Methods for distributing network traffic

Abstract

Network traffic is distributed across two or more different types of radio access network (RAN) by determining a distribution of network traffic to a user equipment (UE) via a first type of RAN and a

Description

FIELD

[001] The disclosure relates generally to telecommunications methods and systems and more specifically, methods for distributing network traffic. BACKGROUND

[002] The radio spectrum is widely used in technology to transmit information across distances. For example, portions of the radio spectrum are used for radio and television broadcasting, satellite communications, wireless networking and telecommunications. In view of the ever growing number of radio spectrum uses, spectrum management is an increasingly relevant issue.

[003] To minimise interference between different uses or applications, certain portions of the radio spectrum are assigned to particular uses or applications by artificially dividing the spectrum into different frequency bands. This process is known as allocation, frequency allocation or spectrum allocation. Guard bands (narrow, intentionally unused frequency bands placed between adjacent frequency bands) are typically used to minimise interference between different frequency bands. This allocation is typically regulated by individual governments, although the allocation may be coordinated to a certain extent across country boundaries by international bodies, such as the International Telecommunication Union, ITU.

[004] However, the radio spectrum, with frequencies from 3Hz to 3000GHz, is a finite resource. Indeed, some portions of the spectrum may not be usable due to practical and physical limitations (for example, lowest frequencies used are limited by the increasing size of transmitting antennas required and at highest frequencies, atmospheric absorption becomes an issue). Thus, as more technologies use portions of the radio spectrum, the spectrum becomes increasingly congested, even with this spectrum management.

[005] As 5G and 6G technologies continue to develop, it is envisaged that different types of network technology will become more integrated, possibly converging into one network. For example, cellular, Wi-Fi and satellite communication technologies may be served by the same core network (which currently only supports cellular communication technologies). This convergence could assist with addressing the finite and limited radio frequencies available. However, implementations to achieve or further this convergence and make efficient use of radio resources are still to be determined.

[006] Therefore, methods and systems that overcome these issues is desirable. SUMMARY

[007] A first type of radio access network, RAN, is able to communicate network traffic to a device. A second type of RAN is also able to communicate network traffic to the device. The network traffic can accordingly be communicated via both the first and second types of RAN. The device may typically be a user equipment, UE, although other devices able to communicate via a network may be used.

[008] Since the RANs are different types, the performance of each RAN when communicating the traffic may differ. For example, one type of RAN may be experiencing high traffic due to a large number of users in the area, whilst the other may have relatively low traffic. The performance of the RANs may be indicated by one or more network performance metrics. Network performance metrics may include, among other things, a throughput, delay, packet loss, or jitter.

[009] A distribution of the network traffic between the first and second type of RAN for communicating with the UE is determined. In other words, the proportion of traffic to be sent via the first type of RAN and the proportion of traffic to be sent via the second type of RAN is determined. This determination is based on the one or more network performance metrics, which may indicate the current or expected performance of the network. The traffic is then communicated to the device via the first and second types of RAN according to the determined distribution. In other words, both the first and second types of RAN will communicate a portion of the network traffic.

[010] This allows the network to make effective use of both types of RAN. For example, if one type of RAN is experiencing delays, but the communications to be sent and / or received from the device are not time-critical, the communications may be sent via this RAN. Other communications can be sent via the other type of RAN. Since the non-time-critical communications are sent via the first type of RAN, the other type of RAN may be able to process the other communications, which may be more time-sensitive, more efficiently (since the non-time-critical communications do not also need to be processed by that RAN). Other possibilities to improve or maintain network performance are possible.

[011] In accordance with a first aspect, there is provided a method for distributing network traffic across two or more different types of radio access network, RAN, the method comprising steps of: determining a distribution configuration for distribution of network traffic to a user equipment, UE, via a first type of RAN and a second, different type of RAN based on one or more network performance metrics, the distribution configuration defining a non-zero network traffic distribution factor for each of the first and second types of RAN; and distributing the network traffic between the first type of RAN and the second, different type of RAN according to the distribution configuration.

[012] This can allow more effective network resource use, since network traffic is split across different types of RAN. This can allow the relative benefits of the different types of RAN (which may be reflected in different network performances at particular times, for example) to be leveraged, which can in turn improve network performance.

[013] Optionally, determining the distribution configuration may be further based on one or more of: a type of network traffic, a network traffic use and one or more network traffic distribution rules. These may represent factors affecting how network resources can be used effectively. For example, certain types of network traffic may be more time-sensitive and so should preferably be sent via a type of RAN having high throughput and small delays. In the opposite case, non-time-sensitive network traffic may be sent via a type of RAN having poorer network performance (which can in turn allow more time-sensitive or otherwise important data to be sent via a better performing RAN).

[014] In some examples, the type of network traffic may comprise a type of data to be transmitted. Similarly to as discussed above, the type of data to be transmitted may impact the determination of the distribution configuration to allow efficient network resource use and improved network performance. The type of data may be voice, video, audio, data, metadata, and so on.

[015] In further examples, the network traffic use may comprise an extended reality use, voice over IP use or video-conferencing use. Other uses are also possible. Again, the use of the network traffic may impact the determination of how network traffic can be split or distributed effectively. Extended reality, voice over IP and video conferencing may involve more complex forms of network traffic or multiple streams of network traffic and may therefore be useful uses to identify to more effectively distribute network traffic.

[016] Preferably, a spectrum allocation of the first type of RAN and a spectrum allocation of the second type of RAN may be within a same or an overlapping frequency band. Implementing network traffic distribution on a shared spectrum may further improve the efficiency of network resource use.

[017] In some examples, determining the distribution configuration may be further based on the spectrum allocation of the first and second types of RAN within the same or overlapping frequency band. Splitting or distributing network traffic across two different types of RAN that are assigned to the same or an overlapping frequency band can involve additional complexities. For example, portions of the spectrum may need to be reallocated before the distribution configuration can be implemented, or assigning too much traffic to one type of RAN may cause it to request additional spectrum, thus possibly causing issues with another type of RAN (which may also be receiving network traffic from the UE).

[018] In some implementations, the non-zero distribution factor for each of the first and second type of RAN may be based on a relative proportion of the spectrum allocation of the first and second types of RAN within the same or overlapping frequency band. This may be a straightforward manner of distributing network traffic that can reduce the determining power or time required to determine the distribution configuration.

[019] Optionally, the method may further comprise sending a communication indicating that the spectrum allocation of the first type of RAN and / or the second type of RAN should be adjusted. This can allow further flexibility in determining the distribution configuration, since the radio spectrum allocation need not remain static.

[020] In some examples, the communication may be sent to a server storing spectrum allocation information comprising information identifying portions of the same or overlapping frequency band allocated to the first type of RAN and the second type of RAN. This may be a straightforward way of instructing or requesting spectrum reallocation.

[021] In examples where a communication indicating that the spectrum allocation should be adjusted, the communication may be sent prior to determining the distribution configuration or based on the distribution configuration. This can allow the distribution configuration to be determined based on more up-to-date network performance metrics, which can allow more effective network traffic distribution.

[022] In further examples, one of first and second type of RAN may be a cellular network and the other of the first and second type of RAN may be a wireless network. These types of RAN may be particularly useful for sharing network traffic, as there may already exist some convergence between these two types of network.

[023] In yet further examples, the cellular network may comprise one or more of a 3G, 4G, 5G and 6G network and the wireless network may comprise a Wi-Fi network. These types of RAN may be particularly useful for sharing network traffic, as there may already exist some convergence between these two types of network.

[024] In some implementations, the method may further comprise, prior to determining the distribution configuration, determining, based on the one or more network performance metrics, whether to distribute the network traffic between the first type of RAN and the second type of RAN. For example, it can be identified whether distributing the network traffic (or distributing the network traffic between those particular types of RAN) is expected to improve network performance or another factor. This expectation may be based on an expected change in the one or more network performance metrics.

[025] In some examples, determine whether to distribute the network traffic between the first and second types of RAN may comprise identifying that this combination of RAN types is an optimal RAN combination (for example, expected to optimise network performance). In other examples, the determination may comprise identifying that the combination of RAN types is expected to improve network performance.

[026] In further implementations, the one or more network performance metrics may comprise a traffic flow rate. This may be a useful indicator of network performance that can allow an effective distribution configuration to be determined. In some examples, the traffic flow rate may comprise a network throughput. Other types of traffic flow rate may be used including, but not limited to, a goodput, bandwidth, and so on.

[027] Optionally, the one or more network performance metrics may comprise a delay parameter. This may be a useful indicator of network performance that can allow an effective distribution configuration to be determined. The delay parameter may comprise a network delay, which may include, for example, a round-time trip, latency, and so on. Other delay parameters that may be used include a buffer delay, jitter and so on.

[028] In some examples, determining the distribution configuration may comprise identifying nonzero network distribution factors for each of the first and second types of RAN that are expected to meet one or more quality of service, QoS, criteria. This may allow network performance to be enhanced whilst also providing sufficient user experience.

[029] In further examples, the one or more QoS criteria may include a traffic flow rate above a threshold traffic flow rate value and / or a delay parameter below the threshold delay parameter value. High traffic flow rate and small delays may typically provide a desirable use experience for typical user cases.

[030] In yet other examples, determining the distribution configuration may be further based on a QoS parameter established based on the one or more network performance metrics. This may further assist in providing a good user experience whilst also providing good network performance.

[031] Optionally, the non-zero network distribution factors may be expected to increase the QoS parameter. This may further assist in providing a good user experience whilst also providing good network performance.

[032] In some implementations, determining the distribution configuration may comprise identifying a first network path of the first type of RAN and a second network path of the second type of RAN that are expected to increase the QoS parameter. Different paths for network traffic may be possible and some may be more effective than others. For example, some network paths may involve communications between more terminals than others and / or some terminals may be experiencing high buffer delays. Determining a network path that can avoid or limit these and other factors can further enhance user experience and improve network performance.

[033] Optionally, the QoS parameter, QaS^t), at time tfor a user / on network j over path k may be determined as I K QoS^j (t) = [my + w2D^ ] i = l JERAN1,RAN2 k = l or i = l JERAN1,RAN2 k = l wherein Tijk represents a first network performance metric and Dijk represents a second performance for the user i on network j over path k, RANa represents the first type of RAN, RAN2 represents the second type of RAN, and and w2 are weighting coefficients. These formulations may allow the determination of a distribution configuration that balances network performance metrics well.

[034] In some examples, the method may further comprise, prior to determining the distribution configuration, communicating network traffic to the UE via one of the first and second type of RAN, and wherein determining the distribution configuration is responsive to identifying that the other one of the first and second type of RAN is available to communicate network traffic to the UE in addition to the one of the first and second type of RAN. In other words, one of the RAN types may not initially be available, and the method involves identifying that another type of RAN has become available. The network can thus react to changes to make more efficient use of network resources available.

[035] In an exemplary aspect, the methods described above may be implemented in a system configured to implement the method steps. The system may comprise a scheduler or intelligent agent configured to implement the method steps.

[036] The methods described above may be implemented as a computer program comprising instructions to operate a computer or computer system. The computer program may be, or may be implemented by, a scheduler or intelligent agent. The computer program may be stored on a non-transitory computer-readable medium.

[037] The computer system may include a processor, such as a central processing unit (CPU). The processor may execute logic in the form of a software program. The computer system may include a memory including volatile and non-volatile storage medium. The different parts of the system may be connected using a network (e.g. wireless networks and wired networks). The computer system may include one or more interfaces. The computer system may contain a suitable operating system such as UNIX (RTM) (including Linux (RTM)), Windows (RTM), for example.

[038] It should be noted that any feature described herein may be used with any particular aspect or embodiment of the invention. Moreover, the combination of any specific apparatus, structural or method features is also provided, even if that combination is not explicitly disclosed.

[039] The invention will now be described with reference to the attached drawings depicting different embodiments thereof, the drawings being provided purely by way of example and not limitation. BRIEF DESCRIPTION OF DRAWINGS

[040] The invention may be put into practice in a number of ways, and preferred embodiments will now be described by way of example only and with reference to the accompanying drawings, in which:

[041] Figure 1 shows a schematic message flow for dynamic spectrum allocation according to an embodiment of the disclosure;

[042] Figure 2 illustrates a system for implementing spectrum reallocation according to an embodiment of the disclosure;

[043] Figure 3 shows a schematic system for dynamic spectrum allocation according to an embodiment of the disclosure;

[044] Figures 4 and 5 illustrate the schematic system of Figure 3 following spectrum reallocation according to an embodiment of the disclosure;

[045] Figure 6 illustrates a flow chart of a method for spectrum reallocation according to an embodiment of the disclosure;

[046] Figures 7A to 7C illustrate flow charts of methods for spectrum reallocation according to embodiments of the disclosure;

[047] Figure 8 shows a schematic diagram of a device configured to operate using two different radio access technologies according to an embodiment of the disclosure;

[048] Figure 9 illustrates a flow chart of a method for distributing network traffic across two or more different types of RAN;

[049] Figure 10 shows a schematic system including an extended reality device in communication with a core network, in which a method for distributing network traffic according to an embodiment of the disclosure may be implemented;

[050] Figure 11 illustrates a schematic system of two overlapping regions, in one of which one RAN technology is used to communicate with end devices, and in one of which a different type of RAN technology is used to communicate with end devices;

[051] Figure 12 illustrates a system architecture for implementing network splitting;

[052] Figure 13 illustrates a system architecture for implementing network splitting in accordance with an embodiment of the disclosure; and

[053] Figure 14 illustrates a scheduler or intelligent agent according to an embodiment of the disclosure.

[054] It should be noted that the Figures are illustrated in schematic form for simplicity and are not necessarily drawn to scale. Like features are provided with the same (or similar) reference numerals. DESCRIPTION OF PREFERRED EMBODIMENTS

[055] Figure 1 illustrates a message flow for dynamic spectrum allocation in accordance with the present disclosure. A system comprises a first type of radio access technology device 110, a second type of radio access technology device 120 and a server 130. The first type of RAT device 110 and the second type of RAT device 120 are both configured to operate using the same, or an overlapping, frequency band using different radio access technologies. For example, the first and second types of RAT devices 110, 120 may provide communications to user equipment, UEs, via radio access technologies.

[056] The first type of RAT device 110 initiates a spectrum reallocation procedure 100 by sending 101 a reallocation request message to the server 130. The server 130 may then identify that a portion of the shared frequency band is available for use (that is, not being used by or is not assigned for use by the second type of RAT device 120). In this case, the server 130 may return 104 a response to the first type of RAT device 110 indicating that communications over the portion of the shared frequency band by the second type of RAT device 120 will not occur. The first type of RAT device 110 can thus begin transmitting using the portion of the shared frequency band without or with a low risk of interference from a device using another, different type of RAT to communicate.

[057] Otherwise, the server 130 may identify that a communication channel over which the second type of RAT device 120 is communicating or is configured to communicate may be used by the first type of RAT device 110. This may comprise identifying that one or more communication channels assigned for use by the first type of RAT device 110 are within a threshold frequency range of the communication channel assigned for use by the second type of RAT device 120, for example. Other criteria may be used to identify that the communication channel could be assigned by the first type of RAT device 110 instead of the second type of RAT device 120.

[058] The server 130 sends 102 an instruction to the second type of RAT device 120 to cease communications over the communication channel assigned to the second type of RAT device 120. The second type of RAT device 120 then sends 103 a message confirming that it has ceased communications over the communication channel and / or will not communicate over the communication channel (for instance, if it was not previously using that communication channel). Then, as described above, the server 130 sends 104 a communication to the first type of RAT device 110 that indicates that communications over the communication channel by the second type of RAT device 120 will not occur.

[059] There may be cases where it is not possible to extend the range of frequencies assigned to the first type of RAT device 110. For example, the radio access technology of the second type of RAT device may also be congested and therefore be unable to offer any channels. The server 130 may store network information so that it can determine when this is the case and return 104 a rejection message in response to the reallocation request (without instructing the second type of RAT device 120 to stop transmitting). In other examples, the server 130 may send a query to the second type of RAT device 120 for (updated) network information. In response to receiving network information from the second type of RAT device indicating that the second type of RAT device 120 has spare resources (for example, indicating that a channel is allocated but not being used or that the network load is lower than expected), the server 130 may then send 103 the instruction to cease communicating over a communication channel. In yet further examples, the server 130 may receive a rejection message from the second type of RAT device 120 and may then send a rejection message on to the first type of RAT device 110.

[060] It will be appreciated that the reallocation procedure 100 illustrated in Figure 1 is exemplary and other procedures will be apparent to the skilled person based on the disclosure herein, particularly based on the descriptions with reference to Figures 6 and 7A to 7C. For example, it will be understood that that the server 130 may initiate the reallocation procedure 100 or that the first type of RAT device 110 may send the reallocation request to second type of RAT device 120 (rather than to the server 130).

[061] Figure 2 illustrates a system implementing spectrum allocation between two, different types of RAT for a shared frequency band. In the example illustrated in Figure 2, the two different types of RAT are Wi-Fi 6E (defined in the IEEE standard 802.11ax) and International Mobile Technology, IMT (and specifically, 5G) but it will be appreciated that the methods may be implemented for other types of RAT and other combinations of two or more RATs. IMT is a generic term used to designate broadband mobile (cellular) systems and encompasses, for example, the standards defined by IMT-2000, IMT- Advanced and IMT-2020.

[062] The system comprises a server 230 that stores spectrum allocation information. The spectrum allocation information includes identifiers of devices configured to use a type of RAT and information identifying portions of the frequency band that are allocated to those devices, if any. If no portion of the frequency band has been allocated to a device, this may be indicated by the information.

[063] The IMT has been allocated a first portion 214 of the shared frequency band and the Wi-Fi has been allocated a second portion 224 of the shared frequency band. In Figure 2, the first portion 214 is a larger portion of the shared frequency band (spectrum slice) than the second portion 224, with the IMT operating on channel 103 with a bandwidth (BW) of 80MHz and the Wi-Fi 6E operating on a 40MHz spectrum slice on channel 115. This information (the channel numbers and / or the range of frequencies over which the RATs may operate) is stored by the server 230.

[064] In Figure 2, the IMT has been allocated a larger spectrum, which may be at least in part because the IMT UEs 212 and Wi-Fi UEs 222 are operating in an outside region (that is, outside of a building). IMT may be more often used or provide a higher quality of service in outside regions. Wi-Fi may correspondingly be allocated a larger spectrum for indoor (inside building) use. In other words, IMT may be given priority outdoors, whilst Wireless Access Systems including Radio Local Area Network (WAS / RLAN) may be given priority indoors. Higher priority may correspond to a higher number of spectrum slices being given to a particular RAT. However, in other scenarios, other frequency allocation criteria may be used.

[065] For example, a frequency range may be made exclusively available for one RAT as a spectrum slice and the frequency range size can be dynamically changed based on, for example, load requirements. This may be referred to as a band-split approach.

[066] In another example, one RAT may be allocated a larger or smaller frequency band range in a particular geographic area. In other words, the spectrum allocation may be separated based on a geographical region. For instance, one RAT may be given priority or assigned lower priority depending on the geographical area. Higher priority may correspond to a higher number of spectrum slices allocated to a particular RAT. The geographical area may be determined based on a region type (for example, city, town, rural, and so on), a base station location (for instance, within a certain geographical area or distance from the base station), historical information (for example, information indicating that IMT network traffic is typically or occasionally higher within a particular geographical region), and so on.

[067] The move towards 6G is showing that IMT requires more spectrum though, as applications are requiring greater amounts of data. If a base station (such as the base station 210 illustrated in Figure 2) is not able to handle the network load, then it will not be able to provide the necessary network services. IMT may thus be at least initially allocated a large portion of spectrum regardless of other factors.

[068] Both IMT and WAS / RLAN may have the highest spectrum requirements in densely populated areas. The present scenario aims at enabling IMT connectivity outside buildings whilst also enabling WAS / RLAN connectivity inside buildings. The methods and systems discussed herein also more generally allow a network to manage network loads efficiently and effectively to maintain or limit loss of quality of service. This may be achieved by a Wi-Fi router 220 or base station 210 (or another suitable RAT device) being configured to implement any of the methods discussed herein.

[069] If the number of UEs 212 accessing the IMT increases (for example, beyond a threshold limit), the IMT may require a greater portion of the spectrum to maintain the level of network services or limit the reduction in network services provided to each UE 212 (for example, maintaining a data-rate, loss, jitter, and so on). A request may therefore be sent to the base station 210 (for example, by any devices that is part of the IMT network structure) to increase the spectrum allocation of the IMT. For example, a low power base station serving a microcell may send the request to a base station serving a larger area. The base station 210 then sends a request to the server 230 (labelled 1 in Figure 2). In other examples, the base station 210 may send the request to the server 230 without receiving a request from another entity.

[070] Based on the stored information, the server 230 determines that the Wi-Fi portion 224 of the spectrum can be reduced without impacting service (or at least not significantly impacting service) to the devices operating using Wi-Fi. The server 230 thus sends an instruction to a Wi-Fi router 220 to stop transmitting on channel 115 (labelled 2 in Figure 2). In the example illustrated in Figure 2, the server 230 also instructs the Wi-Fi router 220 to shrink its bandwidth and start transmitting on channel 117. It will be appreciated that these additional instructions may be optional. For example, if the Wi-Fi is sharing another portion of the spectrum with a different RAT and can use that portion (for instance, by submitting its own request for spectrum reallocation), the Wi-Fi router 220 may not need to shrink its bandwidth. Similarly, if the remaining channel(s) allocated to the Wi-Fi technology are sufficient to handle the number of UEs 222 in the area, an additional channel may not be required.

[071] In response to receiving the instruction to cease transmissions over channel 115, the Wi-Fi router 220 ceases transmissions over the communication channel. The Wi-Fi router 220 also sends an acknowledgement to the server 230 to acknowledge the instructions of the server 230 and / or to confirm that the channel is now empty. This step is labelled 3 in Figure 2. The server 230 then communicates to the base station 210 that the channel is empty (labelled 4 in Figure 2), indicating that the base station 210 may now begin communications over the channel 117. After the spectrum reallocation, the IMT is thus allocated a third portion 215 of the shared frequency band (which in this case, includes the first portion 214) and the Wi-Fi has been allocated a fourth portion 225 of the shared frequency band (which in this case, overlaps with, but is different to, the second portion 214).

[072] In summary, Figure 2 illustrates how two different radio access technologies may adaptably manage spectrum allocation, with a defined procedure for a first radio access technology to request additional spectrum and a second radio access technology to release spectrum as and when necessary.

[073] An exemplary exchange of messages between the database server 230, Wi-Fi router 230 and base station 210 is provided in Table 1 below. It will be understood that where different RAT technologies are used, the messages exchanged may be between other entities or other message content may be used. Source Destination Bit (Meaning) Action 5G / 6G base station (BS) with base station ID (BS-ID) Server 1 (require new spectrum) Ask Wi-Fi to change spectrum. The server is aware of the spectrum availability of the given geographic area Server Wi-Fi 6E 1 (change channel number) Wi-Fi to change channel number Wi-Fi 6E Server 1 (acknowledgement) Confirm channel is empty Server 5G / 6G BS 1 (acknowledgement) Confirm channel is empty

[074] Table 1: An exemplary exchange of messages between three different entities with source, destination, bit and action

[075] In the scenario illustrated in Figure 2, the server 230 instructs the Wi-Fi router 220 to reduce its bandwidth because the original frequency band range (6.425 to 6.545GHz) remains the same. For example, there may not be additional spectrum spare for the Wi-Fi RAT to maintain its previous full bandwidth, as it may be allocated to another RAT that still requires use of that spectrum slice. However, in other scenarios, other actions may be taken.

[076] Tables 2 to 4 below illustrate exemplary information that may be stored by the server to facilitate the dynamic spectrum allocation shown in Figure 2. Base Station Wi-Fi 6E BS-ID Channel No. Frequency Active? Wi-Fi ID Channel No. Frequency Active? bsXooo 103 6.425- 6.505 Yes wiXooo 115 6.505- 6.545 Yes

[077] Table 2: Exemplary spectrum allocation information stored by a server for dynamic spectrum allocation corresponding to steps 1 and 2 in Figure 2.

[078] Table 2 shows the stored information that may be stored by server 230 for a given geographic area. The information comprises an identifier of the base station 210 that is configured to transmit communications using I MT (which may specifically be 5G / 6G communications) and an identifier of the Wi-Fi router 220 that is configured to transmit communications using Wi-Fi (which may specifically be Wi-Fi 6E or Wi-Fi 7). The spectrum allocation information also includes information identifying portions of the frequency bands allocated to the base station 210 and Wi-Fi router. In Table 2, this information is provided by both a channel number and a frequency band range, but in other examples, only one of the channel number and range of frequencies may be provided.

[079] The information in Table 2 indicates that the base station having a base station identifier of bsXooo is transmitting over channel 103 (6.425-6.505 GHz), whilst the Wi-Fi router having a base station identifier of wiXooo is transmitting over channel 115 (6.505-6.545 GHz). This corresponds to the situation at process steps 1 and 2 indicated in Figure 2. After the Wi-Fi router 220 has confirmed that channel 115 is empty (step 3 in Figure 2), the server 230 may update the stored spectrum allocation information as shown in Table 3 below. Base Station Wi-Fi 6E BS-ID Channel No. Frequency Active? Wi-Fi ID Channel No. Frequency Active? bsXooo 103 6.425- 6.505 Yes wiXooo 117 6.525- 6.545 Yes

[080] Table 3: Exemplary spectrum allocation information stored by a server for dynamic spectrum allocation corresponding to step 3 in Figure 2.

[081] Similarly, when (or after) confirming to the base station 210 that the previous Wi-Fi channel is empty, the server 230 may update the stored spectrum allocation information as shown in Table 4 below. Base Station Wi-Fi 6E BS-ID Channel No. Frequency Active? Wi-Fi ID Channel No. Frequency Active? bsXooo 103, 113 6.425- 6.505, 6.505- 6.525 Yes wiXooo 117 6.525- 6.545 Yes

[082] Table 4: Exemplary spectrum allocation information stored by a server for dynamic spectrum allocation corresponding to step 4 in Figure 2.

[083] The updated frequency ranges may be stored as separate ranges - for example, to correspond to the channel numbers indicated by the information, or to indicate that one or more frequency ranges are temporary or reallocated frequency ranges. This may allow the spectrum allocation information to be updated in a more straightforward manner.

[084] Figures 1 and 2 have been described above with reference to a server 130, 230 that may be used to provide communications between a first device configured to communicate using a first RAT and a second device configured to communicate using a second RAT. In other embodiments, the server 130, 230 may specifically be a database server storing transmission parameters (spectrum allocation information). In this case, devices or radio access technologies may access the database server for the transmit parameters and may communicate with a different device or radio access technology based on the stored spectrum allocation information (rather than the server facilitating the communications between the devices or radio access technologies). The transmit parameters may include a list of one or more channels or a range of frequencies that are available for transmission. The device may send a database query to access the database server, the database query including a geographic location or region of the requesting device. The database response can then include spectrum allocation information relevant or most relevant to the requesting device.

[085] Figure 3 illustrates a system 300 comprising a UE 310 and a user plane function (UPF) 320. A UPF implements packet routing and forwarding functions in a radio access technology network. The UE 310 is configured to communicate over a first frequency range using a first radio access technology (in this case, a radio access technology defined by 3GPP standards, such as, for example, 3G, 4G, 5G, 6G and so on) and communicate over a second frequency range using a second radio access technology (in this case, a non-3GPP technology, such as Wi-Fi, LTE-U or fixed line, for instance). The 3GPP RAT has been allotted a 40MHz spectrum slice between 4.465 and 6.505GHz, whilst the non-3GPP RAT has been allocated a 40MHz spectrum slice between 4.425 and 4.465GHz.

[086] Communications from the UE 310 are sent to the UPF 320 using the second radio access technology over channel 99. In this case, the spectrum slice comprises two communications channels, each having respective Quality of Service, QoS, parameters. Communications using the first radio access technology are sent from the UE 310 to the UPF 320 using channel 107.

[087] Referring now to Figure 4, the range of frequencies allocated to the non-3GPP access technologies has been reconfigured compared to Figure 3. In particular, the non-3GPP spectrum slice has been extended to allow the non-3GPP access technologies to communicate over the frequency range 4.425 to 4.485GHz (a 60MHz range) by splitting the 40MHz channel 107 into two 20MHz channels (channels 105 and 109), one of which is allocated to the non-3GPP RAT and the other of which is allocated to the 3GPP RAT. The non-3GPP access technologies thus may communicate over three different communication channels, whilst the 3GPP access technologies may communicate over a single communication channel.

[088] Figure 5 illustrates a similar scenario where the range of frequencies allocated to the 3GPP access technologies has been extended. The non-3GPP access technologies may thus communicate over the over the frequency range 4.425 to 4.445GHz (a 20MHz range), whilst the 3GPP access technologies may communicate over the frequency range 4.445 to 6.505GHz by splitting the 40MHz channel 99 into two 20MHz channels (channels 97 and 101), one of which is allocated to the non-3GPP RAT and the other of which is allocated to the 3GPP RAT.

[089] Figure 6A illustrates a flow chart of a method for spectrum reallocation of a frequency band assigned for use by two or more different types of radio access technology. At step 601, a request to increase a spectrum allocation of a first type of RAT is communicated. Communicating, as used herein, may refer to sending or receiving. For example, the request may be sent by a first device configured to operate over one or more communication channels of the frequency band using the first type of RAT to a second device configured to operate over a communication channel of the frequency band using a second, different type of RAT. In another example, request may be sent from the first device to the second device via a server. In a further example, the request may be sent by the server to the second device without the server receiving a request from the first device. In yet another example, the request may be received by the second device (for instance, from the server or the first device).

[090] At step 602, a communication indicating that communications using the second type of RAT will not occur is transmitted or communicated. Again, this may, in accordance with the methods and systems described herein, refer to receiving or sending the communication. For instance, the first device may receive the communication indicating that the communications using the second type of RAT will not occur. This communication may be received from the server or the second device or another device. In another example, the second device may send the communication indicating that the communications using the second type of RAT will not occur. This may be received by the first device or server or another device. In a further example, the server may send the communication, which may be received by the first device or another device.

[091] Figures 7A to 7C illustrate flow charts of some of the various methods discussed above with reference to Figure 6. It will be appreciated that these methods may be, but need not be, interrelated. Figures 7A-C illustrate a possible correspondence between the various method steps illustrated in the flowcharts of Figures 7A to 7C. Other correspondences may be possible.

[092] Figure 7A illustrates a flow chart of a method for spectrum reallocation a frequency band assigned for use by two or more different types of radio access technology. At step 701, a first device configured to operate over one or more communication channels of the frequency band using a first type of RAT sends a request to increase the spectrum allocated to the first type of RAT. The request may be sent to, for example, a server (Figure 7C) or a second device configured to operate over at least one communication channel of the frequency band using a second type of RAT (Figure 7B).

[093] At step 702, a communication indicating that communications over at least one communication channel using the second type of RAT will not occur is received by the first device. This communication may be received from a server (Figure 7C) or a second device configured to operate over at least one communication channel of the frequency band using a second type of RAT (Figure 7B). The communication may, for example, indicate that the communications using the second type of RAT have ceased or that the channel is not being, was not being and / or will not be used.

[094] In step 703, having received the communication indicating that communications using the second type of RAT will not occur, the first device begins to communicate using the first type of RAT over the at least one communication channel.

[095] Figure 7B illustrates a flow chart of a method for spectrum reallocation a frequency band assigned for use by two or more different types of radio access technology. At step 711, a second device configured to operate over one or more communication channels of the frequency band using a second type of RAT receives a communication indicating that the spectrum allocation of the second type of RAT should be adjusted is received. The communication may, for example, comprise a request to increase a spectrum allocation of a first, different type of RAT operating over one or more communication channels of the frequency band or an instruction to cease communications over at least one communication channel of the one or more communication channels by the second type of RAT. The communication may be received from, for example, a server (as will be discussed with reference to Figure 7C) or a first device configured to operate over at least one communication channel of the frequency band using a first, different type of RAT (Figure 7A).

[096] At step 712, the second device ceases communications over at least one communication channel of the one or more communication channels using the second type of RAT.

[097] In step 713, the second device sends a communication indicating that the communications using the second type of RAT have ceased. This communication may be sent to a server (Figure 7C) or the first device (Figure 7A).

[098] As discussed above with reference to Figure 1, in some examples, steps 712 and 713 may not occur and the second device may instead send a rejection message indicating that it cannot or will not cease communications over the at least one communication channel (for example, because the second radio access technology is congested and the at least one communication channel is still required).

[099] Figure 7C illustrates a flow chart of a method for spectrum reallocation a frequency band assigned for use by two or more different types of radio access technology. At step 721, a server receives a request to increase a spectrum allocation of a first type of RAT. This request may be received from a first device configured to operate over one or more communication channels of the frequency band using the first type of RAT. The request may include an identifier of the first device, a geographic location of the first device and / or any other information that may help identify the first device and help the server determine how the spectrum may be reallocated.

[0100] The server may store spectrum allocation information comprising an identifier of the first device, an identifier of a second device configured to operate over at least one communication channel of the frequency band using a second, different type of RAT, and information identifying portions of the frequency band allocated to each of the first and second devices. The server may thus be able to identify that the second device is not transmitting over a particular communication channel or set of communication channels. The server may thus proceed to step 722.

[0101] In other examples, the spectrum allocation information may indicate that there are no suitable communication channels free. The server may thus identify that the second device should therefore stop communicating over at least one communication channel or inform the server that it is not using at least one communication channel allocated to it so that this can instead be used by the first device. The server may thus instruct the second device to stop communicating over a communication channel of the set of communication channels allocated to the second type of RAT. In response to the instruction, the second device may cease communications. In other examples where the second device is already not using a communication channel of the communication channels allocated to the second type of RAT, the second device may return a response indicating that the channel is unused and will not be used.

[0102] At step 722, the server sends a communication to the first device indicating that communications using the second type of RAT over a communication channel of the frequency band will not occur. The first device may then proceed with method steps 702 and 703 in Figure 7A.

[0103] It will be appreciated that, in other examples, the steps illustrated in Figure 7C may be performed by the second device. For example, the second device may have memory or a database storing spectrum allocation information locally, such that it can adjust the spectrum allocation of the second type of RAT and update the spectrum allocation locally. The second device may in this case send a message to an external server or database to indicate that the spectrum allocation of other devices using the second type of RAT should be adjusted. The external server or database may then send instructions to the other devices to adjust their spectrum allocation as well.

[0104] Figure 8 illustrates a device 800 configured to operate over one or more communication channels of a frequency band using a first type of RAT and at least one communication channel of the frequency band using a second type of RAT. For instance, the device may server as both a Wi-Fi router 820 and a 5G / 6G base station 810 and may operate in the sub-6GHz frequency band and converge to a unified core network. It will be appreciated that the device may operate as one or more additional RAT devices though or as a different combination of RAT devices.

[0105] The device may be configured to use access traffic steering, switching and splitting, (ATSSS) and / or Multipath Quick UDP Internet Connection (MPQUIC). 5G, for example, enables multiple data paths - for instance, one set of traffic using a radio access technology defined by 3GPP standards and one using a non-3GPP technology. The non-3GPP technology connection may be further divided into a trusted and untrusted or non-trusted connection.

[0106] ATSSS control is normally handled by a policy control function (PCF) that provides ATSSS rules to devices and a session management function (SMF) based on service provider policy. ATSSS rules dictate how a device should utilise 3GPP and non-3GPP access networks it may have available, specifically with respect to sending uplink traffic. For downlink traffic, rules can be provided to a UPF, which dictate which access network should be used for which traffic flow.

[0107] The device and / or a UPF may be configured to implement ATSSS at a lower layer (ATSSS-LL). ATSS-LL functionality can be applied to all higher layer traffic types and can include the following conditions: Active / Standby. For a given traffic flow, one of the access networks (3GPP or non-3GPP) is categorised as active and the other as standby. The standby access network is only used for the traffic flow if the active network is unavailable. Smallest delay. This steers a particular network traffic flow to the access network having the smallest delay. Load balancing. This splits a traffic flow across both access networks, based on a percentage split (for example, 50% to each). Priority. This will steer a traffic flow to the higher or highest priority access network. When the higher priority access network becomes congested, traffic can also be split towards the lower priority network.

[0108] Other conditions local to the device (for instance, a network interface availability or signal loss conditions) may also be considered. Higher layer steering functionality (ATSSS-HL) may be used as well or instead (for instance, where multipath QUIC is not used - for example, where multipath TCP (MPTCP) or other multiple traffic flow implementations are used).

[0109] The combination of ATSS-LL with MPQUIC may allow seamless and efficient data transmission across different network interfaces. By integrating these technologies in the device 800, the device 800 can improve or optimise traffic management and enhance reliability and performance. Thus, the dynamic spectrum slicing, which may allow efficient allocation of spectrum resources to meet the diverse needs of various applications, can be enhanced.

[0110] The device 800 may further comprise a local database 830 that stores information about how the spectrum is allocated between the first RAT device 810 and second RAT device 820, which may be accessible to both RAT devices 810, 820. The local database may not store separate identifiers of the first and second RAT devices 810, 820, but may store an identifier of the dual function device 800 (for example, for sending to an external database, such as database 230 illustrated in Figure 2). In other examples, the local database 820 may not be present and the device 800 may be configured to communicate with an external database, as described herein.

[0111] The convergence and advanced spectrum management provided by the device can allow unparalleled connectivity by combining the advantages of each respective RAT (in this case, Wi-Fi and cellular technologies) and future-proof, high performance networking solutions.

[0112] It will be appreciated that, whilst the present disclosure is discussed mainly with reference to sending requests for increasing a spectrum allocation, the methods and systems discussed herein may instead operate to offer a communication channel that is allocated to one type of RAT but that is not being and will not be used (for example, there are fewer UEs in a geographic region than expected) to a different type of RAT. This offer may be transmitted via the server, which may update the spectrum allocation information accordingly.

[0113] It will be appreciated that the channel numbers discussed herein are exemplary and that other channels and frequency ranges or frequency bands may be used.

[0114] Figures 1 to 8 described above have been described mainly from a radio access network, RAN, or RAT perspective. However, spectrum allocation and / or ATSSS may also be considered from the perspective of a core network, which may implement its own (different or additional) methods and systems for spectrum allocation and / or ATSSS.

[0115] Whilst 3GPP standards have considered some steering modes for network traffic being transmitted via non-3GPP and 3GPP accesses (in particular, in the Release 17 (Rel-17) report), these considerations do not take into account how to effectively distribute or split this network traffic. Indeed, most relate solely to steering or switching traffic, and do not consider splitting network traffic at all.

[0116] For example, the active / standby mode discussed above sends all network traffic via one access at a time. The access network used is switched when the active access becomes unavailable. When the active access network becomes available again, traffic is sent again via this network. This fails to take advantage of the possibilities offered by network traffic splitting.

[0117] Similarly, priority-based steering sends all traffic via a high priority access network until congestion arises, whereafter new data flows (overflow traffic) are sent to the low priority access (and if the high priority access becomes unavailable, all traffic is switched to the low priority access). Again, this fails to take advantage of the possibilities offered by network traffic splitting, since traffic is only steered or switched. It is also acknowledged in current 3GPP standards documentation that identifying when congestion arises on the high priority access may be a complex task, and no solution is proposed to address this.

[0118] Best-access steering is the same as priority-based steering, but the high priority access is the one that can provide the best performance. Again, this sends all traffic via one access network only: via the network identified as the best-access network until congestion is identified on the bestaccess network, whereafter the overflow traffic is sent via the other access network. As with prioritybased steering, Rel-17 acknowledges that identifying when congestion has arisen may be a complex task and offers no solution to address this.

[0119] Redundant steering also exists, whereby all traffic is sent via both networks to provide high data reliability (low packet error rate). However, this can be inefficient and complex for a receiving side, which should detect and discard duplicate packets. As such, this steering fails to take advantages of the possibilities of traffic splitting. In particular, access traffic splitting involves sending some traffic of a data flow via one access and some other data of the same data flow via another access. Redundant steering instead sends all data via both accesses.

[0120] Load balancing, as described above, splits traffic across both access networks based on a percentage split. However, this percentage split is predefined and thus fails to take account of changing network conditions.

[0121] In any case, whilst these steering modes are described Rel-17, Rel-17 notes that these steering modes should be further evaluated in terms of complexity before deciding to support them for ATSSS. Thus, whilst these modes are described, Rel-17 does not actually describe how to implement them.

[0122] Furthermore, it could be assumed previously that different access networks were mostly or completely separate, which entailed its own way of splitting traffic. Now, however, it is expected that the same core network may be used for different types of access network, which may change how network splitting should be implemented.

[0123] Figure 9 illustrates a flow chart of a method for distributing or splitting network traffic across two or more different types of RAN. At step 901, a distribution configuration for distribution of network traffic to a UE via a first type of radio access network (RAN) and a second, different type of RAN is determined based on one or more network performance metrics. The distribution configuration defines a non-zero network traffic distribution factor for each of the first and second types of RAN. In other words, both the first and second type of RAN are to be allocated a portion of a data flow according to the distribution configuration.

[0124] This may be useful in various scenarios including, for example, in geographical regions or locations where connection to one RAN is poor. Some locations (for example, city buildings, opera houses, underground locations, and so on) may typically provide limited connection via some RANs (for example, due to building materials, high network traffic or other factors). However, another RAN may be available to provide connection in such locations. The networks can accordingly deliver a first percentage of network traffic over one RAN (for example, non-time-sensitive data) and a second percentage of network traffic over a different RAN.

[0125] The distribution configuration may also define or comprise other factors for distributing network traffic, in addition to the ratio-based distribution. For example, the distribution configuration may be further based on a type of network traffic. A type of network traffic may be a type of transmitted data - for example, voice or audio, video, data, metadata, and so on. For example, the distribution configuration may define that voice data should be allocated to the first type of RAN and that video data should be allocated to the second type of RAN. Other configurations are possible.

[0126] In another example, the distribution configuration may also or instead be further based on a network traffic use. Networks are being used for an increasing number of varied applications, and different applications or uses may provide better user experience or more efficient network resource use with a different distribution configuration. Some uses that may involve more complex forms of network traffic or multiple forms of network traffic, and therefore may benefit more greatly from appropriate network traffic splitting, may include extended reality (XR), voice over IP (VoIP), videoconferencing and so on. Other uses can also be considered in the determination of the distribution configuration, including web-surfing, audio calls, or any other network use.

[0127] In yet a further example, the distribution configuration may also or instead be further based on one or more network traffic distribution rules. For instance, data identified as time-critical (for example, video data, data requiring low latency or jitter, and so on) may be transmitted via the first type of RAN and metadata (which may be less time-critical) may be transmitted via the second type of RAN, or vice versa.

[0128] At step 902, the network traffic is distributed between the first type of RAN and the second type of RAN according to the distribution configuration. Thus, the available network resources are used and can be used more effectively. This may in turn improve network performance.

[0129] The method discussed with reference to Figure 9 may be capable of providing improved network reliability and user experience, as well as enabling the efficient management of available network resources. This may particularly be the case where, as discussed with reference to Figures 1 to 8, the first and second types of RAN are assigned to the same or an overlapping frequency band. Determining how to split or distribute network traffic across two different types of RAN that are assigned to at least a portion of the same radio spectrum can entail additional complexities. For example, portions of the spectrum may need to be reallocated before the distribution configuration can be implemented, or being aware that the RANs are operating using a shared or overlapping frequency band can impact the determination of an appropriate distribution configuration - for instance, by assigning more traffic to a RAN with low network performance metrics under the assumption that the frequency band allocated to that RAN will be extended to improve the network performance metrics.

[0130] Accordingly, determining the distribution configuration at step 901 may further be based on a spectrum allocation of the first type of RAN and the second type of RAN, the first and second types of RAN being assigned to use a same or overlapping frequency band. The operation of this same or overlapping frequency band, as well as how the two different types of RAN may implement sense and avoid techniques, are discussed above with reference to Figures 1 to 8. Other operations and sense and avoid techniques may be implemented, however.

[0131] The distribution configuration may be based on the spectrum allocation according to the relative proportion of the frequency band that is assigned for use by the first type of RAN and the second type of RAN. For example, where 70% of the frequency band is assigned to the first type of RAN and the remaining 30% is assigned to the second type of RAN, this may indicate a relative proportion of traffic for the two types of RAN (which may be a current, previous or expected proportion of traffic). The first RAN may accordingly be assigned a 70% distribution factor, while the second type of RAN is assigned a 30% distribution factor. These distribution factors may be further adjusted based on the one or more network performance metrics or other factors.

[0132] The method may further comprise sending a communication indicating that the spectrum allocation of the first type of RAN and / or the second type of RAN should be adjusted. This communication may be similar to those discussed with reference to Figures 1 to 8.

[0133] The communication may be sent prior to step 901. For example, the communication may be used to initiate spectrum reallocation prior to determining the distribution configuration. In other words, the communication may not specify how the spectrum allocation should be adjusted but merely be a prompt for adjusting spectrum allocation, if necessary (and thus, no spectrum reallocation may occur if the existing spectrum allocation is appropriate). This may allow the distribution configuration to be determined based on a more up-to-date or appropriate spectrum allocation. For example, the first type of RAN may be operating sufficiently well with the current spectrum allocation and so may not need to request any spectrum from the second type of RAN. However, the second type of RAN may also not be operating across the full spectrum allocated to it (for example, because the current network load does not require it). Performing the reallocation before determining the network splitting configuration may provide a more accurate assessment of the current network performance and thus allow a more appropriate distribution configuration to be determined.

[0134] In other examples, the communication may be sent following step 901. For example, the distribution configuration may suggest distributing more traffic to one type of RAN based on the one or more network performance metrics, but it may be determined that the existing spectrum allocation is insufficient or expected to be insufficient to maintain the one or more network performance metrics for the distribution configuration. The communication may thus indicate that the spectrum allocation of the first type of RAN and / or the second type of RAN should be adjusted, based on the distribution configuration. This communication may indicate a proportional adjustment based on the distribution configuration. For instance, if the distribution factor for the first type of RAN is 60% and the second type of RAN is 40%, the first type of RAN may be allocated to 60% of the shared or overlapping spectrum and the second type of RAN may be allocated to 40% of the shared or overlapping frequency band.

[0135] The communication may be sent to a server storing spectrum allocation information as described with reference to Figures 1 to 8. The server may then initiate spectrum reallocation of the first and / or second type of RAT as described above with reference to Figures 1 to 8.

[0136] Still referring to Figure 9, the method may comprise, prior to step 901, determining, based on the one or more performance metrics, whether to distribute the network traffic between the first type of RAN and the second type of RAN. This may comprise a first and second RAN across which to distribute the network traffic. For example, the network traffic may be 3GPP traffic, but it is determined that another type of RAN (non-3GPP) is also available and that the network traffic should thus be distributed across the two types of RAN.

[0137] In another example, two different types of RAN may be available, but it may be determined that one type of RAN cannot support a network traffic distribution. This may be, for example, because a buffer occupancy for one type of RAN is sufficiently high that it is determined that the one or more network performance metrics would not be improved by (or may be worsened by) attempting to send additional network traffic via that RAN. In another example, a signal strength or quality for one type of RAN may be worse than for another type of RAN. In yet a further example, it may be determined that attempting to use both types of RAN may cause undue interference. Other possibilities for determining that the network traffic should not be distributed are possible. In these cases, step 902 may not take place.

[0138] In other examples, these factors may be taken into consideration in determining the distribution configuration. For instance, where a buffer occupancy for one type of RAN is high, it may be determined that metadata (which may not be time-critical) can be sent via that type of RAN.

[0139] It will be appreciated that, although the distribution factor has been referred to mostly in percentage terms, other equivalent forms of factor may be used. For example, the factor may be a weighting factor.

[0140] As discussed above, the distribution configuration is determined based on one or more performance metrics. One example of a network performance metric that may be used is a traffic flow rate. A traffic flow rate may be a measure of how traffic is being delivered. For example, the traffic flow rate may be a network throughout (also referred to herein simply as throughput), which is a rate of message delivery over a communication channel in a network. Throughput may be measured in bits per second, packets per second or data packets per time slot. The throughput can be impacted by various factors, including physical medium limitations, available processing power of system computer, end user behaviour and so on. It may therefore be a useful factor based on which to determine a distribution configuration.

[0141] The traffic flow rate may also or instead comprise another rate, such as a goodput or bandwidth. A goodput is the amount of useful information bits delivered by the network to a particular destination per unit of time (which may exclude protocol overheads and retransmitted packets, for example). A bandwidth may be a maximum rate of data that can be transferred across a given route. It may define the physical layer useful bit rate, which is the capacity of a communication channel excluding the physical layer protocol overhead.

[0142] Another example of a network performance metric that may be used is a delay parameter. The delay parameter may comprise a network delay, for instance. Examples of network delays include a round-time trip (RTT, also known as a round-trip delay), latency and so on. The RTT is the amount of time it takes for a signal (typically a data packet) to be sent across a channel and to receive an acknowledgement that the signal has been received. The RTT may be related to the latency or end-to-end delay (typically approximated as half the RTT, although this assumes that the two communication paths have the same congestion, number of hops and / or quality of service).

[0143] Another type of delay parameter that may be used is a buffer delay. A buffer delay is the time period for which a data packet must wait in a buffer or queue until it is forwarded. A buffer delay may therefore provide an indication of network performance that may be useful for determining a distribution configuration.

[0144] Yet a further delay parameter that may be used is jitter. Jitter refers to the delay-difference between packets in the same traffic flow. Typically, packets are sent at evenly spaced intervals. However, this interval may fluctuate and cause an irregularity of delay-difference between the packets. This irregularity is known as a jitter. Jitter negatively affects real-time services, such as voice and video services, by creating noticeable intermittence. Other delay parameters are also possible.

[0145] Another network performance parameter that may be considered is an error rate. This may include, for example, a bit error rate (a number of bit errors per unit time), a bit error ratio (a number of bit errors divided by the total number of transferred bits during a particular time interval), a packet error ratio (the number of incorrectly received data packets divided by the total number of received packets), and so on. An error rate may thus be useful for determining a distribution configuration, particularly for real-time application such as XR and VoIP.

[0146] A further network performance parameter that may be used is a packet loss. Packet loss occurs when one or more data packets fail to reach their destination. Packet loss may be caused by errors in transmission (usually more common for transmission across wireless networks), or network congestion. Packet loss is typically measured as a percentage of packets lost compared to packets sent. Packet loss can particularly impact a user QoE for real-time applications such as media streaming or online games. A packet loss can thus be useful for determining a distribution configuration for network traffic.

[0147] Determining the distribution configuration at step 901 may comprise identifying non-zero network distribution factors for each of the first and second types of RAN that are expected to meet one or more quality of service, QoS, criteria. One criteria may be a traffic flow rate above a threshold traffic flow rate value. Another criteria may be a delay parameter below a threshold delay parameter. A combination of these criteria may be used. Typically, high QoS and desirable user experience may be achieved by having both a traffic flow rate above a threshold traffic flow rate value and a delay parameter below a threshold delay parameter value (high traffic flow rate, low delays).

[0148] Similarly, determining the distribution configuration at step 901 may be further based on a QoS parameter established based on one or more network performance metrics. For example, the non-zero distribution factors may be expected to increase the QoS parameter. This may be, but need not be, related to the QoS criteria.

[0149] Determining the distribution configuration may thus comprise identifying a first network path of the first type of RAN and a second network path of the second type of RAN that are expected to increase the QoS parameter. For instance, the QoS parameter may be defined as QoSi j^t) at time t for a user i on network j over path k and determined as 1 K (1) QoS^Q £ £ [W1. + ] i = l j£RAN1,RAN2 fc=l or I K (2) i = l jERAN±,RAN2 k = l where Ti,jk represents a first network performance metric and Dijk represents a second performance for the user i on network j over path k, RANa represents the first type of RAN, RAN2 represents the second type of RAN, and and w2 are weighting coefficients for each metric. That is, 0 <wm <1 for all m G {1,2} and Xm=i wm = 1 ■

[0150] Without taking away from the general formulation above, the methods described herein will now be discussed with reference to a specific implementation.

[0151] In particular, the methods discussed herein may be applied to dynamically distribute traffic across one or more available network paths to enhance QoS for XR applications. For example, the approach can leverage Wi-Fi 6E and 5G and / or 6G networks in a multi-path configuration, managed through an ATSSS framework, which can facilitate seamless traffic distribution across different RATs. At the transport layer, the MP-QUIC protocol may ensure efficient utilization of these paths.

[0152] XR traffic, which may comprise video, audio, and metadata streams, can be split between two different types of RAN (for example, Wi-Fi and 5G networks) based on real-time network performance metrics, such as throughput, delay, jitter, and packet loss, for example. The method may operate to balances these or other metrics to enable high reliability and good user experience.

[0153] The splitting mode may employ a Q-Actor Critic Reinforcement Learning Approach, where the network state and traffic allocation actions guide the policy towards maximizing the QoS. This formulation may be defined as follows.

[0154] For a total number of users I with J different network paths with K different paths, the State S at a time t can be defined as St = {!*, D^}, i G {1,2,3 ... I}, j G {1,2,3 ... J}, K G {1,2,3 ... K} (3) where T / y represents a first network performance metric for a user i on network j over path k and represents a second network performance metric for a user i on network j over path k. Preferably, the first network performance metric is a throughput and the second network performance metric is a delay.

[0155] The action space at G A at time t may consist of traffic splitting ratios for each network and path. For each user i, the action space can be defined as at = H / O) (4) where represents a portion of traffic allocated to path k over network / for user / at time t.

[0156] A reward function R(St, at) representing a QoS at time t can be formulated as a weighted combination of the network performance metrics as R(St, at) = QoSi j^t) (according to Equations 1 or 2 above, where RANI and RAN2 are Wi-Fi and 5G, in this example).

[0157] Typically, high QoS and desirable user experience may be provided by high throughput and low latency. Where Dijk represents a delay, a negative value of w2 may be needed for Equation 1. However, the latency can also be considered in a different way. For example, the delay may be defined as the inverse of a threshold or maximum delay allowed. In this case, Equation 2 may be used.

[0158] The optimisation problem can then be formulated as maxE[yf QoS^j(t)], where y is an “t adjustment parameter that can take values from (0,1].

[0159] Although reference has been made to using a Q-Actor Critic Reinforcement Learning Approach, it will be appreciated that other approaches may be implemented.

[0160] Figure 10 illustrates a schematic system in which the methods discussed herein may be implemented. An XR device 1001 is in communication with a UE 1002. In Figure 10, the communication is provided by a wired (USB 3.0) connection, but other wired connections may be used, or a wireless connection (including, for example, Bluetooth or another short-range communication protocol) may be used.

[0161] The UE 1002 transmits data received from the XR device 1001 to a telecommunications network. XR network traffic may include multiple types of data - for example, video, audio and metadata. In accordance with the present disclosure, some of the XR network traffic is sent via a first RAN device (in Figure 10, a 5G base station 1010) and another portion of this XR network traffic is sent via a second RAN device (in Figure 10, a Wi-Fi router 1020). The distribution of the network traffic between the first and second RAN is determined and allocated as described above with reference to Figure 9. The RAN devices may be also referred to herein as RAT devices.

[0162] The first RAN and second RAN are in communication with a core network 1040. This communication may be in accordance with any 3GPP standards. For example, although Figure 10 illustrates a Y2 interface between Wi-Fi sever 1034 and the core network 1040, and an N2 interface between the 5G server 1032 and the core network 1040, other or additional interfaces may be present.

[0163] Preferably, the core network 1040 determines the distribution configuration of the network traffic. A core network may typically receive or determine network performance metrics. For example, a core network may typically include core network monitoring systems such as, for example, an operational support system or traffic probing system. Network performance metrics may thus be identified at the core network 1040. In other examples, network performance metrics may be determined at the core network 1040 based on information reported by a network node (for example, a RAN device) or a UE. In yet further examples, network performance metrics may be reported by a network node or by a UE.

[0164] The distribution configuration may be determined based on a spectrum allocation of the two RATs or adjusted based on the spectrum allocation, as described above with reference to Figure 9.

[0165] Although Figure 10 has been described with reference to a 5G base station 1010 and a Wi-Fi router 1020, it will be appreciated that other RAN devices may be used. For example, the base station may be a different a RAT device defined by 3GPP standards, such as, for example, 3G, 4G, 6G and so on. Likewise, the Wi-Fi router 1020 may be a different non-3GPP technology, such as LTE-U or fixed line, for instance. In yet further examples, both technologies may be 3GPP RAT technologies or non-3GPP RAT technologies.

[0166] Figure 11 illustrates a schematic system in which the methods discussed herein may be implemented. In a first geographic area or region 1113, a base station 1110 enables communication between one or more UEs 1102a, 1102b and a core network. Although four UEs 1102a are shown (RB1-RB4), it will be appreciated that more or fewer UEs 1102a, 1102b may be present in the first geographic area 1113.

[0167] In a second geographic area 1123, one or more Wi-Fi routers 1120 enable communication between one or more UEs 1102c and a network. The network may be the Internet or may be a core network - for example, in 6G and beyond, it is anticipated that cellular, Wi-Fi and satellite networks may share a common core network. Again, more or fewer UEs 1102c may be present in the second geographic area 1123 than is illustrated in Figure 11. Similarly, although a plurality of Wi-Fi routers 1120 is shown in the second geographic region 1123, it will be appreciated that one Wi-Fi router 1120 may serve each of the UEs (for example, where the second geographic region 1123 is a local geographic area, such as a building or room, for example).

[0168] As shown in Figure 11, one UE 1102b (RB4) may move within the first geographic region 1113. The movement may be autonomous or semi-autonomous. For example, the UE 1102b may be or may be comprised within an autonomous robot or vehicle (for instance, an unmanned aerial vehicle). In other examples, the UE 1102b may otherwise be moveable - for example, being comprised within an automotive vehicle (for example, a car, bus, motorcycle, lorry and so on).

[0169] The UEs 1102a, 1120b are illustrated in Figure 11 as robotic devices (which may be able to navigate independently or semi-independently), but it will be appreciated that the issues and implementations that will be discussed with reference to Figure 11 could apply equally to other types of UE. For example, the movement discussed above may be movement by a user in possession of a UE.

[0170] As the UE 1102b moves, it moves towards the boundary of the first geographic region 1113 that the base station 1110 can serve. Traditionally, this may result in handover or traffic steering or switching once the UE is able to connect to a different network. However, these scenarios fail to consider the possibility that the UE may be able to transmit over both networks for a period of time - for example, in a region of overlap 1133 between the first geographic region 1113 and the second geographic region 1123. Traffic splitting can be used in such regions to provide more efficient network resource use, as well as provide a more seamless transmission between different RANs.

[0171] For example, it may be identified (for instance, at the core network) that, although the UE 1102b is communicating network traffic via a first type of RAN, a second type of RAN is available to the UE for communicating network traffic. A distribution configuration may be determined responsive to identifying that this is the case. As the UE 1102b continues to move through the overlap region 1133, the distribution configuration may be updated. For example, the updates may be determined at regular intervals, based on relative distances between the first and second RAN devices, or based on one or more updated network performance metrics. In some instances, this may mean that the distribution configuration is updated in real-time or near-real-time.

[0172] Figure 12 illustrates a system architecture for implementing network splitting (also referred to herein as network traffic distribution). As shown in Figure 10, ATSSS rules are implemented at the lower layer for non MPTCP flows including, for example, UDP, TCP and Ethernet flows. ATSSS rules for MPTCP flows on the other hand are implemented at the high layer. This division results in different flows being treated in different manners, which adds complexity to the architecture.

[0173] Figure 13 illustrates a system architecture in which ATSSS rules (including traffic splitting) for all flows is handled at the lower layer. This can be achieved using a multipath QUIC tunnelling function (MQTF) operating at the lower layer. A scheduler or intelligent agent can operate at this layer to distribute QUIC packets between two or more RANs (for example, a 3GPP access and a non-3GPP access as show in Figure 13) according to a distribution configuration of the ATSSS rules. The scheduler can be configured to implement any of the methods discussed herein. As can be seen from Figure 13, this system architecture can significantly simplify the implementation of ATSSS rules.

[0174] Figure 14 illustrates an exemplary layer architecture that may be used to implement the methods discussed herein. As shown in Figure 14, a scheduler (which may be as discussed above with reference to Figure 13) may reside between the application layer and the transport layer. The scheduler receives data streams via a multipath protocol, which may include MPTCP and / or MPQUIC, as shown in Figure 14. The scheduler can be configured to implement any of the methods discussed herein and in particular, implements the ATSSs rules discussed herein to apply the distribution configuration. The scheduler may be implemented by a server arranged within the core network.

[0175] The methods described herein may be implemented with computer system configurations including hand-held devices, microprocessor systems, microprocessor- based or programmable consumer electronics, minicomputers, mainframe computers and the like. The embodiments can also be practiced in distributed computing environments, where tasks are performed by remote processing devices that are linked through a network.

[0176] Certain embodiments can also be embodied as computer-readable code on a non-transitory computer-readable medium. The computer readable medium is any data storage device than can store data, which can thereafter be read by a computer system. Examples of the computer readable medium include hard drives, network attached storage (NAS), read-only memory, random-access memory, CD-ROMs, CD-Rs, CD-RWs, magnetic tapes, and other optical and non-optical data storage devices. The computer readable medium can also be distributed over a network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion. Although embodiments according to the disclosure have been described with reference to particular types of devices and applications (particularly telecommunications and network devices) and the embodiments have particular advantages in such case, as discussed herein, approaches according to the disclosure may be applied to other types of device and / or application. Each feature disclosed in this specification, unless stated otherwise, may be replaced by alternative features serving the same, equivalent or similar purpose. Thus, unless stated otherwise, each feature disclosed is one example only of a generic series of equivalent or similar features.

[0177] All of the aspects and / or features disclosed in this specification may be combined in any combination, except combinations where at least some of such features and / or steps are mutually exclusive. In particular, the preferred features of the disclosure are applicable to all aspects and embodiments of the disclosure and may be used in any combination. Likewise, features described in non-essential combinations may be used separately (not in combination).

[0178] It will be appreciated that there is an implied "about" prior to temperatures, concentrations, times, pressures, flow rates, cross-sectional areas, voltages, currents, etc. discussed in the present teachings, such that slight and insubstantial deviations are within the scope of the present teachings. Furthermore, values referred to as being "equal" may in fact differ by less than a threshold amount. The threshold amount may be 5%, for example. The threshold may also be greater than 5% (e.g., 10%, 20% or 50%) or less than 5% (for example, 2% or 1 %).

[0179] As used herein, including in the claims, unless the context indicates otherwise, singular forms of the terms herein are to be construed as including the plural form and vice versa. For instance, unless the context indicates otherwise, a singular reference herein including in the claims, such as "a" or "an" (such as a communication channel, for instance) means "one or more" (for instance, one or more communication channels). It will also be appreciated that the terms "one or more" and "at least one" are interchangeable.

[0180] Throughout the description and claims of this disclosure, the words "comprise", "including", "having" and "contain" and variations of the words, for example "comprising" and "comprises" or similar, mean "including but not limited to", and are not intended to (and do not) exclude other components. Also, the use of "or" is inclusive, such that the phrase "A or B" is true when "A" is true, "B is true", or both "A" and "B" are true.

[0181] The use of any and all examples, or exemplary language ("for instance", "such as", "for example" and like language) provided herein, is intended merely to better illustrate the disclosure and does not indicate a limitation on the scope of the disclosure unless otherwise claimed. No language in the specification should be construed as indicating any non-claimed element as essential to the practice of the disclosure.

[0182] The terms "first" and "second" may be reversed without changing the scope of the invention. That is, an element termed a "first" element (e.g. a first device) may instead be termed a "second" element (e.g. a second device) and an element termed a "second" element (e.g. a second device) may instead be considered a "first" element (e.g. a first device). It will also be appreciated that other labelling may be used without changing the scope of the invention (for example, referring to the device by its function or use, such as, for instance, an IMT device, XR device or Wi-Fi RAT).

[0183] Any steps described in this specification may be performed in any order or simultaneously unless stated or the context requires otherwise. Moreover, where a step is described as being performed after a step, this does not preclude intervening steps being performed.

[0184] It is also to be understood that, for any given component or embodiment described herein, any of the possible candidates or alternatives listed for that component may generally be used individually or in combination with one another, unless implicitly or explicitly understood or stated otherwise. It will be understood that any list of such candidates or alternatives is merely illustrative, not limiting, unless implicitly or explicitly understood or stated otherwise.

[0185] In this detailed description of the various embodiments, for the purposes of explanation, numerous specific details are set forth to provide a thorough understanding of the embodiments disclosed. One skilled in the art will appreciate, however, that these various embodiments may be practiced with or without these specific details. Furthermore, one skilled in the art can readily appreciate that the specific sequences in which methods are presented and performed are illustrative and it is contemplated that the sequences can be varied and still remain within the scope of the various embodiments disclosed herein.

[0186] All literature and similar materials cited in this application, including but not limited to patents, patent applications, articles, books, treaties and internet web pages are expressly incorporated by reference in their entirety for any purpose. Unless otherwise described, all technical and scientific terms used herein have a meaning as is commonly understood by one of ordinary skill in the art to which the various embodiments described herein belongs.

Claims

1. A method for distributing network traffic across two or more different types of radio access network, RAN, the method comprising steps of:determining a distribution configuration for distribution of network traffic to a user equipment, UE, via a first type of RAN and a second, different type of RAN based on one or more network performance metrics, the distribution configuration defining a non-zero network traffic distribution factor for each of the first and second types of RAN; anddistributing the network traffic between the first type of RAN and the second, different type of RAN according to the distribution configuration.

2. The method according to claim 1, wherein determining the distribution configuration is further based on one or more of: a type of network traffic, a network traffic use and one or more network traffic distribution rules.

3. The method according to claim 2, wherein the type of network traffic comprises a type of data to be transmitted.

4. The method according to claim 2 or claim 3, wherein the network traffic use comprises an extended reality use, voice over IP use or video-conferencing use.

5. The method according to any previous claim, wherein a spectrum allocation of the first type of RAN and a spectrum allocation of the second type of RAN are within a same or an overlapping frequency band.

6. The method according to claim 5, wherein determining the distribution configuration is further based on the spectrum allocation of the first and second types of RAN within the same or overlapping frequency band.

7. The method according to claim 6, wherein the non-zero distribution factor for each of the first and second type of RAN is based on a relative proportion of the spectrum allocation of the first and second types of RAN within the same or overlapping frequency band.

8. The method according to any of claims 5 to 7, further comprising sending a communication indicating that the spectrum allocation of the first type of RAN and / or the second type of RAN should be adjusted.

9. The method according to claim 8, wherein the communication is sent to a server storing spectrum allocation information comprising information identifying portions of the same or overlapping frequency band allocated to the first type of RAN and the second type of RAN.

10. The method according to claim 8 or claim 9, wherein the communication is sent prior to determining the distribution configuration or based on the distribution configuration.

11. The method according to any previous claim, wherein one of first and second type of RAN is a cellular network and the other of the first and second type of RAN is a wireless network.

12. The method according to any previous claim, further comprising, prior to determining the distribution configuration, determining, based on the one or more network performance metrics, whether to distribute the network traffic between the first type of RAN and the second type of RAN.

13. The method according to any previous claim, wherein the one or more network performance metrics comprises a traffic flow rate.

14. The method according to claim 13, wherein the traffic flow rate comprises a network throughput.

15. The method according to any previous claim, wherein the one or more network performance metrics comprises a delay parameter.

16. The method according to claim 15, wherein the delay parameter comprises a network delay.

17. The method according to any previous claim, wherein determining the distributionconfiguration comprises identifying non-zero network distribution factors for each of the first and second types of RAN that are expected to meet one or more quality of service, QoS, criteria.

18. The method according to claim 17, wherein the one or more QoS criteria include a traffic flow rate above a threshold traffic flow rate value and / or a delay parameter below the threshold delay parameter value.

19. The method according to any previous claim, wherein determining the distribution configuration is further based on a QoS parameter established based on the one or more network performance metrics.

20. The method according to claim 19, wherein the non-zero network distribution factors are expected to increase the QoS parameter.

21. The method according to claim 19 or claim 20, wherein determining the distribution configuration comprises identifying a first network path of the first type of RAN and a second network path of the second type of RAN that are expected to increase the QoS parameter.

22. The method according to any of claims 19 to 21, wherein the QoS parameter, QaS^t), at time t for a user i on network j over path k is determined asI KQoS^j (t) = [my + w2D^ ]i = l JERAN1,RAN2 k = l ori = l JERAN1,RAN2 k = lwherein Tijk represents a first network performance metric and Dijk represents a second performance for the user i on network j over path k, RANa represents the first type of RAN, RAN2 represents the second type of RAN, and and w2 are weighting coefficients.

23. The method according to any previous claim, further comprising, prior to determining the distribution configuration, communicating network traffic to the UE via one of the first and second type of RAN, andwherein determining the distribution configuration is responsive to identifying that the other one of the first and second type of RAN is available to communicate network traffic to the UE in addition to the one of the first and second type of RAN.

24. A computer program comprising instructions which, when executed by a computer, cause the computer to perform the method of any of claims 1 to 23.

25. A computer-readable storage medium comprising instructions which, when executed by a computer, cause the computer to perform the method of any of claims 1 to 23.

Images