Controlling Traffic and Interference in a Communication Network

Abstract

Methods, nodes, computer programs, and communication networks are disclosed. The present disclosure provides a method for controlling traffic in a communication network, performed by a network node, the communication network including a radio access node adapted to serve a plurality of wireless devices. The method includes obtaining network operation data of at least one of the plurality of wireless devices, the network operation data including the location of the respective wireless device and performance data associated with the location, the network operation data indicating a plurality of attributes corresponding to the operation of the respective wireless device in the communication network. The method further includes grouping at least one of the plurality of wireless devices by dividing the network operation data into one or more service clusters, wherein each service cluster defines a geographical area based on the obtained location. The method further includes scheduling network traffic between the radio access node and at least one of the wireless devices depending on the location of the at least one wireless device relative to the one or more service clusters.

Description

Technical Field

[0001] The present disclosure relates to methods, nodes, computer programs, and communication networks. More specifically but not exclusively, the present disclosure relates to controlling traffic and interference mitigation in a communication network. Background Art

[0002] A communication network is a collection of nodes in which links are connected so that information can be transmitted between the nodes. A specific example of a communication network is a cellular or mobile network, where the last link is wireless. A cellular network can be distributed over a geographical area called a "cell", and each area is served by at least one wireless transceiver or wireless access node. These transceivers provide network coverage to the cell, which can be used for the transmission of voice, data, and other types of content from and to wireless devices such as mobile terminals.

[0003] One of the problems encountered in communication networks is interference. When interference propagates along the channel between a source and a receiver, it can modify the signal in a destructive way. In a wireless communication network, wireless signals can be transmitted between wireless terminals and wireless transceivers. Interference in a cellular or mobile network can have adverse consequences for both the users of the network and the network operator. For users of wireless terminals, interference can reduce the quality of service QoS or the quality of experience QoE, which can be manifested as, for example, a reduced channel bandwidth, a reduced channel throughput, or an increased session and call drop rate.

[0004] In a cellular network, interference can typically be observed near the edge of the cell and can depend on, for example, the number of active neighboring wireless devices, the interference between neighboring wireless access nodes, and the type and amount of data processed by the wireless devices. It is also common to observe interference inside tunnels and buildings.

[0005] One of the methods to reduce the impact of interference and maintain QoS and QoE in a cellular network is to increase the transmission Tx downlink power of the radio signals sent from a wireless base station or wireless access node. At the same time, the uplink power of the radio signals from the wireless terminals being served by the wireless access node also increases. This can lead to increased battery consumption of the wireless terminals and increased energy usage of the wireless base stations.

[0006] In a fifth-generation new radio 5G NR network, the impact of interference can be reduced by assigning a wireless device called a user equipment UE to a primary cell PCell in which the UE operates at a first frequency or a secondary cell SCell in which the UE operates at a second frequency. Alternatively, a UE in a 5G network can be configured to operate in a dual-band. When one carrier is interfered with, the UE or the wireless base station can assign data to another carrier within the cell coverage area.

[0007] As the deployment of the new generation of networks (such as 5G mobile networks) increases, the number of physical sites will also increase. The 5G network infrastructure can include both macro cells that provide large coverage and require more powerful and efficient wireless transceivers, and small cells that cover smaller areas. The overall increase in physical network nodes results in overlapping cell coverage. This situation will require careful consideration from the perspective of network operation, as there is an increasing need to optimize energy usage and manage interference in an efficient manner while minimizing the impact on the QoS and QoE of network users.

[0008] WO 2009 / 152097 A1 describes wireless communication devices, and more particularly, describes apparatus and methods for generating performance measurements in a wireless network. Summary of the Invention

[0009] Accordingly, an object of the present invention is to propose a solution to the problems caused by interference and increased energy usage in a communication network.

[0010] The following presents a simplified summary of the present disclosure to provide a basic understanding to those skilled in the art. This summary is not an extensive overview of the present disclosure and is not intended to identify key / important elements in the embodiments of the present invention or to delineate the scope of the present invention. The sole purpose of this summary is to present some concepts disclosed herein in a simplified form as a prelude to the more detailed description that is presented later.

[0011] One aspect of the present disclosure provides a method for controlling traffic in a communication network performed by a network node, the communication network including a wireless access node adapted to serve a plurality of wireless devices. The method includes obtaining network operation data of at least one wireless device among the plurality of wireless devices, the network operation data including the location of the corresponding wireless device and performance data associated with the location, the network operation data indicating a plurality of attributes corresponding to the operation of the corresponding wireless device in the communication network. The method further includes grouping at least one wireless device among the plurality of wireless devices by dividing the network operation data into one or more service clusters, wherein each service cluster defines a geographical area based on the obtained location. The method further includes scheduling network traffic between the wireless access node and at least one wireless device depending on the location of the at least one wireless device relative to the one or more service clusters.

[0012] Another aspect of the present disclosure provides a network node of a communication network, the communication network including a radio access node adapted to serve a plurality of wireless devices. The network node includes processing circuitry and a memory containing instructions executable by the processing circuitry, whereby the network node is operative to obtain network operation data of at least one of the plurality of wireless devices, the network operation data including the location of the respective wireless device and performance data associated with the location, the network operation data indicating a plurality of attributes corresponding to the operation of the respective wireless device in the communication network. The network node is further operative to group at least one of the plurality of wireless devices by dividing the network operation data into one or more service clusters, wherein each service cluster defines a geographical area based on the obtained location. The network node is further operative to schedule network traffic between the radio access node and at least one of the plurality of wireless devices depending on the location of the at least one wireless device relative to the one or more service clusters.

[0013] Another aspect of the present disclosure provides a computer program for controlling traffic in a communication network, the computer program including computer code which, when run on the processing circuitry of a network node, causes the network node to obtain network operation data of at least one of a plurality of wireless devices, the network operation data including the location of the respective wireless device and performance data associated with the location, the network operation data indicating a plurality of attributes corresponding to the operation of the respective wireless device in the communication network. The computer code further causes the network node to group at least one of the plurality of wireless devices by dividing the network operation data into one or more service clusters, wherein each service cluster defines a geographical area based on the obtained location. The computer code further causes the network node to schedule network traffic between the radio access node and at least one of the plurality of wireless devices depending on the location of the at least one wireless device relative to the one or more service clusters.

[0014] Another aspect of the present disclosure provides a communication network. The communication network includes a first radio access node and a second radio access node, wherein the first access node and the second access node are adapted to serve a plurality of wireless devices. The communication network further includes a computer program according to another aspect. The computer program includes computer code which, when run on the processing circuitry of the first radio access node, causes the first radio access node and the second radio access node to approximate the location of at least one wireless device based on control plane reference signals and data plane reference signals received by the respective wireless devices. The control plane reference signals originate from the first radio access node, and the data plane reference signals originate from the second radio access node.

[0015] Advantageously, embodiments of the present invention allow identification of geographical areas with high interference and high power usage and which are not optimized with respect to the power consumption and mobility of wireless devices.

[0016] As another advantage, embodiments of the present invention allow different capabilities to be assigned to a service cluster, such as low power consumption, optimal throughput, QoS, and based on this, perform optimized, improved, and intelligent scheduling and control on network service data transmitted between a wireless device and a wireless access node.

[0017] Another advantage of embodiments of the present invention is to reduce power consumption and improve network energy performance from the perspective of the UE and the network.

[0018] Another advantage of embodiments of the present invention is improved network planning, such as network planning performed by field service operation (FSO) engineers or network operation center (NOC) personnel when using embodiments of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] To better understand the examples of the present disclosure and to more clearly show how the examples can be effectively implemented, reference will now be made, by way of example only, to the accompanying drawings, in which:

[0020] Figure 1 is a schematic diagram of an example of a communication network.

[0021] Figure 2 is a flowchart of an example of a method for controlling traffic in a communication network performed by a network node.

[0022] Figure 3 is a block diagram showing another example of a communication network.

[0023] Figure 4 is a flowchart showing some example steps of a method for controlling traffic in a communication network performed by a network node.

[0024] Figure 5 is obtained by executing Figure 2 a schematic diagram of an example service cluster.

[0025] Figure 6 is a flowchart showing an example network traffic scheduling policy.

[0026] Figure 7 is a message sequence diagram showing an example of communication in a communication network.

[0027] Figure 8 is a schematic diagram of an example of a network node in a communication network.

[0028] Figure 9 schematically shows a telecommunication network connected to a host computer via an intermediate network.

[0029] Figure 10 is an overall block diagram of a host computer communicating with a user equipment via a base station through a partial wireless connection.

[0030] Figure 11 and Figure 12 is a flowchart showing a method implemented in a communication system including a host computer, a base station, and a user equipment. Detailed implementation

[0031] The following sets forth specific details, such as specific embodiments or examples for purposes of explanation and not limitation. Those skilled in the art will understand that other examples may be employed in addition to these specific details. In some instances, detailed descriptions of well-known methods, nodes, interfaces, circuits, and devices are omitted so as not to obscure the description with unnecessary details. Those skilled in the art will appreciate that the described functionality may be implemented in one or more nodes using hardware circuits (e.g., analog and / or discrete logic gates interconnected to perform a specialized function, ASICs, PLAs, etc.), and / or using software programs and data in conjunction with one or more digital microprocessors or general purpose computers. Nodes that communicate using an air interface also have appropriate radio communication circuitry. Moreover, where appropriate, the technology may also be regarded as being implemented entirely in any form of computer-readable memory (e.g., solid-state memory, magnetic disk, or optical disk) containing a suitable collection of computer instructions that would cause a processor to execute the techniques described herein.

[0032] Hardware implementations may include or incorporate, but are not limited to, digital signal processor (DSP) hardware, reduced instruction set processors, hardware (e.g., digital or analog) circuits including, but not limited to, application specific integrated circuits (ASICs) and / or field programmable gate arrays (FPGAs), and state machines (where applicable) capable of performing such functions.

[0033] Figure 1 An example of a wireless communication network 100 is shown, and the wireless communication network 100 may be a 5G NR network. The wireless communication network 100 may include a first radio access node 110 and a second radio access node 120, such as an evolved Node B (eNB) or a next generation Node B (gNB). Both the first radio access node 110 and the second radio access node 120 have corresponding radio coverage areas 111 and 121, which may correspond to the geographical extent of the cells served by the respective nodes 110, 120. Since there are typically multiple cells 111, 121 in the wireless communication network 100, there is usually an overlap area 130 between adjacent cells 111, 121.

[0034] The first radio access node 110 and the second radio access node 120 can serve at least one wireless device 140, 150, such as UE 140, UE 150. For the first UE 140 located in the first cell 111 and not far from the first radio access node 110, the interference from different radio signals may be low, and the radio output power of the first radio access node is also low. For the second UE 150 located at the edge of cells 111 and 121 in the overlapping area 130 where there is an overlap of radio signals from both the first radio access node 110 and the second radio access node 120, the interference experienced due to the mutual influence of these signals may be very high, and the radio output powers of the first radio access node and the second radio access node may also be high because they attempt to establish reliable communication with the second UE 150. This situation usually results in a degradation of the QoS and QoE of the UE 150.

[0035] Figure 2 is a flowchart of an example of a method 200 for controlling traffic in a communication network. In some examples, the method 200 can be performed by network nodes 110, 120, 320, 330, 720, 750, 800. Specifically, in some examples, the method can be performed by the processing circuit 804 of the network node 800 (such as the baseband unit BBU 720). In some examples, the method 200 can also be performed by a virtual node such as a virtual BBU. The method 200 can also be performed by the computer code of a computer program (when executed by the processing circuit 804). In some examples, the communication networks 300, 700 can be wireless communication networks such as fourth-generation 4G or fifth-generation 5G cellular networks. The communication network includes radio access nodes (such as gNBs) adapted to serve multiple wireless devices (such as UEs).

[0036] The method 200 can include: in step S202, obtaining network operation data 410 of at least one wireless device among a plurality of wireless devices. The network operation data 410 can include the location 410a of the corresponding wireless device and performance data 410b. The network operation data 410 preferably indicates a plurality of attributes corresponding to the operation of the corresponding wireless device in the communication network. Specifically, in some examples, obtaining the network operation data 410 can include receiving the network operation data 410 at the radio access node, which is directly sent from at least one of the wireless devices. In another example, obtaining the network operation data 410 can include receiving the network operation data 410 from at least one UE via the gNB at the fifth-generation core network 5G CN node. In another example, obtaining the network operation data 410 can include receiving the network operation data 410 from a radio access node (such as gNB) at the 5G CN node.

[0037] The network operation data 410 can be a set of observed values or samples of parameters corresponding to the operation of a communication network. Each observed value can be an N-dimensional vector, where N corresponds to the multiple parameters included in the network operation data. The vector can be generated by filling the vector with parameters collected from the core network, the UE, and / or the radio access node itself at the radio access node. The location 410a of the corresponding wireless device can be represented as a latitude and longitude tuple. The location data 410a can be calculated at the UE using, for example, a global navigation satellite system, a GNSS receiver. In another example, the location data 410a can be obtained by approximating the location of at least one wireless device based on the elevation angle and azimuth angle of the radio beam transmitted from the radio access node to the corresponding wireless device.

[0038] Figure 3A block diagram showing an example communication network 300 (e.g., a 5G mobile network 300) is presented. In the example communication network 300, location data 410a can be obtained by approximating the location of at least one wireless device 310 based on a first signal 323 and a second signal 333 received by the corresponding wireless device 310, where the first signal 323 and the second signal 333 each originate from separate radio access nodes 320, 330. In this case, the UE 310 can operate in a dual-connectivity / split architecture mode, whereby two radio access nodes 320, 330, namely the first radio access node 320 and the second radio access node 330, are simultaneously connected to the UE 310 via the air interface. The first radio access node 320 can include an eNB 321 and a first radio interface 322. The second radio access node 330 can include a gNB 331 and a second radio interface 332. In this case, a control plane CP 323 signal carrying control information facilitating the management of the connectivity of the UE 310 in the network can be transmitted between the UE 310 and the evolved packet core EPC 340. Between the EPC 340 and the distributed unit DU 324 of the eNB 321, the CP signal 323 can be sent via the S1 interface. Using the first radio interface 322, the CP signal 323 is then sent to and from the UE 310. A data plane DP 333 signal carrying data can also be transmitted between the EPC 340 and the UE 310. The DP 333 signal can be transmitted between the EPC 340 and the radio processing unit RPU 334 or the baseband processing unit BPU 334 of the gNB 331. The DP signal 333 can then be transmitted to and from the UE 310 via the second radio interface 332. In this example, the UE 310 receives two signals 323 and 333 from different directions and different radio access nodes 320, 330. The location of the UE 310 can then be estimated by measuring the power of the received radio signals 323 and 333 (e.g., by calculating and comparing the received radio signal strength indicator RSSI, which is a measure of the power present in the radio signals received by the UE 310, or by comparing the reference signal received power RSRP).

[0039] Using UE positioning (e.g., network-assisted GNSS mechanisms, downlink positioning, and enhanced cell ID mechanisms), location data 410a can also be obtained in the case of a 4G network with the participation of the mobility management entity MME, or in the case of a 5G network with the participation of the access and mobility management function AMF.

[0040] In step S202, obtaining network operation data 410 may further include obtaining performance data 410b. Performance data 410b may include wireless device specific data, such as the strength of the radio signal received by the UE, which is represented by RSSI 412. RSSI data 412 may be transmitted to the radio access node via the radio resource control RRC protocol. Performance data 410b may further include energy data 413 indicating the energy consumption or battery consumption of the wireless device. Performance data 410b may further include the data throughput 416 between the wireless device and the radio access node. The data throughput 416 may be measured in both the uplink UL and the downlink DL, and may be measured, for example, in bits per second.

[0041] Performance data 410b may further include radio access node specific data. This data may include the number 415 of wireless devices served by the radio access node. Data 410b may further include energy consumption data 414 indicating the energy consumption of the radio access node.

[0042] Method 200 may further include: in step S204, grouping at least one wireless device among the plurality of wireless devices by dividing the network operation data 410 into one or more service clusters 510, 520, 530, 540, where each service cluster 510, 520, 530, 540 includes a plurality of data samples of the network operation data 410 showing similarity across multiple dimensions, and where each service cluster defines a geographical area. The network operation data 410, which may include a set of observations as described above, may be input into a clustering algorithm 420 (such as the K-means algorithm), which is an example of an unsupervised machine learning technique. This step is further shown in Figure 4 and Figure 5 which. Based on the input 410, the K-means algorithm may establish or train a model 430. The input to the algorithm 420 may be a set of observations X = {x 1 …x n}. In some examples, each observation may be a 3D vector that has, as a first parameter, the location of the observation, such as a latitude and longitude tuple, or a set of latitude and longitude tuples indicating a geographical zone or region. Another parameter in the vector may be the observed signal strength, e.g., an RSSI or RSRP measurement, or the average of RSSI or RSRP measurements from a bounded location. Another parameter in the vector may be the average throughput in both the UL and DL directions.

[0043] To determine the number of clusters in a dataset, the Elbow method can be used. The Elbow method calculates the percentage variance or rate of change of the clusters as a function of the number of clusters. The minimum number of clusters that does not cause a significant change is selected as the number of selected clusters or the number of serving clusters. The set "S" of serving clusters is thus {S 1 …S k}, where k is the number of serving clusters.

[0044] The K-means clustering algorithm thus takes X, S as input and generates k clusters of all the observations in X. Initially, the K-means or "centroids" can be randomly generated. Subsequently, by calculating the Euclidean distance of each observation from each of the k means and mapping the observation to the mean with the minimum Euclidean distance, each observation in X can be associated with one of the k means. The means are then recalculated and become the "new" means. Then, the association of observations and the recalculation of the means are repeated until the algorithm converges. The K-means algorithm can be run periodically (e.g., daily, weekly, or monthly), or it can be triggered by an external entity, such as by a Network Operations Center NOC.

[0045] Figure 5 An example set of serving clusters 510, 520, 530, 540 is shown, each covering a set of samples or observations 511, 521, 531, 541 of network operation data 410. For example, the K-means algorithm 420 can be used to obtain the result partitioned into serving clusters 510, 520, 530, 540. Each serving cluster covers a corresponding set of observations. Since each observation includes the location data of a wireless device, the serving clusters 510, 520, 530, 540 each correspond to geographical regions 510, 520, 530, 540 that have boundaries linking the outermost observations in the corresponding clusters in such a way that each observation within the cluster is covered by the geographical region. In this way, the wireless devices being served by the radio access nodes can be grouped.

[0046] The method 200 may further include: in step S206, associating a quality of service level with each of the service clusters 510, 520, 530, 540. The quality of service level may indicate the interference experienced by the wireless devices belonging to the corresponding service cluster. The quality of service level may be a set of metrics that may correspond to the overall QoS or QoE experienced by the wireless devices covered by the corresponding service cluster. For example, a first data throughput range and a first RSSI range may be predefined for the wireless devices to indicate "high" interference. When the average vector of any service cluster indicates that the values of throughput and RSSI parameters fall within the first throughput range and the first RSSI range respectively, then that cluster may be labeled as a "high" interference cluster. Correspondingly, a second throughput range and a second RSSI range may be defined to indicate "low" interference. In this case, if the throughput and RSSI parameters of the average vector of any service cluster fall within the corresponding second throughput range and second RSSI range, then that service cluster may be labeled as a "low" interference cluster. Depending on the type of parameters in the observations or the type of network operation data being clustered, different labels may be envisioned for the clusters, which corresponds to grouping the wireless devices according to different quality of service metrics. For example, multiple labels and corresponding parameters may be defined. In some examples, the service cluster may include wireless devices whose battery energy consumption and radio access node energy consumption are both below a predetermined threshold.

[0047] Method 200 may further include: in step S208, scheduling network traffic between a radio access node and at least one wireless device depending on the position of the at least one wireless device relative to one or more service clusters 510, 520, 530, 540. Having information about the service clusters 510, 520, 530, 540 into which the wireless devices are divided may also allow triggering a set of rules for appropriate network actions considering the impact of the movement of the wireless devices relative to the service clusters 510, 520, 530, 540. Scheduling network traffic between a radio access node and at least one wireless device may be performed based on a decision tree model that organizes the set of rules in a tree structure. The decision tree model may be obtained by training a decision tree function based on network operation data and one or more service clusters using a second machine learning algorithm. Scheduling network traffic between a radio access node and at least one wireless device may also be performed according to a network traffic scheduling policy that depends on the position of the at least one wireless device relative to one or more service clusters 510, 520, 530, 540. The network traffic scheduling policy may indicate a quality of service level prioritized for the communication network. For example, a network operator may define a policy that optimizes a selected quality of service level for the communication network. For example, the policy may optimize power consumption. For example, the policy may also optimize the data throughput of the wireless device. In any example, the policy may organize the network traffic between the wireless device and the serving radio access node to obtain the result desired by the network operator.

[0048] Figure 6An example of network service scheduling is shown. In step S602, the location of a wireless device is monitored by, for example, periodically performing UE positioning or by directly obtaining the UE location from the device. The mobility pattern can be recorded or analyzed in real time and can indicate movement within the boundaries of a serving cluster. From a geographical perspective, the mobility pattern can also indicate that the serving cluster is about to change, i.e., indicate that the UE is approaching the edge of the current serving cluster. If the indication is that the serving cluster is not likely to change, then no further action is taken. Alternatively, if the mobility pattern indicates that the cluster is about to change, then in step S604, it is checked whether there is data available for transmission, such as network resource-intensive data. Network resource-intensive data can include, for example, high-definition (HD) video data, latency-sensitive data such as real-time video or audio calls. Information about the availability of data for transmission can be provided by the UE or by using a regression model in which the availability information can be estimated based on the previous traffic flow patterns of the respective UE. The availability information relates to data available for UL or DL transmission, i.e., data from the radio access node to the wireless device or from the wireless device to the radio access node. If there is no data to be transmitted in step S604, then there is no action to be performed. If there is data available for transmission, then in step S606, it is determined whether the network resource-intensive data can be transmitted before the wireless device switches to another cell. In some examples, the handover information can be obtained directly by using a mobility model for calculating the location, speed, and direction of movement of the UE, or indirectly by counting the number of handovers performed by the UE per unit of time. When the handover information indicates that the network resource-intensive data cannot be transmitted before the handover occurs, then the network service scheduling proceeds to step S612, in which the network resource-intensive data is buffered and its transmission is delayed. When step S606 indicates that the network resource-intensive data can be safely transmitted before the handover occurs, then the scheduling proceeds to step S608, in which it is determined whether the network data or the network resource-intensive data is mission-critical. For example, the data includes emergency data such as emergency calls or broadcasts. The communication network can also maintain a custom definition of data classified as mission-critical. Another example of an indication of the criticality of data available for transmission is the Quality Class Indicator (QCI). This parameter can be set for the current Packet Data Network (PDN) connection session of the UE. For example, a QCI value of 9 can mean that the network service is not critical or best-effort, while QCI values of 2 or 3 can mean that the network service is on a prioritized bearer. The value of the QCI can be retrieved from the core network, for example, from the Policy Control and Resource Function (PCRF) node of a given UE. If the QCI is below 9, this indicates that the data is critical and should be received and / or sent as soon as possible. In this case, the scheduling proceeds to step S614, in which the initiation of data transmission is accelerated.Alternatively, when the data to be transmitted is classified as non-critical data, step S610 is performed. In this step, when the UE is in the first service cluster and is about to move to the second service cluster, the quality of service levels of the two clusters are compared. For example, when the first service cluster is a "low" interference cluster and the second service cluster is a "high" interference cluster, step S614 is initiated by accelerating the transmission of data so that network resource-intensive data is transmitted while the UE is still in the cluster with better performance. Advantageously, this results in maintaining the QoS by maximizing the likelihood of successful data transmission. Another advantage is reducing the energy consumption in the network by avoiding excessive radio signal transmission power increase on both the UE side and the radio access point side.

[0049] Figure 7 An example of communication in network 700 during a specific example implementation of a method for controlling traffic is shown. In step S702, the radio access network RAN 710 may send information about a handover of a specific UE 740 to the baseband unit BBU 720 of the eNB, which may be the first radio access node serving the UE 740. It is emphasized that there may be more than one UE involved in the process. In step S704, the RAN 710 notifies the BBU 720 of the eNB that there is network traffic data available for transmission between the UE and the radio access node. Then in step S706, the data to be sent to the UE 740 is sent to the eNB radio interface 730 together with instructions for measuring the power usage, interference, and location or positioning of the UE 740. Then in step S708, the data is sent to the UE together with the instructions of step S706, which then responds in step S710 by sending back to the eNB radio interface 730 control information such as channel state information reference signal CSI-RS used by the UE to estimate the channel and report channel quality information CQI to the eNB, and radio signal information such as RSSI measured at the UE 740. Then, as part of step S712, the data collected from step S710 is fed back to the BBU 720 of the eNB and is thus processed in step S714. The process defined by steps S702 to S712 may be repeatedly executed.

[0050] In step S716, the task of the BBU 720 of the eNB is to identify high interference areas. The BBU 720 of the eNB then sends a control plane CP signal to the RAN 710 in step S718 to request information from an adjacent site, which may be a second radio access point. In step S720, the RAN 710 instructs the radio processing unit RPU 750 of the second radio access point gNB to collect information about a specific UE 740. When the RPU 750 of the gNB has processed the request in step S722, a control signal is sent to the gNB radio interface 760 in step S724, which replies in step S726 with information about the measured interference and battery power usage based on the calculated location of the UE 740. Then, the RPU 750 of the gNB forwards the collected data back to the RAN 710 in step S728. The process defined by steps S716 to S728 can be repeatedly executed.

[0051] In step S730, the RAN 730 sends the collected data to a machine learning ML component 770, which may be a clustering function such as the K-means algorithm described previously. In step S732, the ML component 770 specifies service clusters based on a sample set of the acquisition data collected from the previous step. The ML component 770 may specify a service cluster suitable for data transmission and a service cluster suitable for voice transmission in step S734. In some examples, due to the presence of a high level of interference in the service cluster, the service cluster for voice transmission may correspond to a network traffic scheduling policy that only allows mission-critical traffic in these clusters. The service cluster for data transmission may correspond to a low-interference service cluster that can accommodate network resource-intensive data or traffic without significantly increasing the interference in the service cluster. In step S736, the machine learning component 770 may assign multiple quality of service levels in order to obtain multiple service clusters related to different characteristics of the quality of service. For example, a service cluster including wireless devices showing low battery consumption may be created. Different service clusters including wireless devices showing high data throughput, etc., may be created. Then different network traffic scheduling policies can be created and used, which allow for complex processing of network traffic based on defined priorities, goals, or policies. Then, in step S738, information about the created service clusters is available to the RAN 710. Steps S730 to S738 can be executed periodically (e.g., daily, weekly, or monthly), or can be triggered by an external entity, such as by the NOC.

[0052] In step S740, information regarding different service clusters is propagated from the RAN 710 to the BBU 720 of the eNB, and in step S742, from the RAN 710 to the RPU 750 of the gNB. Then, in steps S744 and S746 respectively, the BBU 720 of the eNB and the RPU 750 of the gNB locally store the received data regarding the service clusters. Steps S740 to S746 can be repeatedly executed according to the specific requirements and configurations of the communication network.

[0053] In steps S748 to S758, an example network traffic scheduling policy is shown. In step S748, as a result of the UE 740 being located in a service cluster assigned a high interference quality of service level, the BBU 720 of the eNB collaborates with the eNB radio interface 730 to decide to buffer the data available for transmission between the UE 740 and the eNB. When the UE 740 changes its location indicating that the current service cluster changes to a service cluster assigned a low interference quality of service level, then the eNB radio interface 730 decides in step S750 to accelerate the transmission of the data available for transmission between the UE 740 and the eNB. A similar process can be performed when considering the gNB, as shown in steps S752 and S754. The UE 740 can then feedback information to the eNB or gNB in steps S756 and S758, which can then be used, for example, to recalculate or update the service cluster.

[0054] Figure 8 is a schematic diagram of an example of a network node 800 of a communication network. In some embodiments, the network node 800 can be an electronic device communicatively connected to other electronic devices on the network (e.g., other network devices, UEs, radio base stations, etc.). In certain embodiments, the network node 800 can include radio access features that provide radio network access to other electronic devices such as UEs (e.g., a "radio access network device" can refer to such a network device). For example, the network node 800 can be a base station, such as a gNodeB in 5G, an eNodeB in Long Term Evolution LTE, a NodeB in Wideband Code Division Multiple Access WCDMA, or other types of base stations, as well as a Radio Network Controller RNC, a Base Station Controller BSC, or other types of control nodes. As Figure 8 shown, the example network node 800 includes a processing circuit or processor 804, a memory 806, an interface 802, and may also include an antenna. These components can work together to provide various network device functions as disclosed herein.

[0055] The processing circuit 804 can be a microprocessor, a controller, a microcontroller, a central processing unit, a digital signal processor, an application specific integrated circuit, a field programmable gate array, any other type of electronic circuit, or any combination of one or more of the foregoing. The processor 804 can include one or more processor cores. In certain embodiments, some or all of the functionality described herein as being provided by the network node 800 can be implemented by the processor 804 executing software instructions alone or in combination with other network node 800 components (e.g., the memory 806).

[0056] The memory 806 can use non-transitory machine-readable (e.g., computer-readable) media to store code (which consists of software instructions and is sometimes referred to as computer program code or a computer program) and / or data, the non-transitory machine-readable media being, for example, machine-readable storage media (e.g., magnetic disks, optical disks, solid state drives, read only memory (ROM), flash devices, phase change memory) and machine-readable transmission media (e.g., electrical, optical, radio, acoustic or other forms of propagated signals, such as carrier waves, infrared signals). For example, the memory 806 can include non-volatile memory containing code to be executed by the processor 804. Where the memory 806 is non-volatile, the code and / or data stored therein can persist even when the network device is powered off (when power is removed). In some instances, when the network node 800 is powered on, the portion of the code to be executed by the processor 804 can be copied from the non-volatile memory to volatile memory (e.g., dynamic random access memory DRAM, static random access memory SRAM) of the network node 800.

[0057] Interface 802 can be used for wired or wireless communication to send and / or receive signaling and / or data to / from network device 800. For example, interface 802 can perform any formatting, encoding, or translation to allow network node 800 to send and receive data over wired and / or wireless connections. In some embodiments, interface 802 can include radio circuitry capable of receiving data from other electronic devices in the network via a wireless connection and / or sending data out to other devices via a wireless connection. The radio circuitry can include a transmitter, a receiver, and / or a transceiver suitable for radio frequency communication. The radio circuitry can convert digital data into radio signals with appropriate parameters (e.g., frequency, timing, channel, bandwidth, etc.). Then, the radio signals can be transmitted via an antenna to an appropriate recipient. In some embodiments, interface 802 can include a network interface controller NIC (also known as a network interface card), a network adapter, a local area network, LAN adapter, or a physical network interface. The NIC can facilitate connecting network node 800 to other devices, allowing them to communicate wired by plugging a cable into a physical port connected to the NIC. As described above, in certain embodiments, processor 804 can represent a part of interface 802, and some or all of the functions described as provided by interface X103 can be more specifically provided by processor 804.

[0058] For the reason of simplifying the description of certain aspects and features of network node 800 disclosed herein, the components of network node 800 are all depicted as multiple separate boxes located within a single larger box. However, in practice, one or more of the components shown in example network node 800 can include multiple different physical elements (e.g., interface 802 can include terminals for coupling the wiring for wired connections to a radio transceiver for wireless connections).

[0059] Although the modules are shown to be implemented as software stored in memory 806, other embodiments implement some or all of each of these modules in hardware.

[0060] Network node 800 can include a wireless access node 800. In some examples, the network node can include a baseband unit BBU 800, or an E-nodeB (eNB), or a next-generation node B (gNB). In another example, network node 800 can include a virtual node 800 such as a virtual BBU 800.

[0061] BBU 730 can be a unit that processes the baseband in a communication system. The wireless access node can include a BBU and a radio frequency RF processing unit or a remote radio unit RRU. The BBU can be placed in an equipment room and connected to the RRU via an optical fiber. For example, the BBU can be responsible for communicating via physical interface 802.

[0062] The BBU in a cellular phone cell site may include a digital signal processor (DSP) that processes a forward voice signal for transmission to a mobile unit and processes a reverse voice signal received from the mobile unit.

[0063] Although the interface 802, the processing circuit 804, and the memory 806 are shown as being connected in series, they may alternatively be interconnected in any other manner (e.g., via a bus).

[0064] In one example, the memory 806, which may include a non-transitory computer-readable medium 806, contains instructions such as a computer program executable by the processing circuit 804, such that the network node 800 can be used to obtain network operation data of at least one wireless device among a plurality of wireless devices, the network operation data including the location of the corresponding wireless device and performance data associated with the location, and the network operation data indicating a plurality of attributes corresponding to the operation of the corresponding wireless device in a communication network. The processing circuit 804 also causes the network node 800 to group at least one wireless device among the plurality of wireless devices by dividing the network operation data into one or more service clusters, where each service cluster defines a geographical area based on the obtained location. The processing circuit 804 also causes the network node 800 to schedule network traffic between a wireless access node and at least one wireless device depending on the location of the at least one wireless device relative to the one or more service clusters.

[0065] In some examples, the network node 800 or the BBU / RPU 720, BBU / RPU 750 of the network node 800 can be used to associate a quality of service level with one or more service clusters. The quality of service level can indicate the interference or energy consumption experienced by the radio devices belonging to the corresponding service cluster or the radio access nodes serving the radio devices belonging to the corresponding service cluster. In some examples, the BBU or RPU 720, BBU or RPU 750 of the network node 800 can be used to schedule network traffic between a radio access node and at least one radio device according to a network traffic scheduling policy indicating the quality of service level prioritized for the communication network. For example, the network traffic scheduling policy can be generated or stored in the BBU or RPU 720, BBU or RPU 750, or can be generated in the core network and uploaded to the BBU or RPU. One or more service clusters 510, 520, 530, 540 can include data samples showing similarities of network operation data 410 across multiple dimensions. The data samples can be obtained or collected by the BBU or RPU 720, BBU or RPU 750 from the radio devices. The data samples can also be received by a virtual BBU in the core network, where the data samples have been forwarded from an eNB or gNB that can be used to divide the network operation data into one or more service clusters by training a clustering function using a first machine learning algorithm based on the network operation data. The performance data 410b can include at least one of the following: received radio signal quality data 412 (e.g., RSSI or RSRP measured at the corresponding radio device and sent to the network node 800 via, for example, the RRC protocol), data throughput 416, and battery data 413 indicating the energy consumption of the corresponding radio device. The performance data 410b can also include at least one of the following: the number of radio devices served by the radio access node and energy consumption data indicating the energy consumption of the radio access node. In some examples, the network node 800 and specifically the BBU or RPU 720, BBU or RPU 750 of the network node 800 can be used to schedule network traffic between a radio access node and at least one radio device by training a decision tree function using a second machine learning algorithm based on the network operation data 410 and one or more service clusters 510, 520, 530, 540. In some examples, the network node 800 or the BBU / RPU 720, BBU / RPU 750 of the network node 800 can be used to: when at least one radio device belongs to a first service cluster of a first quality of service level, such as a low interference cluster, a low power consumption cluster, a high throughput cluster, schedule network traffic between a radio access node and at least one radio device by initiating the acceleration of the transmission of network resource-intensive data, such as the transmission of voice / data.In some examples, when at least one wireless device belongs to a second service cluster of a second quality of service level, such as a high interference cluster, a high power consumption cluster, or a low throughput cluster, scheduling services may include, for example, buffering network resource-intensive data at the BBU / RPU 720, BBU / RPU 750; and wherein, the first quality of service level is higher than the second quality of service level. When at least one wireless device belongs to the second service cluster, the network node 800 or the BBU / RPU of the network node 800 may, for example, prioritize mission-critical network services. For example, in some examples, when a wireless device belongs to a low quality of service level cluster, the BBU / RPU 720, BBU / RPU 750 may buffer and block the transmission of network resource-intensive data, except for mission-critical services such as emergency calls or messages.

[0066] The network nodes 320, 330, 720, 750 can be used to obtain the location of at least one wireless device by approximating the location of at least one wireless device based on a first signal such as a control plane reference signal and a second signal such as a data plane reference signal received by the corresponding wireless device, wherein the first signal and the second signal each originate from a separate radio access node 320, 330, 720, 750. In other examples, the network node 800 can be used to obtain the location of at least one wireless device by approximating the location of at least one wireless device based on the elevation angle and azimuth angle of a radio beam transmitted from a radio access node to the corresponding wireless device.

[0067] Reference Figure 9 , according to an example, a communication system includes: a telecommunication network 3210 (e.g., a 3GPP type cellular network), which includes an access network 3211 (e.g., a radio access network) and a core network 3214. The access network 3211 includes a plurality of base stations 3212a, 3212b, 3212c, such as NB, eNB, gNB, or other types of radio access points or nodes, such as network nodes 110, 120, 320, 330, 720, 750, 800, and each base station defines a corresponding coverage area 3213a, 3213b, 3213c. Each base station 3212a, 3212b, 3212c is connected to the core network 3214 through a wired or wireless connection 3215. A first wireless device or UE 3291 located in the coverage area 3213c is configured to be wirelessly connected to the corresponding base station 3212c or paged by the corresponding base station 3212c. A second UE 3292 in the coverage area 3213a is wirelessly connected to the corresponding base station 3212a. Although a plurality of UEs 3291, 3292 are shown in this example, the disclosed embodiments are equally applicable to the case where a single UE is located in the coverage area or a single UE is connected to the corresponding base station 3212.

[0068] The telecommunications network 3210 is itself connected to a host computer 3230 which may be embodied in the hardware and / or software of a stand-alone server, a cloud-implemented server, a distributed server, or as processing resources in a server farm. The host computer 3230 may be owned or under the control of a service provider, or may be operated by or on behalf of a service provider. The connections 3221, 3222 between the telecommunications network 3210 and the host computer 3230 may extend directly from the core network 3214 to the host computer 3230, or may pass through an optional intermediate network 3220. The intermediate network 3220 may be one or more combinations of a public network, a private network, or a serving network; the intermediate network 3220 (if any) may be a backbone network or the Internet; specifically, the intermediate network 3220 may include two or more sub-networks (not shown).

[0069] Figure 9 The communication system as a whole, realizes connectivity between one of the connected wireless devices or UEs 3291, 3292 and the host computer 3230. This connection may be described as an over-the-top (OTT) connection 3250. The host computer 3230 and the connected UEs 3291, 3292 are configured to transmit data and / or signaling via the OTT connection 3250 using the access network 3211, the core network 3214, any intermediate network 3220, and possibly other intermediate infrastructure (not shown). The participating communication devices through which the OTT connection 3250 passes are not aware of the routing of the uplink and downlink communications, and in this sense, the OTT connection 3250 may be transparent. For example, a base station 3212 such as a network node 800 or a radio access node may not be informed or need not be informed about the past routing of incoming downlink communications having data originating from the host computer 3230 and destined to be forwarded (e.g., handed over) to the connected UE 3291. Similarly, the base station 3212 need not know the future routing of uplink communications originating from the UE 3291 and destined for the output to the host computer 3230.

[0070] Reference will now be made to Figure 10Describe an example implementation of a UE, a base station, and a host computer according to an embodiment discussed in the above paragraph. In a communication system 3300, a host computer 3310 includes hardware 3315, and the hardware 3315 includes a communication interface 3316 configured to establish and maintain a wired or wireless connection with an interface of different communication devices of the communication system 3300. The host computer 3310 further includes a processing circuit 3318, which may have storage and / or processing capabilities. Specifically, the processing circuit 3318 may include one or more programmable processors suitable for executing instructions, application specific integrated circuits, field programmable gate arrays, or a combination of such devices (not shown). The host computer 3310 also includes software 3311, which is stored in or accessible by the host computer 3310 and can be executed by the processing circuit 3318. The software 3311 includes a host application 3312. The host application 3312 can be operated to provide services to a remote user, such as a UE 3330 connected via an OTT connection 3350, and the OTT connection 3350 terminates at the UE 3330 and the host computer 3310. When providing services to a remote user, the host application 3312 can provide user data transmitted using the OTT connection 3350.

[0071] The communication system 3300 further includes a base station 3320 disposed in a telecommunication system, and the base station 3320 includes hardware 3325 enabling it to communicate with the host computer 3310 and the UE 3330. The hardware 3325 may include: a communication interface 3326 for establishing and maintaining a wired or wireless connection with an interface of different communication devices of the communication system 3300; and a radio interface 3327 for at least establishing and maintaining a wireless connection 3370 with the UE 3330, and the UE 330 is located in a coverage area ( Figure 10 not shown) served by the base station 3320. The communication interface 3326 may be configured to facilitate the connection 3360 with the host computer 3310. The connection 3360 may be direct, or it may pass through the core network of the telecommunication system ( Figure 10 not shown) and / or through one or more intermediate networks located outside the telecommunication system. In the illustrated embodiment, the hardware 3325 of the base station 3320 further includes a processing circuit 3328, and the processing circuit 3328 may include one or more programmable processors suitable for executing instructions, application specific integrated circuits, field programmable gate arrays, or a combination thereof (not shown). The base station 3320 also has internal storage or software 3321 accessible via an external connection.

[0072] The communication system 3300 also includes the UE 3330 that has been mentioned. The hardware 3335 of the UE 3330 may include a radio interface 3337, which is configured to establish and maintain a wireless connection 3370 with a base station serving the coverage area where the UE 3330 is currently located. The hardware 3335 of the UE 3330 also includes processing circuitry 3338, which may include one or more programmable processors suitable for executing instructions, application-specific integrated circuits, field-programmable gate arrays, or a combination of such devices (not shown). The UE 3330 also includes software 3331, which is stored in or accessible by the UE 3330 and can be executed by the processing circuitry 3338. The software 3331 includes a client application 3332. The client application 3332 can be operated to provide services to a human or non-human user via the UE 3330 with the support of the host computer 3310. In the host computer 3310, the executing host application 3312 can communicate with the executing client application 3332 via the OTT connection 3350, which terminates at the UE 3330 and the host computer 3310. When providing services to the user, the client application 3332 can receive request data from the host application 3312 and provide user data in response to the request data. The OTT connection 3350 can transmit both the request data and the user data. The client application 3332 can interact with the user to generate the user data it provides.

[0073] It should be noted that the host computer 3310, the base station 3320, and the UE 3330 shown in Figure 10 may be equivalent to the host computer 3230, one of the base stations 3212a, 3212b, 3212c, and one of the UEs 3291, 3292 in Figure 9 respectively. That is to say, the internal working modes of these entities can be as shown in Figure 10 and, independently, the surrounding network topology can be the network topology of Figure 9 .

[0074] In Figure 10 , the OTT connection 3350 is abstractly depicted to illustrate the communication between the host computer 3310 and the user equipment 3330 via the base station 3320, without explicitly referring to any intermediate devices and the exact routing of the messages via these devices. The network infrastructure can determine the routing, which can be configured to be hidden from the UE 3330 or the service provider operating the host computer 3310 or both. When the OTT connection 3350 is active, the network infrastructure can further make decisions to dynamically change the routing (e.g., based on load balancing considerations or network reconfiguration).

[0075] The wireless connection 3370 between the UE 3330 and the base station 3320 is consistent with the teachings of the embodiments described throughout this disclosure. One or more of the various embodiments improve the performance of the OTT services provided to the UE 3330 using the OTT connection 3350, in which the wireless connection 3370 forms the final part. More precisely, the teachings of these embodiments can improve data rate, latency, power consumption, and thus benefits such as reduced user waiting time, better responsiveness, and extended battery life.

[0076] A measurement process can be provided for monitoring data rate, latency, and other factors that are the object of improvement in one or more embodiments. There can also be optional network functions for reconfiguring the OTT connection 3350 between the host computer 3310 and the UE 3330 in response to changes in the measurement results. The measurement process and / or the network functions for reconfiguring the OTT connection 3350 can be implemented in the software 3311 of the host computer 3310 or in the software 3331 of the UE 3330 or in both. In an embodiment, sensors (not shown) can be deployed in or associated with the communication devices through which the OTT connection 3350 passes; the sensors can participate in the measurement process by providing values of the monitored quantities exemplified above, or other physical quantities from which the software 3311, 3331 can calculate or estimate the monitored quantities. The reconfiguration of the OTT connection 3350 can include message format, retransmission settings, preferred routing, etc.; the reconfiguration does not need to affect the base station 3320, and the base station 3320 may be unaware or unperceivable of this. Such processes and functions can be known and practiced in the art. In certain embodiments, the measurement can involve proprietary UE signaling that facilitates the host computer 3310's measurement of throughput, propagation time, latency, etc. The measurement can be achieved by the software 3311, 3331 sending messages (especially empty messages or "virtual" messages) using the OTT connection 3350 while monitoring propagation time, errors, etc.

[0077] Figure 11 is a flowchart showing a method implemented in a communication system according to an embodiment. The communication system includes: a host computer, a base station, and a UE, which can be those host computers, base stations, and UEs described with reference to Figure 9 and Figure 10 described. To simplify this disclosure, only Figure 11. In a first step 3410 of the method, a host computer provides user data. In an optional sub-step 3411 of the first step 3410, the host computer provides the user data by executing a host application. In a second step 3420, the host computer initiates a transmission to the UE, the transmission carrying the user data. In an optional third step 3430, in accordance with the teachings of the embodiments described throughout the present disclosure, the base station sends the user data carried in the transmission initiated by the host computer to the UE. In an optional fourth step 3440, the UE executes a client application, the client application being associated with the host application executed by the host computer.

[0078] Figure 12 is a flow chart showing a method implemented in a communication system according to an embodiment. The communication system includes: a host computer, a base station and a UE, which can be a reference Figure 9 and Figure 10 In order to simplify the present disclosure, only the host computers, base stations and UEs described herein will be included in this section. Figure 12 . In a first step 3510 of the method, a host computer provides user data. In an optional sub-step (not shown), the host computer provides the user data by executing a host application. In a second step 3520, the host computer initiates a transmission to the UE, the transmission carrying the user data. According to the teachings of the embodiments described throughout the present disclosure, the transmission can be delivered via a base station. In an optional third step 3530, the UE receives the user data carried in the transmission.

[0079] It should be noted that the above examples illustrate rather than limit the present invention, and those skilled in the art will be able to design many alternative examples without departing from the scope of the attached claims. The word "comprising" does not exclude the presence of elements or steps other than those listed in the claims, and "one" or "an" does not exclude multiple, and a single processor or other unit can perform the functions of several units recorded in the following claims. In the case of using the terms "first", "second", etc., they should only be understood as labels for convenient identification of specific features. Specifically, unless otherwise explicitly stated, they should not be interpreted as describing the first or second features of multiple such features (i.e., the first or second features occurring in time or space among these features). Unless otherwise stated, the steps in the method disclosed herein can be performed in any order. Any figure marks in the statement should not be understood as limiting its scope.

Claims

1. A method for controlling traffic in a communication network, performed by network nodes (110, 120, 320, 330, 720, 750, 800), the communication network including radio access nodes adapted to serve a plurality of wireless devices, the method comprises the following steps: obtaining (S202) network operation data (410) of at least one wireless device among the plurality of wireless devices, the network operation data including the location (410a) of the corresponding wireless device and performance data (410b) related to the location, the network operation data indicating a plurality of attributes corresponding to the operation of the corresponding wireless device in the communication network; grouping (S204) at least one wireless device among the plurality of wireless devices by dividing the network operation data into one or more service clusters (510, 520, 530, 540), wherein each service cluster defines a geographical area based on the obtained location; associating a quality of service level with the one or more service clusters (S206); scheduling (S208) network traffic between the radio access node and the at least one wireless device depending on the location of the at least one wireless device relative to the one or more service clusters, wherein scheduling network traffic between the radio access node and the at least one wireless device includes: when the at least one wireless device belongs to a first service cluster of a first quality of service level, accelerating (S614) the initiation of transmission of network resource-intensive data; when the at least one wireless device belongs to a second service cluster of a second quality of service level, buffering (S612) network resource-intensive data; and wherein the first quality of service level is higher than the second quality of service level.

2. The method according to claim 1, wherein, the network nodes include the radio access nodes (110, 120, 320, 330, 720, 750).

3. The method according to claim 2, wherein, the quality of service level indicates the interference experienced by wireless devices belonging to the corresponding service cluster.

4. The method according to any one of claims 1 to 3, comprises: scheduling (S208) network traffic between the radio access node and the at least one wireless device according to a network traffic scheduling policy indicating a quality of service level prioritized for the communication network.

5. The method according to any one of claims 1 to 3, wherein, the one or more service clusters include data samples of the network operation data showing similarity across multiple dimensions.

6. The method according to claim 5, wherein, dividing the network operation data into one or more service clusters includes: training a clustering function (430) using a first machine learning algorithm (420) based on the network operation data.

7. The method according to any one of claims 1 to 3, wherein, The performance data includes at least one of the following: received radio signal quality data (412) measured at the corresponding wireless device, data throughput (416), and battery data (413) indicating the energy consumption of the corresponding wireless device.

8. The method according to any one of claims 1 to 3, wherein, the performance data further includes at least one of the following: the number of wireless devices served by the wireless access node (415) and energy consumption data (414) indicating the energy consumption of the wireless access node.

9. The method according to any one of claims 1 to 3, wherein, scheduling network traffic between the wireless access node and the at least one wireless device includes: using a second machine learning algorithm to train a decision tree function based on the network operation data and the one or more service clusters.

10. The method according to any one of claims 1 to 3, wherein, scheduling network traffic between the wireless access node and the at least one wireless device includes: when the at least one wireless device belongs to the second service cluster, prioritizing mission-critical network traffic (S608, S614).

11. The method according to claim 10, further including: obtaining (S608) a quality class identifier QCI of the at least one wireless device, wherein the mission-critical service is determined based on the value of the QCI.

12. The method according to any one of claims 1 to 3, wherein, the performance data includes handover data indicating the likelihood of handover of the at least one wireless device, and wherein scheduling network traffic between the wireless access node and the at least one wireless device includes: buffering (S606, S612) network data when the handover data indicates a possible handover of the corresponding wireless device.

13. The method according to any one of claims 1 to 3, wherein, obtaining the network operation data includes: approximating the location of the at least one wireless device based on a first signal and a second signal received by the corresponding wireless device, wherein the first signal and the second signal each originate from a separate wireless access node.

14. The method according to any one of claims 1 to 3, wherein, obtaining the network operation data includes: calculating the location by means of a GNSS receiver of the at least one wireless device.

15. The method according to any one of claims 1 to 3, wherein, obtaining the network operation data includes: approximating the location of the at least one wireless device based on the elevation angle and azimuth angle of a radio beam transmitted from the wireless access node to the corresponding wireless device.

16. A network node (110, 120, 320, 330, 720, 750, 800) of a communication network, the communication network including a wireless access node adapted to serve a plurality of wireless devices, the network node including a processing circuit (804) and a memory (806) containing instructions executable by the processing circuit, whereby the network node is operative to: Obtain network operation data of at least one wireless device among the multiple wireless devices, where the network operation data includes the location of the corresponding wireless device and performance data related to the location, and the network operation data indicates multiple attributes corresponding to the operation of the corresponding wireless device in the communication network; Group at least one wireless device among the multiple wireless devices by dividing the network operation data into one or more service clusters, where each service cluster defines a geographical area based on the obtained location; Associate a quality of service level with the one or more service clusters; Schedule network traffic between the wireless access node and the at least one wireless device depending on the location of the at least one wireless device relative to the one or more service clusters, where the network node is operative to schedule network traffic between the wireless access node and the at least one wireless device by: When the at least one wireless device belongs to a first service cluster of a first quality of service level, accelerating the initiation of the transmission of network resource-intensive data; When the at least one wireless device belongs to a second service cluster of a second quality of service level, buffering network resource-intensive data; and where the first quality of service level is higher than the second quality of service level.

17. The network node according to claim 16, wherein, The quality of service level indicates the interference experienced by the wireless devices belonging to the corresponding service cluster.

18. The network node according to any one of claims 16 to 17, operative to schedule network traffic between the wireless access node and the at least one wireless device according to a network traffic scheduling policy indicating a quality of service level prioritized for the communication network.

19. The network node according to any one of claims 16 to 17, wherein, The one or more service clusters include data samples showing similarity of the network operation data across multiple dimensions.

20. The network node according to claim 19, operative to divide the network operation data into one or more service clusters by training a clustering function using a first machine learning algorithm based on the network operation data.

21. The network node according to any one of claims 16 to 17, wherein, The performance data includes at least one of the following: received radio signal quality data measured at the corresponding wireless device, data throughput, and battery data indicating the energy consumption of the corresponding wireless device.

22. The network node according to any one of claims 16 to 17, wherein, The performance data further includes at least one of the following: the number of wireless devices served by the wireless access node and energy consumption data indicating the energy consumption of the wireless access node.

23. The network node according to any one of claims 16 to 17, operative to schedule network traffic between the wireless access node and the at least one wireless device by training a decision tree function using a second machine learning algorithm based on the network operation data and the one or more service clusters.

24. The network node according to claim 16, operative to schedule network traffic between the radio access node and the at least one wireless device by: When the at least one wireless device belongs to the second service cluster, giving priority to mission-critical network traffic.

25. The network node according to claim 24, operative to obtain a Quality of Service Class Identifier (QCI) of the at least one wireless device, wherein, the mission-critical traffic is determined based on the value of the QCI.

26. The network node according to any one of claims 16 to 17, wherein, the network operation data includes handover data indicating the likelihood of handover of the at least one wireless device, and wherein the network node is operative to schedule network traffic between the radio access node and the at least one wireless device by buffering network data when the handover data indicates a possible handover of a corresponding wireless device.

27. The network node according to any one of claims 16 to 17, operative to obtain the location of the at least one wireless device by approximating the location of the at least one wireless device based on a first signal and a second signal received by the corresponding wireless device, wherein the first signal and the second signal each originate from a separate radio access node.

28. The network node according to any one of claims 16 to 17, operative to obtain the location of the at least one wireless device by calculating the location using a GNSS receiver of the at least one wireless device.

29. The network node according to any one of claims 16 to 17, operative to obtain the location of the at least one wireless device by approximating the location of the at least one wireless device based on the elevation angle and azimuth angle of a radio beam transmitted from the radio access node to the corresponding wireless device.

30. The network node (110, 120, 320, 330, 720, 750, 800) according to any one of claims 16 to 17, wherein, the network node is a radio access node.

31. A computer program product for controlling traffic in a communication network, the computer program product comprising computer code which, when run on a processing circuit of a network node, causes the network node to: Obtain network operation data of at least one wireless device among a plurality of wireless devices, the network operation data including the location of the corresponding wireless device and performance data associated with the location, the network operation data indicating a plurality of attributes corresponding to the operation of the corresponding wireless device in the communication network; Group at least one wireless device among the plurality of wireless devices by partitioning the network operation data into one or more service clusters, wherein each service cluster defines a geographical area based on the obtained location; Associate a quality of service level with the one or more service clusters; Schedule network traffic between the radio access node and the at least one wireless device depending on the location of the at least one wireless device relative to the one or more service clusters, wherein the network node schedules network traffic between the wireless access node and the at least one wireless device in the following manner: when the at least one wireless device belongs to a first service cluster of a first quality of service level, accelerating the initiation of the transmission of network resource-intensive data; when the at least one wireless device belongs to a second service cluster of a second quality of service level, buffering network resource-intensive data; and wherein the first quality of service level is higher than the second quality of service level.

32. A communication network, the network comprising: a first wireless access node and a second wireless access node, wherein the first wireless access node and the second wireless access node are adapted to serve a plurality of wireless devices; a computer program product according to claim 31, the computer program product comprising computer code which, when run on the processing circuitry of the first wireless access node, causes the first wireless access node and the second wireless access node to approximate the location of the at least one wireless device based on a first signal and a second signal received by the respective wireless devices, wherein the first signal originates from the first wireless access node and the second signal originates from the second wireless access node.

Images