Method for controlling load distribution in a wireless network having multiple access points

Abstract

The invention relates to a method for load balancing in a communication network, the method comprising determining (310) a load level of a first access point (AP1) of the communication network based on at least parameters of services provided by the access point to terminal devices (D1, D2, D3); and preparing (350, 360) a handover of at least one service to a second access point of one or more neighboring access points (AP2, AP3) if the load level of the first access point (AP1) exceeds a predefined first threshold (320), wherein the coverage area (120, 130) of the neighboring access points (AP2, AP3) at least partially overlaps with the coverage area (110) of the first access point (AP1) and thus forms an overlap area (140).

Description

Technical Field

[0001] The present invention relates to a method for load balancing in a communication network, and a computer program product and computing unit for performing the method. Background Technology

[0002] Wireless networks typically consist of multiple access points (Zugangspunkt) over a wide area, through which terminal devices obtain access to the network. This can refer, for example, to a WLAN network or a mobile radio network, such as one based on LTE or 5G standards. Because each access point has only a limited capacity (Kapazität), such as a limited capacity in terms of the bandwidth provided for these connections, situations may arise where the access point reaches its maximum capacity and can no longer accept other connection or service requests. This may result in, for example, newly arriving devices in the area being unable to establish a connection or experiencing connection interruptions.

[0003] Each access point has a defined coverage area or radius within which terminal devices can establish a reliable wireless connection to provide service. This area is also called a radio cell or cell. Typically, multiple access points are arranged in a network to create different overlapping areas within which terminal devices are located in the coverage areas of two or more access points. Therefore, sufficient coverage can be achieved for terminal devices moving from one area to another with uninterrupted connectivity. In different cases, the active connection is transferred from one access point to another, described using the terms handover or handoff. Different variations are possible, such as in a soft handover where the connection to the first access point is maintained until the connection to the second access point is fully established, significantly improving network connectivity. However, handover can also be configured as a hard handover without parallel connections to two access points. Similarly, it is also possible to change the radio cell (i.e., the responsible access point) for terminal devices that are logged into a cell but do not have an active connection with the access point (cell reselection).

[0004] The switching of active or inactive terminal devices to another access point is common across various radio standards, such as mobile radio networks according to 3GPP standards like GSM, UMTS, LTE, and 5G, but also in WLAN networks according to IEEE 802.11, where it is often referred to as roaming. However, the exact configuration and other details of the implementation scheme for the necessary steps of connection handover or cell replacement for the network elements being switched depend on the individual standards and will not be discussed further here. Furthermore, even the handover of connections between networks of different standards or different radio technologies (e.g., from 4G to 5G) is possible. The difficulties described regarding overloaded access points can arise in any wireless network. Summary of the Invention

[0005] According to the present invention, a method for load balancing in a communication network, having the features of the independent patent claims, and a computer program and computing unit for performing the method are proposed. Advantageous configurations are the subject of the dependent claims and the following description.

[0006] In particular, a method for load balancing in a communication network is proposed here, wherein the load level (Lastniveau) of a first access point of the communication network is first determined. This load level is determined at least based on parameters of the services provided by the access point to terminal devices. Here, for example, each telecommunications service according to the 3GPP standard can be understood as a service. If the load level of the first access point is above a pre-given first threshold, then at least one service is prepared to be transferred to a suitable second access point, wherein the second access point is one or more adjacent access points whose coverage areas at least partially overlap with the coverage area of ​​the first access point, thus forming an overlapping area. Therefore, the load can be distributed more evenly within the overlapping coverage area.

[0007] Preferably, the load level can be determined as a normalized load level, i.e., a quotient consisting of the absolute load of the first access point and the maximum load of the first access point. A normalized load metric based on service-based information can be advantageous, for example, for time-sensitive communications within the network.

[0008] To perform the service handover, the method may further include receiving a message containing information about the load levels of one or more adjacent access points. This message may be sent, for example, by the central unit of the network, or by each access point at regular intervals to its adjacent access points.

[0009] In one exemplary embodiment, the method may further include: detecting mobility data of at least one terminal device, wherein the mobility data indicates one or more future locations of the terminal device within one or more predetermined time periods. Based on the detected mobility data for the at least one terminal device and parameters based on the services that the access point should expect to provide to the at least one terminal device at a future time, a portion of the future load of the access point at a future time can be determined, and thus the total load level of the first access point at a future time can also be determined as a sum. By taking into account the mobility data indicating the expected locations of the terminal devices, predictions about the resources needed at the access point can be made and load balancing can be performed in a timely manner, thereby also enabling reliable future service to terminal devices newly entering the coverage area.

[0010] Such movement data is particularly possible when the terminal device is, for example, a driverless vehicle or another terminal device whose location and movement route are at least partially known based on control data.

[0011] Optionally, the mobility data may further include at least one probability value indicating the probability that the terminal device will be in one of the one or more locations at a predetermined time. Thus, determining the future load level may further include weighting the future load caused by the terminal device using this probability value. In this way, a distinction can be made between locations that the terminal device is deterministically or with a high probability of taking and other possible locations, for example, in the route planning of an autonomous vehicle where the next travel location has been determined, while the subsequent travel route is predetermined with a lower probability.

[0012] In preparing for service handover for load balancing, one could, for example, select an access point with the lowest current load level from among adjacent access points as the second access point, and then proceed by actually handing over at least one service to the second access point. This ensures that the service is not handed over to an access point that is also nearing full load and guarantees the most even distribution and reliable connections possible.

[0013] In another embodiment, preparation for service handover may further include forming a combined load level within the overlapping area by summing the load levels of all access points whose coverage areas at least partially overlap. It can then be checked whether the combined load level within the overlapping area exceeds a pre-defined second threshold, wherein handover is not performed if the combined load level exceeds the second threshold. This check can, for example, prevent service handover from continuing in an area where all responsible access points are already close to their maximum load levels. Since typically no access point can provide a load level greater than its maximum, the largest combined load level indicates that all individual load levels are similarly nearly fully loaded.

[0014] The aforementioned method steps can be performed, for example, directly by an access point in the communication network or by a suitable control unit of that access point. Alternatively, the method can be performed in parallel or alternately by a central control unit in the communication network, which is capable of monitoring the load levels of each access point.

[0015] The important applications of wireless networks, or this invention, are not limited to personal communication but are particularly evident in industrial environments where networked control and information processing are increasingly used across all aspects of production and management. In these cases, local area networks (LANs) or private networks are often used, and the same difficulties arise regarding the maximum capacity of access points for these LANs or private networks.

[0016] One area where wireless networks are increasingly being used is driverless transport vehicles (FTF, also known as Automated Guided Vehicles, AGVs). This includes not only fully automated vehicles controlled remotely via commands from a control unit, but also autonomous vehicles that independently undertake at least a portion of the control and route planning. Such vehicles are used, for example, in industrial settings to transport materials, such as goods in a warehouse. By changing their location, such vehicles communicate via different access points to maintain continuous network connectivity when moving, for example, from one coverage area to another, thus utilizing handover and cell switching along the travel route. High demands are placed on both reliable connectivity and low latency.

[0017] The computing unit according to the invention, such as the control unit of the access point, is particularly configured in terms of programming technology to execute the method according to the invention.

[0018] An implementation of the method according to the invention in the form of a computer program or computer program product having program code for executing all method steps is also advantageous because it results in particularly low cost, especially when the control unit performing the execution is also used for other tasks and therefore always exists. Suitable data carriers for providing the computer program are, in particular, magnetic, optical, and electrical storage devices, such as hard disks, flash memory, EEPROM, DVDs, etc. Downloading the program via a computer network (Internet, intranet, etc.) is also possible.

[0019] Other advantages and configurations of the present invention will become apparent from the specification and the accompanying drawings.

[0020] It should be understood that the features mentioned above and described below can be applied not only in the combinations described separately, but also in other combinations or individually, without departing from the scope of protection of this invention.

[0021] The invention is schematically illustrated in the accompanying drawings according to embodiments and is described in detail below with reference to the drawings. Attached Figure Description

[0022] Figure 1 A schematic diagram is shown of a situation in a wireless network in which embodiments of the present invention can be applied. Figure 2 An example is shown in accordance with Figure 1 Application of the implementation method under the following circumstances; Figure 3 The method flow of an exemplary implementation of load balancing is shown; Figure 4a and 4b Different time points are shown in the following implementation, wherein mobile data exists for a portion of the terminal device; and Figure 5 The following method flow is illustrated, wherein mobile data exists for a portion of a terminal device. Detailed Implementation

[0023] exist Figure 1 The diagram illustrates a situation in a wireless network in which different embodiments of the invention can be applied.

[0024] Here, three access points, AP1, AP2, and AP3, are shown, through which terminal devices can wirelessly connect to network 100. Generally, any access point can be involved; in particular, these wireless access points can also be base stations of a mobile radio network, i.e., Node B, for example. Through these access points, suitable terminal devices can communicate with the actual network or core network. For simplicity, intermediate network elements between these access points AP1, AP2, AP3 and network 100, such as switching centers, controllers, etc., are omitted here; these intermediate network elements can be configured differently depending on the network.

[0025] Each wireless access point can be assigned a coverage area (radio cell), which describes a local area where sufficient reception exists for a reliable connection to the terminal device. Therefore, for... Figure 1 The diagram shows access points AP1, AP2, and AP3, each of which results in additional coverage areas 110, 120, and 130 within a radius surrounding the corresponding access point. These additional coverage areas are shown as circles in the diagram. These coverage areas can, and at least partially overlap, as long as the multiple access points are arranged sufficiently close to each other. In the case of overlapping coverage areas 140, a terminal device located within the overlapping area can theoretically establish a wireless connection with each of its respective access points. Therefore, it can be ensured that a connection to the corresponding network can be maintained for the terminal device at any time, even when moving from one coverage area to another or in the event of access point failure.

[0026] Figure 1 The following scenario is illustrated: three terminal devices D1, D2, and D3 are located within the overlapping coverage area 140 of all three access points and are all connected to network 100 via the same first access point AP1. The connection or communication with the network via the access point is shown by the direction of the dashed arrows in the figure. This unilateral distribution ensures that access point AP1 is under high load until it reaches its load limit, as shown by the schematic load scale 151 attached to the side of the figure. If another terminal device D4 now enters the coverage area 110 of access point AP1, which is already almost fully loaded, and attempts to establish a connection, access point AP1 will no longer be able to provide sufficient resources for service, thus the connection or service request will be rejected. Conversely, access points AP2 and AP3 are only relatively less loaded at the same time, as shown by the corresponding load scales 152 and 153.

[0027] Figure 2 This schematically illustrates how, according to an exemplary implementation, such as Figure 1 The overload situation described in the previous example. Here, we present again: three access points AP1, AP2, and AP3 located in different spatial positions and with different, but partially overlapping, coverage areas 110, 120, and 130. Because these three terminal devices D1, D2, and D3 are all within the overlapping area 140, as in the first example, these terminal devices can establish connections with any one of these access points AP1, AP2, and AP3. Now, to avoid overloading individual access points, communication traffic is therefore actively distributed to one of the other access points within the overlapping area, thus distributing the load more evenly across these access points and ensuring that there is sufficient capacity for other terminal devices or other service requests. This can be called load balancing. Here, prior to load balancing, there may be... Figure 1 In this situation, all three terminal devices are connected to access point AP1.

[0028] Thus, in the example shown, the network connection of the first terminal device D1 to the core network 100 can be routed, for example, via the first access point AP1, the network connection of the second terminal device D2 via the second access point AP2, and the network connection of the third terminal device D3 via the third access point AP3. The communication flow is again illustrated by the direction of the dashed arrows between the core network and the terminal devices via the respective access points. As shown by the schematic load scales 251, 252, and 253 of the access points, the load is now more evenly distributed across these access points, and each access point also has free capacity for other connection or service requests. If a fourth terminal device D4 now arrives within the coverage area 210 of the first access point, a reliable connection to that access point can be established. It can be determined here which communication flows should preferably be routed to other access points to achieve not only even load distribution but also continuous provision of the requested service.

[0029] For example, in one implementation, it is preferable to transfer services with a lower priority level to another access point. Optionally, it is acceptable for the connection quality for that service to deteriorate, provided that the minimum prerequisites for providing the service are met.

[0030] Alternatively, if all available access points are overloaded to the point that there are insufficient resources available even under load balancing, consider temporarily abandoning the service or interrupting the communication connection. This decision can also be made, for example, based on priority values.

[0031] According to an exemplary implementation, load balancing can be performed based on the relative current load of each access point. For this purpose, a metric (Metrik) can be defined for this load, which can be based on various service-related information.

[0032] To define a preferably standardized load metric, various parameters can be considered. In particular, for this purpose, for example, Quality of Service (QoS) parameters or descriptors (e.g., TSN flow descriptors) can be used, as defined in different mobile radio and communication standards. However, similarly, all other suitable service-related or connection-related parameters can be utilized.

[0033] For example, the time frequency of the corresponding communication streams or services that generate the load can be used as a parameter, where, for example, a distinction can be made between isochronous, cyclic, and acyclic services. Priority parameters can specify the priority at which the data must be transmitted. For example, security-related data can be transmitted to vehicles with high priority. The amount of data to be transmitted can also be incorporated into the normalized load, for example, in the form of the required number of frames or other appropriate metrics for the data volume. Furthermore, the expected latency or survival time limits for data packets can be included. Other conceivable parameters that can be incorporated into the load metric are jitter in terms of runtime, service availability, or packet loss rate.

[0034] The standardized load can be derived from these and other parameters by applying various arithmetic and / or statistical operations. For example, two of the parameters mentioned above can be multiplied together.

[0035] The actual determination of the load using these parameters, and the selection of which parameters should be incorporated into the load determination, depends on the specific application and is therefore not detailed here. The load determined in this way ultimately forms the basis for the load balancing method. Wherever the term "load" is used in the following text, it can preferably be a standardized load.

[0036] exist Figure 3 In, it is shown that in such Figure 2 The flowchart illustrates an exemplary method for load balancing in a situation like the one described above. Here, the coverage areas of the three wireless access points again form a common overlapping area, and the three mobile terminal devices are located in this common overlapping area. At the start of the method, access point AP1 manages three services for the three terminal devices.

[0037] In response, each access point determines its current standardized load level continuously or at pre-given intervals in step 310 according to the following equation: in It is at the nth access point (AP) n The current load level at that location, It is at the access point (AP) n The current absolute load at that location, and It is at the access point (AP) n The maximum absolute load at that location.

[0038] In step 320, it is checked whether the determined load level of the access point is below a pre-given value. If it is determined that the load level is below the pre-given threshold, the method can be terminated for that time step and the load level can be recalculated according to equation (1) at the next time step. Conversely, the high full load of the corresponding access point, that is, where the current absolute load is close to the maximum load or the normalized load level tends to 1, i.e., The preferred trigger should be to reallocate at least a portion of the services offered by that access point to other access points. Alternatively, a threshold below the maximum load level of that access point can also be specified.

[0039] To this end, the access point can further first verify whether there are sufficient resources within the overlapping area, that is, within the area where multiple coverage areas of multiple access points overlap, by adding the current loads of the N access points in step 330 to form a combined, standardized load level within the overlapping area. , in It is the largest combined absolute load within the overlapping area, which can be formed by the sum of the individual largest loads.

[0040] This can be understood as follows: To determine the combined load level, for each access point, there should be information about the coverage area of ​​access points in the environment and their current load level. For this purpose, adjacent access points can, for example, regularly transmit their current load level directly to other access points or to elements of the network that then transmit or forward the current load levels of the relevant access points to each individual access point. Here, as long as it is clear for each access point which surrounding access points form an overlapping area and therefore which access points can be handed over to, the access point's coverage area does not need to be directly forwarded as a parameter. Alternatively or additionally, for the current standardized load level, the access point can also state its absolute load value and its own limit value for maximum load, so that the access point being evaluated or other control unit can then determine the standardized load level of other access points based on this data. Information about the load can be transmitted in its own protocol message or along with other data in a suitable format.

[0041] If then in step 340 it has been determined that there are sufficient resources in total within the overlapping area, i.e., the combined load level within the overlapping area has not been exceeded. Given a pre-defined limit, an overloaded access point can prepare to hand over one or more services to other access points within the overlapping area. This limit can be, as already described, at the maximum load of these access points or a lower threshold.

[0042] To address this, a suitable access point can first be selected from adjacent access points in step 350. This selection can be based, for example, on choosing access points within the overlapping area that have the lowest current load level. Preferably, for the handover access point, not only is information about the standardized load levels of surrounding access points available, but also information about the absolute load value, thus allowing verification that the receiving access point can adequately meet the load parameters of the service to be handed over. Next, service handover is performed in step 360, for example, through corresponding protocol messages to the involved terminal devices and / or other access points (e.g., the receiving access point) or control units in the network.

[0043] Conversely, if in step 340 it is determined by examining the combined load levels that the combined load of all access points within the overlapping area is approaching or (currently or expected to) exceed the defined load limit, i.e., for the case... Service handover is currently neither meaningful nor possible. Therefore, the assessment of the current load level is temporarily suspended and restarted from scratch for the next time step.

[0044] Alternatively, in this case, service handover can still occur between more than two access points, thereby ensuring a more even distribution of load within the overlapping area and thus guaranteeing reliable communication, even if additional resources are not rendered idle for other terminal devices. Here, as previously shown, the respective load levels of the respective handover and receiving access points can also be used as a basis for selecting the access point to which service should be handed over.

[0045] For service handover, alternatively or additionally, it may be stipulated that in cases of excessively high combined loads in overlapping areas, where handover to other access points is impossible or becomes meaningless, one of the existing services at the access point can be abandoned. This may be particularly meaningful in situations where a high-priority service is requested or anticipated, thereby temporarily abandoning a very low-priority service to allow sufficient resources to be provided at the access point.

[0046] To facilitate service handover to other access points, additional conditions and checks can be specified to prevent frequent access point changes, for example. Alternatively, such conditions can be considered by appropriately selecting parameters introduced into load determination.

[0047] In addition to the parameters already mentioned, location information can also be considered in load balancing methods in other implementations. The location of the terminal device can be determined either by a suitable positioning method, such as triangulation, by a network, or by notification from other control units that have the aforementioned information.

[0048] Therefore, when selecting suitable access points for service handover or transfer, in addition to minimizing load, considerations should also be given to: how close the terminal device is to the edge of each coverage area, i.e., its current distance from each access point. This prevents services within overlapping areas from being transferred to access points whose relative orientation would necessitate a switch to another access point even with minor changes in the location of the corresponding terminal device.

[0049] In other implementations, location information that is expected to change at future points in time or over future periods, i.e., mobile data of terminal devices, can be considered. Here, statistical data, which exists in the network and contains information about the probability that a large number of terminal devices or requested services should be expected at the access point at a given time. This information can be obtained, for example, from data stored over past time periods.

[0050] However, alternatively or additionally, mobility information that is at least partially known, for example, due to control data inside or outside the network, may also be used to determine load balancing.

[0051] Figure 4a and 4b Two points in time are shown in this exemplary method. Here, a total of four terminal devices are re-shown, with motion information for a future time period existing for at least one of these terminal devices. The remaining three terminal devices, D1 to D4, can be any other terminal devices, and no information exists regarding their motion. Furthermore, as in... Figure 1 and Figure 2 The diagram shows three access points with different but overlapping coverage areas. These access points are thus positioned to form areas where connections to at least two access points are established, respectively.

[0052] The movement information of the fourth terminal device D4 is shown and described using nodes K1 to K4, indicating its current location and projected future location. Figure 4aThe current location of terminal device D4 at the first node K1 is shown, where this first node is outside all the coverage areas considered herein for these access points. The terminal device may be connected to or logged into an access point (not shown), or may not have network access to date. However, information exists regarding the expected movement of the terminal device in the future, indicating that the terminal device will move along the routes of nodes K2, K3, and K4. In the current example, nodes K2 and K3 are within the coverage area of ​​the first access point AP1 and do not reach the coverage areas of access points AP2 and AP3. Conversely, the next node K4 is outside the coverage areas of all three observed access points. Preferably, specific time points can be described for one or more nodes along the terminal device's route, at which the terminal device D4 will arrive at the corresponding node. These time points can also be described by indicating the time elapsed since the notification until the terminal device is expected to arrive at the node (Zeitdauer). Alternatively, instead of specifying a specific time point, a wider time interval or the probability that the terminal device will be located at a specific node at a given time point can be stated. As another alternative, a time-dependent function for the terminal device at the given location can be stated, describing the probability and thus having a maximum value at the most likely point of stay (or over a longer period), while this probability decreases again before and after that period.

[0053] Using this information, the network can make predictions about the expected need or timing of resource requirements for connecting the terminal device D4 to the access point AP1, i.e., predicting at least one time point when the terminal device is located within the coverage area of ​​the first access point AP1. In particular, the mobility information can be considered for load balancing even before the terminal device is within the coverage area of ​​a participating access point. Optionally, the distance traveled between nodes can also be considered, for example, by interpolating the travel distance and the terminal device's determined location between these nodes, since the terminal device D4 may have already begun entering the coverage area before reaching the corresponding node. Alternatively, without special calculations, a buffer period can be set for each terminal device before the arrival time at the first node within the coverage area, wherein preferably, network resources for the entering terminal device should remain idle until at least that time point. The buffer period can be selected uniformly for all terminal devices or based on the terminal device type.

[0054] Additionally, regarding the determination of the combined load level for each access point and the overlapping area, as already described, the nth node AP at time t can be... n The expected additional load on the access point is combined with the load of the static terminal device (or terminal device without known mobility information) at time t, and then normalized again based on the maximum load of the access point to obtain the expected future load level for the nth access point at time t. : .

[0055] Because this also includes potential anticipated connections, the value for the anticipated future load level can also be greater than 1; therefore, it is necessary to transfer the service to another access point, abandon the service, and / or reject the requested service, because the maximum load of that access point is expected to be exceeded.

[0056] exist Figure 4a Zhongyu Figure 1 The situation in Figure 4 exemplifies the instantaneous high load level of the first access point AP1, as this first access point manages services for all terminal devices D1, D2, and D3. Furthermore, mobility information regarding terminal device D4 is now available for access point AP1, where terminal device D4 has not yet connected to the access point and is not yet within the access point's coverage area. Therefore, performing the load balancing method as described above, considering the anticipated future load according to equation (5), results in the conclusion that since terminal device D4 is expected to enter the coverage area 110 of the first access point AP1, at least a portion of the service should currently be transferred to other access points. Optionally, an optimal time point can also be determined here, by which the service transfer should end. It is also possible to predict, using the mobility data for terminal device D4 along nodes K1 to K4, how long it is expected that additional resources must be provided for this terminal device. In the exemplary situation in Figure 4, this can therefore be used as a result to compare with the previous approach for… Figure 2 Similar load balancing has already been described.

[0057] Figure 4b Show Figure 4aThis refers to a slightly later point in time in the situation described. As predicted using mobile data, terminal device D4 had moved to node K2 and was therefore within the coverage area of ​​access point AP1 during this period. Using the described load balancing method, services for terminal devices D2 and D3 have been transferred from access point AP1 to access points AP2 and AP3 respectively, thus giving access point AP1, which was previously at the edge of load, sufficient resources to provide services for terminal device D4.

[0058] Similarly, access points AP2 and AP3 may have already received notification of the mobile data of terminal device D4, enabling these access points to predict, based on the notified route, that terminal device D4 is not expected to enter the coverage area of ​​these access points and therefore need not be considered when determining the load for future time periods.

[0059] In addition to the time points when the location is occupied, information can also be provided along with the mobile data used for the terminal device: for example, the estimated load values ​​that the terminal device will use at these times; and / or information about the importance and relevance of the requested service. Alternatively, standard values ​​of assumed load for the terminal device predefined by the access point or control unit can be used, where these standard values ​​can also be set differently based on other parameters, such as those transmitted using location information or based on the type of terminal device.

[0060] In other implementations, terminal devices that are logged into an access point but currently do not have active connections to the network via that access point can also be included in the load balancing method. Here, for example, information about the periodic service requests of the logged-in terminal devices can be used to similarly conclude about the future load requests at the determined access point AP1. If it is expected with high probability that newly established services by the logged-in terminal devices exceed a pre-defined threshold for the load level of the access point, one or more of the inactive logged-in terminal devices can be instructed to select another cell (cell reselection). When, for example, the determined terminal devices request services and establish connections at periodic intervals, the future load caused by the logged-in inactive terminal devices can be predicted again, for example, using statistical or historical data. These implementations can, of course, also be combined with mobile data applications, where not only mobile data from terminal devices with active connections but also mobile data from inactive or connectionless terminal devices can be considered.

[0061] The application of the exemplary load balancing method when using mobile data is in the context of communication within automated or autonomous vehicles, especially driverless transport vehicles. Figure 5 A portion of the illustrative method flow for this example is shown. These vehicles may each include at least one wireless communication unit that enables communication with the desired network and corresponds to a terminal device generally described to date. Information known from vehicle control can be forwarded in various ways to these access points or other elements of the network and incorporated into the determination or load balancing of current and future load distribution. This extended method can be particularly applied when the majority of services and anticipated communication flows in a network, such as a local WLAN network or a 5G network with driverless transport vehicle control, are derived via mobile terminal devices with current mobility information. Here, a control center may exist responsible for controlling the transport vehicles and thus possessing route information for each transport vehicle or its terminal devices. The route information can be specified by other units or by a predefined value and / or actively determined and calculated by the control center. Whenever vehicles are mentioned below, they should be understood as networked vehicles with corresponding terminal devices that can establish connections with these access points.

[0062] The vehicle's route can be defined in different ways. One possible implementation utilizes different nodes, as in the example of Figure 4, along which the route extends. The vehicle can actually move in a straight line between these nodes; alternatively, these nodes can form an approximation of the vehicle's actual route, where the number of nodes can be arbitrarily defined along the route. Such a system is described in detail, for example, in publication VDA5050 "Interface for communication between driverless transport vehicles (FTF) and master stations (Leitsteuerung)". Alternatively, a distinction can be made between predefined nodes (Base Nodes) and expected nodes (Horizon Nodes) that have been issued to the vehicle and / or assigned by corresponding driving instructions. Expected nodes can be, for example, nodes along which the vehicle will move to fulfill its task in the absence of traffic jams or other conditions requiring route replanning.

[0063] If a distinction is made between specified and expected nodes, the probability for a specified node can be, for example, close to 1, while the maximum probability for an expected node is lower than the probability for a specified node and decreases, for example, depending on distance or as the number of possible road intersections or route alternatives increases.

[0064] Here, the control center for transport vehicles, such as the main control center, can possess information about the current location of all vehicles, as well as information about the planned routes of all vehicles and therefore information about their future locations. Since most data communication in the environment originates from mobile terminal devices with known mobility information, the control center's data can be included for load balancing decisions.

[0065] In such exemplary situations, two groups of terminal devices can be considered: driverless transport vehicles (or terminal devices with known mobility information in general) and a group of other terminal devices.

[0066] The load caused by the other terminal devices D1, D2, and D3 (i.e., "static" terminal devices or terminal devices that do not have mobile data for themselves). In step 510, preferably as previously stated... Figures 1 to 3 It is processed and evaluated as described.

[0067] Conversely, the load caused by transport vehicles can be handled in a different way. In step 512, the access point sends a request to the vehicle control center, which requests a list of all vehicles that will be within the coverage area at a pre-given future time point (e.g., "in 60 minutes"). This future time point can be specified as a configuration parameter for the load balancing method, and can also be pre-given the same for all access points or notified to all access points. Furthermore, for each transport vehicle listed in the list... The following probability value p(i) describes the probability of the i-th transport vehicle. At a predetermined time point, the requested access point will be within its coverage area with the stated probability.

[0068] Now, in order to determine the total load of all transport vehicles (AGVs) for access point n at time point t in step 514... The probability p(n, i) obtained with respect to the access point n can be used as a weighting factor for the load of each transport vehicle i. At time t, the load of each transport vehicle i... The value can be pre-given by estimating or obtaining the average load for each vehicle from earlier data.

[0069] Therefore, the future load at access point n at time T can then be determined, where the load caused by I transport vehicles (terminal devices with mobile data) and the load caused by N terminal devices without mobile data according to equation (6) are added together: .

[0070] The total future load determined by equation (7) based on mobile data can be used accordingly as in equation (5) to determine the expected future standardized load level of the access points and to evaluate it in step 520 given a pre-given maximum load limit, so that load balancing among the N access points can then be performed on this basis as already described. The other steps for verifying the total load in the overlapping area and for finding suitable access points correspond to... Figure 3 Steps 330 to 360 in the process, wherein preferably the summed load according to equation (7) is used for all access points, and will not be described further here.

[0071] An access point can also offload multiple services to achieve load balancing, and here it can either offload all services to the access point or variably offload these services to different access points. Here, for example, the expected load level after offloading can be calculated, and offloading can be performed based on whether the expected load level is thus lower than the expected threshold for the load level of that access point, or other services can be selected for offloading to other access points.

[0072] The actual steps and protocols that facilitate and execute a handover to another access point, another network, or another cell are system-dependent and known in the field of expertise. If the steps described above have determined that the service transfer to another access point is desired, then for this purpose, corresponding protocol messages can be sent from the access point to the terminal device, for example, notifying the request to hand over to the access point selected according to the aforementioned method. Similarly, as an intermediate step, each access point can be notified, for example, of a list of possible access points for handover (corresponding to a neighbor list), the current load levels of these possible access points can then be announced individually or in the form of this list.

[0073] In the current example, load level assessment and load balancing decisions have been performed decisively through these access points. However, other variations are also possible, such as utilizing a central control unit in the network (e.g., the Access and Mobility Management Function (AMF) for 5G networks), which then collects service-related information about these connections and thereby derives the current load level and manages load balancing. Similarly, combinations of multiple units within the network can be envisioned.

[0074] Mobile data from mobile terminal devices can be provided by different suitable units in a similar manner, or mobile data from mobile terminal devices can be collected from different sources by, for example, a control unit in the network and then forwarded to the relevant access point when necessary.

[0075] It should be understood that the aforementioned steps for the following terminal devices can be applied to various implementations and are not limited to examples of driverless transport vehicles, where the network or other control unit has mobile data for these terminal devices. A similar situation arises, for example, in the case of autonomous or partially autonomous networked rail vehicles, where a reliable and delay-free connection is required at all times for safety reasons, such as for route signals.

[0076] Similarly, routes can be used to evaluate movement data for navigation systems in networked vehicles where the user has selected a predetermined route. The fact that the vehicle remains individually controllable depending on the implementation and that the user can decide not to adopt the selected route can be taken into account by including probability values ​​in the movement data.

[0077] It can also acquire mobile data with a predetermined reduced probability of entry for each user of a mobile terminal device, for example, based on the user's past behavior, such as by identifying regular dwell locations at specific times of day.

Claims

1. A method for load balancing in a communication network, the method comprising: Detecting motion data (512) of at least one terminal device (D4), wherein the motion data describes one or more future locations (K1, K2, K3, K4) of the terminal device (D4) within one or more predetermined time periods, wherein the motion data further includes at least one probability value, the probability value describing the probability that the terminal device (D4) is at one of the one or more locations (K1, K2, K3, K4) at a predetermined time point, and The load level of the first access point (AP1) is determined (310) based at least on parameters of multiple services provided by the first access point (AP1) of the communication network to other terminal devices (D1, D2, D3). Based on detected mobile data for at least one terminal device and parameters based on the services that the first access point (AP1) should provide for the at least one terminal device (D4) at a future time point, the load level of the first access point (AP1) at that future time point is additionally determined (514), wherein the future load caused by the terminal device (D4) is weighted by the probability value of the mobile data, and If (320) the load level of the first access point (AP1) is above a pre-given first threshold at the present or future time, then preparations are made to transfer at least one service for the other terminal device to a second access point among one or more adjacent access points (AP2, AP3), wherein the coverage area (120, 130) of the adjacent access points (AP2, AP3) at least partially overlaps with the coverage area (110) of the first access point (AP1) and thus forms an overlapping area (140).

2. The method of claim 1, wherein the load level is determined as a standardized load level as a quotient of the absolute load of the first access point and the maximum load of the first access point.

3. The method according to claim 1 or 2, wherein the method further comprises: Receive at least one message, the at least one message including information about the load level of one or more adjacent access points.

4. The method of claim 3, wherein preparation for the handover further comprises: Select (350) the access point with the smallest current load level from the adjacent access points (AP2, AP3) as the second access point; and perform: transfer at least one service to the second access point.

5. The method according to any one of claims 3 or 4, wherein preparation for the handover of at least one service further comprises: The sum of the load levels of all access points (AP1, AP2, AP3) whose coverage areas (110, 120, 130) at least partially overlap with each other forms (330) a combined load level in the overlapping area (140); and the combined load level in the overlapping area (140) is checked (340) to see if it exceeds a pre-given second threshold, wherein no handover is performed when the combined load level exceeds the second threshold.

6. The method according to any one of claims 1 or 2, wherein the method is performed in the control unit of the first access point (AP1).

7. The method according to any one of claims 1 or 2, wherein the method is performed in a central control unit in the communication network.

8. The method according to any one of claims 1 to 5, wherein the method is performed in a computing unit.

9. A computer program product comprising instructions that, when executed on a computing unit, cause the computing unit to perform the method according to any one of claims 1 to 5.

10. A machine-readable storage medium having a computer program product according to claim 9 stored on it.

Images