Interference management

Abstract

Example embodiments of the present disclosure relate to a solution for improving interference management (IM) In an aspect, a first device receives, from a second device, inference capability information of the second device for performing an inference task associated with interference management. The first device determines, based on the inference capability information, an inference task indicator of whether to enable or disable the inference task at the second device. Then, the first device transmits the inference task indicator to the second device. In this way, inference task is contextually activated based on the inference task indicator at the second device for improving IM performance and consequently improving the radio resource management performance.

Description

INTERFERENCE MANAGEMENT FIELD

[0001] Various example embodiments relate to the field of communication and in particular,to devices, methods, apparatuses and computer readable storage media for improvinginterference management.BACKGROUND

[0002] A communication network can be seen as a facility that enables communicationsbetween two or more communication devices, or provides communication devices access to a data network. A mobile or wireless communication network is one example of a communication network.

[0003] Such communication networks operate in according with standards such as thoseprovided by 3GPP (Third Generation Partnership Project) or ETSI (European Telecommunications Standards Institute). Examples of standards are the so-called 5G (5th Generation) standards provided by 3GPP. SUMMARY

[0004] In general, example embodiments of the present disclosure provide a solution forimproving interference management, especially, for improving interference management ina hybrid resource allocation (HRA) mode.

[0005] In a first aspect, there is provided a first device. The first device comprises at leastone processor and at least one memory storing instructions that, when executed by the at leastone processor, cause the first device at least to: receive, from a second device, inferencecapability information of the second device for performing an inference task associated with interference management; determine, based on the inference capability information, an inference task indicator of whether to enable or disable the inference task at the second device; and transmit the inference task indicator to the second device.

[0006] In a second aspect, there is provided a second device. The second device comprisesat least one processor and at least one memory storing instructions that, when executed bythe at least one processor, cause the second device at least to: transmit, to a first device,inference capability information of the second device for performing an inference task associated with interference management; and receive, from the first device, an inferencetask indicator of whether to enable or disable the inference task at the second device.

[0007] In a third aspect, there is provided a method implemented at a first device. Themethod comprises: receiving, from a second device, inference capability information of thesecond device for performing an inference task associated with interference management; determining, based on the inference capability information, an inference task indicator ofwhether to enable or disable the inference task at the second device; and transmitting theinference task indicator to the second device.

[0008] In a fourth aspect, there is provided a method implemented at a second device. Themethod comprises: transmitting, to a first device, inference capability information of thesecond device for performing an inference task associated with interference management; and receiving, from the first device, an inference task indicator of whether to enable or disable the inference task at the second device.

[0009] In a fifth aspect, there is provided an apparatus. The apparatus comprises: meansfor receiving, from a second device, inference capability information of the second device for performing an inference task associated with interference management; means for determining, based on the inference capability information, an inference task indicator of whether to enable or disable the inference task at the second device; and means for transmitting the inference task indicator to the second device.

[0010] In a sixth aspect, there is provided an apparatus. The apparatus comprises: meansfor transmitting, to a first device, inference capability information of the second device for performing an inference task associated with interference management; and means for receiving, from the first device, an inference task indicator of whether to enable or disable the inference task at the second device.

[0011] In a seventh aspect, there is provided a non-transitory computer readable mediumcomprising program instructions that, when executed by an apparatus, cause the apparatus atleast to perform the method according to any of the third and fourth aspects.

[0012] In an eighth aspect, there is provided a computer program comprising instructions,which, when executed by an apparatus, cause the apparatus at least to perform the methodaccording to any of the third and fourth aspects.

[0013] In a ninth aspect, there is provided a first device. The first device comprises:receiving circuitry for receiving, from a second device, inference capability information of the second device for performing an inference task associated with interference management; determining circuitry for determining, based on the inference capability information, an inference task indicator of whether to enable or disable the inference task at the second device;and transmitting circuitry for transmitting the inference task indicator to the second device.

[0014] In a tenth aspect, there is provided a second device. The second device comprises:transmitting circuitry for transmitting, to a first device, inference capability information of the second device for performing an inference task associated with interference management; and receiving circuitry for receiving, from the first device, an inference task indicator of whether to enable or disable the inference task at the second device.

[0015] It is to be understood that the summary section is not intended to identify key oressential features of embodiments of the present disclosure, nor is it intended to be used tolimit the scope of the present disclosure. Other features of the present disclosure willbecome easily comprehensible through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Some example embodiments will now be described with reference to theaccompanying drawings, in which:

[0017] Fig. 1A illustrates an example of a network environment in which someembodiments of the present disclosure may be implemented;

[0018] Fig. 1B illustrates an architectural diagram for sub-networks in accordance withsome embodiments of the present disclosure;

[0019] Fig. 2 illustrates an example signaling process for improving interferencemanagement in accordance with some embodiments of the present disclosure;

[0020] Fig. 3 illustrates an example signaling chart for hybrid resource allocation (HRA)model enhancement in accordance with some embodiments of the present disclosure;

[0021] Fig. 4 illustrates a flowchart of an example method implemented at a first device inaccordance with some embodiments of the present disclosure;

[0022] Fig. 5 illustrates a flowchart of an example method implemented at a second devicein accordance with some embodiments of the present disclosure;

[0023] Fig. 6 illustrates a simplified block diagram of a device that is suitable forimplementing some embodiments of the present disclosure; and

[0024] Fig. 7 illustrates a block diagram of an example of a computer-readable medium inaccordance with some embodiments of the present disclosure.

[0025] Throughout the drawings, the same or similar reference numerals represent the sameor similar elements. DETAILED DESCRIPTION

[0026] Principles of the present disclosure will now be described with reference to someexample embodiments. It is to be understood that these embodiments are described only for the purpose of illustration and help those skilled in the art to understand and implement the present disclosure, without suggesting any limitation as to the scope of the disclosure.The disclosure described herein may be implemented in various manners other than the onesdescribed below.

[0027] In the following description and claims, unless defined otherwise, all technical andscientific terms used herein have the same meaning as commonly understood by one of ordinary skills in the art to which this disclosure belongs.

[0028] References in the present disclosure to “one embodiment,” “an embodiment,” “anexample embodiment,” and the like indicate that the embodiment described may include a particular feature, structure, or characteristic, but it is not necessary that every embodiment includes the particular feature, structure, or characteristic. Moreover, such phrases are not necessarily referring to the same embodiment. Further, when a particular feature, structure,or characteristic is described in connection with an embodiment, it is submitted that it iswithin the knowledge of one skilled in the art to affect such feature, structure, or characteristic in connection with other embodiments whether or not explicitly described.

[0029] It shall be understood that although the terms “first” and “second” etc. may be usedherein to describe various elements, these elements should not be limited by these terms.These terms are only used to distinguish one element from another. For example, a firstelement could be termed a second element, and similarly, a second element could be termed a first element, without departing from the scope of example embodiments. As used herein, the term “and / or” includes any and all combinations of one or more of the listed terms.

[0030] The terminology used herein is for the purpose of describing particular embodimentsonly and is not intended to be limiting of example embodiments. As used herein, the singular forms “a”, “an” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms “includes”, “including”, “has”, “having”, “includes” and / or “including”, when used herein, specify the presence of stated features, elements, and / or components etc., but do not preclude the presence or addition of one or more other features, elements, components and / or combinations thereof. As used herein, “at least one of the following: ” and “at least one of ” and similar wording, wherethe list of two or more elements are joined by “and” or “or”, mean at least any one of theelements, or at least any two or more of the elements, or at least all the elements.

[0031] As used in this application, the term “circuitry” may refer to one or more or all ofthe following: (a) hardware-only circuit implementations (such as implementations in only analog and / or digital circuitry) and (b) combinations of hardware circuits and software, such as (as applicable): (i) a combination of analog and / or digital hardware circuit(s) with software / firmware and (ii) any portions of hardware processor(s) with software (including digital signal processor(s)), software, and memory(ies) that work together to cause an apparatus, such as a mobile phone or server, to perform various functions) and (c) hardware circuit(s) and or processor(s), such as a microprocessor(s) or a portion of a microprocessor(s), that requires software (e.g., firmware) for operation, but the software may not be present when it is not needed for operation.

[0032] This definition of circuitry applies to all uses of this term in this application,including in any claims. As a further example, as used in this application, the term circuitry also covers an implementation of merely a hardware circuit or processor (or multiple processors) or portion of a hardware circuit or processor and its (or their) accompanying software and / or firmware. The term circuitry also covers, for example and if applicable to the particular claim element, a baseband integrated circuit or processor integrated circuit for a mobile device or a similar integrated circuit in server, a cellular network device, or other computing or network device.

[0033] As used herein, the term “communication network” refers to a network followingany suitable communication standards, such as Long Term Evolution (LTE), LTE-Advanced (LTE-A), Wideband Code Division Multiple Access (WCDMA), High-Speed Packet Access (HSPA), Narrow Band Internet of Things (NB-IoT) and so on. Furthermore, the communications between a terminal device and a network device in the communication network may be performed according to any suitable generation communication protocols, including, but not limited to, the first generation (1G), the second generation (2G), 2.5G, 2.75G, the third generation (3G), the fourth generation (4G), 4.5G, the future fifth generation (5G) communication protocols, and / or any other protocols either currently known or to be developed in the future. Embodiments of the present disclosure may be applied in various communication systems. Given the rapid development in communications, there will of course also be future type communication technologies and systems with which the present disclosure may be embodied. It should not be seen as limiting the scope of the present disclosure to only the aforementioned system.

[0034] As used herein, the term “network device” refers to a node in a communicationnetwork via which a terminal device accesses the network and receives services therefrom. The network device may refer to a base station (BS) or an access point (AP), for example, a node B (NodeB or NB), an evolved NodeB (eNodeB or eNB), a NR NB (also referred to as a gNB), a Remote Radio Unit (RRU), a radio header (RH), a remote radio head (RRH), a relay, a low power node such as a femto, a pico, and so forth, depending on the applied terminology and technology.

[0035] The term “terminal device” refers to any end device that may be capable of wirelesscommunication. By way of example rather than limitation, a terminal device may also be referred to as a communication device, user equipment (UE), a Subscriber Station (SS), a Portable Subscriber Station, a Mobile Station (MS), or an Access Terminal (AT). The terminal device may include, but not limited to, a mobile phone, a cellular phone, a smart phone, voice over IP (VoIP) phones, wireless local loop phones, a tablet, a wearable terminaldevice, a personal digital assistant (PDA), portable computers, desktop computer, imagecapture terminal devices such as digital cameras, gaming terminal devices, music storage and playback appliances, vehicle-mounted wireless terminal devices, wireless endpoints, mobile stations, laptop-embedded equipment (LEE), laptop-mounted equipment (LME), USB dongles, smart devices, wireless customer-premises equipment (CPE), an Internet of Things (loT) device, a watch or other wearable, a head-mounted display (HMD), a vehicle, a drone,a medical device and applications (e.g., remote surgery), an industrial device and applications (e.g., a robot and / or other wireless devices operating in an industrial and / or an automated processing chain contexts), a consumer electronics device, a device operating on commercial and / or industrial wireless networks, and the like. In the following description, the terms “terminal device”, “communication device”, “terminal”, “user equipment” and “UE” may be used interchangeably.

[0036] With the development of communication technology, short range sub-networks arepromising components to meet the extreme performance requirements in terms of latency,reliability and / or throughput envisioned for certain sixth generation (6G) short-range scenarios. Appropriate interference management (IM) among sub-networks has been identified as one of the major challenges to fully realize the potential of sub-networks, andin some examples, the IM can be achieved through improved estimation and / or resourceallocation.

[0037] For example, the resource allocation for IM can be implemented in the hybridresource allocation (HRA) mode. A major shortcoming in such an approach is that the resource allocations at both an access point (AP) and the gNB rely solely on a functionencapsulating the channel quality indicator (CQI) information, which cannot meet extremelatency and reliability requirements of 6G sub-networks, since the CQI values only convey quantized information.

[0038] To overcome this, the AP can employ an inference task. However, such inferencetasks are typically associated with a training component, which requires additionalcomputational capabilities, thereby leading to increased power expenditure at battery-powered APs. Further, the AP cannot make a decision on whether to combine the CQIinformation with the output of the inference task, since the AP cannot possess overallinformation about the network of the sub-networks. Therefore, for an efficient HRA realization, there is a need for a module at the gNB which can compute and inform the APsabout the decision to interchangeably use CQI information and enable or disable additionalinference tasks to achieve interference management.

[0039] In view of the above, some embodiments of the present disclosure propose a solutionfor improving interference management, especially, for improving interference managementin a hybrid resource allocation (HRA) mode, such as sub-network radio resourcemanagement (RRM) with interference prediction. In some example embodiments of thepresent disclosure, a first device receives, from a second device, inference capabilityinformation of the second device for performing an inference task associated with interference management. The first device determines, based on the inference capability information, an inference task indicator of whether to enable or disable the inference task at the second device. Then, the first device transmits the inference task indicator to the second device. In this way, inference task is contextually activated based on the inference task indicator at the second device for improving IM performance, especially in a hybrid resourceallocation (HRA) mode, thereby improving the radio resource management performance.

[0040] Fig. 1A illustrates an example network environment 100 in which exampleembodiments of the present disclosure may be implemented. The environment 100, which may be a part of a communication network, includes terminal devices and network devices.As illustrated in Fig. 1A, the communication network 100 may include a plurality of terminaldevices 110 (for example, UEs in the subnetwork). The communication network 100 mayfurther include a first network device 130 (hereinafter may also be referred to as a base stationor a gNB, or a first device) and a second network device 120 in the sub-network (hereinafteralso referred to as an AP in the sub-network or a second device, and the sub-network isdenoted by dotted line). The plurality of terminal devices 110 are under the control of the second network device 120, for example, via the WiFi communication, and the secondnetwork device 120 further communicates with the first network device 130 (for example, agNB of the 6G wide area network).

[0041] It is to be understood that the number of network devices and terminal devices isgiven only for the purpose of illustration without suggesting any limitations. The system 100may include any suitable number of network devices and / or terminal devices adapted for implementing embodiments of the present disclosure. Although not shown, it would be appreciated that one or more terminal devices may be located in the environment 100.

[0042] Communications in the network environment 100 may be implemented accordingto any proper communication protocol(s), comprising, but not limited to, the third generation (3G), the fourth generation (4G), the fifth generation (5G) or beyond, wireless local network communication protocols such as institute for electrical and electronics engineers (IEEE) 802.11 and the like, and / or any other protocols currently known or to be developed in the future. Moreover, the communication may utilize any proper wireless communication technology, comprising but not limited to: multiple-input multiple-output (MIMO), orthogonal frequency division multiplexing (OFDM), time division multiplexing (TDM),frequency division multiplexing (FDM), code division multiplexing (CDM), Bluetooth, ZigBee, and machine type communication (MTC), enhanced mobile broadband (eMBB), massive machine type communication (mMTC), ultra-reliable low latency communication(URLLC), carrier aggregation (CA), dual connection (DC), and new radio unlicensed (NR-U) technologies.

[0043] It is to be understood that the number of devices and their connection relationshipsand types shown in Fig. 1A are for illustrative purposes only without suggesting anylimitation. The communication system 100 may include any suitable number of devicesadapted for implementing embodiments of the present disclosure.

[0044] As mentioned above, short range sub-networks are a promising component to meetthe extreme performance requirements in terms of latency, reliability and / or throughput envisioned for certain sixth generation (6G) short-range scenarios. For example, the short- range sub-networks are generally installed in, or on, or around specific entities e.g., in-vehicle, in-body or on-body, in-house, etc., to provide life-critical data service with extreme performances over the local capillary coverage.

[0045] These sub-networks have the following pivotal properties and technical features:support of extreme performance requirements in terms of latency, reliability and / or throughput; low transmit power, which implies limited coverage range (e.g., in the order of few meters); star or tree topology with one sub-network AP and one or more sub-network UEs under AP’s control; overall mobility of AP and associated UEs, but limited mobility across different sub-networks; and part of 6G network and must continue to work also when out of network coverage.

[0046] The system design for a sub-network shall take all the above (or a subset) technicalfeatures into account. In this way, sub-networks can be seen as a potential evolution of the fifth generation (5G) Sidelink (SL). However, many enhancements are needed for air interface and architectural enablers, for example, to allow the AP to simultaneously support more devices as opposed to the D2D operation supported by 5G SL Mode 2; to allow an out- of-coverage AP to sense the channel, get resources and schedule those resources to the sub-network devices (beyond what the 5G SL Mode 2 allows); and to provide improvedauthentication / policy enforcements in such scenarios.

[0047] For example, the use cases of in-robot / in-production module sub-networks and in-vehicle sub-networks have extreme performance requirements in both reliability (up to 6nines or more) and latency (down to the level of 100μs or even below), e.g., for the highdemanding periodic deterministic communication services, which may be the mostchallenging scenarios in 6G system.

[0048] Fig. 1B illustrates an architectural diagram for sub-networks in accordance withsome embodiments of the present disclosure. As illustrated in Fig. 1B, the terminal device110 may be actuators of in-robot sub-network, or actuators of in-production module sub-network, or sensor of in-vehicle sub-network, or sensor of in-body subnetwork, or household appliances of in-house sub-network, and the like. The terminal devices 110 are undercontrol of the second network device 120 (for example, an AP of the sub-network).

[0049] As illustrated in Fig. 1B, the second network device 120 on one hand serves andmanages the terminal devices 110 (for example, the sub-network devices in the capillary sub-network coverage), and on the other hand, the second network device 120 is connected to thefirst network device 130 (for example, a network device in the 6G wide area network (WAN),which to certain extend, is able to control and coordinate different sub-networks). Asillustrated in Fig. 1B, there may be high time critical traffic (for example, less than 1ms),medium time critical traffic (for example, between 1ms to 10ms), and non-critical traffic (forexample, KPI monitoring) between the terminal devices 110, the second network devices 120(for example, APs), and the first network devices 130 (for example, base stations).

[0050] The deployment of the small and mobile 6G sub-networks, which support novel day-to-day and critical application, poses many new engineering challenges. For example, appropriate interference management (IM) among sub-networks has been identified as one of the major challenges to fully realize the potential of sub-networks. That is to say, the IM forms an integral aspect of 6G sub-networks to meet the extreme performance requirements. It can be noted that IM can be achieved through improved estimation and / or resource allocation.

[0051] When sub-networks are connected to a WAN, the gNB can perform centralizedresource allocation (CRA) since the gNB, in theory, is capable of processing the channel and interference measurements experienced by all the sub-networks under its purview. Thus,CRA can potentially achieve better performance with respect to IM associated withdistributed resource allocation (DRA), in which an AP performs resource allocation solelybased on the local measurements. That is to say, the DRA is advantageous when sub-networks go out-of-coverage or when the signalling overhead becomes significant to operate with CRA.

[0052] In some scenarios, the IM may also be achieved in a hybrid resource allocation(HRA) mode, in which the gNB can possess only partial information about the channel orinterference conditions of the sub-networks, whereas some decisions are taken at the gNBand others offloaded to the AP. For instance, the gNB can perform sub-band selection, where each sub-network is allocated with one or more sub-bands, while the AP performs local time or frequency resource allocation tasks. The benefit of HRA is that it at least partially reduces the signalling overhead required to convey the interference measurement information from the sub-networks to the gNB. For example, the information exchange between a sub-network and gNB can be simply restricted to long term interference measurements, while keeping the short-term interference measurements local to the sub- network.

[0053] Resource allocation for IM in the HRA mode can be perceived as a two-tieroperation. As to the first-tier operation, at the sub-network level, an AP periodically or a-periodically collects the buffer status report (BSR) and channel quality indicator (CQI)information from the served UEs to mainly perform time or frequency resource allocation and link adaptation (LA) as part of its layer two (L2) operations. As to the second-tier operation, the gNB then collects long-term observations from either all or a subset of APs to perform sub-band allocation and other related radio resource management (RRM) operations.In other words, in the HRA mode, resource allocation is carried out in a distributed mannerwith some operations performed at the gNB and the remaining operations at the AP.

[0054] Such distributed and nested approach as mentioned above is envisioned to maintainthe inter sub-network interference as low as possible while improving the L2 performance atthe AP. A major shortcoming in such approach is that the resource allocations at both theAP and the gNB rely solely on a function encapsulating the CQI information. This can be counterproductive to meet extreme latency and reliability requirements of 6G sub-networks, since CQI values only convey quantized information.

[0055] In summary, IM has been identified as an important aspect of 6G sub-networks inorder to meet the extreme performance requirements defined in terms of latency and reliability. While it is appealing to simply employ conventional IM techniques at a sub- network AP, these techniques solely rely on CQI information, which is adjusted over timeusing outer loop LA (OLLA) feedback mechanisms. Lack of precise interference information due to quantization in CQI-signal to interference plus noise ratio (SINR) mapping and slow convergence of OLLA makes conventional IM techniques highlyunsuitable to address performance requirements in sub-networks, especially in dense networkconditions.

[0056] In order to overcome this issue, an AP can employ state-of-the-art (SOTA) statisticalinference mechanisms (referred to as an inference task hereafter) on top of CQI reports usinglayer one (L1) information available at the AP. For example, the IM can be achieved through (a) SOTA statistical inference tools from estimation theory / machine learning (ML);(b) compact resource allocation; and / or (c) a combination of both. While the inference taskis almost always implemented at an AP, the resource allocation component can reside either at an AP or a gNB, depending on the underlying mode of operation e.g., the CRA, DRA or the HRA mode. Specifically, in the HRA mode (which comprises of the best aspects of both CRA and DRA modes), resource allocation is carried out in a distributed manner with some operations performed at the gNB and the remaining operations at the AP.

[0057] However, such inference tasks are typically associated with a training component,which requires additional computational capabilities, thereby leading to increased power expenditure at battery-powered APs. Moreover, making the decision to combine CQIinformation with the output of the inference task solely at the AP can be ineffectual, sincethe AP only possesses partial information about the network of sub-networks. Thus, for the purpose of an efficient HRA realization, there is a need for a module (for example, a radio resource management (RRM) module) at the gNB which can compute and inform the APsabout the decision to interchangeably use CQI information and enable or disable additionalinference tasks, and which can also ensure timely and minimal signalling to the APs andavoid any overhead with signalling mechanisms between the gNB and APs.

[0058] Hereinafter, an example signal process 200 for improving interference managementin accordance with some embodiments of the present disclosure will be described withreference to Fig. 2. For the purpose of discussion, the process 200 may be described withreference to Fig. 1A. The process 200 may involve the second network device 120 (forexample, the AP of the sub-network) and the first network device 130 (for example, a basestation, or a gNB) as illustrated in Fig.1A. It would be appreciated that although the process200 has been described in the communication environment 100 of Fig. 1A, this process maybe likewise applied to other communication scenarios with similar issues. It should not thatthe first device 211 may be an example of the first network device 130, and the second device 212 may be an example of the second network device 120.

[0059] As shown in Fig. 2, a second device 212 transmits (205) inference capabilityinformation 201 of the second device 212 for performing an inference task associated withinterference management (IM) to a first device 211, and the first device 211 receives (210)the inference capability information 201. The first device 211 determines (215), based on the inference capability information 201, an inference task indicator (ITI) 202 of whether toenable or disable the inference task at the second device 212, and then transmits (220) theITI 202 to the second device 212. The second device 212 receives (225) the ITI 202.Based on the received ITI 202, the second device 212 may enable or disable the inferencetask.

[0060] In some embodiments, the first device 211 is a base station, for example, a gNB, andthe second device 212 may be an access point of the sub-network. In some embodiments,the inference capability information 201 can be received by the first device 211 after aconnection is established between the first device and the second device. Alternatively, theinference capability information 201 may be received at beginning of a signalling epoch. For example, the sub-network APs may share their device inference capability information(DICI) in scenario (a) right after the first connection is established with the gNB or scenario(b) at the beginning of every signalling epoch t. For example, an AP n may perform theinference task by default after establishing the initial connection with the gNB and mayindicate its DICI value, which is a binary indicator value, to the gNB.

[0061] In some embodiments, the inference capability information 201 may have a firstvalue or a second value. For instance, the first value can indicate that the second device iscapable of performing the inference task, and the second value can indicate that the seconddevice is incapable of performing the inference task. In these embodiments, the inferencecapability information 201 may be a binary indicator value. For example, this embodimentfor DICI characterization can be used when an AP ^ establishes its first connection with thegNB. Here, ^^ is modelled as a binary indicator value according to the equation (1)

[0062] Let ^ ∶= {1,2, … , ^} denote the set of all APs associated with the gNB, where ^is the maximum number of APs a gNB can support. The choice of ^ is gNB specific andcan depend on several sub-network operation features, but it does not represent a limitingfactor. Furthermore, each AP is indexed as ^ ∈ ^. For convenience, the time durationbetween two consecutive signalling time instants is referred to as an epoch (or signallingepoch), where each epoch is denoted by ^ ∈ ℕ^. Unless otherwise specified, it is assumedthat the APs send their signalling information to the gNB periodically.

[0063] In some embodiments, the inference capability information 201 comprises a ratio ofa first quantity to a second quantity for a signalling epoch. The first quantity represents afirst metric of energy required to perform the inference task, and the second quantityrepresents a second metric of energy available at the second device at the signalling epoch. For example, the first metric of energy may be the number of flow the number of floating point operations per second (FLOPs) for performing inference task, and the second metric of energy may be the estimate number of FLOPs as function of remaining batter power at the signalling epoch t. It should be understood that this embodiment for calculating the inference capability information 201 based on two metrics can be used for above-mentioned scenario (a) and scenario (b).

[0064] For example, the inference capability information 201 ^^(^) is modelled as a ratiobetween the following quantities for every signalling epoch ^ according to the equation (2):where ^^^^^^ denotes the number of FLOPs required to perform the inference task, and^^^^^^^(^^^^; ^) denotes an estimate of number of FLOPs as a function of remaining batterypower ^^^^at signalling epoch ^. Since the type of inference task (e.g., parametric / non- parametric approaches) and its associated algorithm are predetermined at an AP by the device manufacturer, ^^^^^^is known beforehand. However, the denominator of the equation (2) changes over the time due to the variation in battery power over time.

[0065] It can be understood that an advantage of updating the inference capabilityinformation 201 at the second device 212 is that it allows to override the ITI decision (ifneeded) from the first device 211 based on its battery conditions or forecast of energyexpenditure. For instance, more battery power may mean the AP can perform the inferencetask, low battery power may mean the AP may not be able to perform the inference task.

[0066] In some embodiments, the inference capability information comprises an indicationof a selected one of a first operation mode and a second operation mode, which can be twodifferent operation modes. For instance, in the first operation mode, the second device performs the inference task by default after establishing a connection with the first device,whereas in the second operation mode, the second device performs the inference task basedon receiving an inference task indicator indicating to enable the inference task, without performing the inference task by default after establishing a connection with the first device.

[0067] In one example, an AP n may perform the inference task by default after establishingthe initial connection with the gNB and indicates its DICI value to the gNB. In anotherexample, an AP n does not perform the inference task by default after establishing the initialconnection with the gNB and indicates its DICI value, and the APs perform the first inferencetask upon receiving ^(^) = 1, which is the ITI value sent from the gNB to AP n during tthsignalling epoch, where t is any arbitrary epoch. In a further example, an AP n chooses apreferred mode of operation from the preceding examples, and reports the selected mode tothe gNB it as part of DICI signalling information.

[0068] It can be seen that the process 200 of Fig. 2 proposes an enhancement for IM in theHRA model when performing distributed resource allocation at the second device and thefirst device, the first device (for example, a gNB) may aid the second device (for example,an AP of a subnetwork) in minimizing the interference among the associated UEs andimproving the performance of L2 operations by gathering relevant information (for example,inference capability information 201) and computing the decision (for example, the ITI 202)based on the information 201 to contextually enable or disable the inference task at the seconddevice 212.

[0069] In the process 200, the first device 211 first gathers the relevant contextualinformation (for example, the inference capability information 201) from the second device212 and then uses the gathered contextual information to determine the ITI value of eachsecond device. Since the first device 211 possesses all information about the network ofsub-networks in which the second device 212 is located, the gathered contextual informationmay be modelled at the first device 211 as a state tuple to calculate the appropriate ITI valuefor each second device 212. Thus, the proposed process 200 or enhancement maycontextually activate, based on the appropriate ITI value, inference tasks at a second device212 for improving IM and consequently the RRM performance. That is to say, the firstdevice 211 may compute and inform the second device 212 about the decision tointerchangeably use CQI information and enable or disable additional inference tasks.Furthermore, the proposed enhancements are designed to have minimal impact in terms ofboth signalling overhead and computational demands, since the information exchange between the first device and the second device only involves the inference capabilityinformation 201 and the inference task indicator 202.

[0070] For example, the proposed solution may have the advantages that the first devicecan contextually activate additional inference tasks at a second device that is capableimplement it only when really needed, thereby saving energy expenditure at the AP bydisabling it when the performance gain is not significant enough: as an example, a scenariowhere the inference task may be disabled is the case with only two sub-networks in a large area, so with very little interference among them, and it can predict that here is not expected to provide large gain, so better to have the gNB indicated the AP to disable the inference task. As a difference example, a scenario where the inference task may be enabled in a dense network scenario, where many sub-networks exist next to each other with relatively slow movement in space.

[0071] In some embodiments, the inference capability information 201 (for example, theDICI) may act as the first decision checkpoint at the first device 211 to determine whether ornot the second device 212 can perform additional inference tasks. In some embodiments,the above mentioned ITI is a first ITI, and the first device 211 further receives, from thesecond device 212, a performance value indicating performance achieved by including orexcluding the inference task for the interference management. In some embodiments, thefirst device 211 further determines a second ITI based on the inference capability informationand the performance value. Then, the first device 211 may transmit the second ITI to thesecond device 212.

[0072] In some embodiments, the performance value may be periodically received fromthe second device, and may be radio resource management (RRM) performance value (RPV).In some embodiments, the second device 212 may determine the performance value asfollows. First, if the inference task indicator indicates to enable the inference task, thesecond device 212 may perform the inference task to obtain inference output. Afterwards,the second device 212 can combine the inference output and channel quality indicator (CQI)information received from at least one user equipment (UE) associated with the seconddevice to obtain combined information. Then, the second device 212 can perform at leastone layer 2 (L2) operation based on the combined information. Next, the second device 212may compute the performance value based on the at least one L2 operation.

[0073] Hereinafter, the calculation of the RPV will be described with referent to Fig. 3,which illustrates a signalling chart 300 for hybrid resource allocation (HRA) modelenhancement in accordance with some embodiments of the present disclosure. As shownin Fig. 3, at 301, the AP 322 (which is an example of the second device 212) transmits itsdevice inference capability information (DICI) to the gNB 321 (which is an example of thefirst device 211). At 302, the gNB 321 transmits the ITI determined based on the receivedDICI to the AP 322. Then, at 303, the AP 322 receives CQI information from served UEs323 associated with this AP to perform inference task on top of CQI reports using layer one(L1) operation.

[0074] At 304, if the ITI is equal to 1, the AP 322 performs inference task based on the CQIreport using L1 operation. At 305, the AP 322 combines the CQI with the output of the inference task. At 306, the AP 322 performs other L1 operations. At 307, the AP 322 sends the SINR information to the Layer 2. At 308, the AP 322 performs the L2 operation. At 309, the AP 322 computes the RRV based on L2 operation. At 310, the AP 322 transmits the L1 / L2 decisions to the UE 323. At 311, the AP 322 transmits the RPV to the gNB 321.

[0075] For example, at 309, this RPV is computed at L2 of an AP and is periodically sharedwith the gNB as part of the signalling information at 311. RPV incorporates the effect ofperforming additional L1 inference tasks on RRM operations. Formally, the RPV of an AP^ is denoted as ^^ and characterized by using the following embodiment.

[0076] Letdenotes the set of ^ L2operations at an AP ^, where, ^^,^^^^; ^^,^^, ∀ ^ ∈ {1,denotes the ^^^L2 operation. Furthermore, ^^denotes the output of combing the CQI and inference output and ^^,^denote any additional arguments required to perform ^^^L2 operation. For instance, ifrepresents MCS selection, then ^^,^ corresponds to the set of UEs associatedselected for transmission. Similarly, if ^^,^^^^; ^^,^^ represents PRB selection, then ^^,^corresponds to the subset of PRBs available to AP ^ to perform resource allocation.

[0077] In some embodiments, the performance value for a current signalling epoch may bedetermined based on the following one or more factors. The first factor may be theperformance value of an L2 operation as a result of incorporating the inference task at aprevious signalling epoch. The second factor may be the inference task indicatortransmitted from the first device to the second device during the previous signalling epoch. The third factor can be a performance score of the L2 operation averaged between epochsstarting from a last signalling epoch in which the second device performed the inference taskto the previous signalling epoch. The fourth factor may be the number of L2 operations forthe interference management.

[0078] For example, assuming ^^,^(^) denotes the performance value of the ^^^L2 operation as a result of incorporating the additional L1 inference task at the signalling epoch^, the RPV, ^^(^) can be computed according to the equation (3):where, ^^(^ − 1) ∈ {0,1} denotes the ITI value sent from gNB to AP ^ during the (^ −1)^^ signalling epoch; ^^^,^(^) is the performance score of the ^^^L2 operation averagedbetween the epochs starting from the last signalling epoch when AP ^ performed additionalinference tasks to (^ − 1)^^ signalling epoch. If ^^ denotes the last signalling epochwhen ^(^^) = 1, then ^^^,^(^) can be formally expressed according to equation (4):

[0079] In some embodiments, the second inference task indicator is determined furtherbased on coverage statistics information, a time since last indication (TSLI) value, or both ofthem. As used herein, the coverage statistics information may correspond to an estimate ofuser equipment (UE) density and mobility statistics of UEs associated with at least onesecond device. In these embodiments, when combined with coverage statistics and theTSLI value, the RPV functions as important feedback to compute the ITI value at the gNB, since the combination jointly encompasses the overall impact of the proposed IM enhancements on both L1 and L2 procedures at an AP.

[0080] For example, as illustrated in Fig. 3, at 312, the gNB 321 (for example, the RRMmodule) may update time since last indication (TSLI), and at 313, the gNB 321 (for example, the RRM module) may update the coverage statistics. Then, at 314, the gNB 321 (forexample, the RRM module) computes the ITI value based on the received RPV, the updatedTSLI and the updated coverage statistics. At 315, the gNB 321 transmits the determinedITI to the AP 322.

[0081] Hereinafter, some embodiments for calculating the coverage statistics informationwill be described. In general, the first device 211 can obtain the coverage statisticsinformation in various manners. For example, the first device 211 may estimate thecoverage statistics information through at least one signalling exchange between the firstdevice and the at least one second device. In some other example, the first device 211 canreceive the coverage statistics information from a subset of the at least one second device periodically.

[0082] For example, this coverage statistics information corresponds to an estimate of UEdensity and mobility statistics of the UEs associated with each AP. The gNB can either estimate this quantity through signalling exchanges between the gNB-APs or request a selected subset of APs to share this information periodically.

[0083] In some embodiments, the coverage statistics information for the second device 212is represented as a weighted sum of a first number of UEs associated with the second deviceand a second number of active UEs windowed over a number W of signalling epochs.

[0084] For example, the coverage statistics is modelled in terms of number of UEs at eachAP, since the probability of channel usage and data exchange is directly proportional to thenumber of UEs associated to an AP and it could be beneficial to activate additional inferencetasks for improved IM. Formally, the coverage statistics information of AP n is denotedwith ^^, which is modelled by modeling ^^ as a weighted sum of the number UEs (in activestate, or in inactive state, or in idle state) associated with each AP and the number of activeUEs windowed over a set of past signalling epochs. Formally,^ can be written according to equation (5):where, ^^(^) denotes the number of UEs associated with AP ^ , ^^(^) ≔denotes the average number of active UEs windowed over the last ^signalling epochs and ^ is a predetermined hyper-parameter. It can be noted that when^ = 1, simply reduces to the number of

[0085] In some embodiments, the first number has a weight of α and the second numberhas a weight of (1-α). In these embodiments, as an example, the number W and the weightα can be predetermined. As another example, the number W and the weight α can bedetermined by the second device and indicated to the first device. Alternatively, they maybe determined by the first device and indicated to the second device.

[0086] For example, following options are possible for the selection of ^ and ^: (1) theyare specified as a range of parameters in 3GPP specifications and hence, ^ and ^ can bechosen by both gNB and AP device manufacturers; (2) they are left up to the AP to decide asan implementation aspect and can be optionally reported back to the gNB; and (3) they areleft up the gNB to decide as an implementation aspect and can be indicated to the associated set of APs e.g., after the first connection is established.

[0087] Hereinafter, some embodiments for calculating the TSLI will be described. Insome embodiments, the TSLI value of the second device at a current signalling epoch maybe determined based on the TSLI value of the second device at a previous signalling epoch,and a random variable for the previous signalling epoch. In some examples, a value of therandom variable can be determined based on the inference task indicator at the previoussignalling epoch and a performance value at the current signalling epoch.

[0088] For example, in order to ensure that the interference information obtained byperforming additional inference tasks is as fresh as possible at APs, the gNB can maintain asimple linear counter, which may be succinctly referred to as the TSLI. The TSLI value ofan AP ^ directly impacts RPV and DICI quantities, provided DICI is computed at beginningof each signalling epoch. TSLI characterization is presented in the following embodiment.

[0089] In this embodiments, the TSLI value of an AP ^ is denoted by Δ^(^) at thesignalling epoch ^ and is characterized by the following update equation (6):Δ^(^) ≔ Δ^(^ − 1)^1 − ^^(^ − 1)^ + 1 (6)where, ^^(^ − 1) is a joint Bernoulli random variable which is characterized by thefollowing equation (7):where, ^^,^^^^^^corresponds to a predetermined threshold value. It can be noted that theRPV, ^^(^) lies in the range [−1,1] due to normalization and hence, ^^,^^^^^^ ∈ [−1,1]as well. For the selection of ^^,^^^^^^value, it is either left for gNB to update over the time treating it as a hyper-parameter or is specified in 3GPP standards depending on the type of sub-network operation under consideration.

[0090] Hereinafter, the embodiments of algorithm for determining the ITI will be described.It should be noted that this algorithm mentioned above is just an example simple heuristic tocompute the ITI, other algorithms can also be implemented. In some embodiments, thegNB 321 determines the second ITI based on the gathered information, such as RPV, updatedtime since last information (TSLI), and updated coverage statistics. That it so say, once the gNB gathers all the relevant information required to form the state tuple, it computes the ITI value for each AP or a subset of APs at the beginning of each signalling epoch ^. Tothis extent, the following embodiment may be used to calculate the ITI value:

[0091] In the above algorithm, the choice of the utility function ^(. ), can either bespecified in 3GPP specifications or is left for the device manufacturers for custom realization.An example of such utility function could simply be as the following equation (8):where, ℙ(. ) denotes the probability measure.

[0092] As shown in Fig. 3, in the process 300, the gNB first gathers the relevant contextualinformation from APs, which is then used to jointly determine the inference task indicator(ITI) value of each AP. The contextual information, modelled using a state tuple at the gNB,comprises observations of the APs and other relevant information gathered through signallinginformation. For example, the RRM module residing at the gNB utilizes the deviceinference capability information (DICI) to compute the binary inference task indicator (ITI)value, and the ITI is then sent by the gNB to the AP (for instance at 302) to enable (in caseITI==1) or disable (in case ITI==0) the inference task at the AP.

[0093] Then, the RRM module further gathers the information about coverage statistics,TSLI, or RRM performance value (RPV) to compute a further binary inference task indicator(ITI) value. The further ITI is then sent by the gNB to the AP (for instance at 315) to enable(in case ITI==1) or disable (in case ITI==0) the inference task at the AP. The TSLI is considered along with other inputs to ensure timely signalling to the APs and to reduce the overall signalling overhead between the gNB-AP. On a sub-network level, the AP combines the locally received CQI information with the inference output (when ITI isenabled) and passes it on to perform L2 operation or RRM operation. Finally, RPVindicates the consolidated performance achieved by including or excluding the inference taskoutput in L2 operation or RRM operation at the AP.

[0094] Therefore, the proposed enhancements are intended to contextually activateinference tasks at an AP for improving IM and consequently the RRM performance. Secondly,the proposed enhancements are designed to have minimal impact in terms of both signalling overhead and computational demands. The proposed enhancements introduce minimal modifications to AP’s L1-L2 interactions, which in-turn assists the RRM module at the gNB to perform resource allocation efficiently.

[0095] Fig.4 illustrates a flowchart of an example method 400 implemented at a first devicein accordance with some other embodiments of the present disclosure. For the purpose ofdiscussion, the method 400 will be described from the perspective of the first network device130 with reference to Fig. 1A.

[0096] At block 410, the first device 130 receives, from a second device, inferencecapability information of the second device for performing an inference task associated with interference management. At block 420, the first device 130 determines, based on the inference capability information, an inference task indicator of whether to enable or disablethe inference task at the second device. At block 430, the first device 130 transmits theinference task indicator to the second device.

[0097] In some embodiments, the inference capability information is received at least oneof the following: after a connection is established between the first device and the second device; or at beginning of a signalling epoch. In some embodiments, the inference capability information comprises: a first value indicating that the second device is capable of performing the inference task; or a second value indicating that the second device is incapable of performing the inference task.

[0098] In some embodiments, the inference capability information comprises a ratio of afirst quantity to a second quantity for a signalling epoch; the first quantity represents a firstmetric of energy required to perform the inference task; and the second quantity represents asecond metric of energy available at the second device at the signalling epoch.

[0099] In some embodiments, the first device 130 further receives, from the second device,a performance value indicating performance achieved by including or excluding the inferencetask for the interference management. In some embodiments, the performance value is periodically received from the second device.

[0100] In some embodiments, the inference task indicator is a first inference task indicator,and the first device 130 further determines, based on the inference capability information andthe performance value, a second inference task indicator; and transmits, to the second device, the second inference task indicator. In some embodiments, the second inference task indicator is determined further based on at least one of the following: coverage statistics information corresponding to an estimate of user equipment (UE) density and mobilitystatistics of UEs associated with at least one second device; or a time since last indication(TSLI) value.

[0101] In some embodiments, the first device 130 estimates the coverage statisticsinformation through at least one signalling exchange between the first device and the at leastone second device; or receives the coverage statistics information from a subset of the at leastone second device periodically.

[0102] In some embodiments, the coverage statistics information for the second device isrepresented as a weighted sum of a first number of UEs associated with the second device and a second number of active UEs windowed over a number W of signalling epochs. In some embodiments, the first number has a weight of α and the second number has a weight of (1-α), and wherein the number W and the weight α are one of the following: predetermined; determined by the second device and indicated to the first device; or determined by the first device and indicated to the second device.

[0103] In some embodiments, the TSLI value of the second device at a current signallingepoch is determined based on: the TSLI value of the second device at a previous signalling epoch; and a random variable for the previous signalling epoch, wherein a value of therandom variable is determined based on the inference task indicator at the previous signallingepoch and a performance value at the current signalling epoch.

[0104] In some embodiments, the inference capability information comprises an indicationof a selected one of a first operation mode and a second operation mode; in the first operation mode, the second device performs the inference task by default after establishing aconnection with the first device 130; and in the second operation mode, the second deviceperforms the inference task based on receiving an inference task indicator indicating to enable the inference task, without performing the inference task by default after establishing aconnection with the first device 130. In some embodiments, the first device is a base station;or the second device is an access point of the sub-network.

[0105] Fig. 5 illustrates a flowchart of an example method 500 implemented at a secondnetwork device in accordance with some other embodiments of the present disclosure. Forthe purpose of discussion, the method 500 will be described from the perspective of thesecond network device 120 with reference to Fig. 1A.

[0106] At block 510, the second device 120 transmits, to a first device, inference capabilityinformation of the second device for performing an inference task associated with interference management. At block 520, the second device 120 receives, from the first device, an inference task indicator of whether to enable or disable the inference task at the second device.

[0107] In some embodiments, the inference capability information is transmitted at leastone of the following: after a connection is established between the first device and the second device; or at beginning of a signalling epoch.

[0108] In some embodiments, the inference capability information comprises: a first valueindicating that the second device 120 is capable of performing the inference task; or a secondvalue indicating that the second device 120 is incapable of performing the inference task.In some embodiments, the inference capability information comprises a ratio of a firstquantity to a second quantity for a signalling epoch; the first quantity represents a first metricof energy required to perform the inference task; and the second quantity represents a secondmetric of energy available at the second device at the signalling epoch.

[0109] In some embodiments, the second device 120 further determines a performancevalue indicating performance achieved by including or excluding the inference task for theinterference management; and transmits the performance value to the first device. In someembodiments, the performance value is periodically transmitted to the first device.

[0110] In some embodiments, the second device 120 determines the performance value by:based on determining that the inference task indicator indicates to enable the inference task, performing the inference task to obtain inference output; combining the inference output and channel quality indicator (CQI) information received from at least one user equipment (UE) associated with the second device to obtain combined information; performing at least one layer 2 (L2) operation based on the combined information; and computing the performance value based on the at least one L2 operation.

[0111] In some embodiments, the performance value for a current signalling epoch isdetermined based on: the performance value of an L2 operation as a result of incorporating the inference task at a previous signalling epoch; the inference task indicator transmitted from the first device to the second device during the previous signalling epoch; a performance score of the L2 operation averaged between epochs starting from a last signalling epoch in which the second device performed the inference task to the previous signalling epoch; and the number of L2 operations for the interference management.

[0112] In some embodiments, the inference task indicator is a first inference task indicator,and the second device 120 further receives, from the first device, a second inference taskindicator, which is determined based on the inference capability information and the performance value.

[0113] In some embodiments, the second device 120 further performs the inference taskbased on a first operation mode in which the second device performs the inference task bydefault after establishing a connection with the first device; performs the inference task basedon a second operation mode in which the second device performs the inference task based on receiving an inference task indicator indicating to enable the inference task, without performing the inference task by default after establishing a connection with the first device;or selects one of the first operation mode and the second operation mode and an indicationof the selected operation mode is included in the inference capability information. In someembodiments, the first device is a base station; or the second device is an access point of thesub-network.

[0114] In some embodiments, an apparatus (for example, the first network device 130)capable of performing the method 400 may comprise means for performing the respectivesteps of the method 400. The means may be implemented in any suitable form. For example, the means may be implemented in a circuitry or software module.

[0115] In some embodiments, the apparatus comprises: means for receiving, from a seconddevice, inference capability information of the second device for performing an inference task associated with interference management; means for determining, based on the inference capability information, an inference task indicator of whether to enable or disable theinference task at the second device; and means for transmitting the inference task indicatorto the second device.

[0116] In some embodiments, the inference capability information is received at least oneof the following: after a connection is established between the apparatus and the seconddevice; or at beginning of a signalling epoch. In some embodiments, the inference capability information comprises: a first value indicating that the second device is capable of performing the inference task; or a second value indicating that the second device is incapable of performing the inference task.

[0117] In some embodiments, the inference capability information comprises a ratio of afirst quantity to a second quantity for a signalling epoch; the first quantity represents a firstmetric of energy required to perform the inference task; and the second quantity represents asecond metric of energy available at the second device at the signalling epoch.

[0118] In some embodiments, the apparatus further receives, from the second device, aperformance value indicating performance achieved by including or excluding the inferencetask for the interference management. In some embodiments, the performance value is periodically received from the second device.

[0119] In some embodiments, the inference task indicator is a first inference task indicator,and the apparatus further determines, based on the inference capability information and theperformance value, a second inference task indicator; and transmits, to the second device, the second inference task indicator. In some embodiments, the second inference task indicator is determined further based on at least one of the following: coverage statistics information corresponding to an estimate of user equipment (UE) density and mobility statistics of UEs associated with at least one second device; or a time since last indication (TSLI) value.

[0120] In some embodiments, the apparatus estimates the coverage statistics informationthrough at least one signalling exchange between the apparatus and the at least one seconddevice; or receives the coverage statistics information from a subset of the at least one seconddevice periodically.

[0121] In some embodiments, the coverage statistics information for the second device isrepresented as a weighted sum of a first number of UEs associated with the second device and a second number of active UEs windowed over a number W of signalling epochs. In some embodiments, the first number has a weight of α and the second number has a weight of (1-α), and wherein the number W and the weight α are one of the following: predetermined; determined by the second device and indicated to the apparatus; or determined by theapparatus and indicated to the second device.

[0122] In some embodiments, the TSLI value of the second device at a current signallingepoch is determined based on: the TSLI value of the second device at a previous signalling epoch; and a random variable for the previous signalling epoch, wherein a value of the random variable is determined based on the inference task indicator at the previous signalling epoch and a performance value at the current signalling epoch.

[0123] In some embodiments, the inference capability information comprises an indicationof a selected one of a first operation mode and a second operation mode; in the first operation mode, the second device performs the inference task by default after establishing aconnection with the apparatus; and in the second operation mode, the second device performsthe inference task based on receiving an inference task indicator indicating to enable the inference task, without performing the inference task by default after establishing aconnection with the apparatus. In some embodiments, the apparatus is a base station; or thesecond device is an access point of the sub-network.

[0124] In some embodiments, the apparatus further comprises means for performing othersteps in some embodiments of the method 400. In some embodiments, the means comprises at least one processor and at least one memory including computer program code, the at least one memory and computer program code configured to, with the at least one processor, cause the performance of the apparatus.

[0125] In some embodiments, an apparatus (for example, the second network device 120 orsecond device) capable of performing the method 500 may comprise means for performingthe respective steps of the method 500. The means may be implemented in any suitable form. For example, the means may be implemented in a circuitry or software module.

[0126] In some embodiments, the apparatus comprises: means for transmitting, to a firstdevice, inference capability information of the apparatus for performing an inference taskassociated with interference management, and means for receiving, from the first device, an inference task indicator of whether to enable or disable the inference task at the apparatus.

[0127] In some embodiments, the inference capability information is transmitted at leastone of the following: after a connection is established between the first device and the apparatus; or at beginning of a signalling epoch.

[0128] In some embodiments, the inference capability information comprises: a first valueindicating that the apparatus is capable of performing the inference task; or a second value indicating that the apparatus is incapable of performing the inference task. In someembodiments, the inference capability information comprises a ratio of a first quantity to asecond quantity for a signalling epoch; the first quantity represents a first metric of energyrequired to perform the inference task; and the second quantity represents a second metric ofenergy available at the apparatus at the signalling epoch.

[0129] In some embodiments, the apparatus further determines a performance valueindicating performance achieved by including or excluding the inference task for theinterference management; and transmits the performance value to the first device. In someembodiments, the performance value is periodically transmitted to the first device.

[0130] In some embodiments, the apparatus determines the performance value by: based ondetermining that the inference task indicator indicates to enable the inference task, performing the inference task to obtain inference output; combining the inference output and channel quality indicator (CQI) information received from at least one user equipment (UE)associated with the apparatus to obtain combined information; performing at least one layer2 (L2) operation based on the combined information; and computing the performance value based on the at least one L2 operation.

[0131] In some embodiments, the performance value for a current signalling epoch isdetermined based on: the performance value of an L2 operation as a result of incorporating the inference task at a previous signalling epoch; the inference task indicator transmitted fromthe first device to the apparatus during the previous signalling epoch; a performance score ofthe L2 operation averaged between epochs starting from a last signalling epoch in which theapparatus performed the inference task to the previous signalling epoch; and the number ofL2 operations for the interference management.

[0132] In some embodiments, the inference task indicator is a first inference task indicator,and the apparatus further receives, from the first device, a second inference task indicator, which is determined based on the inference capability information and the performance value.

[0133] In some embodiments, the apparatus further performs the inference task based on afirst operation mode in which the apparatus performs the inference task by default afterestablishing a connection with the first device; performs the inference task based on a secondoperation mode in which the apparatus performs the inference task based on receiving aninference task indicator indicating to enable the inference task, without performing theinference task by default after establishing a connection with the first device; or selects oneof the first operation mode and the second operation mode and an indication of the selectedoperation mode is included in the inference capability information. In some embodiments,the first device is a base station; or the apparatus is an access point of the sub-network.

[0134] In some embodiments, the apparatus further comprises means for performing othersteps in some embodiments of the method 500. In some embodiments, the means comprises at least one processor and at least one memory including computer program code, the at least one memory and computer program code configured to, with the at least one processor, cause the performance of the apparatus.

[0135] Fig. 6 is a simplified block diagram of a device 600 that is suitable for implementingembodiments of the present disclosure. The device 600 may be provided to implement thecommunication device, for example the terminal device 110, the first and second networkdevices 120, 130 as shown in Fig. 1A. As shown, the device 600 includes one or moreprocessors 610, one or more memories 620 coupled to the processor 610, and one or morecommunication modules 640 coupled to the processor 610.

[0136] The communication module 640 is for bidirectional communications. Thecommunication module 640 has at least one antenna to facilitate communication. Thecommunication interface may represent any interface that is necessary for communication with other network devices.

[0137] The processor 610 may be of any type suitable to the local technical network andmay include one or more of the following: general purpose computers, special purpose computers, microprocessors, digital signal processors (DSPs) and processors based on multicore processor architecture, as non-limiting examples. The device 600 may have multiple processors, such as an application specific integrated circuit chip that is slaved in time to a clock which synchronizes the main processor.

[0138] The memory 620 may include one or more non-volatile memories and one or morevolatile memories. Examples of the non-volatile memories include, but are not limited to,a read only memory (ROM) 624, an electrically programmable read only memory (EPROM),a flash memory, a hard disk, a compact disc (CD), a digital video disk (DVD), and other magnetic storage and / or optical storage. Examples of the volatile memories include, but arenot limited to, a random access memory (RAM) 622 and other volatile memories that maynot last in the power-down duration.

[0139] A computer program 630 includes computer executable instructions that areexecuted by the associated processor 610. The program 630 may be stored in the ROM 624.The processor 610 may perform any suitable actions and processing by loading the program 630 into the RAM 622.

[0140] The embodiments of the present disclosure may be implemented by means of theprogram so that the device 600 may perform any process of the disclosure as discussed withreference to Figs. 2, 3, 4, and 5. The embodiments of the present disclosure may also beimplemented by hardware or by a combination of software and hardware.

[0141] In some embodiments, the program 630 may be tangibly contained in a computerreadable medium which may be included in the device 600 (such as in the memory 620) or other storage devices that are accessible by the device 600. The device 600 may load theprogram 630 from the computer readable medium to the RAM 622 for execution. Thecomputer readable medium may include any types of tangible non-volatile storage, such as ROM, EPROM, a flash memory, a hard disk, CD, DVD, and the like.

[0142] Fig. 7 illustrates an example of the computer readable medium 700 in form of CDor DVD in accordance with some embodiments of the present disclosure. The computerreadable medium has the program 730 stored thereon. It is noted that although the computer-readable medium 700 is depicted in form of CD or DVD, the computer-readable medium 700 may be in any other form suitable for carry or hold the program 630.

[0143] Generally, various embodiments of the present disclosure may be implemented inhardware or special purpose circuits, software, logic or any combination thereof. Some aspects may be implemented in hardware, while other aspects may be implemented in firmware or software which may be executed by a controller, microprocessor or other computing device. While various aspects of embodiments of the present disclosure are illustrated and described as block diagrams, flowcharts, or using some other pictorialrepresentations, it is to be understood that the block, apparatus, system, technique or methoddescribed herein may be implemented in, as non-limiting examples, hardware, software, firmware, special purpose circuits or logic, general purpose hardware or controller or other computing devices, or some combination thereof.

[0144] The present disclosure also provides at least one computer program product tangiblystored on a non-transitory computer readable storage medium. The computer programproduct includes computer-executable instructions, such as those included in programmodules, being executed in a device on a target real or virtual processor, to carry out themethod 400 or 500 as described above with reference to Fig.4 to Fig.5. Generally, programmodules include routines, programs, libraries, objects, classes, components, data structures, or the like that perform particular tasks or implement particular abstract data types. The functionality of the program modules may be combined or split between program modules as desired in various embodiments. Machine-executable instructions for program modules may be executed within a local or distributed device. In a distributed device, program modules may be located in both local and remote storage media.

[0145] Program code for carrying out methods of the present disclosure may be written inany combination of one or more programming languages. These program codes may beprovided to a processor or controller of a general purpose computer, special purposecomputer, or other programmable data processing apparatus, such that the program codes,when executed by the processor or controller, cause the functions / operations specified in theflowcharts and / or block diagrams to be implemented. The program code may execute entirely on a machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

[0146] In the context of the present disclosure, the computer program codes or related datamay be carried by any suitable carrier to enable the device, apparatus or processor to perform various processes and operations as described above. Examples of the carrier include a signal, computer readable medium, and the like.

[0147] The computer readable medium may be a computer readable signal medium or acomputer readable storage medium. A computer readable medium may include but notlimited to an electronic, magnetic, optical, electromagnetic, infrared, or semiconductorsystem, apparatus, or device, or any suitable combination of the foregoing. More specificexamples of the computer readable storage medium would include an electrical connectionhaving one or more wires, a portable computer diskette, a hard disk, a random access memory(RAM), a read-only memory (ROM), an erasable programmable read-only memory(EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory(CD-ROM), an optical storage device, a magnetic storage device, or any suitable combinationof the foregoing. The term “non-transitory,” as used herein, is a limitation of the mediumitself (i.e., tangible, not a signal) as opposed to a limitation on data storage persistency (e.g.,RAM vs. ROM).

[0148] Further, while operations are depicted in a particular order, this should not beunderstood as requiring that such operations be performed in the particular order shown orin sequential order, or that all illustrated operations be performed, to achieve desirable results. In certain circumstances, multitasking and parallel processing may be advantageous. Likewise, while several specific implementation details are contained in the above discussions, these should not be construed as limitations on the scope of the present disclosure, but rather as descriptions of features that may be specific to particular embodiments. Certain features that are described in the context of separate embodimentsmay also be implemented in combination in a single embodiment. Conversely, variousfeatures that may be described in the context of a single embodiment may also beimplemented in multiple embodiments separately or in any suitable sub-combination.

[0149] Although the present disclosure has been described in languages specific tostructural features and / or methodological acts, it is to be understood that the present disclosure defined in the appended claims is not necessarily limited to the specific featuresor acts described above. Rather, the specific features and acts described above may bedisclosed as example forms of implementing the claims.

Claims

WHAT IS CLAIMED IS:

1. A first device comprising:at least one processor; and at least one memory storing instructions that, when executed by the at least oneprocessor, cause the first device at least to:receive, from a second device, inference capability information of the second device for performing an inference task associated with interference management; determine, based on the inference capability information, an inference task indicator of whether to enable or disable the inference task at the second device; and transmit the inference task indicator to the second device.

2. The first device of claim 1, wherein the inference capability information isreceived at least one of the following: after a connection is established between the first device and the second device; or at beginning of a signalling epoch.

3. The first device of claim 1 or 2, wherein the inference capability informationcomprises: a first value indicating that the second device is capable of performing the inference task; or a second value indicating that the second device is incapable of performing the inference task.

4. The first device of claim 1 or 2, wherein:the inference capability information comprises a ratio of a first quantity to a secondquantity for a signalling epoch;the first quantity represents a first metric of energy required to perform the inferencetask; and the second quantity represents a second metric of energy available at the seconddevice at the signalling epoch.

5. The first device of any of claims 1-4, wherein the first device is further caused to:receive, from the second device, a performance value indicating performanceachieved by including or excluding the inference task for the interference management.

6. The first device of claim 5, wherein the performance value is periodicallyreceived from the second device.

7. The first device of claim 5 or 6, wherein the inference task indicator is a firstinference task indicator, and the first device is further caused to: determine, based on the inference capability information and the performance value, a second inference task indicator; and transmit, to the second device, the second inference task indicator.

8. The first device of claim 7, wherein the second inference task indicator isdetermined further based on at least one of the following: coverage statistics information corresponding to an estimate of user equipment (UE)density and mobility statistics of UEs associated with at least one second device; ora time since last indication (TSLI) value.

9. The first device of claim 8, wherein the first device is further caused to at leastone of the following: estimate the coverage statistics information through at least one signalling exchangebetween the first device and the at least one second device; or receive the coverage statistics information from a subset of the at least one seconddevice periodically.

10. The first device of claim 8 or 9, wherein the coverage statistics information forthe second device is represented as a weighted sum of a first number of UEs associated withthe second device and a second number of active UEs windowed over a number W ofsignalling epochs.

11. The first device of claim 10, wherein the first number has a weight of α and thesecond number has a weight of (1-α), and wherein the number W and the weight α are oneof the following: predetermined;determined by the second device and indicated to the first device; or determined by the first device and indicated to the second device.

12. The first device of any of claims 8-11, wherein the TSLI value of the seconddevice at a current signalling epoch is determined based on: the TSLI value of the second device at a previous signalling epoch; anda random variable for the previous signalling epoch, wherein a value of the randomvariable is determined based on the inference task indicator at the previous signalling epochand a performance value at the current signalling epoch.

13. The first device of any of claims 1-12, wherein:the inference capability information comprises an indication of a selected one of afirst operation mode and a second operation mode;in the first operation mode, the second device performs the inference task by defaultafter establishing a connection with the first device; andin the second operation mode, the second device performs the inference task based onreceiving an inference task indicator indicating to enable the inference task, withoutperforming the inference task by default after establishing a connection with the first device.

14. The first device of any of claims 1-13, wherein at least one of the following:the first device is a base station; or the second device is an access point of a sub-network.

15. A second device comprising:at least one processor; and at least one memory storing instructions that, when executed by the at least oneprocessor, cause the second device at least to:transmit, to a first device, inference capability information of the second device for performing an inference task associated with interference management; and receive, from the first device, an inference task indicator of whether to enableor disable the inference task at the second device.

16. The second device of claim 15, wherein the inference capability information istransmitted at least one of the following:after a connection is established between the first device and the second device; or at beginning of a signalling epoch.

17. The second device of claim 15 or 16, wherein the inference capabilityinformation comprises: a first value indicating that the second device is capable of performing the inference task; or a second value indicating that the second device is incapable of performing the inference task.

18. The second device of claim 15 or 16, wherein:the inference capability information comprises a ratio of a first quantity to a secondquantity for a signalling epoch;the first quantity represents a first metric of energy required to perform the inferencetask; and the second quantity represents a second metric of energy available at the seconddevice at the signalling epoch.

19. The second device of any of claims 15-18, wherein the second device is furthercaused to: determine a performance value indicating performance achieved by including orexcluding the inference task for the interference management; andtransmit the performance value to the first device.

20. The second device of claim 19, wherein the performance value is periodicallytransmitted to the first device.

21. The second device of claim 19 or 20, wherein the second device is caused todetermine the performance value by: based on determining that the inference task indicator indicates to enable the inference task, performing the inference task to obtain inference output; combining the inference output and channel quality indicator (CQI) information received from at least one user equipment (UE) associated with the second device to obtain combined information;performing at least one layer 2 (L2) operation based on the combined information; and computing the performance value based on the at least one L2 operation.

22. The second device of any of claims 19-21, wherein the performance value for acurrent signalling epoch is determined based on:the performance value of an L2 operation as a result of incorporating the inferencetask at a previous signalling epoch;the inference task indicator transmitted from the first device to the second deviceduring the previous signalling epoch;a performance score of the L2 operation averaged between epochs starting from a lastsignalling epoch in which the second device performed the inference task to the previoussignalling epoch; and the number of L2 operations for the interference management.

23. The second device of any of claims 19-22, wherein the inference task indicatoris a first inference task indicator, and the second device is further caused to: receive, from the first device, a second inference task indicator, which is determined based on the inference capability information and the performance value.

24. The second device of any of claims 15-23, wherein the second device is furthercaused to: perform the inference task based on a first operation mode in which the second deviceperforms the inference task by default after establishing a connection with the first device;perform the inference task based on a second operation mode in which the seconddevice performs the inference task based on receiving an inference task indicator indicatingto enable the inference task, without performing the inference task by default afterestablishing a connection with the first device; orselect one of the first operation mode and the second operation mode and an indicationof the selected operation mode is included in the inference capability information.

25. The second device of any of claims 15-24, wherein at least one of the following:the first device is a base station; or the second device is an access point of a sub-network.

26. A method comprising:receiving, by a first device from a second device, inference capability information of the second device for performing an inference task associated with interference management; determining, by the first device based on the inference capability information, an inference task indicator of whether to enable or disable the inference task at the second device; and transmitting, by the first device, the inference task indicator to the second device.

27. A method comprising: transmitting, by a second device to a first device, inference capability information of the second device for performing an inference task associated with interference management; and receiving, by the second device from the first device, an inference task indicator ofwhether to enable or disable the inference task at the second device.

28. An apparatus comprising:means for receiving, from a second device, inference capability information of the second device for performing an inference task associated with interference management; means for determining, based on the inference capability information, an inference task indicator of whether to enable or disable the inference task at the second device; and means for transmitting the inference task indicator to the second device.

29. An apparatus comprising: means for transmitting, to a first device, inference capability information of the second device for performing an inference task associated with interference management; and means for receiving, from the first device, an inference task indicator of whether toenable or disable the inference task at the second device.

30. A non-transitory computer readable medium comprising program instructionsthat, when executed by an apparatus, cause the apparatus at least to perform the method ofclaim 26 or 27.

Images