of network nodes, and transmitting information relating to such a transmission schedule to the network nodes in the group; and at the group of network nodes, transmitting a data item over a plurality of wireless communications links according to the transmission schedule.
6. A method as claimed in claim 5, wherein the transmission schedule is determined in order to reduce latency of data packet transfer across the network.
7. A method as claimed in claim 5 or 6, further comprising, at one of the network nodes, modifying the transmission schedule received from the controller for a communications link used by that network node in dependence upon conditions local to the network node concerned.
8. A method as claimed in claim 7, wherein such local conditions are determined with reference to delay measurements.
9. A method of scheduling transmission of a data item across a plurality of wireless communications links of a wireless mesh communications network which includes a plurality of network nodes, each node having at least one radio frequency communications means operable to transmit and receive data items over a radio frequency wireless communications link, the plurality of network nodes thereby defining a plurality of radio frequency wireless communications links between adjacent nodes of the network, the method comprising: at a central controller, determining a transmission schedule relating to timing of transmission of a data item over a plurality wireless communications links for a predetermined group of the plurality of network nodes, and transmitting information relating to such a transmission schedule to the network nodes in the group.
10. A method as claimed in claim 9, wherein the transmission schedule is determined in order to reduce latency of data item transfer across the network.
11. A method as claimed in claim 9 or 10, further comprising, at one of the network nodes, modifying the transmission schedule received from the controller for a communications link used by that network node in dependence upon conditions local to the network node concerned.
12. A method as claimed in claim 11, wherein such local conditions are determined with reference to delay measurements.
13. A wireless mesh network substantially as hereinbefore described with reference to Figures 3 to 9 of the accompanying drawings.
14. A method of transmitting a data packet across a plurality of wireless communication links in a wireless mesh communications network substantially as hereinbefore described with reference to Figure 3 to 9 of the accompanying drawings.
15. A method of scheduling transmission of a data packet across a plurality of wireless communication links in a wireless mesh communications network substantially as hereinbefore described with reference to Figure 3 to 9 of the accompanying drawings.Amended claims have been filed as follows:-CLAIMS: 1. A wireless mesh communications network comprising: a plurality of network nodes, each node having at least one radio frequency communications means operable to transmit and receive data items over a radio frequency wireless communications link, the plurality of network nodes thereby defining a plurality of radio frequency wireless communications links between adjacent nodes of the network; and CO 2.CD 3. 4. 5.a controller operable to determine a transmission schedule relating to timing of transmission of a data item over a plurality of wireless communications links for a predetermined group of the plurality of network nodes, and to transmit information relating to such a transmission schedule to the network nodes in that group.A network as claimed in claim 1, wherein the controller is operable to determine the transmission schedule such that data transfer latency for the data item concerned across the network is minimized.A network as claimed claim 1, wherein one of the plurality of network nodes is operable to modify a transmission schedule received from the controller for a communications link used by that network node in dependence upon conditions local to that at least one network node.A network as claimed in claim 3, wherein the network node concerned is operable to determine such local conditions with reference to delay measurements.A method of transmitting a data item across a plurality of wireless communications links of a wireless mesh communications network which includes a plurality of network nodes, each node having at least one radio frequency communications means operable to transmit and receive data items over a radio frequency wireless communications link, the plurality of network nodes thereby defining a plurality of radio frequency wireless communications links between adjacent nodes of the network, the method comprising: at a central controller, determining a transmission schedule relating to timing of transmission of a data item over a plurality wireless communications links for a predetermined group of the plurality of network nodes, and transmitting information relating to such a transmission schedule to the network nodes in the group; and at the group of network nodes, transmitting a data item over a plurality of wireless communications links according to the transmission schedule.6. A method as claimed in claim 5, wherein the transmission schedule is determined in order to reduce latency of data packet transfer across the network.7. A method as claimed in claim 5 or 6, further comprising, at one of the network nodes, modifying the transmission schedule received from the controller for a communications link used by that network node in dependence upon conditions local to the network node concerned.8. A method as claimed in claim 7, wherein such local conditions are determined with reference to delay measurements.9. A method of scheduling transmission of a data item across a plurality of wireless communications links of a wireless mesh communications network which includes a plurality of network nodes, each node having at least one radio frequency communications means operable to transmit and receive data items over a radio frequency wireless communications link, the plurality of network nodes thereby defining a plurality of radio frequency wireless communications links between adjacent nodes of the network, the method comprising: at a central controller, determining a transmission schedule relating to timing of transmission of a data item over a plurality wireless communications links for a predetermined group of the plurality of network nodes, and transmitting information relating to such a transmission schedule to the network nodes in the group.10. A method as claimed in claim 9, wherein the transmission schedule is determined in order to reduce latency of data item transfer across the network.11. A method as claimed in claim 9 or 10, further comprising, at one of the network nodes, modifying the transmission schedule received from the controller for a communications link used by that network node in dependence upon conditions local to the network node concerned.12. A method as claimed in claim 11, wherein such local conditions are determined with reference to delay measurements.13. A wireless mesh network substantially as hereinbefore described with reference to Figures 3 to 9 of the accompanying drawings.14. A method of transmitting a data packet across a plurality of wireless communication links in a wireless mesh communications network substantially as hereinbefore described with reference to Figure 3 to 9 of the accompanying drawings.15. A method of scheduling transmission of a data packet across a plurality of wireless communication links in a wireless mesh communications network substantially as hereinbefore described with reference to Figure 3 to 9 of the accompanying drawings.