A method and associated device for use in a wireless network

Abstract

A new method is disclosed for use by a device as part of a wireless network for maintaining a list of other nodes in the wireless network which that device communicates with (also known as a neighbour list). The method comprises: receiving a request from a first node; and searching for a replaceable node in the list of other nodes in the wireless network which that device can currently communicate with which is eligible for replacement. A decision as to whether node in the list of other nodes is eligible for replacement is based on criteria which include: (a) a category of messages that the device has exchanged with the current nodes in the list and / or (b) a time since a last successful message was exchanged with the current nodes. If the criteria are met, the method further comprises removing the replaceable node which is eligible for replacement and replacing it with the first node.

Description

[0001] A METHOD AND ASSOCIATED DEVICE FOR USE IN A WIRELESS NETWORK

[0002] The present invention relates to a method for use by a device as part of a wireless network. In particular, it relates to a method is for maintaining a list of other nodes in the wireless network which that device can communicate with. The present invention also relates to any associated apparatus, device or node which is arranged to carry out the method.

[0003] It is known to construct smart infrastructure from a plurality of internet of things (loT) enabled devices with wireless capabilities. For example, such an infrastructure may be used to enable smart meters and the like to communicate information regarding utilities usage across a network to a centrally located server for monitoring and analysis. Metering devices may be deployed at businesses, homes, and other premises for measuring consumption of resources, such as electricity, water, and gas. While some metering devices may provide only basic metering functions, other metering devices, known in the art as “smart meters”, may provide more advanced functionality, such as control and communications functionality

[0004] In an example, some metering devices may be configured to communicate information relating to consumption of resources. In another example, some metering devices may be configured to receive information such as billing information and control signals, such as service disconnect control signals or the like. In examples, transmission of information relating to consumption of metered resources may simplify automated billing, reduce operational costs, and may enable advanced analytics of resource consumption.

[0005] While in some examples a metering device may communicate directly with a router or gateway device, in other examples a wireless mesh network may be formed from multiple metering devices, wherein each metering device operates as an interconnected node and effective endpoint within the wireless mesh network.

[0006] When implemented in a wireless mesh network, a metering device operating as a network node may relay messages to or from a gateway device or router, or a head-end system. In an example, messages may be routed along a path by hopping from node to node, e.g. from metering device to metering device until said messages reaches a target destination, e.g. the gateway / router or a target metering device. The proximity of such devices to each other in terms of distance can vary from very close to very distant depending on the application requirements. Such devices can be powered up using an AC power supply or can be powered using a battery end point. These end points can form Wireless Mesh network to communicate with a central coordinator like an access point or a collector device to send sensor data (for example indicative of utilities usage).

[0007] Typically, the communication is via a parent node in the mesh network. A device transmits all its information data to a centrally located server routing through its parent. The parent for each device is chosen based on a list of neighbours with whom it can transmit and receive data packets. There can be multiple parameters based on which a node can select a parent. Each device or node maintains a list of neighbour devices and information received from those neighbours. This neighbour list can be useful, for example, in scenarios where a node can perform peer-to-peer communication with other nearby neighbours for sharing critical data gathered from sensors on nodes from the mesh network.

[0008] In a wireless RF mesh network, a node does not rely on a single neighbour (parent) to reach to the collector device. A fully-functioning node (FFN) maintains a list of neighbour nodes and updates information for each neighbour node on the list as this information is received from node itself or by establishing a communication with a neighbouring node to check reliable communication. A FFN may be in communication with some nodes frequently to obtain updated information as soon as possible so that in case of failure of communication with its parent, a node can quickly switch to another parent (and be confident that good communication with the new parent is available).

[0009] In general, nodes may maintain certain data for each neighbour on its list such as, for example, a unique IPv6 address, a local area network (LAN) address, packet success rate, signal strength, path cost and other parameters that may be useful for to making decisions which help in maintaining reliable communication.

[0010] Each node within a wireless network may send various messages including broadcast messages and unicast messages. A broadcast message may be sent as a broadcast to be received by a plurality of nodes. As an example, some nodes may send as a broadcast message a data packet containing status information about those nodes to any nodes within their communication range. A unicast message may be a message sent from one node to another. As an example, each node may send (or relay) sensor data (for example indicative of utilities usage) to a specific end point as a unicast signal or data packet.

[0011] It may be desirable to provide an alternative methods for operating devices on a wireless network and / or apparatus or devices for use with such a wireless network which at least partially addresses one or more of the problems of the prior art, whether such problems are identified herein or elsewhere.

[0012] According to a first aspect of the present disclosure invention there is a provided a method for use by a device as part of a wireless network for maintaining a list of other nodes in the wireless network which that device communicates with, the method comprising: receiving a request from a first node; searching for a replaceable node in the list of other nodes in the wireless network which that device can currently communicate with which is eligible for replacement based on criteria which include: (a) a category of messages that the device has exchanged with the current nodes in the list and / or (b) a time since a last successful message was exchanged with the current nodes; and if the criteria are met, removing the replaceable node which is eligible for replacement and replacing it with the first node.

[0013] Other nodes in the wireless network which that device communicates with may be referred to as neighbour nodes. The list of the other nodes may be referred to as a neighbour list.

[0014] As used herein, unless stated to the contrary, “node” and “device” refer to a physical entity that can, or is, participating in a mesh network. These terms may be used interchangeably. A node or device may be powered by either mains (AC) electricity or by a battery.

[0015] Each node or device may comprise an associated parent node and one or more associated child nodes. For any given node or device, its associated parent node is a node having capability to route traffic from that given node or device. Any node or device can have a primary parent or multiple back up parents. At any given time, the current parent node of a given node or device is a node via which sensor data (for example indicative of utilities usage) is routed from that given node or device (ultimately to a collector device and then to a centrally located server) as a unicast signal or data packet. At any given time, a child of any given node may be any node whose information is being routed through that given node.

[0016] As used herein, unless stated to the contrary, for any given node or device, a “neighbour node” is intended to mean any node that the given node or device can exchange information with (i.e. any node within its communication range).

[0017] As used herein, unless stated to the contrary, for any given node or device, a “critical neighbour” is intended to mean any node that the given node or device exchanges unicast messages with. Critical neighbours include a parent node, any child nodes and nodes running any critical applications. A critical neighbour can exchange information with a node or device by sending information in either a medium access control (MAC) layer packet or an application layer packet. A system administrator can also declare a node as critical and communicate this information to a node or device (via its parent). In other aspects, a node can determine a neighbour as critical based on a number of high priority quality of service (QoS) packets generated by a child node.

[0018] Similarly, as used herein, unless stated to the contrary, for any given node or device, a “non-critical neighbour” is intended to mean a node that the given node or device can exchange information with but which is not currently important for that node or device to perform any activity of node such as transmitting its own data or data from its child node data to a central server. The performance of a node or device is not dependent on non- critical neighbours. Note that a non-critical neighbour can become a critical neighbour, for example if it becomes a child or the parent of the node or device.

[0019] Typically internet of things (loT) enabled devices with wireless capabilities may have rather limited memory or storage capabilities. In a wireless network with a large number of devices present in close proximity to each other, and which are able to listen to messages transmitted by one another, there may not be sufficient storage capabilities for each device to add all of the devices within its communication range in its neighbour list. Furthermore, even if there was sufficient memory or storage capability, in such a densely populated wireless network the computational overhead of maintaining a list of all of the devices within its communication range would be overly burdensome (and may have a significant impact of battery lifetime for battery operated devices). The method according to the first aspect of the present disclosure is advantageous as it allows for each device to manage its neighbour list in a smart, simple, efficient manner, as now discussed.

[0020] Since the criteria for determining whether a node is eligible for replacement may include: a category of messages that the device has exchanged with the current nodes in the list, the method according to the first aspect allows for critical and non-critical neighbours to be treated differently. In particular, it may allow the device to maintain a greater number of critical neighbours and preferentially replace non-critical neighbours on the neighbour list. Furthermore, since the criteria for determining whether a node is eligible for replacement may include: a time since a last successful message was exchanged with the current nodes, the method according to the first aspect allows for obsolete neighbours to be preferentially replaced.

[0021] Fully-functioning nodes (FFNs) transmit periodic messages to advertise their presence in the network. These messages are usually heard by other nodes to add or update node information in their neighbour lists. Hence, the neighbour list of any given node may have many nodes and, typically, many of these will be non-critical neighbours.

[0022] A node can only become critical when that node has joined the network (and, for example, when other nodes start selecting that node as a parent). Therefore, a node which is trying to join a network is non-critical. The neighbours of such a node may not have any available space in their neighbour lists and therefore those neighbour nodes may start rejecting the request of any new joining node. However, the inventors have realized that some such neighbour nodes may have at least one non-critical node or obsolete node on their neighbour list. Advantageously, the method according to the first aspect of the present disclosure allows for such neighbour nodes to replace such a non- critical or obsolete entry in its neighbour list with the new node trying to join the network in order to support joining of this new node.

[0023] Searching for a replaceable node in the list of other nodes in the wireless network which that device communicates with which is eligible for replacement may comprise: searching for a broadcast node in the list which the device has only exchanged broadcast type messages with. As used herein a broadcast node is intended to mean a node with which the device has only exchanged broadcast messages with. Broadcast nodes can be replaced as they are not critical.

[0024] If the device only receives broadcast messages from a node on the current neighbour list this signifies either: (a) that node has already joined the network and is doing its own broadcast transmissions as per defined specifications (e.g. IEEE 802.15.4 of some other standard or protocol); or (b) that node is trying to find devices in vicinity. If the device receives only such broadcast messages from one or more nodes on the neighbour list then these nodes will be considered to be non-critical neighbours.

[0025] If any such broadcast node is found, the method may further comprise deciding whether it is eligible for replacement based on a time since a last successful message was exchanged with the broadcast node.

[0026] The time since a last successful message was exchanged with the broadcast node gives a better idea which nodes should be preferred over others for replacement.

[0027] This allows for non-critical broadcast nodes to be replaced to allow the first node to join the list whilst also taking into account the time since the last communication with such broadcast nodes. Advantageously, this additional criterion may reduce the frequency at which nodes are replaced on the neighbour list. In particular, it can prevent a situation in which the device is constantly switching neighbours, for example replacing node A with node B and then, subsequently, replacing node B with node A. Such an arrangement would use excessive processing power and may consume battery life, which is undesirable.

[0028] In principle, any such broadcast node that is deemed to be eligible for replacement may be identified as the replaceable node (and replaced by the first node).

[0029] Such a broadcast node may be deemed to be eligible for replacement if the time since a last successful message was exchanged with the broadcast node is greater than a threshold value.

[0030] For example, the threshold value may be an average time, for all nodes on the neighbour list, since a last successful message was exchanged with those nodes. Advantageously, such an arrangement allows for the device to self-regulate the threshold. For example, an appropriate absolute threshold value to be used by any given device may depend on how close it is to its neighbours.

[0031] If any such broadcast node is found before the end of the list has been reached, the rest of the list may not be searched, and that broadcast node may be identified as the replaceable node.

[0032] That is, as far as broadcast nodes are concerned, the algorithm is a first found, first replaced method. For example, it may be that there are two or more broadcast nodes in the list which the device has only exchanged broadcast type messages with that may be eligible to be replaced as non-critical neighbours. In alternative embodiments, the algorithm may continue so as to identify the broadcast node that is most eligible for replacement. However, operating on a first found, first replaced is advantageous over such alternative embodiments as it saves computation resources.

[0033] Searching for a replaceable node in the list of other nodes in the wireless network which that device communicates with which is eligible for replacement may comprise: searching for any unicast nodes in the list which the device has exchanged unicast type messages with.

[0034] If the device exchanges any successful unicast messages with a node from the neighbour list, then that node is a critical node or critical neighbour. It could, for example be a parent node of the device or a child node of the device.

[0035] Such a unicast message may, for example, comprise a medium access control (MAC) layer packet or an application layer packet.

[0036] If any such unicast node is found, the method may further comprise deciding whether it is eligible for replacement based on a time since a last successful message was exchanged with the unicast node.

[0037] In principle, any such unicast node that is deemed to be eligible for replacement may be identified as the replaceable node (and replaced by the first node). In some embodiments, such a unicast node that is deemed to be eligible for replacement may only be identified as the replaceable node (and replaced by the first node) if no broadcast nodes have been identified that are deemed to be eligible for replacement.

[0038] The time since a last successful message was exchanged with the unicast node gives a better idea which unicast nodes should be preferred over others for replacement.

[0039] Any such unicast node may be deemed to be eligible for replacement if the time since a last successful message was exchanged with the unicast node is greater than a threshold value.

[0040] For example, the threshold value may be an average time, for all nodes on the neighbour list, since a last successful message was exchanged with those nodes. Advantageously, such an arrangement allows for the device to self-regulate the threshold. For example, an appropriate absolute threshold value to be used by any given device may depend on how close it is to its neighbours.

[0041] If any such unicast nodes are found on the neighbour list which have not successfully exchanged a message with the device for a time period greater that the threshold value it may be indicative that such nodes have been lost (for example are no longer connected to the network) or have changed its parent node in the network (and so its unicast messages are now being send via a different route). Such nodes may be considered to be obsolete and may be eligible for replacement.

[0042] The method may comprise searching for a unicast node that is eligible for replacement for which the time since a last successful message was exchanged between this unicast node and the device is greater than the time since a last successful message was exchanged between all other such unicast nodes that are eligible for replacement and the device. If no broadcast nodes have been identified that are eligible for replacement then that unicast node may be identified as the replaceable node.

[0043] That is, the method may involve finding the oldest unicast node or a unicast node which has gone for the longest time without exchanging a successful message with the device.

[0044] In some embodiments, if there is at least one broadcast node in the list which the device has only exchanged broadcast type messages with and which is eligible for replacement and there is at least one unicast node in the list which the device has exchanged unicast type messages with which is eligible for replacement, the replaceable node may be selected from the at least one broadcast node.

[0045] That is, the algorithm may preferentially select broadcast nodes to be replaced over unicast nodes.

[0046] The criteria for selecting a replaceable node in the list of other nodes in the wireless network which that device can currently communicate with may include a category of messages that the request comprises.

[0047] For example, the request from the first node may comprise a broadcast message or a unicast message. The criteria for selecting a replaceable node may be different depending on what type of message the request is.

[0048] A first set of criteria may be used for searching for a replaceable node when the request comprises a broadcast message and a second, different set of criteria may be used for searching for a replaceable node when the request comprises a unicast message.

[0049] The first criteria may be more restrictive than the second criteria.

[0050] That is, it may be more likely select a replaceable node when the second criteria are applied than when the first criteria are applied. In other words, for such embodiments, the algorithm tries harder to accommodate on its neighbour list a first node which sends a unicast message as a request.

[0051] When searching for a replaceable node, the first set of criteria may be applied to identify a replaceable node. If no such replaceable node is identified and the request comprises a unicast message, the second criteria may be subsequently applied in order to identify a replaceable node.

[0052] If the request from the first node comprises a unicast message and the list contains at least one broadcast node which the device has only exchanged broadcast type messages with then one of said at least one broadcast node may be identified as the replaceable node. That is, if the request from the first node is a unicast message and there is at least one broadcast node on the current neighbour list then the device will accommodate the first node and allow it to replace one of the broadcast nodes.

[0053] The method may comprise finding, from the at least one broadcast node which the device has only exchanged broadcast type messages with, a broadcast node which has gone for the longest time without exchanging a successful message with the device and that broadcast node may be identified as the replaceable node.

[0054] The method may further comprise, for each node in the list of other nodes in the wireless network which that device can currently communicate with, maintaining a time since the last message was successfully exchanged with that node.

[0055] As discussed above, this parameter may be used when searching for a replacement node that is eligible for replacement. One method for maintaining these times for all nodes on the neighbour list is as follows. For each node on the neighbour list a time associated with that neighbour may be retained (for example stored in memory). When a node is added to the neighbour list, the method may retain (for example store in memory) a time that it is so added. When a node already on the list successfully exchanges a message with the device, the current time stored for that node may be updated to the time of that successful message.

[0056] The method may further comprise, for each node in the list of other nodes in the wireless network which that device can currently communicate with, maintaining a status type that is indicative of the type of messages that the device has exchanged with that node.

[0057] The method may further comprise sending a reply message to the first node informing it whether or not it has successfully accommodated the first node on the list.

[0058] The criteria for finding a replaceable node in the list of other nodes in the wireless network which that device can currently communicate with which is eligible for replacement based may include a power status of the first node and / or a power status of the current nodes in the list.

[0059] As used here, a power status of a device or node may mean whether the device or node is battery operated or powered by another source such as, for example, by mains electricity. Additionally or alternatively, as used here, a power status of a device or node may include a remaining lifetime of a battery or power supply of the device or node. The algorithm may, for example try harder to accommodate a request from a first node that is battery powered and / or has limited battery life remaining than, for example, a first node that is mains powered or which has significant battery lifetime remaining. The algorithm may even allow the device to replace an existing critical (unicast) neighbour that is mains powered with a first device that is battery powered.

[0060] The criteria for searching for a replaceable node in the list of other nodes in the wireless network which that device can currently communicate with which is eligible for replacement may include: whether or not the node has received a central message indicating that one or more nodes on the list are critical nodes.

[0061] For example, a system administrator may declare a node as critical and communicate this information to a node or device (via its parent).

[0062] The criteria for searching for a replaceable node in the list of other nodes in the wireless network which that device can currently communicate with which is eligible for replacement may include: a number of high priority quality of service (QoS) packets generated by a child node.

[0063] In other aspects, a node can determine a neighbour as critical based on a number of high priority quality of service (QoS) packets generated by a child node.

[0064] According to a second aspect of the present disclosure there is provided a computer- readable medium having instructions for carrying out the method according to the first aspect of the present disclosure.

[0065] The computer-readable medium may be a non-transitory computer-readable medium.

[0066] According to a third aspect of the present disclosure there is provided an apparatus for use in a wireless network, the apparatus comprising: a transceiver; at least one processor; and storage media, the storage media having instructions to cause the at least one processor to carrying out the method according to the first aspect of the present disclosure. The list of other nodes in a wireless network which the apparatus has communicated with may be stored on the storage media and the at least one processor may be operable to read from and / or write to the storage media so as to maintain the list.

[0067] Various aspects and features of the disclosure set out above or below may be combined with various other aspects and features of the disclosure as will be readily apparent to the skilled person.

[0068] Embodiments of the disclosure will now be described, by way of example only, with reference to the accompanying schematic drawings, in which:

[0069] Figure 1 depicts a block diagram of an example of a mesh network;

[0070] Figure 2 is a schematic illustration of a new method according to an embodiment of the present disclosure for use by a device as part of a wireless network for maintaining a list of other nodes in the wireless network which that device communicates with;

[0071] Figure 3 is a flowchart representing an embodiment of the method shown in Figure 2; and

[0072] Figure 4 is a schematic illustration of an apparatus according to an embodiment of the present disclosure, which may represent a node of the mesh network shown in Figure 1 and which may implement the method(s) illustrated in Figures 2 and 3.

[0073] Figure 1 depicts an example of mesh network 100. The example network 100 may be a time-synchronized channel hopping (TSCH) network, as defined by IEEE 802.15.4. The mesh network 100 may be a network of (radio-frequency) operated wireless mesh network nodes. The mesh network 100 is an example of a mesh network upon which the present invention may be implemented. In examples, the mesh network 100 may comprise a full mesh topology, e.g. where any node can communicate with any node within range, or a partial mesh topology with more limited or selected connectivity between nodes.

[0074] The example mesh network comprises a head-end system 110. The head-end system 110 may function as a central processing system that transmits and / or receives data, streams of data, data packets and / or messages from nodes 105a - 105h of the mesh network 100, as described in more detail below. The head-end system 110 may generate and / or process the data. In examples, head-end system 1 10 may be communicably coupled to a further system, such as a cloud based system (not shown), for generating and / or processing the data.

[0075] The example mesh network 100 may comprise a root node 1 15. The root node 115 of the mesh network 100 may be configured for communicating with the nodes 105a - 105h to perform operations such as retrieving data from the nodes 105a - 105h and / or transmitting data to the head-end system 110. In some examples, the root node 115 may also operate as a node similar to other nodes 105a-105h. In examples, the root node 1 15 may comprise a gateway device. The root node 1 15 may be configured to transmit and receive data to / from the head-end system 110 via backhaul 120, such as the Internet, an intranet, or any other data communication network.

[0076] Although only a single root node 1 15 is depicted for illustrative purposes, it will be understood that the mesh network 100 may comprise more than one root node 1 15.

[0077] Similarly, the depicted mesh network 100 comprises multiple nodes 105a-105h. Each node 105a-105h may be an end-node. For purposes of example, only nodes 105a-105h are depicted, but it will be understood that substantially more than eight nodes may be implemented. For example, in a mesh network of metering device, thousands, hundreds of thousands or even millions of devices may be implemented in the mesh network.

[0078] In examples, each node 105a-105h may comprise a device for metering and / or controlling a resource. For example, the device may comprise circuitry and / or components for metering a consumption of a resource, and / or controlling access to the resource, such as by a service disconnect switch or the like. In an example, the mesh network 100 may be associated with a utility network. In such an example, the nodes 105a-105h may comprise circuitry and / or components for metering the resource, for determining various operating characteristics of the utility network, and / or for transmitting collected data through the mesh network 100 to the head-end system 1 10 via the root node 115.

[0079] Furthermore, the nodes 105a-105h may be further configured to communicate with each other such that data may be exchanged between the nodes 105a-105h. That is, the nodes 105a-105h may comprise local processing capabilities, enabling a degree of autonomy over communication with other nodes in the network 100. The nodes 105a-105h forming the mesh network 100 may be effectively provided in layers, as annotated in Figure 1 . In this example, the root node 1 15 forms layer 0. Nodes 105a, 105b that are communicably coupled directly to the root node 115 form a first layer, “Layer 1 ”, of the mesh network 100. Similarly, nodes 105c - 105e that are communicably coupled to the mesh network 100 through a “Layer 1 ” node form a second layer, “Layer 2”, of the mesh network 100. Similarly, nodes 105f - 105h that are communicably coupled to the mesh network 100 through a “Layer 2” node form a third layer, “Layer 3’’, of the mesh network 100. For data to propagate from the root node 1 15 to a Layer 3 of the mesh network 100 would require three “hops” of the data transmission.

[0080] As used herein, unless stated to the contrary, “node” 105a-105h and “device” refer to a physical entity that can, or is, participating in a mesh network 100. These terms may be used interchangeably. A node 105a-105h or device may be powered by either mains (AC) electricity or by a battery.

[0081] Each node 105a-105h or device may comprise an associated parent node and one or more associated child nodes. For any given node 105a-105h or device, its associated parent node is a node 105a-105h having capability to route traffic from that given node 105a-105h or device. Any node 105a-105h or device can have a primary parent or multiple back up parents. At any given time, the current parent node of a given node 105a-105h or device is a node via which sensor data (for example indicative of utilities usage) is routed from that given node or device (ultimately to a collector device such as the root node 115 and then to a centrally located server such as the head-end system 1 10) as a unicast signal or data packet. At any given time, a child of any given node 105a-105h may be any node 105a-105h whose information is being routed through that given node 105a-105h.

[0082] As used herein, unless stated to the contrary, for any given node 105a-105h or device, a “neighbour node” is intended to mean any node 105a-105h that the given node 105a- 105h or device can exchange information with (i.e. any node 105a-105h within its communication range).

[0083] As used herein, unless stated to the contrary, for any given node 105a-105h or device, a “critical neighbour” is intended to mean any node 105a-105h that the given node 105a- 105h or device exchanges unicast messages with. Critical neighbours include a parent node, any child nodes and nodes 105a-105h running any critical applications. A critical neighbour can exchange information with a node 105a-105h or device by sending information in either a medium access control (MAC) layer packet or an application layer packet. A system administrator can also declare a node 105a-105h as critical and communicate this information to a node 105a-105h or device (via its parent). In other aspects, a node 105a-105h can determine a neighbour as critical based on a number of high priority quality of service (QoS) packets generated by a child node.

[0084] Similarly, as used herein, unless stated to the contrary, for any given node 105a-105h or device, a “non-critical neighbour” is intended to mean a node 105a-105h that the given node 105a-105h or device can exchange information with but which that node 105a- 105h or device is not currently dependent on to perform any activity of node 105a-105h such as transmitting its own data or data from its child node data to a central server (for example head-end system 110). The performance of a node 105a-105h or device is not dependent on non-critical neighbours. Note that a non-critical neighbour can become a critical neighbour, for example if it becomes a child or the parent of the node 105a-105h or device.

[0085] Some embodiments of the present disclosure relate new methods for use by a device as part of a wireless network. For example, the new methods may be implemented by the nodes 105a-105h in the network 100 of the type shown schematically in Figure 1 . In particular, the new methods are for maintaining a list of other nodes in the wireless network 100 which the device implementing the methods communicates with. An example of such a new method 200 for use by a device as part of a wireless network 100 for maintaining a list of other nodes in the wireless network which that device communicates with is shown schematically in Figure 2 and is now described.

[0086] The new method 200 shown in Figure 2 comprises a step 210 of receiving a request from a first node (which may, for example, comprise a new node that is trying to join the network 100). The request from the first node may be referred to as a first request.

[0087] The new method 200 shown in Figure 2 further comprises a step 220 of searching for a replaceable node in the list of other nodes in the wireless network 100 which that device can currently communicate with, which is eligible for replacement. A decision as to whether or not a node 105a-105h from the list of other nodes in the wireless network 100 which that device can currently communicate with is eligible for replacement is based on criteria which include: (a) a category of messages that the device has exchanged with the current nodes 105a-105h in the list and / or (b) a time since a last successful message was exchanged with the current nodes 105a-105h.

[0088] The new method 200 shown in Figure 2 further comprises a step 230 of, if the criteria are met, removing the replaceable node which is eligible for replacement and replacing it with the first node (which sent the request at step 210).

[0089] Other nodes 105a-105h in the wireless network 100 which that device communicates with may be referred to as neighbour nodes. The list of the other nodes 105a-105h may be referred to as a neighbour list.

[0090] The nodes 105a-105h in a mesh network 100 may comprise internet of things (loT) enabled devices. Typically such devices with wireless capabilities may have rather limited memory or storage capabilities. In a wireless network 100 with a large number of devices 105a-105h present in close proximity to each other, and which are able to listen to messages transmitted by one another, there may not be sufficient storage capabilities for each device 105a-105h to add all of the devices within its communication range in its neighbour list. Furthermore, even if there was sufficient memory or storage capability, in such a densely populated wireless network 100 the computational overhead of maintaining a list of all of the devices 105a-105h within its communication range would be overly burdensome (and may have a significant impact of battery lifetime for battery operated devices).

[0091] The method 200 shown schematically in Figure 2 is advantageous as it allows for each device 105a-105h to manage its neighbour list in a smart, simple, efficient manner, as now discussed.

[0092] Since the criteria for determining whether a node 105a-105h is eligible for replacement (at step 220) may include: a category of messages that the device has exchanged with the current nodes 105a-105h in the list, the method 200 shown in Figure 2 allows for critical and non-critical neighbours to be treated differently. In particular, it may allow the device to maintain a greater number of critical neighbours and preferentially replace non- critical neighbours on the neighbour list. Furthermore, since the criteria for determining whether a node 105a-105h is eligible for replacement may include: a time since a last successful message was exchanged with the current nodes 105a-105h, the method 200 shown in Figure 2 allows for obsolete neighbours to be preferentially replaced.

[0093] Fully-functioning nodes (FFNs) transmit periodic messages to advertise their presence in the network 100. These messages are usually heard by other nodes 105a-105h to add or update node information in their neighbour lists. Hence, the neighbour list of any given node 105a-105h may have many nodes and, typically, many of these will be non- critical neighbours.

[0094] A node can only become critical when that node has joined the network 100 (and, for example, when other nodes start selecting that node as a parent). Therefore, a node which is trying to join a network 100 is non-critical. The neighbours of such a node may not have any available space in their neighbour lists and therefore those neighbour nodes may start rejecting the request of any new joining node. However, the inventors have realized that some such neighbour nodes may have at least one non-critical node or obsolete node on their neighbour list. Advantageously, the method 200 shown in Figure 2 allows for such neighbour nodes to replace such a non-critical or obsolete entry in its neighbour list with the new node trying to join the network in order to support joining of this new node.

[0095] A flowchart for an embodiment of the method 200 shown in Figure 2 is shown in Figure 3 and is now described.

[0096] The flowchart starts with the receipt 210 of a request from the first node. Upon receipt of the request, the type or category of the message that the request comprises is determined and stored using flag “Add Request”. For example, the request from the first node may comprise a broadcast message or a unicast message. As discussed below, the criteria for selecting a replaceable node may be different depending on what type of message the request is.

[0097] Next, at step 240, the device determines whether or not the neighbour list is full. If the neighbour list is not full then, at step 242, the first node is added to the neighbour list and its neighbour type is updated. If the neighbour list is full then, the method proceeds to the step 220 of searching for a replaceable node in the list of other nodes in the wireless network 100 which that device can currently communicate with, which is eligible for replacement. As a first step, at step 244, the method involves a loop over all N members of the (full) neighbour list. Before the loop 244 over all N members of the neighbour list, the flags Eligible Unicast and Old Broad may be set to 0.

[0098] For each entry in the neighbour list, the method first determines at step 246 whether that neighbour is a broadcast type neighbour. That is, searching for a replaceable node in the list of other nodes in the wireless network which that device communicates with which is eligible for replacement comprises: searching 246 for any broadcast nodes in the list which the device has only exchanged broadcast type messages with.

[0099] As used herein a broadcast node is intended to mean a node with which the device has only exchanged broadcast messages with. Broadcast nodes can be replaced as they are not critical.

[0100] If the device only receives broadcast messages from a node on the current neighbour list this signifies either: (a) that node has already joined the network and is doing its own broadcast transmissions as per defined specifications (e.g. IEEE 802.15.4 of some other standard or protocol); or (b) that node is trying to find devices in vicinity. If the device receives only such broadcast messages from one or more nodes on the neighbour list then these nodes will be considered to be non-critical neighbours.

[0101] If, at step 246, the current node is such a broadcast node, then the method further comprises deciding at step 248 whether it is eligible for replacement. This may be, for example, based on a time since a last successful message was exchanged with that broadcast node in the neighbour list. The time since a last successful message was exchanged with the broadcast node gives a better idea which nodes should be preferred over others for replacement.

[0102] This allows for broadcast neighbour nodes and / or non-critical nodes to be replaced to allow the first node to join the list whilst also taking into account the time since the last communication with such broadcast nodes. Advantageously, this additional criterion may reduce the frequency at which nodes are replaced on the neighbour list. In particular, it can prevent a situation in which the device is constantly switching neighbours, for example replacing node A with node B and then, subsequently, replacing node B with node A. Such an arrangement would use excessive processing power and may consume battery life, which is undesirable.

[0103] In principle, any such broadcast node that is deemed to be eligible for replacement (at step 248) may be identified as the replaceable node (and replaced by the first node).

[0104] In some embodiments, any such a broadcast node is deemed to be eligible for replacement if the time since a last successful message was exchanged with the broadcast node is greater than a threshold value. For example, the threshold value may be an average time, for all nodes on the neighbour list, since a last successful message was exchanged with those nodes. Advantageously, such an arrangement allows for the device to self-regulate the threshold. For example, an appropriate absolute threshold value to be used by any given device may depend on how close it is to its neighbours on average.

[0105] If any such broadcast node (that is eligible for replacement) is found (at step 248) before the end of the list has been reached the method proceeds to the step 230 of removing the broadcast node which is eligible for replacement and replacing it with the first node. In particular, the rest of the list is not searched, and that broadcast node is identified as the replaceable node (and replaced by the first node at step 230).

[0106] That is, as far as broadcast nodes are concerned, the algorithm 200 is a first found, first replaced method. For example, it may be that there are two or more broadcast nodes in the list which the device has only exchanged broadcast type messages with that may be eligible to be replaced as non-critical neighbours. In alternative embodiments, the algorithm may continue so as to identify the broadcast node that is most eligible for replacement. However, operating on a first found, first replaced is advantageous over such alternative embodiments as it saves computation resources.

[0107] If at step 248, the current neighbour on the neighbour list is not deemed to be eligible for replacement then, at step 250, the flag Old Broad may be set to 1 to indicate that at least one such non-eligible broadcast type neighbour has been found. Furthermore, if this is the oldest non-eligible broadcast neighbour identified so far in the loop 244 over the neighbour list then its details are saved as the oldest broadcast neighbour. As used here, the oldest broadcast node or oldest broadcast neighbour is intended to mean a broadcast node (which is not eligible for replacement) which has gone for the longest time without exchanging a successful message with the device.

[0108] After step 248, the loop 244 over the neighbour list continues to the next neighbour in the list.

[0109] For any entry in the neighbour list, if the method determines at step 246 that that neighbour is not a broadcast type neighbour then it is a unicast type neighbour, as indicated at 252. Therefore, in some embodiments, searching for a replaceable node in the list of other nodes in the wireless network which that device communicates with which is eligible for replacement may be considered to comprise: searching 246, 252 for any unicast nodes in the list which the device has exchanged unicast type messages with.

[0110] If the device exchanges any successful unicast messages with a node from the neighbour list, then that node is a critical node or critical neighbour. It could, for example be a parent node of the device or a child node of the device. Such a unicast message may, for example, comprise a medium access control (MAC) layer packet or an application layer packet.

[0111] If the any such unicast nodes are found then, at step 254, the method comprises deciding whether it is eligible for replacement based. This decision may be based on a time since a last successful message was exchanged with the unicast node. In principle, any such unicast node that is deemed to be eligible for replacement may be identified as the replaceable node (and replaced by the first node). In some embodiments, such a unicast node that is deemed to be eligible for replacement may only be identified as the replaceable node (and replaced by the first node) if no broadcast nodes have been identified that are deemed to be eligible for replacement.

[0112] The time since a last successful message was exchanged with the unicast node gives a better idea which unicast nodes should be preferred over others for replacement.

[0113] At step 254, any such unicast node may be deemed to be eligible for replacement if the time since a last successful message was exchanged with the unicast node is greater than a threshold value. For example, the threshold value may be an average time, for all nodes on the neighbour list, since a last successful message was exchanged with those nodes. Advantageously, such an arrangement allows for the device to self- regulate the threshold. For example, an appropriate absolute threshold value to be used by any given device may depend on how close it is to its neighbours on average.

[0114] If any such unicast nodes are found on the neighbour list which have not successfully exchanged a message with the device for a time period greater that the threshold value it may be indicative that such nodes have been lost (for example are no longer connected to the network) or have changed its parent node in the network (and so its unicast messages are now being send via a different route). Such nodes may be considered to be obsolete and may be eligible for replacement.

[0115] If at step 254, a neighbour on the neighbour list is deemed to be not eligible for replacement then, the loop 244 over the neighbour list continues to the next neighbour in the list.

[0116] If at step 254, a neighbour on the neighbour list is deemed to be eligible for replacement then, at step 256, the flag Eligible Unicast may be set to 1 to indicate that at least one such eligible unicast type neighbour has been found. Furthermore, if this is the oldest eligible unicast neighbour identified so far in the loop 244 over the neighbour list then its details are saved as the oldest unicast neighbour. As used here, the oldest unicast node or oldest unicast neighbour is intended to mean a unicast node which has gone for the longest time without exchanging a successful message with the device.

[0117] That is, the method 200 may be considered to comprises searching for a unicast node that is eligible for replacement for which the time since a last successful message was exchanged between this unicast node and the device is greater than the time since a last successful message was exchanged between all other such unicast nodes that are eligible for replacement and the device.

[0118] After step 256, the loop 244 over the neighbour list continues to the next neighbour in the list.

[0119] If, after the loop 244 over the neighbour list is complete, if no broadcast nodes have been identified that are eligible for replacement then the method proceeds to step 258. The oldest eligible unicast node has been identified (if it exists, as indicated by flag Eligible Unicast) and the oldest (not eligible) broadcast neighbour has been identified (if it exists, as indicated by flag Old Broad). As now discussed, it is only if no broadcast nodes have been identified that are eligible for replacement after the loop 244 over the neighbour list is complete that an eligible unicast neighbour may be considered as the replaceable node.

[0120] The method 200 preferentially replaces broadcast neighbours over unicast neighbours. In particular, if there is at least one broadcast node in the neighbour list which the device has only exchanged broadcast type messages with and which is eligible for replacement and there is at least one unicast node in the list which the device has exchanged unicast type messages with which is eligible for replacement, the replaceable node is selected from the at least one broadcast node. In particular, as soon as such a broadcast node in the neighbour list which the device has only exchanged broadcast type messages with and which is eligible for replacement is identified (at step 248) it is replaced by the first node.

[0121] If no broadcast nodes have been identified that are eligible for replacement after the loop 244 over the neighbour list is complete then, at step 258, if an eligible unicast neighbour has been identified (as indicated by the flag Eligible Unicast being set to 1 ) then the method proceeds to the step 230 of removing the unicast node which is eligible for replacement and replacing it with the first node.

[0122] As mentioned above, in the embodiment of the new method 200 shown in Figure 3, the criteria for selecting a replaceable node in the list of other nodes in the wireless network 100 which that device can currently communicate with (at step 220) includes a category of message that the request from the first node comprises. For example, the request from the first node may comprise a broadcast message or a unicast message. The criteria for selecting a replaceable node may be different depending on what type of message the request is, as now discussed.

[0123] If, at step 258, no eligible unicast neighbour has been identified (as indicated by the flag Eligible Unicast being not equal to 1 ) then the method proceeds to the step 260 of determining whether or not the request comprises a unicast message (i.e. whether or not the flag Add Request is set to Unicast).

[0124] If at step 260 the flag Add Request is not set to Unicast then the method 200 proceeds to step 264 of determining that the device is not able to accommodate the request from the first node and therefore the request may be rejected at step 264. The first node may be accommodated by another node on the network.

[0125] In some embodiments, a first set of criteria are used for searching for a replaceable node when the request comprises a broadcast message and a second, different set of criteria are used for searching for a replaceable node when the request comprises a unicast message. In particular, the first criteria may be more restrictive than the second criteria. That is, the method 200 may be more likely select a replaceable node when the second criteria are applied than when the first criteria are applied. In other words, for such embodiments, the algorithm 200 tries harder to accommodate on its neighbour list a first node which sends a unicast message as a request.

[0126] In this embodiment, the first set of criteria are first applied to identify a replaceable node (in steps 244-258). However, if no such replaceable node is identified (in steps 244-258) and the request comprises a unicast message (i.e. the flag Add Request is set to Unicast, as determined at step 260), then the second criteria are subsequently applied in order to identify a replaceable node. In particular, if, at step 260, the request is determined to comprise a unicast message then the method 200 proceeds to a step 262 of determining whether or not there are any broadcast nodes on the neighbour list. Such a broadcast node is present if the flag Old Broad is set to 1 ). Note that any such broadcast nodes must be not eligible to be replaced (according to the first set of criteria).

[0127] If the request from the first node comprises a unicast message (and so Add Request is set to Unicast) and the list contains at least one broadcast node which the device has only exchanged broadcast type messages with (indicated by flag Old Broad being set to 1 ) then one of said at least one broadcast node may be identified as the replaceable node. That is, if the request from the first node is a unicast message and there is at least one broadcast node on the current neighbour list then the device will accommodate the first node and allow it to replace one of the broadcast nodes.

[0128] Therefore, if the request from the first node comprises a unicast message (and so Add Request is set to Unicast) and the list contains at least one broadcast node which the device has only exchanged broadcast type messages with (indicated by flag Old Broad being set to 1 ) then, at step 262, the oldest of the at least one broadcast node which the device has only exchanged broadcast type messages with is identified as the replaceable node (and replaced by the first node at step 230). However, the list does not contain any broadcast nodes which the device has only exchanged broadcast type messages with (indicated by flag Old Broad being not set to 1 ) then then the method 200 proceeds to step 264 of determining that the device is not able to accommodate the request from the first node and therefore the request may be rejected at step 264. The first node may be accommodated by another node on the network.

[0129] In this embodiment, the method 200 comprises finding, from the at least one broadcast node which the device has only exchanged broadcast type messages with (and which is not eligible for replacement under the first criteria), a broadcast node which has gone for the longest time without exchanging a successful message with the device and that broadcast node may be identified as the replaceable node.

[0130] In some embodiments, the method 200 may further comprise, for each node in the list of other nodes in the wireless network 100 which that device can currently communicate with, maintaining a time since the last message was successfully exchanged with that node. As discussed above, this parameter may be used when searching for a replacement node that is eligible for replacement. One method for maintaining these times for all nodes on the neighbour list is as follows. For each node on the neighbour list a time at which associated with that neighbour may be is retained (for example stored in memory). When a node is added to the neighbour list, the method may retain (for example store in memory) a time associated with that neighbour. When a node already on the list successfully exchanges a message with the device, the current time stored for that node may be updated to the time of that successful message.

[0131] Some embodiments the method 200 may further comprise, for each node in the list of other nodes in the wireless network which that device can currently communicate with, maintaining a status type that is indicative of the type of messages that the device has exchanged with that node.

[0132] Some embodiments the method 200 may further comprise sending a reply message to the first node informing it whether or not it has successfully accommodated the first node on the list. In some embodiments, the criteria for finding a replaceable node in the list of other nodes in the wireless network which that device can currently communicate with which is eligible for replacement may be based on a power status of the first node and / or a power status of the current nodes in the list. As used here, a power status of a device or node may mean whether the device or node is battery operated or powered by another source such as, for example, by mains electricity. Additionally or alternatively, as used here, a power status of a device or node may include a remaining lifetime of a battery or power supply of the device or node. The algorithm 200 may, for example try harder to accommodate a request from a first node that is battery powered and / or has limited battery life remaining than, for example, a first node that is mains powered or which has significant battery lifetime remaining. The algorithm may even allow the device to replace an existing critical (unicast) neighbour that is mains powered with a first device that is battery powered.

[0133] In some embodiments, the criteria for searching for a replaceable node in the list of other nodes in the wireless network which that device can currently communicate with which is eligible for replacement may include: whether or not the node has received a central message indicating that one or more nodes on the list are critical nodes. For example, a system administrator may declare a node as critical and communicate this information to a node or device (via its parent).

[0134] In some embodiments, the criteria for searching for a replaceable node in the list of other nodes in the wireless network which that device can currently communicate with which is eligible for replacement may include: a number of high priority quality of service (QoS) packets generated by a child node. In some embodiments, a node can determine a neighbour as critical based on a number of high priority quality of service (QoS) packets generated by a child node.

[0135] Some embodiments of the present disclosure relate to a computer-readable medium having instructions for carrying out the method 200 shown in Figure 2 and as described above with reference to Figures 2 and 3. The computer-readable medium, for example, may be a non-transitory computer-readable medium.

[0136] Some embodiments of the present disclosure relate to an apparatus for use in a wireless network 100. An example of such an apparatus 300 is shown schematically in Figure 4. The apparatus 300 comprises: a transceiver 310; at least one processor 320; and a storage media 330. The apparatus 300 may, for example, represent any of the nodes 105a-105h of a wireless network 100 as described above.

[0137] The at least one processor 320 may include any suitable processing device(s) such as, for example, a microprocessor, an application-specific integrated circuit (ASIC), or a field programmable gate array (FPGA). The at least one processor 320 is communicatively coupled to the storage media 330.

[0138] The storage media 330 has instructions to cause the at least one processor 320 to carry out the method 200 shown in Figure 2 and as described above with reference to Figures 2 and 3. The at least one processor 320 can read from and write to the storage media 330.

[0139] The storage media 330 may be any suitable type of computer-readable medium. In some embodiments, the list of other nodes in a wireless network 100 which the apparatus 300 has communicated with may be stored on the storage media 330 and the at least one processor 320 may be operable to read and / or write to the storage media 330 so as to maintain the list.

[0140] The transceiver 310 may be communicatively coupled to the processor 320 and / or the storage media 330. The transceiver 310 may be used by the apparatus 300 to communicate with neighboring nodes 105a-105h within a network 100, for example via an antenna 340.

[0141] While specific embodiments of the invention have been described above, it will be appreciated that the invention may be practiced otherwise than as described. The descriptions above are intended to be illustrative, not limiting. Thus it will be apparent to one skilled in the art that modifications may be made to the invention as described without departing from the scope of the claims set out below.

Claims

CLAIMS:1 . A method for use by a device as part of a wireless network for maintaining a list of other nodes in the wireless network which that device communicates with, the method comprising: receiving a request from a first node; searching for a replaceable node in the list of other nodes in the wireless network which that device can currently communicate with which is eligible for replacement based on criteria which include: (a) a category of messages that the device has exchanged with the current nodes in the list and / or (b) a time since a last successful message was exchanged with the current nodes; and if the criteria are met, removing the replaceable node which is eligible for replacement and replacing it with the first node.

2. The method of claim 1 wherein searching for a replaceable node in the list of other nodes in the wireless network which that device communicates with which is eligible for replacement comprises: searching for a broadcast node in the list which the device has only exchanged broadcast type messages with.

3. The method of claim 2 wherein if any such broadcast node is found, the method further comprises deciding whether it is eligible for replacement based on a time since a last successful message was exchanged with the broadcast node.

4. The method of claim 3 any such a broadcast node is deemed to be eligible for replacement if the time since a last successful message was exchanged with the broadcast node is greater than a threshold value.

5. The method of any one of claims 2 to 4 wherein if any such broadcast node is found before the end of the list has been reached, the rest of the list is not searched, and that broadcast node is identified as the replaceable node.

6. The method of any preceding claim wherein searching for a replaceable node in the list of other nodes in the wireless network which that device communicates with which is eligible for replacement comprises:searching for any unicast nodes in the list which the device has exchanged unicast type messages with.

7. The method of claim 6 wherein if any such unicast node is found, the method further comprises deciding whether it is eligible for replacement based on a time since a last successful message was exchanged with the unicast node.

8. The method of claim 7 wherein any such unicast node is deemed to be eligible for replacement if the time since a last successful message was exchanged with the unicast node is greater than a threshold value.

9. The method of any one of claims 6 to 8 wherein the method comprises searching for a unicast node that is eligible for replacement for which the time since a last successful message was exchanged between this unicast node and the device is greater than the time since a last successful message was exchanged between all other such unicast nodes that are eligible for replacement and the device; and wherein if no broadcast nodes have been identified that are eligible for replacement then that unicast node is identified as the replaceable node.

10. The method of any preceding claim wherein if there is at least one broadcast node in the list which the device has only exchanged broadcast type messages with and which is eligible for replacement and there is at least one unicast node in the list which the device has exchanged unicast type messages with which is eligible for replacement, the replaceable node is selected from the at least one broadcast node.11 . The method of any preceding claim wherein the criteria for selecting a replaceable node in the list of other nodes in the wireless network which that device can currently communicate with includes a category of messages that the request comprises.

12. The method of any preceding claim wherein a first set of criteria are used for searching for a replaceable node when the request comprises a broadcast message and a second, different set of criteria are used for searching for a replaceable node when the request comprises a unicast message.

13. The method of claim 12 wherein the first criteria are more restrictive than the second criteria.

14. The method of claim 12 or claim 13 wherein when searching for a replaceable node, the first set of criteria are applied to identify a replaceable node and wherein, if no such replaceable node is identified and the request comprises a unicast message, the second criteria are subsequently applied in order to identify a replaceable node.

15. The method of any preceding claim if the request from the first node comprises a unicast message and the list contains at least one broadcast node which the device has only exchanged broadcast type messages with then one of said at least one broadcast node is identified as the replaceable node.

16. The method of claim 15 wherein the method comprises finding, from the at least one broadcast node which the device has only exchanged broadcast type messages with, a broadcast node which has gone for the longest time without exchanging a successful message with the device and wherein that broadcast node is identified as the replaceable node.

17. The method of any preceding claim further comprising, for each node in the list of other nodes in the wireless network which that device can currently communicate with, maintaining a time since the last message was successfully exchanged with that node.

18. The method of any preceding claim further comprising, for each node in the list of other nodes in the wireless network which that device can currently communicate with, maintaining a status type that is indicative of the type of messages that the device has exchanged with that node.

19. The method of any preceding claim further comprising sending a reply message to the first node informing it whether or not it has successfully accommodated the first node on the list.

20. The method of any preceding claim wherein the criteria for finding a replaceable node in the list of other nodes in the wireless network which that device can currently communicate with which is eligible for replacement based includes a power status of the first node and / or a power status of the current nodes in the list.21 . The method of any preceding claim wherein the criteria for searching for a replaceable node in the list of other nodes in the wireless network which that device can currently communicate with which is eligible for replacement includes: whether or not the node has received a central message indicating that one or more nodes on the list are critical nodes.

22. The method of any preceding claim wherein the criteria for searching for a replaceable node in the list of other nodes in the wireless network which that device can currently communicate with which is eligible for replacement includes: a number of high priority quality of service (QoS) packets generated by a child node.

23. A computer-readable medium having instructions for carrying out the method of any one of claims 1 to 22.

24. An apparatus for use in a wireless network, the apparatus comprising: a transceiver; at least one processor; and storage media, the storage media having instructions to cause the at least one processor to carrying out the method of any one of claims 1 to 22.

25. The apparatus of claim 24 wherein the list of other nodes in a wireless network which the apparatus has communicated with is stored on the storage media and wherein the at least one processor is operable to read from and / or write to the storage media so as to maintain the list.

Images