APPARATUS

FIELD OF THE INVENTION

The present invention relates to the field of communication network, with for example some application in outdoor lighting. More specifically, the invention relates to methods for configuring a network, configuration apparatus, methods for configuring a node device and node devices.

This invention is, for example, relevant but not limited to outdoor lighting network, like traffic lighting network.

BACKGROUND OF THE INVENTION

In a network, data rate and risk of loss of data packets are dependent mostly to the quality of the links between the nodes of the network. If a route to deliver packet to a destination node device is not reliable, it can cost a large amount of resource because of the loss of packets and the required retransmission. This is in particular true for low power networks for which the nodes may be limited in computation power.

EP 2 592 870 shows a mesh network comprising several nodes linked together. An optimization of the routing is made inside the network. For long range communication of the mesh network, one node comprises a long range interface in such a way that this particular node becomes the router of the mesh network for long range communication.

In a particular solution shown on Figure 1 , for example described in the published patent application WO2015/000803, a network 1 comprises a plurality of node devices 10 or 10a, here light controllers (CityTouch Client 2.0, or CTC2.0), which have two communication interfaces: a local communication interface 101 , for example a wireless mesh interface, and a long range communication interface 102, for example a cellular interface. Each node device 10 or 10a uses the local interface 101 to communicate with other node devices in the network while the long range interface 102 is used to allow the node device 10 or 10a to communicate with a backend 1 1 .

However, in order to reduce operating costs, the long range (cellular) interfaces 102 of the majority of the node devices 10 are disabled after configuration. Indeed, during the configuration phase, a configuration apparatus 104 typically in the backend configures this network by segmenting it into a plurality of subnetworks. In this phase, each node 10 uses the cellular interface 102 to receive the configuration data enabling a fast and efficient communication with the configuration apparatus. In each subnetwork, the configuration apparatus 104 selects at least one node device 10a to become a subnetwork router, the remaining node devices 10 taking the role of subnetwork node devices 10. The subnetwork router 10a keeps both long range and local interfaces activated, while the other node devices 10 deactivates the long range interface, and only maintain the local interface active.

During operation, these subnetwork node devices may still connect to the backend indirectly, first over the mesh network and then, via a subnetwork router which still has its cellular interface enabled. One aim of this invention is to propose a technique to configure the network by deciding which node devices in the network should deactivate their long range interfaces. Thus, during network configuration it has to be decided which role to assign to each node device: a subnetwork router (also referred to as Border Router or BR) and which should be assigned as a subnetwork node device (also referred to as Client). It should also be decided which subnetwork served by a subnetwork router a particular node belongs to. For example in CTC 2.0, every BR might be assigned 100 Clients. However, segmentation solutions based only on node device location or hop count do not provide with the best results requiring the network to be reconfigured during operation, requiring the reactivation of the long range interface and the inherent costs due to this reactivation.

SUMMARY OF THE INVENTION

It is an object of the invention to propose a method for configuring the network and a configuration apparatus which alleviates the above mentioned problems.

An object of the invention is to provide with a method for configuring a network which optimizes the segmentation of the network to the real conditions experienced by the node devices.

Another object of the invention is to enable a close to optimal first configuration of the network to reduce the risk of communication issues which would require a complete reconfiguration of the network. To this end, in accordance with a first aspect of the invention, it is proposed a method of a configuration apparatus configuring nodes of a network, the method comprising the steps of, at the configuration apparatus,

(a) receiving link quality information indicative of the quality of a communication link between the nodes;

(b) determining a link exclusion set of at least one blacklisted communication link based on the received link quality information;

(c) dividing the network into a plurality of subnetworks based on the link exclusion set, by including a subset of nodes of the network in each subnetwork, and assigning the nodes of each subset of nodes with a role including subnetwork router and subbnetwork node, so that each subnetwork comprises a plurality of subnetwork nodes and at least one subnetwork router interfacing the plurality of subnetwork nodes of the subnetwork with a backend.

As a consequence, the configuration apparatus can base the configuration of the network (exclusively or typically in accordance with other parameters) on quality measurements in the network carried out by the node devices with their communication interfaces (This could be done on their local interface and/or their long range interface). Further, in order to build the subnetworks and determine to which subnetwork each node devices belong to, and which node device(s) will be elected as subnetwork router of a subnetwork, the configuration apparatus creates an exclusion list, which includes all the links between node devices that should not be used. The segmentation is performed taking into account these links, for example such that such links cannot be used for communication within a subnetwork, or such that alternative links not listed in the exclusion list exist.

Further, in an embodiment of the invention, the subnetwork node devices and/or the subnetwork routers receives in the configuration data an indication of links that are blacklisted. For example, the method of the first aspect of the invention further comprises a step (d) of sending configuration data to the node devices, said configuration data being indicative of at least some of the blacklisted communication links of the exclusion set. In operation or in a route determination phase, the subnetwork router and the subnetwork node devices should avoid to route data packets along a route including a link in the exclusion list.

In variants of this embodiment, the configuration apparatus may send to a considered node device the full exclusion list, or the list of blacklisted links that relate to node devices of the subnetwork the considered node device belongs to, or the list of links involving directly the considered node device. These various amounts of information may depend on the role assigned to the node device. It may be beneficial that a subnetwork router knows the full exclusion list or at least the list of blacklisted links that relate to node devices belonging to the subnetwork the subnetwork router serves. Thus, the information about blacklisted links can be shared with more granularity, to allow a better usage of the memory storage of the node devices of the network.

It is to be noted that in examples of this embodiment, the configuration apparatus may include in the configuration data one or more of the following:

a list of subnetwork routers in the network,

a list of subnetworks,

a list of subnetwork nodes,

a list of subnetwork nodes belonging to a single subnetwork.

In accordance with another embodiments that can be combined with the previously mentioned exemplary embodiments, the method comprises the step of the configuration apparatus monitoring the network after configuration of the network during a watching phase, and causing the subnetwork nodes to deactivate a direct communication interface with the backend at the end of the watching phase. Thus, operation can be monitored during some time during which, for example, measurements at the node devices can be performed and reported to the configuration apparatus using the long range interface. Communication issues may also be detected and reported so that the configuration apparatus may decide on a widespread resegmentation of the network. A widespread resegmentation can be understood as a resegmentation involving one or more subnteworks but that may not be solved locally. For example, it can only be solved by the configuration apparatus in the backend itself since it involves reconfiguring some network nodes (eg. role election, subnetwork list). For example, a widespread resegmentation involves a plurality of subnetworks. Another example of a widespread resegmentation involves a single subnetwork which needs to divided in two or more subnetworks, thus requiring to change the role of network nodes and the list of network nodes in each subnetwork. In this example, the watching phase may end if no communication issue is detected during a predetermined test period. However, it may also be decided by an installer. In all the preceding examples of this aspect of the invention, a communication link is blacklisted at step (b) if at least one of the conditions is met:

- the communication link quality is part of a lower quality percentile of the whole set of communication links;

- the average communication link quality is below a first predetermined threshold;

- occurrence of instantaneous communication link quality being below a second predetermined threshold;

- the blacklisting of the communication link does not cause the number of subnetworks to be above a third predetermined threshold.

In an example of the invention, the configuration apparatus is part of the backend. For example, it is a dedicated module or a computer program which when executed on a computer operates in accordance with the steps of the method of the first aspect of the invention.

Further, in all the previous examples, the method of the first aspect may occur only at the initialization of the network, during a configuration phase. However, in another embodiment of the invention which can be combined with all the previous alternatives discussed above, the configuration method can also be initiated upon detection of a widespread communication issue during operation of the network.

For example, a widespread communication issue may be a communication issue which cannot be repaired solely within a subnetwork of the network. Thus, a widespread communication issue is, in the sense of this application, a communication issue which involves two or more subnetworks, but not necessarily all the subnetworks. In particular, if a subnetwork router attempts to solve a communication issue, but cannot find a route to all the subnetwork node devices of the subnetwork without using a blacklisted link, then the subnetwork router initiates the configuration method of the first aspect to cause a global or an inter-subnetworks resegmentation.

In accordance with another aspect of the invention, it is proposed a configuration apparatus in accordance to claim 13.

In accordance with a further aspect of the invention, it is proposed a method for operating a node device in a network having a plurality of nodes, the method comprising the steps of, at the node device: (a) performing link quality measurements for estimating the link quality between the node device and neighboring nodes;

(b) generating link quality information based on the link quality measurements;

(c) transmitting the link quality information to a configuration apparatus;

(d) receiving from the configuration apparatus configuration data including a list of blacklisted links and a role defining a behavior of the node in the subnetwork, said role including subnetwork router and subnetwork node.

In accordance with this further aspect of the invention, the configuration data may also include an indication of whether the configuration apparatus selected the node device as a router of a subnetwork including a plurality of end nodes or as an end node. In case the node device is selected as a router, the node device may form or compute a routing tree to route packets between the end nodes and the node device while preventing to use the blacklisted communication links.

Another further aspect of the invention includes a node device in accordance with the features recited in claim 14.

It is to be noted that in an embodiment of all these aspects of the invention, it is also possible to include measurements of the direct interface link quality (e.g. cellular interface) between the node device. Thus, the segmentation can take into account a low cellular link quality that would be suboptimal for a subnetwork during operation and communication with the backend.

These and other aspects of the invention will be apparent from and will be elucidated with reference to the embodiments described hereinafter.

BRIEF DESCRIPTION OF THE DRAWINGS

The present invention will now be described in more detail, by way of example, with reference to the accompanying drawings, wherein:

- Fig. 1 , already described is a block diagram of a network in which the invention can be implemented.

- Fig. 2A-2B are block diagrams of a network configured in accordance with a first embodiment of the invention.

- Fig. 3A-3B are block diagrams of a network configured in accordance with a second embodiment of the invention. Fig. 4 is a graph showing an example of a decision process for creating the exclusion list.

Fig. 5 is a flowchart of a method in accordance with an embodiment of the first aspect of the invention.

Fig. 6 is a flowchart of a method in accordance with an embodiment of the third aspect of the invention.

Fig. 7 is a block diagram of a configuration apparatus in accordance with an embodiment of the invention.

Fig. 8 is a block diagram of a node device in accordance with another aspect of the invention.

DETAILED DESCRIPTION OF THE INVENTION

The present invention relates to a network including a plurality of node devices, and in particular to configuration of such network.

As explained in the preamble in the application, in a network in accordance with Figure 1 , it is required to segment or divide the network into a plurality of subnetworks. Each subnetwork includes one or more subnetwork router interfacing the other subnetwork node devices with a backend. Configuration of such network and in particular the segmentation is critical since it will have a high influence on the efficiency of the network.

To perform network configuration, various approaches have been looked at by the inventors.

A first approach is to use location information of all the nodes in the network. This aspect works fine for the application of outdoor lighting where the node devices are lighting controllers whose position is known from the installer and/or obtained by GPS units in the lighting controllers. With this approach, the idea here is to segment the whole region of deployment into equal small regions and simply place the BR (Border Router) in the centre of each region.

However, some issues arise with this approach:

(i) It does not take into account a basic well known characteristic of lower power radio signals, i.e. the distance between two nodes is not directly proportional to the quality of a radio link between the two nodes. In other words, if two nodes are very close together, it does not mean that the signal strength between these two nodes would be high enough to allow proper communication. Similarly two nodes far apart might be able to communicate very well. This characteristic occurs due to environmental factors, e.g. reflections, attenuation, etc. and is hard to accurately model.

(ii) It does not take into account properties of the network graph, such as hop count. Every hop that a message is transmitted in a multihop network increases the chance of packet loss.

Another approach would be to use both location and hop count information. The advantage of this approach over the first approach is that it minimizes the hop count when performing the segmentation. This results in a more robust network segmentation. However, it still assumes a fixed communication distance between nodes and in reality, as described above, the relationship between the distance between two nodes and their link quality is very often not very well defined.

Both the first and second approaches do not take into account the actual network conditions on the ground. Thus effectively, the segmentation (and thus network configuration) performed using these approaches as such can result in network configurations which simply render certain nodes unreachable. Another consequence is that nodes might be forced to use poor quality links to communicate. This would not only result in excessive packet loss, but also increase control overhead (due to the in-built network repair operations in RPL) and finally potentially increase cellular costs due to more data usage in the case of an embodiment like described with Figure 1 . Naturally excessive control overhead could also reduce the available bandwidth to execute the applications meant to run on the network.

In accordance with a first embodiment of the invention, it is proposed to base the configuration method on some link quality measurements between node devices and to create an exclusion list including communication links that are "blacklisted". In other words, these communication links included in the exclusion list are to be avoided in operation. For example, the routing process may be carried out while carefully avoiding to route a packet along a route that would use one of these links.

It is to be noted that this approach encompassed in this embodiment, the following embodiments and their variants may be combined with the approaches detailed above. In other words, a method of configuring the network may take into account, in addition to the exclusion list, other parameters like the geographic coordinates of the node devices, for example given by an installation map or GPS data, and/or the hop count in the subnetwork. Further parameters may also beneficially be taken into account like a model type of node device, presence of a centre of interest in the vicinity or traffic level (e.g. primary road vs secondary road, crossing). As the behaviour of the node devices may be different between primary roads where heavier traffic is expected, it may be relevant to group the subnetwork also taking into account this aspect.

With reference to Figures 2A and 2B, this first embodiment will now be detailed. Figure 2A shows a network 200 including a plurality node devices 201 a-

201 j at various fixed positions of the map.

During a configuration phase, the node devices 201 a-201j may carry out some measurements to collect link quality information from between node devices using their local communication interface 201 1 a-201 1 j. In a variant, these measurements may also take into account the link quality of the long range communication (using their long range communication interfaces respectively 2012a- 2012j) between each node device and the network. Once the measurements are performed, the node devices 201 a-201 j send over by means of their respective long range communication interface 2012a-2012j the results to a configuration apparatus

202 in a remote backend 203. This measurement transmission can be done in a manner which minimizes bandwidth usage. The measurement report may be under the shape of a discovery matrix, where for each node device, all the detected neighbors are listed with the respective link quality.

In an illustrative example for node device 201 a, the discovery matrix could be

be suitable in connection with exemplary embodiments of this invention.

The measurements can be carried out by through various measurements, for example Bit Error Rate or Loss of Packet Rate, Signal over Interference Ratio, or Signal over Interference plus Noise Ratio. The sampling of the measurements may be over a long duration, since it enables to detect temporary issues that a single measurement over time would not have allowed to detect. At the configuration apparatus 202, a processing of the measurement data received from the node devices is performed in the backend, in order to decide which links should be blacklisted. This blacklisting may be done using global information (i.e. if a particular link is blacklisted, the pair of nodes sharing this link should not actively use it for exchanging any data). The decision process and the criterion used will be discussed at a later stage of the application.

Based on the exclusion list of blacklisted links, the configuration apparatus creates a new connectivity matrix. In this exemplary embodiment, the set of blacklisted list includes links AE (between 201 a and 201 e), EG (between 201 e and 201 g) and IJ (between 201 i and 201j).

Then, the configuration apparatus segments the network and outputs (i) which nodes should be subnetwork routers and which ones should be subnetwork nodes, (ii) which subnetwork nodes should belong to particular subnetwork. In this example, the segmentation may be done such that all the blacklisted links are shared between a pair of subnetworks. In other words, if a blacklisted link is between a pair of node devices, eg 201 a and 201 e, the segmentation is done to assign each node of the pair to a different subnetwork. As explained earlier, this segmentation may be done using further criterion than link quality and exclusion list. It may take into account in addition to the exclusion list, other parameters like the geographic coordinates of the node devices and/or the hop count in the subnetwork. Further parameters may also be beneficially taken into account a model type of node device, the presence of a centre of interest in the vicinity or traffic level (e.g. primary road vs secondary road, crossing).

These elements may be included in a configuration data to be sent to the node devices. Typically, the configuration data may include:

List of subnetwork routers.

List of subnetwork nodes in a subnetwork.

List of blacklisted links.

However, with this segmentation of the first embodiment, it may not be required to provide with this exclusion list to the node devices since these links are between different subnetworks. This configuration data may be sent to the subnetwork routers only over the long range communication interface. Then, these routers may spread the configuration data to each subnetwork node of their respective subnetwork. In the example of Figure 2B, the configuration data is sent to Node 201 a, 201 g, and 201 h as these have been elected as subnetwork routers. However, in another variant, configuration data is sent directly from the backend to all node devices of the network.

Based on this configuration data, the subnetwork routers 201 a, 201 g and 201 h uses a routing algorithm to determine the routes in each subnetwork. For example, RPL to form a routing graph. The blacklisted links are thus not used for routing packets based on the sole segmentation of the subnetworks.

However, this segmentation based on assigning different subnetworks to node devices of blacklisted links may not be always suitable depending on the network, although it enables not to have to include the exclusion list in the configuration data.

A second embodiment will now be described with reference to Figures 3A and 3B. For the sake of conciseness only the parts that differ from the first embodiment will detailed. Figure 3A shows a network 300 including a plurality node devices 301 a- 301 j at various fixed positions of the map.

During a configuration phase, as in the first embodiment, the node devices 301 a-301j carry out some measurements to collect link quality information from between node devices, and send the results to a configuration apparatus 302 in a remote backend 303.

Then, the configuration apparatus 302 segments the network 300 and outputs

(i) which nodes should be subnetwork routers and which ones should be subnetwork nodes, (ii) which subnetwork nodes should belong to particular subnetwork. In this example, the segmentation may be done in a different manner as the first embodiment. It may be done regardless of whether all the blacklisted links are shared between a pair of subnetworks. As explained earlier, this segmentation may be done using further criterion than link quality and exclusion list. It may take into account in addition to the exclusion list, other parameters like the geographic coordinates of the node devices and/or the hop count in the subnetwork. Further parameters may also be beneficially taken into account a model type of node device, the presence of a centre of interest in the vicinity or traffic level (e.g. primary road vs secondary road, crossing).

With this segmentation of the second embodiment, it is beneficial to provide with at least part of this exclusion list, or the whole exclusion list to the node devices, or at least to the subnetwork routers. This configuration data including at least part of the exclusion list may be sent to the subnetwork routers only over the long range communication interface. Then, these routers may spread the configuration data to each subnetwork node of their respective subnetwork. In the example of Figure 3B, the configuration data is sent to Node 301 b, 301 c, and 301 h as these have been elected as subnetwork routers. However, in another variant, configuration data is sent directly from the backend to all node devices of the network.

Based on this configuration data, the subnetwork routers 301 a, 301 c and 301 j uses RPL to form the routing graph, such that the Objective function ignores links that have been blacklisted by the backend. The blacklisted links have to be avoided in the routing graphs as shown on Figure 3B.

Moreover, if a communication issue occurs in operation, a subnetwork router and/or the subnetwork nodes may attempt to repair locally the communication link by creating a new route but still avoiding the blacklisted links.

If it appears that the communication issue cannot be repaired while still avoiding to use blacklisted links, the subnetwork router or a subnetwork router (via the subnetwork router) may decide to warn the backend of the occurrence of a widespread communication issue which would require a potential partial re- segmentation.

The decision process for blacklisting nodes may take into account different criterion, like one or more of the following criterion:

1 . the communication link quality is part of a lower quality percentile of the whole set of communication links. As shown on Figure 4, the results can be classified by link quality Q. The horizontal axis represents the values of link quality in an increasing order from left to right. On the vertical axis, the number of links for a given link quality q is represented by the curve 40. In accordance with this variant, the link quality threshold Qth is selected so that a percentage or a predetermined number of communication links present a link quality being less than Qth. This is represented by the striped area under the curve 40. This can be estimated based on average link quality or on instantaneous link quality.

2. the average communication link quality is below a first predetermined threshold. The first predetermined threshold can be determined arbitrarily by the designer or from testing. It is to be noted that criterion 1 may be combined with criterion 2 such that the higher of the threshold determined by criterion 1 and the predetermined threshold of criterion 2 is used.

3. occurrence of instantaneous communication link quality being below a second predetermined threshold. In accordance with this criterion 3, the blacklisted links are those that were the longer or the higher number of times below a predetermined threshold. This can enable to detect links experiencing temporary interference or fading, for example because of reflection or source of interference. Again, this criterion can be combined with the previous ones, for example by using a weighted combined determination.

4. the blacklisting of the communication link does not cause the number of subnetworks to be above a third predetermined threshold. This criteria is in particular adapted with the first embodiment where the segmentation is done such that each node devices of the pair of node devices forming the blacklisted links are in different subnetworks. The more blacklisted links, the more subnetworks.

Another criterion that may also be taken into account is the geographic spreading of the links. To achieve that, different thresholds may selected depending on the location in the network to ensure that not too many links are blacklisted locally. For example, in the network of Figure 2A, if links between node devices 201 h, 201 i and 201 j are of relatively low quality, it should be avoided to exclude all the links by using locally a lower quality threshold.

As mentioned earlier, performing network configuration with only location and hop count information can result in sub-optimal segmentation that can lead to high packet loss and control overhead. This would have for consequence high cellular costs. To mitigate these problems, the solution presented in the various embodiments of the invention uses a combination of location, hop count and measured link quality.

In the illustrated embodiments, link quality information is used to decide which links should be blacklisted. However, it is to be noted that a node device may not be able to perform blacklisting by itself. Indeed, a node device may not have a global view on the links in the network or over a long sampling duration. Therefore, it can be beneficial that the configuration apparatus is taken care on a specific device, for example in the backend. It is important to note that while link quality between a pair of nodes may be very good for several days in a row, it could deteriorate rapidly due to environmental changes, e.g. in an outdoor network, a parked car, weather conditions, etc.

Thus, in the illustrated embodiments of the invention, it is proposed for the node devices to send link quality information to the configuration apparatus on a periodic basis, during the configuration phase and optionally also during operation to maintain the exclusion list up to date. This data could be collected over an extended period of time, e.g. several weeks. This would give a better idea of the typical variations that one could expect of a particular link. The backend uses this information to blacklist the link.

With reference to the Figure 5, a method of operating a configuration apparatus for configuration of the network in accordance with an exemplary embodiment of the invention may comprise the steps.

Step 51 : The configuration apparatus, for example at the backend, receives for example from the node devices link quality information.

Step 52: the configuration apparatus blacklists certain links and generates a connectivity matrix. It can also use further parameters to perform the segmentation as mentioned before.

Step 53: the configuration apparatus performs network segmentation using the connectivity matrix.

Step 54: the configuration apparatus generates configuration data based on the network segmentation. This configuration data may comprise a list of subnetwork routers and subnetwork nodes, and blacklisted links.

Step 55: The configuration apparatus transmits to some or all the node devices in the network, for enforcing the generated configuration.

Figure 6 is a flowchart illustrating a method for operating a node device for the configuration of the network.

Step 61 : the node device collects link quality information from all its neighbours.

Step 62: The node device sends link quality information to the configuration apparatus, for example at the backend.

Step 63: The node device then receives the configuration data prepared by the configuration apparatus based on a generated network segmentation, for enforcing the generated configuration. This configuration data may comprise a list of subnetwork routers and subnetwork nodes, and blacklisted links. Step 64: If the node device is elected as a subnetwork router (SR), it switches to this role and generates routing information, for instance by forming DODAGs (Destination Oriented Directed Acyclic Graphs) using standard RPL (Routing Protocol for Low power and lossy network). During DODAG formation, the node device never uses links that have been blacklisted by the configuration apparatus.

Step 65: If the node device is elected as a subnetwork node device (SND), it switches to this role but still keep cellular interface active, during a watching period.

Step 66: At the end of the watching period, cellular interface of the node device is disabled after proper network operation when the new network configuration has been confirmed.

Then, in accordance to an example of this embodiment, once network configuration has been performed, the whole network enters the maintenance phase where node devices continue to send link quality updates so that the exclusion list at the configuration apparatus is maintained up to date. Thus, only good quality links may be used during operation.

In accordance with this variant, if certain links deteriorate momentarily, RPL at the subnetwork router uses the standard Local Repair procedure to ensure that a node device continues to be reachable from the backend by connecting to another node using a link which has not been blacklisted.

If the node device being the subnetwork router detects that Local Repair is not operational to repair the network, the subnetwork router requests a Widespread Repair for example to the configuration apparatus. In a preferred example, this Widespread Repair is typically performed at the configuration apparatus along a similar configuration method that was explained in the embodiments so far. The link quality used can be based on the data that was kept up to date by the continuous measurements carried out during the operation. Re-segmentation at the configuration apparatus is performed if at least one DODAG suffer from poor performance but cannot be repaired by Local Repair for example if the subnetwork requires a resegmentation, or to perform a Widespread Repair.

In another variant, the Widespread Repair may be performed by contacting one or more other subnetwork routers. In this variant, the subnetwork routers maintain each an exclusion sublist, for example at the level of their subnetwork. This can be obtained directly from the measurements at the node devices or from updated exclusion lists gathered by the backend. On Figure 7 is shown a configuration apparatus 70 in accordance with an embodiment of the invention. This configuration apparatus 70 comprises a communication unit 71 . This communication unit 71 may be a wireless transceiver, for example connected to Internet via a cellular interface (UMTS, LTE), or wired, like an Ethernet port. The configuration apparatus further comprises a memory 72, for example a ROM, that can be used for storing the measurement reports and the configuration data. The configuration apparatus 70 comprises a microprocessor 73 which is dimensioned for carrying out the configuration process of the embodiments of the invention.

In another embodiment, the configuration apparatus may be implemented by a computer having a computer program stored on a storage medium and which comprises instructions for carrying the method of the invention when the program is executed on the computer.

On Figure 8 is shown a node device 80 in accordance with an embodiment of the invention. This node device 80 comprises a communication unit 81 with at least two interfaces: a long range interface 81 1 and a short range interface 812. Typically, the long range interface is a cellular communication module, like a UMTS or LTE module. The short range interface is typically a wireless mesh communication unit, e.g. a 6L0WPAN communication unit. The node device further comprises a microcontroller 83 which controls the long range interface such that this long range interface 81 1 can be deactivated if the role of the node device is determined to be a subnetwork node. The node device 80 further comprises a non volatile memory 82, for example an EEPROM or a flash memory. Further, the node device further comprising a Routing entity 813, for example a RPL entity, that can be implemented by a software for example. This RPL entity is adapted to generate a routing graph based on received configuration data which includes a list of blacklisted links.

In another embodiment, the node device may be implemented by a computer or a microcontroller of a wireless communication device having a computer program stored on a storage medium and which comprises instructions for carrying the method of the invention when the program is executed on the computer.

This invention is not only applicable to Lighting networks but could also apply for any types of networks, like home automation networks, smart meter networks, sensor networks... Although in the examples throughout the description, the network is a wireless network and the communication devices are wireless devices, it is to be noted that the invention applies for wired networks. In particular, it is possible that the second telecommunication mode is wired. For example, it could be a communication through powerline, or Ethernet (which could be used for powering the communication devices).

Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure and the appended claims. In the claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality. A single processor or other unit may fulfill the functions of several items recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.

The foregoing description details certain embodiments of the invention. It will be appreciated, however, that no matter how detailed the foregoing appears in text, the invention may be practiced in many ways, and is therefore not limited to the embodiments disclosed. It should be noted that the use of particular terminology when describing certain features or aspects of the invention should not be taken to imply that the terminology is being re-defined herein to be restricted to include any specific characteristics of the features or aspects of the invention with which that terminology is associated.

A single unit or device may fulfill the functions of several items recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.

The described operations of the components of the network system according to various embodiments can be implemented as program code means of a computer program and/or as dedicated hardware. The computer program may be stored and/or distributed on a suitable medium, such as an optical storage medium or a solid-state medium, supplied together with or as part of other hardware, but may also be distributed in other forms, such as via the Internet or other wired or wireless telecommunication systems.