WO2019001743A1

WO2019001743A1 - Network device for a wireless communication network, and method thereof

Info

Publication number: WO2019001743A1
Application number: PCT/EP2017/066370
Authority: WO
Inventors: Jingyi Liao; Marcus Kahn
Original assignee: Huawei Technologies Co., Ltd.
Priority date: 2017-06-30
Filing date: 2017-06-30
Publication date: 2019-01-03

Abstract

It is an object to provide a network device for a wireless communication network. According to a first aspect, the network device comprises: an at least one antenna configured to receive reference data of an at least one user equipment (UE); and a processor configured to: determine a coherence time for the UE based on the reference data; determine at least one trustable estimation area (TEA) having a starting point at the receipt of the reference data and an ending point separated from the starting point by the coherence time for the UE; and allocate data to be transmitted for the at least one UE into at least one processing queue based on the at least one TEA. The network device may provide better utilization of available channel resources, and accordingly the performance and capacity of the wireless communication network may be improved.

Description

NETWORK DEVICE FOR A WIRELESS COMMUNICATION NETWORK,

AND METHOD THEREOF

TECHNICAL FIELD [0001] The present application relates to the field of wireless communications, and more particularly to a network device for processing user equipment (UE) events in a wireless communication network, and a method thereof.

BACKGROUND [0002] In a wireless communication network, the application of multiple-input and multiple-output (MIMO) based operation, enables a base station (BS) to communicate with multiple user equipment (UE) simultaneously and offers a higher system throughput. One of the main challenges of MIMO is to avoid interference among the multiple UE, so as to allow communication with multiple UE on same radio resource. In the downlink, this may be accomplished by spatial precoding signals at the BS; however, this may only be achieved if the BS has correct information about channel state information (CSI) of the communication link with the UE. Low quality CSI could severely degrade the throughput performances of MIMO.

[0003] The next generation wireless communication networks (like, 5G) would require to fulfill several hundreds of thousands of simultaneous connections as compared to the present communication networks. More the number of connected UEs, the more computational capacity required to process UE requests or the like, and therefore higher the cost of the system. Therefore, the known methods may not be feasible for implementation in all types of wireless communication networks, e.g., if the network has limited computation resources.

[0004] The implementation of MIMO in a wireless communication network necessitates the need of new methods for processing the CSI. SUMMARY

[0005] This summary is provided to introduce a selection of concepts in a simplified form that are further described below in the detailed description. This summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.

[0006] It is an object to provide a network device for a wireless communication network. The object is achieved by the features of the independent claims. Further implementation forms are provided in the dependent claims, the description and the figures.

[0007] According to a first aspect, a network device for a wireless communication network is provided. The network device comprises: an at least one antenna configured to receive reference data of an at least one user equipment (UE); and a processor configured to: determine a coherence time for the UE based on the reference data and an at least one estimation technique for estimating channel state information (CSI); determine at least one trustable estimation area (TEA) having a starting point at the receipt of the reference data and an ending point separated from the starting point by the coherence time for the UE; and allocate data to be transmitted for the at least one UE into at least one processing queue based on the at least one TEA. The network device may provide better utilization of available channel resources, and accordingly the MIMO performance and capacity of the wireless communication network may be increased.

[0008] In a further implementation form of the network device, the processor is configured to determine a beam characteristics of a received beam with the reference data, the beam characteristics including at least one of a beam width and an angular spread, and determine the coherence time based on the beam characteristics. This allows adjusting the estimated channel state according to the spatial information related to UEs. For example, the coherence time may be longer in duration when the beam width and the angular spread is small, and vice- versa.

[0009] In a further implementation form of the network device, the processor is configured to allocate data to be transmitted for at least two UEs into same processing queue when the TEAs for the UEs overlap. This may improve resource utilization in the wireless communication network, for example when multi-user MIMO (MU-MIMO) is utilized.

[0010] In a further implementation form of the network device, the processor is configured to share a transmission resource between UEs allocated in the same processing queue. In this way, the transmission resource is used for two UEs only when both of them have CSI of sufficient quality, which in turn may reduce negative effects arising from interference.

[0011] In a further implementation form of the network device, the processor is configured to determine at least one non-trustable estimation area (NTEA) having a starting point at the end of the TEA and ending point either open or at the next scheduled receipt of reference data for the UE. NTEA information may be utilized to filter out UE events with outdated CSI, thereby reducing the process overhead of the wireless communication network. These UE events may, optionally, be processed in a different way, for example in a single-user MIMO (SU-MIMO) transmission mode so that they do not interfere with the UEs or UE events utilizing MU-MIMO.

[0012] In a further implementation form of the network device, the processor is configured to determine scheduling results for transmission in the wireless communication network or the load of at least one of the wireless communication network and the processor, compare the scheduling results or the load to a threshold and to perform at least one of selecting the estimation technique based on whether the scheduling results or the load exceeds the threshold; and increase the size of the TEA if the scheduling results or the load is below the threshold. This may be done to achieve a trade-off between the performance and the process overhead in the wireless communication network.

[0013] In a further implementation form of the network device, the processor is configured to determine the speed of the UE and perform at least one of decrease the size of the TEA for the UE if the speed of the UE is larger than a threshold speed; and increase the size of the TEA for the UE if the speed of the UE is smaller than the threshold speed. This may be done to regulate the number of UE events in the processing queue, for example, based on a load scenario in the wireless communication network.

[0014] In a further implementation form of the network device, the processor is configured to define at least two TEAs for at least one UE, wherein the different TEAs are associated with at least one of different transmission frequencies; different spatial coordinates for the UE; and different types of data to be transmitted for the UE. This allows optimizing the channel quality estimate with respect of different types of parameters using different types of information available. For example, this may be done to compensate for the possible interference in certain frequency and spatial coordinates or to adjust processing of different types of data to be transmitted for the UE.

[0015] In a further implementation form of the network device, the processor is configured to prioritize the UE events in the at least one processing queue based on the coherence time for the corresponding UE. This may help to prioritize different UEs' events in TEA, e.g., high process priority for UE events with TEA lifetime close to the end, and thus could be utilized before expiration.

[0016] In a further implementation form of the network device, the processor is configured to adjust the size of the TEA by defining at least one of threshold for latency for receiving the reference data (Tl) and threshold for maximum TEA (T2) so that the size of the TEA becomes

size = mm(T2, max(coherence time, Tl)).

By giving a non-zero value for Tl, which by default would be zero, also the UEs with high latency may be admitted a non-zero TEA allowing them to obtain the corresponding processing and/or scheduling benefits. By giving a finite value for T2, which by default would be infinity, scheduling fairness between UEs may be improved so that UEs having small coherence times are not necessarily always favored in comparison to UEs having large coherence times.

[0017] In a further implementation form of the network device, the processor is configured to determine a transmission mode for the UE to be used in at least one of uplink transmission and downlink transmission based on the corresponding one or more TEAs. For example, if the processor divides the measurement period into TEA and NTEA, then MU-MIMO mode may be selected for the UEs in TEA, and SU-MIMO mode may be selected for UEs in NTEA.

[0018] According to a second aspect, a method of processing events of at least one user equipment (UE) in a wireless communication network is disclosed. The method comprises: receiving, by at least one antenna, reference data of the at least one UE; determining a coherence time for the UE based on the reference data and an at least one estimation technique for estimating channel state information (CSI); determining at least one trustable estimation area (TEA) having a starting point at the receipt of the reference data and an ending point separated from the starting point by the coherence time for the UE; and allocating data to be transmitted for the at least one UE into at least one processing queue based on the at least one TEA. The method may provide better utilization of available channel resources, and accordingly the MIMO performance and capacity of the wireless communication network may be increased.

[0019] In a further implementation form of the method, the method comprises determining a transmission mode for the UE (12a, 12b...) to be used in at least one of uplink transmission and downlink transmission based on the corresponding one or more TEAs. For example, if the measurement period is divided into TEA and NTEA, then MU-MIMO mode may be selected for the UEs in TEA, and SU- MIMO mode may be selected for UEs in NTEA.

[0020] In a further implementation form of the method, the method comprises prioritizing the UE events in the at least one processing queue based on the coherence time for the corresponding UE. This may help to prioritize different UEs' events in TEA, e.g., high process priority for UE events with TEA lifetime close to the end, and thus could be utilized before expiration.

[0021] According to a third aspect, a computer program is provided, comprising program code configured to perform a method according to the second aspect when the computer program is executed on a computer.

[0022] Many of the attendant features will be more readily appreciated as they become better understood by reference to the following detailed description considered in connection with the accompanying drawings.

DESCRIPTION OF THE DRAWINGS

[0023] The present description will be better understood from the following detailed description read in light of the accompanying drawings, wherein:

[0024] FIG. 1 illustrates a schematic representation of a network device in a wireless communication network according to an embodiment;

[0025] FIG. 2 illustrates an example of a process sequence;

[0026] FIG. 3 illustrates a plot showing TEA sizes for two exemplary UEs according to a comparative example;

[0027] FIG. 4 illustrates a flowchart according to an embodiment;

[0028] FIG. 5 illustrates a graph showing a comparison between the achieved throughput of the wireless communication network between a legacy network and networks adopting TEA for multi-user pairing according to a comparative example;

[0029] FIG. 6 illustrates a graph showing a comparison between the memory re- usage in the wireless communication network between a legacy network and networks adopting TEA for multi-user pairing according to a comparative example; and

[0030] FIG. 7 illustrates a graph showing an effect on number of scheduled UE events in the wireless communication network between a legacy network and networks adopting TEA for multi-user pairing according to a comparative example.

[0031] Like references are used to designate like parts in the accompanying drawings.

DETAILED DESCRIPTION [0032] The detailed description provided below in connection with the appended drawings is intended as a description of the embodiments and is not intended to represent the only forms in which the embodiment may be constructed or utilized. However, the same or equivalent functions and structures may be accomplished by different embodiments.

[0033] Multiple-input multiple-output (MIMO) technology is a promising candidate for next- generation wireless communication networks. In order to achieve a high data rate over channels, MIMO implements precoding across both spatial and temporal dimensions to maximize the number of simultaneous connections without sacrificing channel bandwidth. The effective use of MIMO relies on good knowledge of the channel characteristics, i.e. channel state information (CSI), to select the optimal precoding for the transmissions to user equipment (UEs). However, good knowledge of the channel may be difficult to obtain in high speed scenarios of the UEs, due to rapid changes in the channel's characteristics. There is always an inherent delay in the CSI reporting to the base station, and in the event of rapid channel changes, such as those that might occur when the UE is moving rapidly, the CSI information might be outdated by the time it is used to pre-code the data. [0034] Similarly, accurate knowledge of the transmission channel may be unavailable in other circumstances as well, such as when UE first goes into active mode and does not have accurate channel estimates, or when the Signal to Noise Ratio (SNR) at the UE is very low, or when the UE is not properly configured to send precoding information to the base station. It may be possible that some incorrect precoding method may thus be applied in any of these circumstances, e.g., precoding optimized for radio channel as it was few milliseconds earlier rather than for the current state of the channel. Applying a precoding based on outdated CSI might actually make the receiving conditions worse, e.g. there might be destructive interference between the radio signals from the multiple antennas.

[0035] According to an embodiment, a network device for a wireless communication network is provided. The network device comprises at least one antenna to receive reference data of at least one UE in the wireless communication network. The network device further comprises a processor which separates a measurement period for a UE into areas depending on whether their CSI may be trusted, hereinafter referred to as trustable estimation area (TEA). For this purpose, the processor determines a coherence time for the UE based on the reference data and an at least one estimation technique for estimating CSI. The TEA is defined as an area having a starting point at the receipt of the reference data and an ending point separated from the starting point by the coherence time for the UE. Further, the processor filters events from the UE into one or more processing queues based on the TEA. The network device of the present disclosure utilizing TEA to filter out UE events may be useful for applications in which the network device's performance is based on the estimation accuracy, and their effective estimation periods is in average less than the measurement period.

[0036] The channel prediction methods and parameters are selected to adjust the system load and tradeoff between performances and system's process overhead. That is, the network device efficiently utilizes CSI information by tuning the estimation techniques and parameters to divide a measurement period into different TEAs for the UE, and have the different pairing solutions for UEs with different TEAs. In a high load scenario, the network device targets for event reduction while making a tradeoff between the performance and algorithm complexity; whereas, in low load scenarios, the network device uses more accurate and high order channel prediction methods even with more complexity, so that the TEA size is increased and more number of UEs may be processed.

[0037] Although primary embodiments have been described relating to the reference data being classified either as TEA or NTEA for handling multi-user (MU) pairing in the wireless communication network, in other embodiments, such reference data may be classified into more than two categories and such information may be used for purposes such as HARQ retransmission without any limitations.

[0038] FIG. 1 schematically describes a wireless communication network 10 (also, sometimes simply referred to as "network") configured for multiple-input and multiple-output, MIMO, operations in which various devices communicates over multiple radio communication channels. In specific embodiments, the wireless communication network 10 is described in the context of systems implementing Long-Term Evolution (LTE) technology, also known as evolved UMTS Terrestrial Radio Access Network (E-UTRAN), as standardized by the membership of the 3rd Generation Partnership Project (3GPP). It will be understood, however, that the present disclosure is not limited to such embodiments and may be embodied generally in various types of communication networks. According to an embodiment, a network device 100 is provided in the wireless communication network 10 which is configured to process events of one or more user equipment (UEs) (12a, 12b...) therein. The use of terminology such as "network device" (which may sometimes be simply referred to as "device", for example) and "User Equipment" (often referred to as "UE," or in some cases as "wireless terminal," "mobile terminal," or "wireless device") shall be considered non-limiting to the disclosure.

[0039] In the present context, the term "user equipment (UE)" may refer to any device enabled to communicate voice or data, via the radio access network, with another entity, such as another receiver or a server, and may include, but are not limited to, a mobile telephone ("cellular" telephone), laptop/portable computer, pocket computer, hand -he Id computer, modem and/or desktop computer. The examples of user equipment (UEs) may also include machine-to-machine type communication devices, used without direct human interaction, such as sensors. In general, UE may be a Station (STA) which is any device that contains an IEEE 802.11 -conformant Media Access Control (MAC) and Physical Layer (PHY) interface to the Wireless Medium (WM). Further, the term "network device" may refer to a base station, a (radio) network node or an access node or an access point or a radio base station (RBS), which in some networks may also be referred to as transmitter, "eNB", "eNodeB", "NodeB" or "B node", depending on the technology and terminology used. The network nodes may be of different classes, such as, e.g., macro eNodeB, home eNodeB or pico base station, based on transmission power and thereby also cell size. The network node may be a Station (STA) which is any device that contains an IEEE 802.1 1 -conformant Media Access Control (MAC) and Physical Layer (PHY) interface to the Wireless Medium (WM). In general, the network device 100 and the user equipment (12a, 12b...) may be considered as examples of respective different communications devices that communicate with each other over a wireless radio channel.

[0040] In particular, the network device 100 comprises an at least one antenna (110a, 110b..) configured to receive reference data of an at least one user equipment, UE (12a, 12b...). In the exemplary illustration of the wireless communication network 10 in FIG. 1, the network device 100 is shown to include three antennas (1 10a, 1 10b, 1 10c) in communication with three UEs (12a, 12b, 12c). It may be understood that the number of the antennas (1 10a, 1 10b..) and the UEs (12a, 12b...) shown in FIG. 1 are exemplary only and shall not be construed as limiting to the disclosure. The arrows in FIG. 1 represent transmissions and/or receptions between the antennas (1 10a, 1 10b, 110c) and the UEs (12a, 12b, 12c). The antennas (1 10a, 110b..) may receive reference data of the UEs (12a, 12b..) either directly or via some access points or intermediate nodes, not shown in FIG. 1. The reference data in the present context may refer to any data corresponding to the UEs (12a, 12b..) which may be processed to determine the channel characteristics, e.g. channel state information (CSI) in the wireless communication network 10. In the present example, the network device 100 is configured to use both the periodic as well as aperiodic reference data without any limitations. The channel state information (hereinafter referred to as "CSI") refers to information that is based on the instantaneous channel state, and includes, for example, channel quality information (CQI), the precoding matrix indicator (PMI), the rank indicator (RI) and so on. For example, Sounding Reference Signal (SRS) is a reference signal transmitted by UE in the uplink direction, and SRS may be used by the eNB to estimate uplink channel quality, and this estimated channel quality may be directly used for MIMO networks implementing Time Division Duplex (TDD), due to the channel reciprocity for TDD. It may be understood that different reference signals, other than SRS, may be used for estimating CSI, including, but not limited to, Cell-Specific RS (C-RS), Channel State Information RS (CSI-RS), pilot reference signal, and further a combination of uplink RS and downlink RS, without any limitations.

[0041] The network device 100 also comprises a processor 120, which may be an example of a signal processing device. Although FIG. 1 illustrates only one processor, the network device 100 may have several distributed processors so that each antenna (1 10a, 1 10b...) has a processor of its own. As illustrated, the processor 120 is electrically coupled to receive multiple feeds of antennas (1 10a, 1 10b...) as shown by the arrows in FIG. 1. The processor 120 receives reference data of one or more UE (12a, 12b..) via the antennas (1 10a, 110b...), processes the reference data for each UE (12a, 12b..), and transmits data for each UE (12a, 12b..). The processor 120 may be embodied as a multi-core processor, a single core processor, or a combination of one or more multi-core processors and one or more single core processors. For example, the one or more processors may be embodied as one or more of various processing devices, such as a coprocessor, a microprocessor, a controller, a digital signal processor (DSP), a processing circuitry with or without an accompanying DSP, or various other processing devices including integrated circuits such as, for example, an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), a microcontroller unit (MCU), a hardware accelerator, a special-purpose computer chip, or the like. Typically, the processor 120 is associated with a memory (not shown) to store the data and instructions for enabling the processor 120 to carry out various functions. The memory may include, but is not limited to, volatile and/or non-volatile memories. For instance, the memory may be volatile memory (e.g., registers, cache, RAM), no n- volatile memory (e.g., ROM, EEPROM, flash memory, etc.), or some combination of the two.

[0042] According to an embodiment, the processor 120 is configured to determine a coherence time for the UE (12a, 12b..) based on the reference data. Coherence time is the time duration over which the channel impulse response is considered to be not varying; or in other words, it is the time interval within which the phase of the receiving signal with the reference data is, on average, predictable. The coherence time may be based on the beam characteristics of a received beam having the reference data. In one example, the beam characteristics may be dependent on at least one of a beam width and an angular spread thereof; however, various other factors may also affect the beam characteristics and taken into consideration for determining the coherence time. As maybe contemplated by a person skilled in the art that larger beam width and/or angular spread may lead to shorter coherence time and vice versa. In some examples, the beam-forming may be controlled to have reduced beam width and angular spread, and thus the coherence time for the UE (12a, 12b...) may be increased. The techniques for controlling the beam-forming are known in the art and are beyond the scope of the present disclosure. It is, however, noted that the coherence time is related to the time over which a propagating wave (especially a laser or maser beam) may be considered coherent. Coherent time reflects the time varying channel effect, and in a system with directional antennas or beamforming method, the coherent time is also correlated to the beam width and the angular spread. Furthermore, the processor 120 is configured to determine an estimation technique for estimating the CSI. The selected estimation technique may be based on multiple factors, including, but not limited to, one or more of Angle of Arrival (AoA), Doppler affect, channel coherent time, spatial channel, etc. In some examples, the estimation technique may also involve prediction for the channel state with respect to one or more parameters, wherein prediction is to be understood as estimating the channel state at one or more future points in time.

[0043] As described above, MIMO based communication networks work well when accurate knowledge of CSI is available. However, to support multiple UEs in a network, the CSI feedback or SRS period have to be prolonged since the radio resource is always a bottleneck for transmission. For example, SRS could only be granted for limited users due to the limitation of uplink (UL) air resources. If a system has to support more users with SRS resources, the SRS periods have to be enlarged. Consider an example, if SRS is only configured in TDD special sub- frames, and the number with CDM-SRS is 4 users, it could only support 32 users with SRS if SRS period is 10ms, and if the system asks for supporting 384 users, the SRS period have to prolong to 120ms. On the other hand, this long SRS period may have already exceeded the effective channel estimation period for UEs (12a, 12b..), especially for UEs moving with high velocities, and therefore such UEs could not have the trustable estimation information during a total SRS measurement period.

[0044] To overcome this, the processor 120 is configured to determine at least one trustable estimation area (TEA) having a starting point at the receipt of the reference data and an ending point separated from the starting point by the coherence time as determined for the UE (12a, 12b...), in accordance with an embodiment. In some examples, the processor 120 is also configured to determine at least one non-trustable estimation area (Non-TEA, or simply "NTEA") having a starting point at the end of the TEA and ending point either open or at the next scheduled receipt of reference data for the UE (12a, 12b...). For example, let's say the processor 120 determines that the coherence time for the UE is 50ms, i.e. the CSI could be considered trustable only for 50ms. Then, the processor 120 may define TEA for time interval 0-50ms, where t=0 is the moment for receipt of the reference data from the UE. Further, NTEA may be defined as an interval 50ms- infinity, which means that after 50ms the reference data is no longer considered as trustable for determining CSI. Also, it may be considered that sometime after period of 50ms, for example at t=120ms, the processor 120 may receive another signal from UE (12a, 12b..) with the reference data; then, TEA may be defined anew for time interval 120-170ms, and a new half-open interval (i.e. one without a pre-specified end point) may be defined for NTEA starting from 170ms, and so on. In one example, if the processor 120 only divides a measurement period into two areas, i.e. TEA and NTEA, the definition of TEA for a UE (12a, 12b..) is decided by equation (1) below, in which Γ is the time of receiving the CSI measurement report, '?2' is the effective time period for the estimation data from tl :

TEA area = {tl, t2}, t2 < T^_easurerrient (1)

[0045] It may be understood that the processor 120 receives the reference data of the UE (12a, 12b..) in any manner. For example, UE (12a, 12b..) may send the reference data independently or the reference data may be obtained after the antennas (1 10a, 110b..), or the base station, requests the reference data from the UE (12a, 12b..). In the latter case, there might be a possibility that UE (12a, 12b..) may never have a valid TEA but only NTEAs, e.g. in a case where a UE has reliable CSI only for 3ms and the measurement delay/latency is 5ms. According to an embodiment, the processor 120 is configured to adjust the size of the TEA by defining at least one of threshold for latency for receiving the reference data (Tl) and threshold for maximum TEA (T2), so that the size of the TEA may be defined as per equation (2) below: size = mm(T2, max(coherence time, 71)) (2)

[0046] According to an embodiment, the processor 120 is configured to define at least two TEAs for each of the UEs (12a, 12b...), wherein the different TEAs are associated with at least one of different transmission frequencies and/or different spatial coordinates for the UEs (12a, 12b...). Further, in some examples, the processor 120 is configured to define at least two TEAs for each of the UEs (12a, 12b...), when different types of data are to be transmitted for the UEs (12a, 12b...). It may be contemplated by a person skilled in the art that this may be done to compensate for the possible interference in certain frequency and spatial coordinates and for different types of data to be transmitted for the UEs (12a, 12b...).

[0047] The processor 120 is further configured to allocate data to be transmitted for the at least one UE (12a, 12b...) into at least one processing queue based on the at least one TEA. In some examples, the subsequent processes for the filtered UE queues could be utilized for multi-user (MU) pairing, precoding and other adaptive signal process, or just a single UE transmission. Assuming that the processor 120 only allows MU pairing for the UEs in TEA group, then the UEs with outdated estimation data (as determined by their TEAs) may be filtered out from the MU pairing events. That is, the network device 100 utilizes the trustable estimation areas (TEAs) information to identify UEs which could be paired or the like by filtering out the UEs (12a, 12b...) which do not have the effective channel information or the UEs (12a, 12b...) with NTEAs information. In other words, if the network device 100 only define a measurement period _McasUr_Eme t' ^mto two areas, i.e. TEA and NTEA; then, when a system process time ^{i f}t_process' is in a UEs TEA, the UE is kept in the TEA group, otherwise, the UE is moved outside from the TEA group as represented in equation (3) below:

UE inside TEA group ^■process G TEA, area

UE outside TEA group ^■process £ TEA, (3)

area [0048] According to an embodiment, the processor 120 is further configured to tune the estimation technique employed for determining the TEA to adjust the event quantities for each area. For this purpose, the processor 120 determines scheduling results for transmission in the wireless communication network 10 or the load of at least one of the wireless communication network 10 and the processor 120. The processor 120 further compares the scheduling results or the load to a threshold. The processor 120 also performs at least one of selecting the estimation technique based on whether the scheduling results or the load exceeds the threshold, and increase the size of the TEA if the scheduling results or the load is below the threshold. It may be understood that the threshold, as used herein, may be based on the computational capacity of the processor 120. In some examples, the processor 120 may tune the TEA to balance the scheduling priority among different UEs with different behaviors. It may be understood that the definition of the effective estimation area itself is decided by the selected estimation technique as well as the corresponding UEs behavior, e.g. whether TEA is being defined by coherent time or Doppler effect. In such case, a UE with a faster movement may have a smaller effective estimation time as compared to a UE with a slow movement. As shown in equation (3), the TEA decides the UE persistent time in the TEA group, therefore, tuning TEA for per UE could decide the UEs pairing and scheduling opportunity in a TEA group.

[0049] According to an embodiment, the processor 120 is configured to allocate data to be transmitted for at least two UEs (12a, 12b...) into same processing queue when the TEAs for the UEs (12a, 12b...) overlap. Furthermore, according to an embodiment, the processor 120 is configured to share a transmission resource between UEs (12a, 12b...) allocated in the same processing queue. In a wireless communication network, such as the network 10 with multiple connected UEs (12a, 12b...), even if only part of UEs (12a, 12b...) have trustable channel information in one of limited trustable channel estimation period, other UEs (12a, 12b...) could have their trustable channel information in another limited effective period. Therefore, the UEs' processing events within their TEAs could be averagely distributed during a measurement reporting period.

[0050] FIG. 2 provides an exemplary schematic 130 of the process sequence of MU pairing showing filtering of UE events for different TEAs in time domain. As illustrated in FIG. 2, if the MU pairing is only allowed for UEs in TEA group, the processor 120 may allocate data in the UE event queues based on the time comparison between the processing time and the measured TEA for each UE (12a, 12b...). Assuming for the same downlink (DL) transmission resource, the channel estimation for UEs: 12a, 12b, 12c, 12d may be based on CSI feedbacks completed at tO, tl, t2 and t3, respectively. Based on an exemplary TEA based UE filtering process; at the scheduling time t4, only UEs 12a, 12b, 12d are in the TEA group and UE 12c is in non-TEA group even if UE 12c's CSI feedback is later than UEs 12a, 12b CSI feedbacks. This may happen, for example, if the processor 120 considers Doppler affect for defining the TEA, and if UE 12c is faster than UE 12a, 12b. The filtered UEs are used for MU pairing process, i.e. UE 12a, 12d are finally paired and pre-coded at t4 (as represented by block 132), and at t5, the data for UEs 12a, 12d is transmitted on the same DL transmission resource.

[0051] The idea of using TEA concept to filter out UE events could be useful for many applications, in which their performances are based on the estimation accuracy, and their effective estimation periods is in average less than the measurement period. Although in the given example, TEA has been defined in terms of time range; alternatively, TEA may be defined in terms of frequency range, spatial range, or any other parameter with respect to which the transmission resource may be divided. A single TEA can be defined so that it is associated with multiple parameter criteria: for example a coherence time and a frequency range. In addition, TEAs may be defined separately for different types of UE events, e.g. retransmission may have differently sized TEA in comparison to initial transmission. Furthermore, TEAs may be defined separately for each UE using the same or different criteria which may be dependent on the type of the UE device, the characteristics of the time-varying channel of the UE, the speed of the UE, etc. It may be understood that if MU pairing method may have paired these UEs but in a non-trustable estimation period, the system performance might have been much reduced, since the UE with non-trustable channel estimation leads to a heavy interference with other MU paired UEs. Therefore, the present solution better utilizes the effective channel estimation and not only reduces the process events, and thereby the system's process overhead, but also improves the system performance, as will be discussed in more detail in the subsequent paragraphs.

[0052] FIG. 3 illustrates an exemplary plot 140 of TEA definition for two UEs. As an example, retransmission (e.g. HARQ) has been considered but the principle of dividing TEAs may be applied to other types of UE events as well. The difference in TEA sizes may result, for example, from different velocities of the UEs but they may also be due to other reasons. In the illustrated example, a measurement/reporting period for reference data for different UEs is shown to be of equal length; however, in other examples such period for considered UEs may be different without any limitations. As illustrated in FIG. 3, the transmission time for the UEs may be divided into multiple transmit time intervals (TTIs). At TTI1 : UE#1 and UE#2 will be added to the same queue of new transmission in TEA. If both are scheduled but both have non-acknowledgement (NACK) for their transmission, they both have the retransmission events at TTI5. At TTI5, their retransmission will be added to two different UE pairing event queues, e.g. UE#1 will be added to the queue of retransmission in NTEA, while UE#2 will be added to the queue of retransmission in TEA. In such cases, MU pairing, for example, could then have different pairing policies for the different UE groups. In some examples, it could only pair the UEs in TEA, but do not pair the UEs in NTEA, this could largely reduce the pairing process events. Further, in some cases, different DL transmissions for the UEs may happen on different scheduling TTIs.

[0053] As discussed earlier, UEs with high velocity, generally, have relatively smaller TEAs compared to a UEs with relatively low velocity. It may be understood that the information about speed of the UE may be included in the reference data for the corresponding UE. According to an embodiment, the processor 120 is configured to determine the speed of the UE (12a, 12b...) and decrease the size of the TEA for the UE (12a, 12b...) if the speed of the UE (12a, 12b...) is larger than a threshold speed. Furthermore, the processor 120 is configured to increase the size of the TEA for the UE (12a, 12b...) if the speed of the UE (12a, 12b...) is smaller than a threshold speed. Alternatively, multiple threshold speeds, or in general a direct relationship between the TEA size and speed of the UE may be defined for such calculations. In this sense, if a system has more UEs with a relative fast velocity, the more UEs' pairing events could be filtered out, and even more the memory for the estimation data may be released. On the other hand, since only UEs in TEA are in the MU pairing list, it may improve the system performance since it could reduce multi-user interference by granting the paired users with more trustable estimation information. This may further guarantee that an appropriate number of UEs be processed, e.g. for MU pairing, while reducing the overall computation complexity and cost.

[0054] According to an embodiment, the processor 120 is configured to prioritize the UE events in at least one processing queue based on the coherence time for the corresponding UE (12a, 12b...). This may help the network device 100 to define different priority processes for different UEs' events in TEA, e.g., high process priority for UE events with TEA lifetime close to the end. If TEA is measured in terms of coherent time, high speed UEs usually have smaller TEA size as compared to low speed UEs. Therefore, the processor 120 may grant a higher process priority for a UE with high speed than it does for a UE with relatively low speed.

[0055] Further, according to an embodiment, the network device 100 may be used in a computation resource limitation scenario by adapting TEA estimation algorithms and parameters for tuning the TEA to adjust the system load and tradeoff between performances and system process overhead. As it is known that complicated effective channel prediction methods e.g. MMSE channel prediction models may improve the channel estimation accuracy and may also prolong the TEA to some extent. But it may be hard to adopt such models for all UEs in a high load scenario, as in such cases, the scheduling resource rather than the number of paired UEs might be the biggest bottleneck for the system improvement. In this sense, adopting a more accurate channel prediction model may not add big contribution to the system performance. But in a low load scenario, the number of paired UEs might be more limited than the scheduling resource. In such load scenario, the processor 120 may adopt more accurate and high order channel prediction models even if it is more computation intensive, as it could increase the TEA size, which could guarantee more UEs be processed, and therefore the system performance could be improved. In short, the processor 120 may select more accurate and high order channel prediction algorithms for low load scenario, but use simple channel estimation algorithms with the parameters to strictly limit the TEA size for UEs during high load scenario.

[0056] Further, as discussed, TEA size determines a UEs persistent time inside of TEA group or outside thereof. Therefore, the definition of TEA may determine the UE event queue size, and impact the system process overhead. According to an embodiment, in a network employing different estimation methods, each having different process complexity and estimation accuracy, the processor 120 may select algorithms or algorithm parameters which could change TEA size for different UE event queues, and therefore adapt to the system load. The processor 120 may adjust TEA parameter, e.g. by defining a threshold to adapt for different load scenarios, as in equation (4) below, wherein the TEA size for each UE is determined not only by coherent time T_{coherent tjme} ■_> but also by the two threshold parameters Thresholdl and Thresholdl, in which Thresholdl is help to add the UE even if T_coherent is smaller than the measurement process latency, and Thresholdl is used to grant a fairness scheduling among UEs and is used to limit the TEA_size for a UE if its T_coherent is quite larger, and thereby avoid the UE staying quite long time in the TEA queues comparing to other UEs.

TEA_size = m (Thresholdl, max(T_{coherent time} Thresholdl)) (4)

[0057] By adapting the estimation technique algorithm and parameter to the system load, the TEA size could be changed which may increase or decrease the UE event quantities of different event queues, and therefore configures the network device 100 for different load scenarios.

[0058] According to an embodiment, the processor 120 is configured to determine a transmission mode for the UE (12a, 12b...) to be used in at least one of uplink transmission and downlink transmission based on the corresponding one or more TEAs. For example, for MU pairing, the defined TEA for UEs may dynamically adopt their preferable MIMO modes in different time periods. As may be understood, if the processor 120 divides the measurement period into TEA and NTEA, then MU-MIMO mode may preferably be selected for the UEs (12a, 12b...) in TEA, and SU-MIMO mode may preferably be selected for UEs (12a, 12b...) in NTEA. Further, when TEAs may be defined in terms of spatial ranges, other MIMO transmission modes, e.g. spatial diversity, spatial multiplexing, etc. may be preferably implemented. Furthermore, each UE (12a, 12b...) may change their favorite MIMO mode at the time when TEA is expiring or new measurement is coming. Also, it may be contemplated by a person skilled in the art that although certain type of mode changes requires signaling in the downlink to indicate change in mode but the present implementation of the network device 100 may be used irrespective of whether the signal is sent or not. [0059] FIG. 4 illustrates a flowchart showing a method 200 of processing events of at least one user equipment (UE) (12a, 12b...) in the wireless communication network 10 according to an embodiment. In general, the method 200 may be implemented to separate a measurement period for a UE (12a, 12b...) into areas depending on whether their channel state information may be trusted and to filter events from the UE (12a, 12b...) based on the said classification.

[0060] In operation 210, the reference data of the at least one UE (12a, 12b...) is received. As discussed earlier, the reference data may be received by the antennas (110a, 1 10b...) either directly or via some access points or intermediate nodes. The said reference data may include information for estimating channel state information (CSI) for the one or more UE (12a, 12b...).

[0061] In operation 220, the coherence time for the UE (12a, 12b...) is determined based on the reference data and the at least one estimation technique for estimating channel state information (CSI). In one example, the coherence time is determined based on the beam characteristics of the received beam with the reference data, wherein the beam characteristics includes information about at least one of the beam width and the angular spread. In some examples, the said estimation technique may be selected based on various factors, including, but not limited to, the load scenario in the wireless communication network 10.

[0062] In operation 230, at least one trustable estimation area (TEA) having a starting point at the receipt of the reference data and an ending point separated from the starting point by the coherence time for the UE (12a, 12b...) is determined. In some examples, the processor 120 is also configured to determine at least one non-trustable estimation area (Non-TEA or simply "NTEA") having a starting point at the end of the TEA and ending point either open or at the next scheduled receipt of reference data for the UE (12a, 12b...).

[0063] Further, in operation 240, the data to be transmitted for the at least one UE (12a, 12b...) is allocated into at least one processing queue based on the at least one TEA. For example, in applications of MU pairing, the UEs (12a, 12b...) with outdated estimation data as determined by its TEA could be filtered out from the MU pairing events. In the wireless communication network 10 with multiple connected UEs, even if only part of UEs (12a, 12b...) have trustable channel information in one limited trustable channel estimation period, other UE (12a, 12b...) could have their trustable channel information in another limited effective period. Therefore, the UEs' processing events within their TEA could be averagely distributed during a measurement reporting period, thus improving the performance of the wireless communication network 10.

[0064] FIGS. 5-7 depict the effects of utilizing the network device 100 of the present disclosure in the wireless communication network 10. In particular, FIGS. 5-7 generally show the improvement in the resource utilization in the wireless communication network 10 using the network device 100 of the present disclosure. Specifically, these figures show the cell throughput performance gain and memory re-usage could be much improved with the TEA based UE filtering as compared to the legacy systems which do not consider the limitation of TEA.

[0065] FIG. 5 shows the effect of implementing the network device 100 or the corresponding method 200 to increase the throughput in the wireless communication network 10 by adopting TEA for MU pairing for new transmission only as well as for both new and retransmission situations. As may be seen from the graph, in some case, MU pairing for both new and retransmission in TEA may obtain up to 136% cell throughput gain over the MU pairing solution which do not consider the limitation of TEA. It may be understood that this could be achieved as the present method 200 may avoid the inter-user interference impact from the aged estimation information.

[0066] FIG. 6 shows the effect of implementing the network device 100 or the corresponding method 200 for efficient memory re-usage in the wireless communication network 10 by adopting TEA for MU pairing. In some examples, the said memory re-usage may be calculated as per the equation (5) below. As may be seen, implementing the network device 100 or the corresponding method 200 could significantly reduce the system process costs. For example, the present method 200 could improve memory usage, i.e. if dynamic memory allocation is used for the channel matrix related process. The stored channel information for each UE could be created and released based on the duration of TEA, and such TEA based UE filtering method could vastly improve the memory usage. Further, the present method 200 may also reduce the number of scheduled UEs, and the less the number of scheduled UEs, the less the post scheduling process overheads, e.g. less UEs for LI processing. This, in turn, may result in less utilization of control channel resources which reduces the costs associated with the computational capacity. (Measurement_Duration \

Memory Reuse

Duration '

I TEA_Durati_on < M easurement Duration (5)

[0067] FIG. 7 shows the effect on the computational resources if different TEA selection models are employed, in comparison with legacy methods. For FIG. 7, the data has been collected based on simulation of 480 connected UEs, each of them having SRS period of 100ms, and assuming that all UEs are moving with a speed of 30 Km/h. In the present example of FIG. 7, "Model 1" is a relatively stricter model for estimating TEA as compared to "Model 2" for defining TEA, for the same UE. It may be seen that the determined TEA based on Model 1 is about 4 times less than the one based on Model 2, both of which have shorter TEA compared to legacy methods. As discussed, the TEA size could determine the process overhead required for implementing, say MU pairing process. Therefore, if desired, the TEA could be tuned to balance the performance and process cost, for example, in limited computational capacity networks.

[0068] The functionality described herein can be performed, at least in part, by one or more computer program product components such as software components. According to an embodiment, the processor 120, in the network device 100, may be configured by the program code to execute the embodiments of the operations and functionality described. Alternatively, or in addition, the functionality described herein can be performed, at least in part, by one or more hardware logic components. For example, and without limitation, illustrative types of hardware logic components that can be used include Field-programmable Gate Arrays (FPGAs), Program-specific Integrated Circuits (ASICs), Program-specific Standard Products (ASSPs), System-on-a-chip systems (SOCs), Complex Programmable Logic Devices (CPLDs), Graphics Processing Units (GPUs).

[0069] Any range or device value given herein may be extended or altered without losing the effect sought. Also any embodiment may be combined with another embodiment unless explicitly disallowed.

[0070] Although the subject matter has been described in language specific to structural features and/or acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as examples of implementing the claims and other equivalent features and acts are intended to be within the scope of the claims.

[0071] It will be understood that the benefits and advantages described above may relate to one embodiment or may relate to several embodiments. The embodiments are not limited to those that solve any or all of the stated problems or those that have any or all of the stated benefits and advantages. It will further be understood that reference to 'an' item may refer to one or more of those items.

[0072] The steps of the methods described herein may be carried out in any suitable order, or simultaneously where appropriate. Additionally, individual blocks may be deleted from any of the methods without departing from the spirit and scope of the subject matter described herein. Aspects of any of the embodiments described above may be combined with aspects of any of the other embodiments described to form further embodiments without losing the effect sought.

[0073] The term 'comprising' is used herein to mean including the method, blocks or elements identified, but that such blocks or elements do not comprise an exclusive list and a method or apparatus may contain additional blocks or elements.

[0074] It will be understood that the above description is given by way of example only and that various modifications may be made by those skilled in the art. The above specification, examples and data provide a complete description of the structure and use of exemplary embodiments. Although various embodiments have been described above with a certain degree of particularity, or with reference to one or more individual embodiments, those skilled in the art could make numerous alterations to the disclosed embodiments without departing from the spirit or scope of this specification.

Claims

1. A network device (100) for a wireless communication network (10), the network device (100) comprising:

an at least one antenna (1 10a, 1 10b..) configured to receive reference data of an at least one user equipment UE (12a, 12b...); and

a processor (120) configured to:

determine a coherence time for the UE (12a, 12b...) based on the reference data and an at least one estimation technique for estimating channel state information (CSI);

determine at least one trustable estimation area (TEA) having a starting point at the receipt of the reference data and an ending point separated from the starting point by the coherence time for the UE (12a, 12b...); and

allocate data to be transmitted for the at least one UE (12a, 12b...) into at least one processing queue based on the at least one TEA.

2. The network device (100) of claim 1, wherein the processor (120) is configured to determine a beam characteristics of a received beam with the reference data, the beam characteristics including at least one of a beam width and an angular spread, and determine the coherence time based on the beam characteristics.

3. The network device (100) of any of the preceding claims, wherein the processor (120) is configured to allocate data to be transmitted for at least two UEs (12a, 12b...) into same processing queue when the TEAs for the UEs (12a, 12b...) overlap.

4. The network device (100) of any of the preceding claims, wherein the processor (120) is configured to share a transmission resource between UEs (12a, 12b...) allocated in the same processing queue.

5. The network device (100) of claim 1, wherein the processor (120) is configured to determine at least one non-trustable estimation area (NTEA) having a starting point at the end of the TEA and ending point either open or at the next scheduled receipt of reference data for the UE (12a, 12b...).

6. The network device (100) of any of the previous claims, wherein the processor (120) is configured to determine scheduling results for transmission in the communication network (10) or the load of at least one of the communication network (10) and the processor (120), compare the scheduling results or the load to a threshold and to perform at least one of

a. selecting the estimation technique based on whether the scheduling results or the load exceeds the threshold; and

b. increase the size of the TEA if the scheduling results or the load is below the threshold.

7. The network device (100) of any of the previous claims, wherein the processor (120) is configured to determine the speed of the UE (12a, 12b...) and perform at least one of

a. decrease the size of the TEA for the UE (12a, 12b...) if the speed of the UE (12a, 12b...) is larger than a threshold speed; and

b. increase the size of the TEA for the UE (12a, 12b...) if the speed of the UE (12a, 12b...) is smaller than a threshold speed.

8. The network device (100) of any of the previous claims, wherein the processor (120) is configured to define at least two TEAs for at least one UE (12a, 12b...), wherein the different TEAs are associated with at least one of a. different transmission frequencies;

b. different spatial coordinates for the UE (12a, 12b...); and

c. different types of data to be transmitted for the UE (12a, 12b...).

9. The network device (100) of any of the preceding claims, wherein the processor (120) is configured to prioritize the UE events in the at least one processing queue based on the coherence time for the corresponding UE (12a, 12b...).

10. The network device (100) of any of the preceding claims, wherein the processor (120) is configured to adjust the size of the TEA by defining at least one of threshold for latency for receiving the reference data (Tl) and threshold for maximum TEA (T2) so that the size of the TEA becomes size = min(72, max(coherence time, Tl)).

1 1. The network device (100) of any of the preceding claims, wherein the processor (120) is configured to determine a transmission mode for the UE (12a, 12b...) to be used in at least one of uplink transmission and downlink transmission based on the corresponding one or more TEAs.

12. A method (200) of processing events of at least one user equipment (UE) (12a, 12b...) in a wireless communication network (10), the method (200) comprising:

a. receiving, by at least one antenna (1 10a, 1 10b..), reference data of the at least one UE (12a, 12b...);

b. determining a coherence time for the UE (12a, 12b...) based on the reference data and an at least one estimation technique for estimating channel state information (CSI);

c. determining at least one trustable estimation area (TEA) having a starting point at the receipt of the reference data and an ending point separated from the starting point by the coherence time for the UE (12a, 12b...); and d. allocating data to be transmitted for the at least one UE (12a, 12b...) into at least one processing queue based on the at least one TEA.

13. The method (200) of claim 12 further comprising, prioritizing the UE events in the at least one processing queue based on the coherence time for the corresponding UE (12a, 12b...).

14. The method (200) of the previous claim further comprising, determining a transmission mode for the UE (12a, 12b...) to be used in at least one of uplink transmission and downlink transmission based on the corresponding one or more TEAs.

15. A computer program comprising program code configured to perform a method according to claims 12-14 when the computer program is executed on a computer.