WO2013087036A1 - 资源调度方法及设备 - Google Patents

资源调度方法及设备 Download PDF

Info

Publication number
WO2013087036A1
WO2013087036A1 PCT/CN2012/086742 CN2012086742W WO2013087036A1 WO 2013087036 A1 WO2013087036 A1 WO 2013087036A1 CN 2012086742 W CN2012086742 W CN 2012086742W WO 2013087036 A1 WO2013087036 A1 WO 2013087036A1
Authority
WO
WIPO (PCT)
Prior art keywords
channel quantization
channel
ues
quantization codebook
scheduled
Prior art date
Application number
PCT/CN2012/086742
Other languages
English (en)
French (fr)
Inventor
王燚
龙毅
李铮铮
蒋伟
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Publication of WO2013087036A1 publication Critical patent/WO2013087036A1/zh

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W72/00—Local resource management
    • H04W72/12—Wireless traffic scheduling
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B7/00—Radio transmission systems, i.e. using radiation field
    • H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
    • H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
    • H04B7/0413—MIMO systems
    • H04B7/0417—Feedback systems

Definitions

  • Resource scheduling method and device The present application claims priority to Chinese patent application filed on December 16, 2011, the Chinese Patent Office, Application No. 201110424343.7, entitled “Resource Scheduling Method and Related Equipment", the entire contents of which are incorporated by reference. In this application.
  • the present invention relates to the field of wireless communication technologies, and in particular, to a resource scheduling method and related devices. BACKGROUND OF THE INVENTION
  • resource scheduling e.g., time, frequency, space, power, etc.
  • the first and second generation mobile communication systems are generally narrowband systems, and multi-user gains are obtained by scheduling different users to time slots of respective corresponding channel gains.
  • the wireless communication system is generally a slow fading system, the gain obtained by time domain scheduling is relatively limited.
  • the third-generation mobile communication system is a broadband system with strong frequency selective fading.
  • the base station of the third-generation mobile communication system can be scheduled in the time-frequency two-dimensional domain, and the multi-user gain is relatively obvious.
  • the MIMO (Multiple Input Multiple Output) system utilizes multiple physical antennas on the transmitter and receiver to increase the multipath fading channel capacity with the number of transceiver antennas without increasing the spectrum and power resources. Or the minimum number of antennas received) Linear growth, a revolutionary technology to improve spectral efficiency.
  • the channel matrix can be considered as a full rank matrix, by precoding in the transmitter (such as V-BLAST, D-BLAST, etc.), or in the receiver for signal detection (eg ML, MMSE, ZF, etc., can eliminate crosstalk between antennas, form multiple parallel independent subchannels, and the total channel capacity is the sum of channel capacities of each subchannel, breaking the limitation of channel capacity only with power logarithm increase, increasing space Degree of freedom.
  • the MIMO system is only for a single user.
  • the physical distance between the array antenna elements is subject to many restrictions, the channel fading irrelevant characteristics may not always be satisfied, so the channel capacity may be affected; and, the user equipment (UE) , User Equipment ) volume, cost, Power consumption is limited, and it is not suitable to configure too many antennas. It is difficult to obtain higher capacity and performance gain on single-user MIMO. Therefore, multi-user multiple input multiple output (MU-MIMO, Multi-User)
  • MU-MIMO multi-user multiple input multiple output
  • MIMO systems are an option to further improve system performance.
  • data of multiple UEs can be spatially multiplexed on the same time-frequency resource, but the number of UEs that can be supported is much smaller than the number of UEs that need to be transmitted. Since the channel fading of different UEs is different, the capacity and performance obtained by transmitting on the same time-frequency resource are also completely different. Therefore, it is necessary to select a suitable UE for pairing, and the paired UE jointly performs data transmission on the same time-frequency resource. To achieve multi-user gain, improve the spectrum utilization efficiency of the system.
  • the base station uses the Greedy algorithm for resource pairing to perform multi- UE pairing.
  • the base station selects one UE according to different scheduling priorities of the UE, and then performs traversal search in the remaining UEs, finds UEs that can be paired with the first selected UE to obtain the highest system capacity, and then continues to be in the remaining UEs. Traverse the search until selected
  • the number of UEs reaches the maximum number of UE pairs. It is found that the Greedy algorithm requires the base station to obtain the channel state information of all UEs. Each UE needs to feed back the channel state information of the entire transmission bandwidth, instead of merely feeding back the channel gain of the corresponding bandwidth of the UE; and the channel fading changes rapidly and randomly, requiring a large amount of Resource overhead. At the same time, each time the base station selects one UE, all other to be scheduled
  • the UE performs search calculation according to the pairing criterion.
  • the complexity and delay of the calculation are too large, and there are certain difficulties in the deployment in the actual system.
  • the embodiments of the present invention provide a resource scheduling method, a related device, and a communication system, so as to reduce the computational complexity of UE pairing and shorten the delay of UE pairing.
  • an embodiment of the present invention provides a resource scheduling method, including:
  • the first channel quantization codebook set of the first channel quantization codebook set includes N channel quantization codebook sets, and each of the first channel quantization codebook sets is selected respectively.
  • the channel quantization codebook corresponds to one of the UEs, and obtains N UEs, and the first UE and the obtained N UEs are paired, wherein the first channel quantization codebook set and the first channel quantization And a sum of correlations between the channel quantization codebooks in the second channel quantization codebook set formed by the codebook, less than or equal to any of the M channel quantization codebooks, including the first channel quantization codebook The sum of the correlations between the two pairs of N+1 channels, wherein N is greater than 1 and less than M;
  • the method further includes:
  • the UEs to be scheduled are respectively divided into candidate users corresponding to the channel quantization codebooks according to the channel quantization codebooks respectively corresponding to the to-be-scheduled UEs.
  • the M channel quantization codebooks respectively create a candidate user queue.
  • the pairing the one of the UEs corresponding to the second channel quantization codebook with the first UE includes: selecting one UE in a candidate user queue corresponding to the second channel quantization codebook The first UE performs pairing;
  • One of the UEs obtains N UEs, and includes: selecting one UE from each of the channel quantization codebook corresponding candidate user queues in the first channel quantization codebook set to obtain N UEs.
  • the allocating transmission resources for the first UE and the UE successfully paired with the first UE includes:
  • a precoding weight value after the pairing of each UE Determining, by using a channel quantization codebook of each UE in the paired UE group formed by the first UE and the UE successfully paired with the first UE, a precoding weight value after the pairing of each UE;
  • the pre-coding weight difference value corresponding to each UE is used to obtain a channel quality indicator CQI adjustment amount corresponding to each UE, and the corresponding UE is adjusted according to the CQI adjustment amount corresponding to each UE.
  • the precoding weight difference corresponding to each UE is a difference between a precoding weight of each UE pairing and a precoding weight before pairing; according to each UE
  • the CQI determines a modulation coding mode corresponding to each UE; and allocates a transmission resource to the paired UE group according to the modulation coding mode of each UE;
  • the first UE is the UE with the highest scheduling priority or any UE among the to-be-scheduled UEs; and/or
  • the pairing the UE corresponding to the second channel quantization codebook with the first UE to obtain a paired UE group including: the UE with the highest scheduling priority among the UEs corresponding to the second channel quantization codebook The first UE is paired;
  • N UEs respectively selecting one of the UEs corresponding to each channel quantization codebook in the first channel quantization codebook set to obtain N UEs, including: respectively selecting each channel in the first channel quantization codebook set The UE with the highest priority is scheduled in the UE corresponding to the quantized codebook, and N UEs are obtained.
  • an embodiment of the present invention provides a resource scheduling apparatus, including:
  • An acquiring module configured to obtain a channel quantization codebook corresponding to each UE to be scheduled UE, where the channel quantization codebook corresponding to each UE in each to-be-scheduled UE passes the pre-built M channel quantization codebook pair The channel of each UE is quantized;
  • a pairing module configured to determine, in the M channel quantization codebook, a second channel quantization codebook with a minimum correlation with a first channel quantization codebook corresponding to a first UE to be paired in each to-be-scheduled UE And pairing one of the UEs corresponding to the second channel quantization codebook with the first UE; or determining, in the M channel quantization codebook, a location corresponding to the first UE
  • the first channel quantization codebook satisfies a first channel quantization codebook set that combines a minimum correlation criterion, where the first channel quantization codebook set includes N channel quantization codebooks, respectively
  • Each of the channel quantization codebooks in the first channel quantization codebook set corresponds to one of the UEs, and the N UEs are obtained, and the first UE and the obtained N UEs are paired, where the first channel quantization And a sum of correlations between the codebook set and the channel quantization codebook in the second channel quantization codebook set formed by the first
  • a resource allocation module configured to allocate a transmission resource to the first UE and a UE that is successfully paired with the first UE.
  • the acquiring module is configured to: perform quantization on a channel of each to-be-scheduled UE according to the pre-configured M channel quantization codebooks to obtain a channel quantization codebook corresponding to each of the to-be-scheduled UEs; And the channel quantization codebook corresponding to each of the UEs that are sent by each UE to be scheduled, to obtain a channel quantization codebook corresponding to each of the to-be-scheduled UEs.
  • the resource scheduling device further includes:
  • a queue module configured to: after obtaining, by the acquiring module, a channel quantization codebook corresponding to each to-be-scheduled UE, according to the channel quantization codebook corresponding to each to-be-scheduled UE, respectively, each to-be-scheduled UE is separately allocated to the channel
  • the resource allocation module includes:
  • a precoding weight calculation subunit configured to determine each of the channel quantization codebooks of each UE in the paired UE group formed by the first UE and the UE successfully paired with the first UE, respectively Precoding weights after UE pairing;
  • a first CQI adjustment unit configured to obtain, according to the precoding weight difference value corresponding to each UE, a CQI adjustment amount corresponding to each UE, and adjust each of the CQI adjustment amounts corresponding to each UE a CQI corresponding to the UE, where a precoding weight difference corresponding to each UE is a difference between a precoding weight value of each UE pairing and a precoding weight value before pairing;
  • a modulation and coding determining unit configured to determine, according to the CQI corresponding to each UE, a modulation and coding mode corresponding to each UE;
  • the resource allocation module includes:
  • a second CQI adjusting unit configured to obtain a correlation between a channel quantization codebook corresponding to each UE in the paired UE group formed by the first UE and the UE successfully paired with the first UE, respectively Adjusting, according to the CQI adjustment amount corresponding to each UE, the CQI corresponding to each UE according to the CQI adjustment amount corresponding to each UE;
  • a modulation and coding determining unit configured to determine, according to the CQI corresponding to each UE, a modulation and coding mode corresponding to each UE;
  • a resource allocation unit configured to allocate a transmission resource to the paired UE group according to the corresponding modulation and coding manner of each UE.
  • the access device is a base station or an access point.
  • the embodiment of the present invention further provides a communication system, including:
  • the channel quantized codebooks of the UEs to be scheduled are obtained by using the pre-configured M channel quantization codebooks to obtain the channel quantized codebooks of the UEs to be scheduled.
  • the paired UE and the channel quantization codebook corresponding to the UE corresponding to the channel quantization codebook are matched with one of the UEs, or the channel codebook corresponding to the to-be-paired UE is used.
  • the N channel quantization codebooks in the channel quantization codebook set that meet the combined minimum correlation criterion respectively correspond to one UE in the UE, and a total of N UEs and the to-be-paired UE perform pairing.
  • FIG. 1 is a schematic diagram of a downlink model of a MU-MIMO system according to an embodiment of the present invention
  • FIG. 2 is a schematic flowchart of a resource scheduling method according to an embodiment of the present invention
  • FIG. 3 is a schematic flowchart of another resource scheduling method according to an embodiment of the present invention
  • FIG. 4 is a schematic flowchart of another resource scheduling method according to an embodiment of the present invention
  • FIG. Schematic diagram of an access device
  • FIG. 5-b is a schematic diagram of another access device according to an embodiment of the present invention
  • FIG. 5-c is a schematic diagram of a resource allocation module of an access device according to an embodiment of the present invention
  • FIG. 5-d is a schematic diagram of the present invention.
  • the computational complexity of UE pairing shortens the delay of UE pairing.
  • the downlink model of the MU-MIMO system can be as shown in Figure 1.
  • the base station transmitter includes M antennas, and the base station selects K UEs for multi-UE communication, wherein the UE k includes N k antennas, and the total number of antennas of the K UEs can be expressed as:
  • the received data corresponding to UE k can be represented as a vector:
  • the subscript L k represents the number of layers of spatial multiplexing of the UE k, and the value of L k is less than or equal to the minimum value of the number of receiving antennas of the base station and the number of receiving antennas of the UE k, that is,
  • the base station needs to pre-code the transmit data, and the precoding matrix corresponding to UE k is P k , and the dimension of the matrix is ⁇ ⁇ , that is, the number of rows of the precoding matrix is the number of base station transmit antennas M.
  • the number of columns is the number of spatial multiplexing layers L k of UE k .
  • the corresponding hair of all paired UEs The radio data is pre-coded and added to form a transmit signal on each transmit antenna of the base station, which can be expressed as M
  • the channel response matrix H k of the transmitting antenna of the base station and the UE k can be expressed as:
  • the element J in the channel response matrix 3 indicates the channel response coefficient between the jth transmit antenna of the base station and the ith receive antenna of the UE k.
  • the channel response can be represented by a single tap coefficient; if the signal transmitted between the antennas is a wideband signal, Orthogonal Frequency Division Multiplexing (OFDM) technology can be utilized.
  • OFDM Orthogonal Frequency Division Multiplexing
  • the channel response coefficient corresponding to each subcarrier of the OFDM symbol can still be represented by a single tap coefficient, and the above analysis can be applied to each subcarrier, and details are not described herein again.
  • the multi-user received signal model can be expressed as equation (1): Formula ( 1 ) , y k 2 H k s + n k
  • Gaussian white noise signal sample values indicates the base station transmitted signal of each antenna, to meet the characteristics of the independent and identically distributed £
  • the first term HkPk X k represents the expected signal of UE k
  • the second term represents interference signals from other UEs.
  • Dirty-Paper Coding is a non-linear coding method that can obtain the upper limit of multi-UE channel capacity.
  • Linear coding methods such as Minimum Mean-squared Error (MMSE), Zero-Forcing (Zero-Forcing), and Block Diagonalized (BD), can also be obtained with appropriate scheduling and pairing algorithms. Near optimal channel capacity.
  • MMSE Minimum Mean-squared Error
  • Zero-Forcing Zero-Forcing
  • BD Block Diagonalized
  • the base station can theoretically make the interference of the paired UE zero by selecting an appropriate precoding matrix.
  • the total channel model including the M transmit antenna and the N receive antennas can be expressed as equation (3):
  • Equation (3) Take the UE received by a single antenna as an example, and equation (2) can be evolved into equation (5):
  • data of multiple UEs can be spatially multiplexed on the same time-frequency resource, but the number of UEs that can be supported is often smaller than the number of UEs that need to perform data transmission. Since different UEs are subject to different channel fading, the capacity and performance obtained by transmitting on the same time-frequency resource are also quite different. Therefore, it is necessary to select a suitable UE for UE pairing, and jointly perform data transmission on the time-frequency resource to obtain multi-UE gain, thereby improving the spectrum utilization efficiency of the system.
  • the signal separation can be performed well, and the relationship can be as shown in the following formula (6): 0
  • U is a predetermined correlation threshold
  • ⁇ may be set to, for example, 0.05, 0.1, 0.15 h n H hm
  • An embodiment of the resource scheduling method of the present invention may include: acquiring a channel codebook corresponding to each UE to be scheduled, where a channel quantization codebook corresponding to each UE in each to-be-scheduled UE passes in advance
  • the M channel quantization codes are configured to quantize the channel of each UE; and the first channel quantization code corresponding to the first UE to be paired in the to-be-scheduled UEs is determined in the M channel quantization codebooks.
  • the least correlated second channel quantization codebook, And the one of the UEs corresponding to the second channel quantization codebook is paired with the first UE; or, determining that the first channel quantization codebook corresponding to the first UE meets the combined minimum correlation in the M channel quantization codebooks a first channel quantization codebook set of the first criterion, the first channel quantization codebook set includes N channel quantization codebooks (ie, the first channel quantization codebook set includes N of the M channel quantization codebooks, but the first The channel quantization codebook set does not include the first channel quantization codebook); one of the UEs corresponding to each channel quantization codebook in the first channel quantization codebook set is respectively selected, and N UEs are obtained, and the first UE and the obtained UE are obtained.
  • the N UEs are paired; wherein, the sum of the correlation between the channel quantization codebooks in the second channel quantization codebook set formed by the first channel quantization codebook set and the first channel quantization codebook is smaller than Or equal to the sum of the correlations between the N+1 channel quantized codebooks of the first channel quantization codebook and the M channel quantization codebook, wherein N is greater than 1 and less than M; UE assigned to successfully pair with the first UE Resources.
  • a resource scheduling method provided by an embodiment of the present invention may include:
  • the access device acquires a channel quantization codebook corresponding to each UE to be scheduled.
  • the channel quantization codebook corresponding to each UE in each to-be-scheduled UE is obtained by quantizing the channel of each UE by using the pre-configured M channel quantization codebooks;
  • the channel quantization codebook corresponding to each UE to be scheduled can be obtained in multiple manners.
  • the channel of each to-be-scheduled UE may be quantized according to the pre-configured M channel quantization codebooks to obtain a channel quantization codebook corresponding to each of the to-be-scheduled UEs (ie, the access device performs channel quantization according to the channel measurement information of the UE).
  • the channel quantization codebook corresponding to the UE reported by each UE in each of the to-be-scheduled UEs may be received (ie, the UE performs channel quantization according to the channel measurement information, and the access device directly acquires the channel quantization code corresponding to the UE from the UE.
  • the channel quantization codebook corresponding to each UE to be scheduled is obtained according to the received channel quantization codebook corresponding to each UE reported by each UE to be scheduled, or may be a third party device
  • the channels of the UEs to be scheduled are quantized according to the pre-configured M channel quantization codebooks, to obtain the channel quantization codebooks respectively corresponding to the to-be-scheduled UEs, and the corresponding to-be-scheduled UEs reported by the third-party devices to the access device are corresponding to the UEs to be scheduled.
  • the channel quantizes the codebook, and the access device obtains the channel quantization codebook corresponding to each of the to-be-scheduled UEs accordingly.
  • the access device may also create a candidate user queue corresponding to the pre-built M channel quantization codebooks (ie, create M candidates). After the channel quantization codebook corresponding to each UE to be scheduled is obtained, the access device may respectively allocate the to-be-scheduled UEs to the corresponding channel according to the channel quantization codebook corresponding to each UE to be scheduled.
  • the candidate user queue corresponding to the quantized codebook in each candidate user queue, for example, each UE may be arranged in a scheduling priority order or randomly arranged or arranged in other manners. In this case, the channel quantization codebook corresponding to each UE in each candidate user queue is the same, and the channel quantization codebook corresponding to each UE in each candidate user queue is the channel quantization codebook corresponding to the candidate user queue.
  • the access device determines, in the M channel quantization codebook, a first UE to be paired with each to-be-scheduled UE, where the first UE may be the UE with the highest scheduling priority among the current UEs to be scheduled, and scheduling a second channel quantized codebook with the least correlation of the first channel quantization codebook corresponding to the second highest priority UE or other UE or any one of the UEs, and one of the UEs corresponding to the second channel quantization codebook (the one)
  • the UE is paired with the first UE, for example, the UE with the highest scheduling priority in the UE corresponding to the second channel quantization codebook, the UE with the second highest scheduling priority, or any other UE or any UE.
  • the access device may use one of the candidate user queues corresponding to the second channel quantization codebook.
  • the UE is paired with the first UE, for example, the UE with the highest scheduling priority among the candidate user queues corresponding to the second channel quantization codebook, the UE with the second highest scheduling priority or any other UE)
  • the method is mainly for a scenario in which the MU-MIMO system supports only two UEs to be paired; or, the access device determines, in the M channel quantization codebook, a first UE corresponding to the first UE to be paired in each to-be-scheduled UE.
  • a channel quantization codebook satisfies a first channel quantization codebook set that combines minimum channel correlation codes, and the first channel quantization codebook set includes N channel quantization codebooks (ie, the first channel quantization codebook set includes M channel quantization codes N of the present, but the first channel quantization codebook set does not include the first channel quantization codebook, and N+1 is the maximum number of paired UEs supported by the access device); respectively, each of the first channel quantization codebook sets is selected One of the UEs corresponding to the channel quantization codebook (for example, the UE with the highest scheduling priority among the UEs corresponding to each channel quantization codebook in the first channel quantization codebook set can be separately selected, and the scheduling priority is the second highest. UE or any other UE or a UE), to obtain N UE
  • the access device may respectively select the corresponding one of the channel quantization codebooks in the first channel quantization codebook set.
  • One UE in the candidate user queue (wherein the UE is, for example, a corresponding candidate user) Aligning the UE with the highest priority, the UE with the second highest priority, or the other UE or the UE to obtain the N UEs, and pairing the first UE with the obtained N UEs; a channel quantization codebook set composed of a channel quantization codebook set and a first channel quantization codebook (wherein the second channel quantization codebook set includes N+1 channel quantization codebooks)
  • the sum of the correlations is less than or equal to the sum of the correlations between any of the M channel quantization codebooks and the N+1 channel quantization codebooks of the first channel quantization codebook, where N is greater than 1 And less than M (wherein this mode is mainly for the scenario where the MU-MIMO
  • the access device allocates a transmission resource to the first UE and the UE that is successfully paired with the first UE.
  • the access device may determine, by using the channel quantization codebook of each UE in the paired UE group that is formed by the first UE and the UE that is successfully paired with the first UE, And a channel quality indicator (CQI, Channel Quality Indicator) adjustment amount corresponding to each UE, and adjusting the CQI adjustment amount corresponding to each UE according to the precoding weight difference corresponding to each UE a CQI corresponding to each UE, where a precoding weight difference corresponding to each UE is a difference between a precoding weight of each UE pairing and a precoding weight before pairing; according to each UE Corresponding CQI determines a modulation coding mode corresponding to each UE; and allocates transmission resources to the paired UE group according to the corresponding modulation and coding mode of each UE;
  • CQI Channel Quality Indicator
  • the access device may obtain a correlation between the channel quantization codebooks of each UE in the paired UE group formed by the first UE and the UE successfully paired with the first UE, respectively.
  • the CQI adjustment amount corresponding to the UE, and the CQI corresponding to each UE is adjusted based on the CQI adjustment amount corresponding to each UE; determining a modulation and coding mode corresponding to the UE according to the CQI corresponding to each UE;
  • the modulation coding mode allocates transmission resources for the paired UE group.
  • the access device in the embodiment of the present invention may be, for example, a base station, an access point, or another entity in the network that has a UE access function.
  • the channel quantized codebooks of the UEs to be scheduled are obtained by using the pre-configured M channel quantization codebooks to obtain the channel quantized codebooks of the UEs to be scheduled. Then, the paired UE and the channel quantization codebook corresponding to the UE corresponding channel quantization codebook are matched to one of the UEs, or the channel quantization codebook corresponding to the to-be-paired UE is satisfied. N in the channel quantization codebook set combining the minimum correlation criteria The channel quantization codebooks respectively correspond to one UE in the UE, and a total of N UEs and the to-be-paired UE are paired.
  • the UE is scheduled to be scheduled according to the correlation between the channel quantization codebooks of the UEs to be scheduled, and the Greedy algorithm is not required to perform search calculation on all the UEs to be scheduled according to the pairing criterion, thereby facilitating greatly reducing the computational complexity of UE pairing. Degrees and delays simplify the UE pairing mechanism and enhance enforceability.
  • a specific application scenario in which the MU-MIMO system supports only two UE pairings is introduced as an example.
  • another resource scheduling method provided by the embodiment of the present invention may include:
  • the access device creates a candidate user queue corresponding to the pre-built M channel quantization codebooks respectively.
  • the access device may create a candidate user queue corresponding to the pre-built M channel quantization codebooks, that is, create M candidate user queues, and each channel quantization codebook corresponds to a different one. Candidate user queue.
  • the access device may also allocate different queue indexes for each candidate user queue that is created. For example, the channel quantization codebook corresponding to each candidate user queue may be directly used as the queue index of the candidate user queue, or may be created for each The candidate user queues are assigned different other queue indexes.
  • the pre-built M channel quantized codebooks may be uniformly distributed as possible, and the correlation change between the M channel quantized codebooks is also as close as possible, and the M channel quantized codebooks are traversed as much as possible. All channels that the UE UE may traverse, so that the M channel quantized codebooks can quantize the channel of the UE as accurately as possible, and reduce the quantization error as much as possible.
  • Constellation map according to the CSI of the UE to be scheduled, constellation mapping according to the minimum Euclidean distance or the maximum correlation criterion, Nt is the transmitting antenna book of the base station, and R is the total number of channel quantization codebooks.
  • the constellation mapping mechanism can be expressed as equation (7): Formula (7) Wherein, the precoding weight corresponding to the UE i to be scheduled, 3 ⁇ 4 is the channel state corresponding to the UE i to be scheduled n)
  • the eigenbeam weight vector of the UE to be scheduled may be constelled according to a minimum Euclidean distance or a maximum correlation criterion to obtain a channel corresponding to the UE to be scheduled.
  • the eigenvector quantization codebook that is, the eigenbeam weight vector in equation (5) is replaced, and the F k is replaced by the eigenvector quantization codebook.
  • the access device obtains a channel quantization codebook corresponding to each to-be-scheduled UE.
  • the access device may obtain a channel quantization codebook corresponding to each to-be-scheduled UE by using multiple methods.
  • the M channel quantization codebooks may be preset in the UE, or the access device may deliver the pre-constructed M channel quantization codebooks to the UE; the UE quantizes the codebooks through the M channels. Performing quantization to obtain a channel quantization codebook corresponding to the UE; each to-be-scheduled UE may report the obtained channel quantization codebook (and/or codebook indication information capable of indicating the channel quantization codebook) to the access The access device may receive the channel quantization codebook corresponding to the UE reported by each UE in each of the to-be-scheduled UEs (or the indication information corresponding to the codebook), according to each of the received UEs to be scheduled.
  • the channel quantization codebook corresponding to each UE that is reported by the UE obtains a channel quantization codebook corresponding to each of the to-be-scheduled UEs, that is, the UE quantizes the channel according to the channel measurement result, and the access device directly acquires the UE from the UE.
  • the UE to be scheduled may report the CSI to the access device; the access device may quantize the channel of each UE to be scheduled according to the pre-configured M channel quantization codebooks to obtain a channel quantization codebook corresponding to each UE to be scheduled ( That is, the access device quantizes the channel of the UE according to the CSI reported by the UE.
  • the channel of each of the to-be-scheduled UEs is quantized by the third-party device according to the pre-configured M channel quantization codebooks, to obtain the channel quantization codebook corresponding to each to-be-scheduled UE, and the third-party device reports the obtained to the access device.
  • a channel quantization codebook corresponding to each to-be-scheduled UE (and/or a codebook indication information capable of indicating the channel quantization codebook)
  • the access device obtains a channel quantization codebook corresponding to each to-be-scheduled UE accordingly (ie, The access device obtains, from the third-party device, a channel quantization codebook corresponding to each UE to be scheduled.
  • the access device divides each to-be-scheduled UE into a candidate user queue.
  • the access device adjusts the codebook according to the channel corresponding to each UE to be scheduled.
  • the UEs are respectively divided into candidate user queues corresponding to their corresponding channel quantization codebooks.
  • the UEs in each candidate user queue are arranged, for example, in a scheduling priority order or arbitrarily arranged or otherwise arranged.
  • the channel quantization codebook corresponding to all UEs in each candidate user queue is the same, and the channel quantization codebook corresponding to each UE in each candidate user queue is the channel quantization codebook corresponding to the candidate user queue.
  • the channel to be scheduled is quantized by the channel quantization codebook, and the UEs to be scheduled are grouped according to the channel quantization codebook.
  • the access device selects one UE from the candidate user queue as the to-be-matched UE.
  • the access device may first select, from the candidate user queue, the UE with the highest scheduling priority (or the UE with the second highest scheduling priority or other UE or any UE) as the UE to be paired.
  • the selected UE to be paired is referred to as UE-1
  • the channel quantization codebook corresponding to UE-1 is referred to as a first channel quantization codebook
  • the candidate user queue to which UE-1 belongs is referred to as a first candidate.
  • the access device may determine, according to the correlation between the quantized codebooks of the respective channels, another channel quantization codebook with the least correlation between the first channel quantization codebook corresponding to the UE-1 in the M channel quantization codebooks.
  • another channel quantization codebook with the smallest correlation of the first channel quantization codebook corresponding to UE-1 in the determined M channel quantization codebooks may be referred to as a second channel quantization codebook.
  • the candidate user queue corresponding to the second channel quantization codebook is referred to as a second candidate user queue.
  • determining another channel quantization codebook ie, the second channel quantization codebook represented by the constellation point with the largest Euclidean distance or the smallest correlation with respect to the first channel quantization codebook
  • the UE with the highest scheduling priority among the second candidate user queue corresponding to the second channel quantization codebook is paired with UE-1.
  • the constellation map of the channel quantization codebook mapping the correlation between the two channel quantization codebooks represented by the two constellation points of the maximum Euclidean distance or the minimum correlation is the smallest, and the two channel quantization codebooks correspond to
  • the interference between UEs in the candidate user queue is also relatively minimal, so UE direct pairing that satisfies this condition has the smallest inter-UE interference loss t
  • the pairing formula can be as in formula (8):
  • a pairing queue mapping table may be established in advance, and the preferred pairing relationship between each candidate user queue is recorded in the pairing queue mapping table, and according to the preferred pairing relationship recorded in the pairing queue mapping table when pairing Pair it.
  • the foregoing UE pairing mechanism is adopted, because the UEs to be scheduled are grouped according to the channel correlation size, and the UEs in the candidate user queue with the smallest channel correlation are paired, which can effectively solve the current greedy algorithm based on the DS domain.
  • the problem of large fluctuation of scheduling delay is to improve the probability of UE pairing success, simplify the scheduling calculation and improve the throughput of the MU-MIMO system.
  • the access device calculates a precoding weight value of each UE paired according to a channel quantization codebook of each UE in the paired UE group that is formed by the UE-1 and the UE that is successfully paired with the UE-1, so as to follow Eliminate interference between multiple UEs;
  • the access device calculates, according to each UE in the paired UE group, a difference between the paired precoding weights and the precoding weights before each UE pairing, to obtain a CQI adjustment amount of each UE, and Adjusting a CQI of each UE based on a CQI adjustment amount of each UE;
  • each UE to be scheduled corresponds to one precoding weight and CQI (which may be reported by the UE).
  • CQI which may be reported by the UE.
  • interference between the paired UEs is introduced, so as to eliminate interference between multiple UEs.
  • the CQI before pairing each paired UE to be scheduled is adjusted, and the adjusted CQI can better adapt to the paired transmission scenario.
  • the access device determines, according to the CQI corresponding to each UE in the paired UE group, a modulation and coding mode (MCS, Modulation and Coding Set) corresponding to each UE.
  • MCS modulation and coding mode
  • the access device performs pairing according to a modulation and coding manner corresponding to each UE in the paired UE group.
  • the UE group allocates transmission resources.
  • the access device may also obtain the correlation between the channel quantization codebooks of each UE in the paired UE group formed by the UE-1 and the UEs successfully paired with the UE-1, respectively.
  • the CQI adjustment amount of each UE (the correlation between the channel quantization codebooks is determined, the CQI adjustment amount of the corresponding codebooks can be determined accordingly), and the each is adjusted based on the CQI adjustment amount of each UE.
  • the CQI of the UE determines the modulation and coding mode corresponding to each UE according to the CQI corresponding to each UE in the paired UE group; and allocates transmission resources to the paired UE group according to the modulation and coding mode corresponding to each UE.
  • the paired UE group includes UE k and UE i, and the CQI of UE k can be expressed by formula (9):
  • the channel quantized codebooks of the UEs to be scheduled are obtained by using the pre-configured M channel quantization codebooks to obtain the channel quantized codebooks of the UEs to be scheduled. Then, the paired UE and the channel quantization codebook corresponding to the corresponding channel quantization codebook of the UE are paired with one of the UEs. The UE is scheduled to be scheduled according to the correlation between the channel quantization codebooks corresponding to the UEs to be scheduled, and the Greedy algorithm is not required to perform search calculation on all the UEs to be scheduled according to the pairing criterion, thereby facilitating greatly reducing the pairing of the two UEs.
  • N is greater than 1 and less than M
  • another resource scheduling method provided by the embodiment of the present invention may include: 401.
  • the access device creates a candidate user queue corresponding to the pre-built M channel quantization codebooks respectively.
  • the access device may create a candidate user queue corresponding to the pre-built M channel quantization codebooks, that is, create M candidate user queues, and each channel quantization codebook corresponds to a different one. Candidate user queue.
  • the access device may also allocate a different queue index for each candidate user queue created. For example, the channel quantization codebook corresponding to each candidate user queue may be directly used as the queue index of the candidate user queue.
  • the pre-built M channel quantized codebooks may be uniformly distributed as possible, and the correlation change between the M channel quantized codebooks is also as close as possible, and the M channel quantized codebooks are traversed as much as possible. All channels that the UE UE may traverse, so that the M channel quantized codebooks can quantize the channel of the UE as accurately as possible, and reduce the quantization error as much as possible.
  • the access device obtains a channel quantization codebook corresponding to each to-be-scheduled UE.
  • the access device may obtain a channel quantization codebook corresponding to each to-be-scheduled UE by using multiple methods.
  • the M channel quantization codebooks may be preset in the UE, or the access device may deliver the pre-constructed M channel quantization codebooks to the UE; the UE quantizes the codebooks through the M channels. Performing quantization to obtain a channel quantization codebook corresponding to the UE; each to-be-scheduled UE may report the obtained channel quantization codebook (and/or codebook indication information capable of indicating the channel quantization codebook) to the access The access device may receive the channel quantization codebook corresponding to the UE reported by each UE in each of the to-be-scheduled UEs (or the indication information corresponding to the codebook), according to each of the received UEs to be scheduled.
  • the channel quantization codebook corresponding to each UE that is reported by the UE obtains a channel quantization codebook corresponding to each of the to-be-scheduled UEs, that is, the UE quantizes the channel according to the channel measurement result, and the access device directly acquires the UE from the UE.
  • the corresponding channel quantized codebook may report channel state information (CSI) to the access device; the access device may quantize the channel of each UE to be scheduled according to the pre-configured M channel quantization codebooks to obtain corresponding to each UE to be scheduled.
  • the channel quantized codebook ie, the access device quantizes the channel of the UE according to the CSI reported by the UE).
  • the channel of each of the to-be-scheduled UEs is quantized by the third-party device according to the pre-configured M channel quantization codebooks, to obtain the channel quantization codebook corresponding to each to-be-scheduled UE, and the third-party device reports the obtained to the access device.
  • Channel quantization codebook corresponding to each UE to be scheduled (and/or capable of indicating the channel amount
  • the access device obtains the channel quantization codebook corresponding to each to-be-scheduled UE according to the access device (that is, the access device obtains the channel quantization codebook corresponding to each to-be-scheduled UE from the third-party device) ).
  • the access device divides each to-be-scheduled UE into a candidate user queue.
  • the access device divides each to-be-tuned UE into a candidate user queue corresponding to the corresponding channel quantization codebook according to the channel quantization codebook corresponding to each to-be-scheduled UE.
  • the UEs in each candidate user queue are arranged, for example, in a scheduling priority order or arbitrarily arranged or otherwise arranged.
  • the channel quantization codebook corresponding to all UEs in each candidate user queue is the same, and the channel quantization codebook corresponding to each UE in each candidate user queue is the channel quantization codebook corresponding to the candidate user queue.
  • the channel to be scheduled is quantized by the channel quantization codebook, and the UEs to be scheduled are grouped according to the channel quantization codebook.
  • the access device selects one UE from the candidate user queue as the to-be-paired UE.
  • the selected UE to be paired may be referred to as UE-1
  • the channel quantization code corresponding to UE-1 is referred to as
  • the candidate user queue to which the UE-1 belongs is referred to as the first candidate user queue);
  • the first channel quantization codebook set Determining, in the M channel quantization codebooks, a first channel quantization codebook set that satisfies a combined minimum correlation criterion with respect to a first channel quantization codebook corresponding to the UE-1, where the first channel quantization codebook set includes N
  • the channel quantization codebook ie, the first channel quantization codebook set includes N of the M channel quantization codebooks, but the first channel quantization codebook set does not include the first channel quantization codebook, and N+1 is the access device.
  • the channel quantization in the second channel quantization codebook set composed of the first channel quantization codebook set and the first channel quantization codebook (where the second channel quantization codebook set includes N+1 channel quantization codebooks) The sum of the correlations between the codebooks is less than or equal to the sum of the correlations between any two of the M channel quantization codebooks including the N+1 channel quantization codebooks of the first channel quantization code.
  • the 455 types do not include a possible combination of the first channel quantization codebook (the first channel quantization codebook set is one of the combinations) plus the first channel quantization codebook is also 455 possible combinations (the second channel quantization codebook set) For one of the combinations;); a combination of the sum of the correlations between the channel quantization codebooks and the least of the correlations is selected as the second channel quantization codebook set.
  • the foregoing UE pairing mechanism is adopted, because the UEs to be scheduled are grouped according to the channel correlation size, and the UEs in the candidate user queue with the smallest channel correlation are paired, which can effectively solve the current greedy algorithm based on the DS domain.
  • the problem of large fluctuation of scheduling delay is to improve the probability of UE pairing success, simplify the scheduling calculation and improve the throughput of the MU-MIMO system.
  • the access device calculates, according to the channel quantization codebook of each UE in the paired UE group that is formed by the UE-1 and the UE that is successfully paired with the UE-1, the precoding weights of each UE after pairing, so as to be followed. Eliminate interference between multiple UEs;
  • the access device calculates, according to each UE in the paired UE group (total N+1 UEs), a difference between the paired precoding weights and the precoding weights before each UE pairing, to obtain the per The CQI adjustment amount of the UE, and adjusting the CQI of each UE based on the CQI adjustment amount of each UE;
  • each UE to be scheduled corresponds to one precoding weight and CQI (which may be reported by the UE).
  • CQI which may be reported by the UE.
  • interference between the paired UEs is introduced, so as to eliminate interference between multiple UEs.
  • the CQI before pairing each paired UE to be scheduled is adjusted, and the adjusted CQI can better adapt to the paired transmission scenario.
  • the access device determines, according to a CQI corresponding to each UE in the paired UE group, a modulation and coding mode (MCS) corresponding to each UE.
  • MCS modulation and coding mode
  • the access device allocates a transmission resource to the paired UE group according to a modulation and coding manner corresponding to each UE in the paired UE group.
  • the access device may also obtain the correlation between the channel quantization codebooks of each UE in the paired UE group formed by the UE-1 and the UEs successfully paired with the UE-1, respectively.
  • the CQI adjustment amount of each UE (the correlation between the channel quantization codebooks is determined, the CQI adjustment amount of the corresponding codebooks may be determined accordingly), and the per-CQI adjustment amount corresponding to each UE is adjusted.
  • a CQI corresponding to each UE determining a modulation and coding mode corresponding to each UE according to a CQI corresponding to each UE in the paired UE group; and allocating transmissions to the paired UE group according to a modulation and coding mode corresponding to each UE in the paired UE group Resources.
  • the channel quantized codebooks of the UEs to be scheduled are obtained by using the pre-configured M channel quantization codebooks to obtain the channel quantized codebooks of the UEs to be scheduled.
  • the N channel quantization codebooks in the channel quantization codebook set that meet the combined minimum correlation criterion respectively correspond to one UE in the UE, and a total of N UEs and the channel quantization codebook corresponding to the to-be-paired UE.
  • the paired UEs are paired.
  • the Greedy algorithm is not required to perform search calculation on all the UEs to be scheduled according to the pairing criterion, thereby facilitating the calculation of greatly reducing multi-UE pairing. Complexity and latency, simplifying multi-UE pairing mechanisms and enhancing enforceability.
  • an access device 500 which may include:
  • the acquisition module 501 the pairing module 502, and the resource allocation module 503.
  • the obtaining module 501 is configured to obtain a channel quantization codebook corresponding to each UE to be scheduled, where the channel quantization codebook corresponding to each UE in each to-be-scheduled UE passes the pre-configured M channel quantization codebooks.
  • the channels of the UEs are quantized;
  • the pairing module 502 is configured to determine, in the foregoing M channel quantization codebooks, a second channel quantization codebook having the smallest correlation between the first channel quantization codebook corresponding to the first UE to be paired in each of the to-be-scheduled UEs, And pairing the UE corresponding to the second channel quantization code with the first UE; or determining that the first channel quantization codebook corresponding to the first UE meets the minimum combination in the M channel quantization codebooks.
  • the first channel quantization codebook set includes N channel quantization codebooks, and each of the channel quantization codebook sets in the first channel quantization codebook set is selected And obtaining, by the N UEs, the first UE and the obtained N UEs, wherein the first channel quantization codebook set and the first channel quantization codebook form a channel in the second channel quantization codebook set
  • the sum of the correlations between the quantized codebooks is less than Or a value equal to a correlation between the N+l channel quantization codebooks of the first channel quantization codebook and the N+l channel quantization codebooks, where N is greater than 1 and less than M;
  • the resource allocation module 503 is configured to allocate a transmission resource to the first UE and the UE successfully paired with the first UE.
  • the obtaining module 501 may be specifically configured to: perform quantization on a channel of each to-be-scheduled UE according to the pre-configured M channel quantization codebooks to obtain a channel quantization codebook corresponding to each of the to-be-scheduled UEs; or And the channel quantization codebook corresponding to each UE that is reported by each of the UEs to be scheduled, and the channel quantization codebook corresponding to each of the to-be-scheduled UEs is obtained.
  • the obtaining module 501 can obtain the channel quantization codebook corresponding to each UE to be scheduled in multiple manners. For example, the obtaining module 501 may quantize the channel of each to-be-scheduled UE according to the pre-configured M channel quantization codebooks to obtain a channel quantization codebook corresponding to each of the to-be-scheduled UEs respectively (ie, the access device 500 according to the UE) The channel measurement information is used for channel quantization.
  • the acquisition module 501 can receive the channel quantization codebook corresponding to the UE reported by each UE in the UE to be scheduled (that is, the UE performs channel quantization according to the channel measurement information, and the obtaining module 501 directly Acquiring, by the UE, a channel quantization codebook corresponding to the UE, according to the received channel quantization codebook corresponding to each UE reported by each UE in the to-be-scheduled UE, obtaining channel quantization corresponding to each to-be-scheduled UE.
  • the codebook or may be quantized by the third-party device according to the pre-configured M channel quantization codebooks, to obtain the channel quantization codebook corresponding to each of the to-be-scheduled UEs, and the third-party device is connected.
  • the channel quantization codebook corresponding to each to-be-scheduled UE is sent to the device 500, and the acquiring module 501 obtains corresponding to each to-be-scheduled UE.
  • the access device 500 may further include:
  • the queue module 504 is configured to: after the acquiring module 501 acquires the channel quantization codebook corresponding to each to-be-scheduled UE, according to the channel quantization codebook corresponding to each of the to-be-scheduled UEs, respectively, the respective to-be-scheduled UEs are respectively divided into their channel quantization In the candidate user queue corresponding to the codebook, wherein the M channel quantization codebooks respectively have a candidate user queue created.
  • the pairing module 502 may determine, in the M channel quantization codebooks, the first UE to be paired with each of the to-be-scheduled UEs, where the first UE may be the scheduling priority in the current UEs to be scheduled.
  • the pairing module 502 may use one of the candidate user queues corresponding to the second channel quantization codebook (the UE) For example, the UE with the highest scheduling priority among the candidate user queues corresponding to the second channel quantization codebook, the UE with the second highest scheduling priority, or other UEs or any one UE is paired with the first UE (wherein, the method is mainly for The MU-MIMO system only supports two UE pairing scenarios);
  • the pairing module 502 may determine, in the M channel quantization codebooks, a first channel quantization that meets a combined minimum correlation criterion with respect to a first channel quantization codebook corresponding to a first UE to be paired in each to-be-scheduled UE.
  • the first channel quantization codebook set includes N channel quantization codebooks (ie, the first channel quantization codebook set includes N of M channel quantization codebooks, but the first channel quantization codebook set does not include
  • the first channel quantized codebook, N+1 is the maximum number of matched UEs supported by the access device; and one of the UEs corresponding to each channel quantization codebook in the first channel quantization codebook set is selected respectively (if respectively, the UE can be selected separately)
  • the module 504 creates M candidate user queues to respectively manage UEs corresponding to different channel quantization codebooks, and the pairing module 502 can respectively select candidates corresponding to each channel quantization codebook in the first channel quantization codebook set.
  • One UE in the queue (wherein the UE is, for example, the UE with the highest scheduling priority among the corresponding candidate user queues, the UE with the second highest scheduling priority, or other UEs or any one of the UEs) to obtain N UEs)
  • the first UE and the obtained N UEs are paired; wherein, the first channel quantization codebook set and the first channel quantization codebook comprise a second channel quantization codebook set (wherein the second channel quantization codebook set includes
  • the sum of the correlation between the channel quantization codebooks in the N+1 channel quantization codebooks is less than or equal to any N+1 channel quantizations in the M channel quantization codebooks including the first channel quantization codebook.
  • the sum of the correlations between the codebooks, where N is greater than 1 and less than M (wherein this approach is primarily for scenarios where the MU-MIMO system supports N+1 UE pairings).
  • the resource allocation module 503 can include:
  • a precoding weight calculation subunit 5031 configured to respectively utilize the first UE and the first a channel quantization codebook of each UE in the paired UE group formed by the UE that is successfully paired with the UE, and determining a precoding weight value after each UE pairing;
  • the first CQI adjusting unit 5032 is configured to obtain, according to the precoding weight difference value corresponding to each UE, a CQI adjustment amount corresponding to each UE, and adjust, according to the CQI adjustment amount corresponding to each UE, the corresponding UE a CQI, where the precoding weight difference corresponding to each UE is a difference between a precoding weight value after each UE pairing and a precoding weight value before pairing;
  • the modulation and coding determining unit 5033 is configured to determine, according to the CQI corresponding to each UE adjusted by the first CQI adjusting unit 5032, a modulation and coding mode corresponding to each of the UEs;
  • the resource allocation unit 5034 is configured to perform the UE according to the modulation and coding determining unit 5033.
  • the resource allocation module 503 may include: a second CQI adjusting unit 5035, configured to respectively Obtaining a CQI adjustment amount of each UE by using a correlation between a channel quantization codebook of each UE in the paired UE group formed by the first UE and the UE successfully paired with the first UE, and based on the CQI of the UE Adjusting the amount to adjust the CQI of the UE;
  • the modulation and coding determining unit 5033 is configured to determine, according to the CQI corresponding to each UE adjusted by the second CQI adjusting unit 5035, the modulation and coding mode corresponding to each of the UEs.
  • the resource allocation unit 5034 is configured to be used by the UE according to the modulation and coding determining unit 5033.
  • the access device 500 in this embodiment may be, for example, a base station, an access point, or another entity in the network that has the UE access function. .
  • the access device 500 in this embodiment may be implemented in the access device in the foregoing method embodiment, and the functions of the respective functional modules may be specifically implemented according to the method in the foregoing method embodiment, and the specific implementation process may be implemented. Reference is made to the related description of the foregoing method embodiments, and details are not described herein again.
  • the channel quantization codebooks of the UEs to be scheduled are obtained by quantizing the channels of the UEs to be scheduled by the pre-configured M channel quantization codebooks, and the resource scheduling apparatus 500 will be a UE in each UE to be scheduled.
  • the paired UE and the channel quantization codebook corresponding to the UE corresponding to the channel quantization codebook are matched with one of the UEs, or the channel corresponding to the to-be-paired UE is quantized.
  • the N channel quantization codebooks in the channel quantization codebook set satisfying the combined minimum correlation criterion respectively correspond to one UE in the UE, a total of N UEs, and the to-be Pair the UEs for pairing.
  • the UE is scheduled to be scheduled according to the correlation between the channel quantization codebooks corresponding to the UEs to be scheduled, and the Greedy algorithm is not required to perform search calculation on all the UEs to be scheduled according to the pairing criterion, thereby facilitating greatly reducing the computational complexity of UE pairing. Degrees and delays simplify the UE pairing mechanism and enhance enforceability.
  • the embodiment of the invention further provides a communication system, which may include: an access device 500.
  • the embodiment of the present invention quantizes the channel of each UE to be scheduled by using the pre-configured M channel quantization codebooks to obtain a channel quantization codebook of each to-be-scheduled UE, and uses a certain UE in each to-be-scheduled UE as the to-be-paired UE. Then, the paired UE and the channel quantization codebook corresponding to the UE corresponding channel quantization codebook are matched to one of the UEs, or the channel quantization codebook corresponding to the to-be-paired UE is satisfied.
  • the N channel quantization codebooks in the channel quantization codebook set of the combined minimum correlation criterion respectively correspond to one UE in the UE, and a total of N UEs and the to-be-paired UE perform pairing.
  • the UE is scheduled to be scheduled according to the correlation between the channel quantization codebooks of the UEs to be scheduled, and the Greedy algorithm is not required to perform search calculation on all the UEs to be scheduled according to the pairing criterion, thereby facilitating greatly reducing the computational complexity of UE pairing. Degrees and delays simplify the UE pairing mechanism and enhance enforceability.
  • the program can be stored in a computer readable storage medium.
  • the storage medium can include: Read-only memory, random access memory, disk or optical disk, etc. , , , , I , , ; ; ;

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本发明实施例公开了一种资源调度方法及相关设备。通过预先构建的M个信道量化码本对各待调度UE的信道进行量化得到各待调度UE的信道量化码本，将各待调度UE中的某UE作为待配对UE后，将该待配对UE和与该UE对应信道量化码本相关性最小的信道量化码本对应UE中的一个进行配对，或，将相对于该待配对UE所对应的信道量化码本，满足组合最小相关性准则的信道量化码本集合中的N个信道量化码本分别对应UE中的一个UE，共N个UE和该待配对UE进行配对。由于是根据各待调度UE对应的信道量化码本间的相关性来配对待调度UE，故而有利于以降低UE配对的计算复杂度，缩短UE配对的时延。

Description

资源调度方法及设备 本申请要求于 2011 年 12 月 16 日提交中国专利局、 申请号为 201110424343.7、发明名称为"资源调度方法及相关设备"的中国专利申请的 优先权， 其全部内容通过引用结合在本申请中。 技术领域 本发明涉及无线通讯技术领域， 具体涉及一种资源调度方法及相关设 备。 背景技术 在无线通信系统中， 资源调度（如时间、 频率、 空间、 功率等）对系 统性能提升有着关键作用。
第一代和第二代移动通信系统一般为窄带系统， 通过将不同用户调度 到各自对应信道增益较大的时隙以获得多用户增益。 但是， 由于无线通信 系统一般为慢衰落（Slow Fading ) 系统， 时域调度能获得的增益相对有限。 第三代移动通信系统为宽带系统， 频率选择性衰落较强， 第三代移动通信 系统的基站可在时频二维域上进行调度， 多用户增益相对较明显。
多入多出 （ MIMO, Multiple Input Multiple Output ) 系统利用发射机和 接收机上的多根物理天线， 可在不增加频谱和功率资源的前提下， 使多径 衰落信道容量随着收发天线数量（发或收天线数量的最小值） 线性增长， 成为提升频谱效率的革命性技术。 在富反射环境下， 信道矩阵可认为是满 秩矩阵， 通过在发射机中进行预编码 (如 V-BLAST、 D-BLAST等）， 或者， 在接收机中进行信号检测 （例如 ML、 MMSE、 ZF等）， 可以消除天线间的 串扰， 形成多路并行的独立子信道， 总的信道容量为各个子信道的信道容 量之和， 打破了信道容量仅随功率对数增长的限制， 增加了空间自由度。
MIMO系统仅针对单用户，但是， 由于阵列天线阵元间的物理距离受到 很多限制， 可能不一定总是满足天线间信道衰落不相关的特性， 因此信道 容量会受到影响； 并且， 用户设备（UE, User Equipment ) 的体积、 成本、 功耗等有限制，也不宜配置过多的天线，很难在单用户 MIMO上获取更高的 容量和性能增益， 因此， 多用户多输入多输出 （MU-MIMO , Multi-User
MIMO ) 系统成为进一步提升系统性能的一个选择。
在 MU-MIMO系统中， 同一个时频资源上可空间复用多个 UE的数据， 但能支持的 UE数量远小于需要进行传输的 UE数量。 由于不同 UE所受到的 信道衰落不同， 因此在同一时频资源上进行传输所获得的容量和性能也截 然不同， 所以需要选取合适的 UE进行配对， 配对 UE联合在同一时频资源上 进行数据传输以获得多用户增益， 提高系统的频谱利用效率。
现有 MU-MIMO系统中， 基站在资源调度时釆用贪婪（Greedy ) 算法 来进行多 UE配对。 基站根据 UE不同的调度优先级， 先选取一个 UE, 然 后在剩下的 UE中进行遍历搜索，找到能与先选取的 UE配对以获取最高系 统容量的 UE进行配对； 而后继续在剩下的 UE 中遍历搜索， 直到选取的
UE数量达到最大 UE配对数量为止。 实践发现， Greedy算法需基站获取所 有 UE的信道状态信息， 每个 UE需要反馈整个传输带宽的信道状态信息， 而不仅仅反馈 UE各自对应带宽的信道增益； 且信道衰落快速随机变化， 需 要大量的资源开销。 同时， 基站每选取一个 UE, 都要对其它所有待调度
UE按照配对准则进行搜索计算， 该计算的复杂度和时延过大， 实际系统中 部署存在一定困难。 发明内容 本发明实施例提供资源调度方法及相关设备和通信系统， 以降低 UE配 对的计算复杂度， 缩短 UE配对的时延。
一方面， 本发明实施例提供一种资源调度方法， 包括：
获取各待调度用户设备 UE分别对应的信道量化码本， 其中， 所述各待 调度 UE中的每个 UE对应的信道量化码本通过预先构建的 M个信道量化码 本对所述每个 UE的信道进行量化得到；
确定出所述 M个信道量化码本中， 与所述各待调度 UE中待配对的第一 UE所对应的第一信道量化码本相关性最小的第二信道量化码本， 将所述第 二信道量化码本所对应 UE中的一个和所述第一 UE进行配对； 或者， 确定出 所述 M个信道量化码本中， 相对于所述第一 UE所对应的所述第一信道量化 码本满足组合最小相关性准则的第一信道量化码本集合， 其中， 所述第一 信道量化码本集合包含 N个信道量化码本，分别选出所述第一信道量化码本 集合中每个信道量化码本对应 UE中的一个， 得到 N个 UE, 将所述第一 UE 和该得到的 N个 UE进行配对， 其中， 所述第一信道量化码本集合和所述第 一信道量化码本所组成的第二信道量化码本集合中的信道量化码本两两之 间相关性的和值，小于或等于所述 M个信道量化码本中任何包含所述第一信 道量化码本的 N+1个信道量化码本两两之间相关性的和值， 其中， N大于 1 且小于 M;
为所述第一 UE及与所述第一 UE成功配对的 UE分配传输资源。
可选的， 所述方法还包括：
所述获取各待调度 UE分别对应的信道量化码本之后， 根据所述各待调 度 UE分别对应的信道量化码本，将该各待调度 UE分别划分到与其信道量化 码本相对应的候选用户队列中，其中，所述 M个信道量化码本分别对应创建 有候选用户队列。
可选的，所述将所述第二信道量化码本所对应 UE中一个和所述第一 UE 进行配对包括： 将所述第二信道量化码本所对应的候选用户队列中的一个 UE和所述第一 UE进行配对；
和 /或 ,
所述分别选出所述第一信道量化码本集合中每个信道量化码本对应的
UE中一个， 得到 N个 UE, 包括： 分别从所述第一信道量化码本集合中的每 个信道量化码本对应候选用户队列中选出一个 UE, 得到 N个 UE。
可选的， 所述为所述第一 UE及与所述第一 UE成功配对的 UE分配传输 资源， 包括：
分别利用所述第一 UE及与所述第一 UE成功配对的 UE所组成的配对 UE 组中的每个 UE的信道量化码本， 确定所述每个 UE配对后的预编码权值； 基 于所述每个 UE对应的预编码权值差值得到所述每个 UE对应的信道质量指 示 CQI调整量， 基于所述每个 UE对应的 CQI调整量调整所述每个 UE对应的 CQI, 其中， 所述每个 UE对应的预编码权值差值为所述每个 UE配对后的预 编码权值和配对前的预编码权值的差值； 根据所述每个 UE对应的 CQI确定 所述每个 UE对应的调制编码方式；根据所述每个 UE对应调制编码方式为所 述配对 UE组分配传输资源；
或者，
分别利用所述第一 UE及与所述第一 UE成功配对的 UE所组成的配对 UE 组中的每个 UE对应的信道量化码本之间的相关性，得到所述每个 UE对应的 CQI调整量， 基于所述每个 UE对应的 CQI调整量调整所述每个 UE对应的 CQI; 根据所述每个 UE对应的 CQI确定所述每个 UE对应的调制编码方式； 可选的， 所述第一 UE为所述各待调度 UE中调度优先级最高的 UE或任 意 UE; 和 /或，
所述将所述第二信道量化码本所对应 UE中一个和所述第一 UE配对得 到配对 UE组， 包括： 将所述第二信道量化码本所对应 UE中调度优先级最高 的 UE和所述第一 UE配对；
和 /或，
所述分别选出所述第一信道量化码本集合中每个信道量化码本对应 UE 中的一个， 得到 N个 UE, 包括： 分别选出所述第一信道量化码本集合中每 个信道量化码本对应的 UE中调度优先级最高的 UE, 得到 N个 UE。
另一方面， 本发明实施例提供一种资源调度装置， 包括：
获取模块， 用于获取各待调度用户设备 UE分别对应的信道量化码本， 其中， 所述各待调度 UE中的每个 UE对应的信道量化码本通过预先构建的 M 个信道量化码本对所述每个 UE的信道进行量化得到；
配对模块， 用于确定出所述 M个信道量化码本中， 与所述各待调度 UE 中待配对的第一 UE所对应的第一信道量化码本相关性最小的第二信道量化 码本， 将所述第二信道量化码本所对应 UE中的一个和所述第一 UE进行配 对； 或者， 确定出所述 M个信道量化码本中， 相对于所述第一 UE所对应的 所述第一信道量化码本满足组合最小相关性准则的第一信道量化码本集 合， 其中， 所述第一信道量化码本集合包含 N个信道量化码本， 分别选出所 述第一信道量化码本集合中每个信道量化码本对应 UE中的一个， 得到 N个 UE, 将所述第一 UE和该得到的 N个 UE进行配对， 其中， 所述第一信道量化 码本集合和所述第一信道量化码本所组成的第二信道量化码本集合中的信 道量化码本两两之间相关性的和值，小于或等于所述 M个信道量化码本中任 何包含所述第一信道量化码本的 N+1个信道量化码本两两之间相关性的和 值， 其中， N大于 1且小于 M;
资源分配模块， 用于为所述第一 UE及与所述第一 UE成功配对的 UE分 配传输资源。
可选的，所述获取模块具体用于，按照预先构建的 M个信道量化码本对 各待调度 UE的信道进行量化以得到所述各待调度 UE分别对应的信道量化 码本； 或根据接收到的各待调度 UE中的每个 UE上报的所述每个 UE对应的 信道量化码本， 得到所述各待调度 UE分别对应的信道量化码本。
可选的， 资源调度装置还包括：
队列模块， 用于在所述获取模块获取各待调度 UE分别对应的信道量化 码本之后， 根据所述各待调度 UE分别对应的信道量化码本， 将该各待调度 UE分别划分到与其信道量化码本相对应的候选用户队列中， 其中， 所述 M 个信道量化码本分别对应创建有候选用户队列。
可选的， 所述资源分配模块包括：
预编码权值计算子单元，用于分别利用所述第一 UE及与所述第一 UE成 功配对的 UE所组成的配对 UE组中的每个 UE的信道量化码本， 确定所述每 个 UE配对后的预编码权值；
第一 CQI调整单元， 用于基于所述每个 UE对应的预编码权值差值得到 所述每个 UE对应的 CQI调整量， 基于所述每个 UE对应的 CQI调整量调整所 述每个 UE对应的 CQI, 其中， 所述每个 UE对应的预编码权值差值为所述每 个 UE配对后的预编码权值和配对前的预编码权值的差值；
调制编码确定单元， 用于根据所述每个 UE对应的 CQI确定所述每个 UE 对应的调制编码方式；
资源分配单元，用于根据所述每个 UE对应调制编码方式为所述配对 UE 组分配传输资源。 可选的， 所述资源分配模块包括：
第二 CQI调整单元， 分别利用所述第一 UE及与所述第一 UE成功配对的 UE所组成的配对 UE组中的每个 UE对应的信道量化码本之间的相关性 , 得 到所述每个 UE对应的 CQI调整量， 基于所述每个 UE对应的 CQI调整量调整 所述每个 UE对应的 CQI;
调制编码确定单元， 用于根据所述每个 UE对应的 CQI确定所述每个 UE 对应的调制编码方式；
资源分配单元，用于根据所述每个 UE对应调制编码方式为所述配对 UE 组分配传输资源。
可选的， 所述接入设备为基站或接入点。
又一方面， 本发明实施例还提供一种通信系统， 包括：
如本发明实施例所述的接入设备。
由上可见，本发明实施例通过预先构建的 M个信道量化码本对各待调度 UE的信道进行量化得到各待调度 UE的信道量化码本， 将各待调度 UE中的 某 UE作为待配对 UE后， 将该待配对 UE和与该 UE对应信道量化码本相关性 最小的信道量化码本对应 UE中的一个进行配对， 或， 将相对于该待配对 UE 所对应的信道量化码本， 满足组合最小相关性准则的信道量化码本集合中 的 N个信道量化码本分别对应 UE中的一个 UE,共 N个 UE和该待配对 UE进行 配对。 由于是根据各待调度 UE对应的信道量化码本间的相关性来配对待调 度 UE, 无需使用 Greedy算法对所有待调度 UE按照配对准则进行搜索计算， 故而有利于降低 UE配对的计算复杂度和时延， 简化 UE配对机制， 增强可实 施性。 附图说明 图 1是本发明实施例提供的 MU-MIMO系统的下行链路模型示意图； 图 2是本发明实施例提供的一种资源调度方法的流程示意图；
图 3是本发明实施例提供的另一种资源调度方法的流程示意图； 图 4是本发明实施例提供的另一种资源调度方法的流程示意图； 图 5-a是本发明实施例提供的一种接入设备的示意图； 图 5-b是本发明实施例提供的另一种接入设备的示意图； 图 5-c是本发明实施例提供的一种接入设备的资源分配模块的示意图； 图 5-d是本发明实施例提供的另一种接入设备的资源分配模块的示意 图。 具体实施方式
UE配对的计算复杂度， 缩短 UE配对的时延。
为使得本发明的发明目的、 特征、 优点能够更加的明显和易懂， 下面 将结合本发明实施例中的附图， 对本发明实施例中的技术方案进行清楚、 完整地描述， 显然， 下面所描述的实施例仅仅是本发明一部分实施例， 而 非全部的实施例。 基于本发明中的实施例， 本领域普通技术人员在没有做 出创造性劳动前提下所获得的所有其它实施例， 都属于本发明保护的范围。
下面先对 MU-MIMO系统的工作方式进行简单介绍。
MU-MIMO系统的下行链路模型可如图 1所示。
基站发射机包括 M根天线， 基站选取 K个 UE进行多 UE通信， 其中， UE k包 Nk根天线， K个 UE总的天线数目可表示为：
Figure imgf000008_0001
UE k对应的接收数据可表示为向量：
Figure imgf000008_0002
其中， 下标 Lk表示 UE k的空间复用的层数， Lk的数值小于或等于， 基 站 发 射 天 线 和 UE k 接 收 天 线 数 量 的 最 小 值 ， 即
L ≤ min(M, Nk )
为了消除 UE间的干扰， 基站需要对发射数据进行预编码， 设 UE k对应 的预编码矩阵为 Pk, 该矩阵的维度为^ ^ , 即， 预编码矩阵的行数为 基站发射天线数 M, 列数为 UE k的空间复用层数 Lk。 所有配对 UE对应的发 射数据进行预编码后相加，形成基站各发射天线上的发射信号，可表示为 M
K
Figure imgf000009_0001
维的列向量： k=l 。
基站的发 天线与 UE k的信道响应矩阵 Hk可以表示为：
Figure imgf000009_0002
h
其中，信道响应矩阵 ¾中的元素 J表示基站的第 j根发射天线与 UE k 的第 i根接收天线间的信道响应系数。
此处， 若天线间传输的信号为窄带信号， 其信道响应可以由单抽头系 数表示； 若天线间传输的信号为宽带信号， 可利用正交频分复用 （OFDM, Orthogonal Frequency Division Multiplexing )技术对抗频率选择 '1"生衰落, OFDM符号的每个子载波对应的信道响应系数仍可由单抽头系数表示，可以 将上述分析应用于每个子载波， 此处不再赘述。
经过无线 后， 多用户接收信号模型可表示为公式（1 ):
Figure imgf000009_0003
公式（ 1 ) , yk 二 H k s + n k
公式（ 1 ) 中 信号;
Figure imgf000009_0004
的高斯白噪声 信号抽样值， s表示基站各天线的发射信号， 满足独立同分布的特性£
将公式（1 )展开： K
Figure imgf000010_0001
PjXj + n,
， 第一项 HkPkXk 表示 UE k的期望信号， 第二项
Figure imgf000010_0002
表示来自其它 UE的干扰信号。
预编码的目的之一就是抑制配对 UE间的干扰信号， Dirty-Paper Coding (脏纸编码）釆用是非线性编码方式， 可获得多 UE信道容量的上限。 最小 均方误差 ( MMSE, Minimum Mean-squared Error ), Zero-Forcing (迫零)、 以及块对角矩阵（BD, Block Diagonalized )等线性编码方式， 在适宜调度 和配对算法配合下， 也能取得接近最优的信道容量。
基站在已知信道状态信息 （CSI, Channel State Information ) 情况下， 通过选取合适的预编码矩阵， 理论上可使得配对 UE的干扰为 0。
因此， 预编码算法的目标函数可表示为公式（2 ): k j 0, j = k
HhP = 0, j≠ k
k J J 公式（2 )
包括 M根发射天线和 N根接收天线的总的信道模型可表示为公式（3 ):
Η Ρ H P
Η Ρ H P
+ yK_ ΗκΡλ ΗΚΡ2 · • · ΗΚΡΚ _χκ _ ηκ _ 公式（3 ) 其中， 选取满足预编码算法目标函数的预编码矩阵后， 等效信道响应 矩阵变成公式（4 ):
Figure imgf000011_0001
公式（4 ) 以单天线接收的 UE为例， 公式（2 )可演化为公式（5 ):
KPj ≠ O， j = k
在 MU-MIMO系统中， 同一个时频资源上可空间复用多个 UE的数据， 但能支持的 UE数量远往往小于需要进行数据传输的 UE数量。 由于不同的 UE所受到的信道衰落不同， 因此， 在同一个时频资源上进行传输所获得的 容量和性能也截然不同。 所以需要选取合适的 UE进行 UE配对， 联合在这个 时频资源上进行数据传输以获取多 UE增益， 提高系统的频谱利用效率。
在选取 UE进行配对时， 若使得其信道空间特征相互正交， 则能很好地 进行信号分离， 其关系可如下公式（6 ) 所示：
Figure imgf000011_0002
0
公式（ 6 )
Ρ0 ρ0
U为一个预设的相关系数门限， υ例如可设置为 0.05、 0.1、 0.15 hn Hh m
或其它值; 越小， 表明两个 UE的信道正交性越好， 信道相 关性越小。 本发明资源调度方法的一个实施例， 可包括： 获取各待调度 UE分别对 应的信道量化码本（ Channel Codebook ), 其中， 该各待调度 UE中的每个 UE 对应的信道量化码本通过预先构建的 M个信道量化码本来对该每个 UE的信 道进行量化得到； 确定出该 M个信道量化码本中， 与该各待调度 UE中待配 对的第一 UE对应的第一信道量化码本相关性最小的第二信道量化码本， 将 该第二信道量化码本所对应 UE中一个和该第一 UE配对； 或， 确定出该 M个 信道量化码本中， 相对于第一 UE所对应的第一信道量化码本满足组合最小 相关性准则的第一信道量化码本集合，该第一信道量化码本集合包含 N个信 道量化码本（即第一信道量化码本集合包含 M个信道量化码本中的 N个， 但 第一信道量化码本集合不包含第一信道量化码本）； 分别选出第一信道量化 码本集合中每个信道量化码本对应的 UE中一个， 得到 N个 UE, 将第一 UE 和该得到的 N个 UE进行配对； 其中， 第一信道量化码本集合和第一信道量 化码本所组成的第二信道量化码本集合中的信道量化码本两两之间相关性 的和值， 小于或者等于 M个信道量化码本中任何包含第一信道量化码本的 N+1个信道量化码本两两之间相关性的和值， 其中， N大于 1且小于 M; 为 第一 UE及与第一 UE成功配对的 UE分配传输资源。
参见图 2, 本发明实施例提供的一种资源调度方法可包括：
201、 接入设备获取各待调度 UE分别对应的信道量化码本；
其中，该各待调度 UE中的每个 UE对应的信道量化码本通过预先构建的 M个信道量化码本来对该每个 UE的信道进行量化得到；
在实际应用中， 可通过多种方式来获取各待调度 UE分别对应的信道量 化码本。 例如， 可按照预先构建的 M个信道量化码本对各待调度 UE的信道 进行量化以得到该各待调度 UE分别对应的信道量化码本（即接入设备根据 UE的信道测量信息进行信道量化）； 或者， 可接收各待调度 UE中的每个 UE 上报的该 UE对应的信道量化码本 (即 UE根据信道测量信息进行信道量化， 接入设备直接从 UE获取该 UE对应的信道量化码本）， 根据接收到的各待调 度 UE中的每个 UE上报的该每个 UE对应的信道量化码本， 得到该各待调度 UE分别对应的信道量化码本， 或者， 也可由第三方设备按照预先构建的 M 个信道量化码本对各待调度 UE的信道进行量化，以得到该各待调度 UE分别 对应的信道量化码本， 第三方设备向接入设备上报得到的各待调度 UE对应 的信道量化码本， 而接入设备则据此获取各待调度 UE分别对应的信道量化 码本。
在一种应用场景下， 为便于管理各待调度 UE, 接入设备还可以创建与 预先构建的 M个信道量化码本分别对应的候选用户队列 (即，创建 M个候选 用户队列；)； 在获取到各待调度 UE分别对应的信道量化码本之后，接入设备 可根据各待调度 UE分别对应的信道量化码本，将各待调度 UE分别划分到与 其对应的信道量化码本相对应的候选用户队列中， 其中， 在每个候选用户 队列中， 例如可将各个 UE按照调度优先级顺序排列或者任意排列或按照其 它方式排列。 如此， 则每个候选用户队列中的所有 UE对应的信道量化码本 相同， 每个候选用户队列中的每个 UE对应的信道量化码本即为该候选用户 队列所对应的信道量化码本。
202、接入设备确定出该 M个信道量化码本中， 与各待调度 UE中待配对 的第一 UE (其中，第一 UE可能是当前各待调度 UE中调度优先级最高的 UE、 调度优先级次高的 UE或其它 UE或任意一个 UE ) 所对应的第一信道量化码 本相关性最小的第二信道量化码本， 将该第二信道量化码本所对应 UE中一 个（该一个 UE例如为第二信道量化码本所对应 UE中调度优先级最高的 UE、 调度优先级次高的 UE或其它 UE或者任意一个 UE )和第一 UE配对。 其中， 若接入设备创建了 M个候选用户队列来分别管理对应不同信道量化码本的 UE,则接入设备可将第二信道量化码本所对应的候选用户队列中的一个 UE
(该 UE例如为第二信道量化码本所对应的候选用户队列中调度优先级最高 的 UE、 调度优先级次高的 UE或其它 UE或者任意一个 UE )和第一 UE进行配 对（其中， 此方式主要针对 MU-MIMO系统只支持两个 UE配对的场景）； 或， 接入设备确定出该 M个信道量化码本中， 相对于各待调度 UE中待 配对的第一 UE所对应的第一信道量化码本满足组合最小相关性准则的第一 信道量化码本集合， 该第一信道量化码本集合包含 N个信道量化码本 (即第 一信道量化码本集合包含 M个信道量化码本中的 N个，但第一信道量化码本 集合不包含第一信道量化码本， N+1为接入设备支持的最大配对 UE数）； 分 别选出第一信道量化码本集合中每个信道量化码本对应的 UE中一个（如可 分别选出第一信道量化码本集合中每个信道量化码本对应的 UE中调度优先 级最高的 UE、调度优先级次高的 UE或其它 UE或任意一个 UE ),得到 N个 UE
(其中，若接入设备创建 M个候选用户队列来分别管理对应不同信道量化码 本的 UE, 则接入设备可分别选出第一信道量化码本集合中每个信道量化码 本所对应的候选用户队列中的一个 UE (其中， 该 UE例如为对应的候选用户 队列中调度优先级最高的 UE、 调度优先级次高的 UE或其它 UE或任意一个 UE ), 以得到 N个 UE ), 将第一 UE和该得到的 N个 UE进行配对； 其中， 第一 信道量化码本集合和第一信道量化码本所组成的第二信道量化码本集合 (其中， 第二信道量化码本集合包括 N+1个信道量化码本）中的信道量化码 本两两之间相关性的和值，小于或者等于 M个信道量化码本中任何包含第一 信道量化码本的 N+1个信道量化码本两两之间相关性的和值， 其中， N大于 1且小于 M (其中， 此方式主要针对 MU-MIMO系统支持 N+1个 UE配对的场 景）。
203、 接入设备为第一 UE及与第一 UE成功配对的 UE分配传输资源。 在一种应用场景下，接入设备可分别利用第一 UE及与第一 UE成功配对 的 UE所组成的配对 UE组中的每个 UE的信道量化码本， 确定该每个 UE配对 后的预编码权值；基于该每个 UE对应的预编码权值差值得到该每个 UE对应 的信道质量指示（CQI, Channel Quality Indicator )调整量， 基于该每个 UE 对应的 CQI调整量调整该每个 UE对应的 CQI, 其中， 该每个 UE对应的预编 码权值差值为该每个 UE配对后的预编码权值和配对前的预编码权值的差 值； 根据该每个 UE对应的 CQI确定该每个 UE对应的调制编码方式； 根据该 每个 UE对应调制编码方式为该配对 UE组分配传输资源；
在另一种应用场景下，接入设备可分别根据第一 UE及与第一 UE成功配 对的 UE所组成的配对 UE组中的每个 UE的信道量化码本之间的相关性 , 得 到每个 UE对应的 CQI调整量， 并基于该每个 UE对应的 CQI调整量调整该每 个 UE对应的 CQI; 根据每个 UE对应的 CQI确定该 UE对应的调制编码方式； 根据每个 UE对应的调制编码方式为配对 UE组分配传输资源。
需要说明的是， 本发明实施例中的接入设备例如可以是基站、 接入点 或网络中其它具有 UE接入功能的实体。
由上可见， 本实施例通过预先构建的 M个信道量化码本对各待调度 UE 的信道进行量化得到各待调度 UE的信道量化码本， 将各待调度 UE中的某 UE作为待配对 UE后， 将该待配对 UE和与该 UE对应信道量化码本相关性最 小的信道量化码本对应 UE中的一个进行配对， 或， 将相对于该待配对 UE所 对应的信道量化码本，满足组合最小相关性准则的信道量化码本集合中的 N 个信道量化码本分别对应 UE中的一个 UE, 共 N个 UE和该待配对 UE进行配 对。 由于是根据各待调度 UE对应的信道量化码本间的相关性来配对待调度 UE, 无需使用 Greedy算法对所有待调度 UE按照配对准则进行搜索计算， 故 而有利于极大降低 UE配对的计算复杂度和时延， 简化 UE配对机制， 增强可 实施性。 为便于理解和实施本发明实施例的上述方案，下面以 MU-MIMO系统只 支持两个 UE配对的一个具体应用场景为例进行介绍。
参见图 3、 本发明实施例提供另一种资源调度方法可包括：
301、 接入设备创建与预先构建的 M个信道量化码本分别对应的候选用 户队列；
其中， 为便于后续管理各待调度 UE, 接入设备可创建与预先构建的 M 个信道量化码本分别对应的候选用户队列， 即创建 M个候选用户队列，每个 信道量化码本对应不同的候选用户队列。 接入设备还可为创建的每个候选 用户队列分配不同的队列索引， 例如可直接将每个候选用户队列对应的信 道量化码本作为该候选用户队列的队列索引， 或可为创建的每个候选用户 队列分配不同的其它队列索引。
在实际应用中，预先构建的 M个信道量化码本可尽量均勾分布， 该 M个 信道量化码本之间的相关性变化量也尽量均勾 ,该 M个信道量化码本尽量充 分遍历当前小区 UE可能遍历的所有信道， 使得该 M个信道量化码本能够尽 可能准确的对 UE的信道进行量化， 尽可能减少量化误差。
在 道量化码本映射成信道
Figure imgf000015_0001
星座图 , 根据待调度 UE的 CSI 按照最小欧式距离或者最大相关性准则进行星座映射， Nt为基站的发射天 线书， R为信道量化码本的总数。
星座映射机制可表示为公式 ( 7 ):
Figure imgf000015_0002
公式（ 7 ) 其中， 为待调度 UE i对应的预编码权值， ¾为待调度 UE i对应的信道 状态 n )„
Figure imgf000016_0001
其中， 若信道量化码本为信道的特征向量（eigen vector ), 则可以将待 调度 UE的特征波束加权向量， 按照最小欧式距离或最大相关性准则进行星 座映射， 得到该待调度 UE对应的信道特征向量量化码本， 即将公式（5 )中 的 用特征波束加权向量替代、 Fk用特征向量量化码本替代。
302、 接入设备获取到各待调度 UE分别对应的信道量化码本； 在实际应用中， 接入设备可能通过多种方式来获取各待调度 UE分别对 应的信道量化码本。
举例来说， 例如可在 UE中预置 M个信道量化码本， 或者接入设备可向 UE下发预先构建的 M个信道量化码本； UE则通过该 M个信道量化码本对其 信道进行量化以得到该 UE对应的信道量化码本；各待调度 UE可将获得的其 对应的信道量化码本（和 /或能够指示出该信道量化码本的码本指示信息） 上报给接入设备； 接入设备则可接收各待调度 UE中的每个 UE上报的该 UE 对应的信道量化码本（或该码本对应的指示信息）， 根据接收到的各待调度 UE中的每个 UE上报的该每个 UE对应的信道量化码本， 得到该各待调度 UE 分别对应的信道量化码本， 即， UE根据信道测量结果对其信道进行量化， 接入设备直接从 UE获取该 UE对应的信道量化码本。 或待调度 UE可向接入 设备上报 CSI; 接入设备可按照预先构建的 M个信道量化码本， 对各待调度 UE的信道进行量化以得到各待调度 UE分别对应的信道量化码本（即， 接入 设备根据 UE上报的 CSI对该 UE的信道进行量化）。 或， 由第三方设备按照预 先构建的 M个信道量化码本对各待调度 UE的信道进行量化， 以得到各待调 度 UE分别对应的信道量化码本， 第三方设备向接入设备上报得到的各待调 度 UE对应的信道量化码本 (和 /或能够指示出该信道量化码本的码本指示信 息）， 接入设备则据此获取各待调度 UE分别对应的信道量化码本（即， 接入 设备从第三方设备获取各待调度 UE分别对应的信道量化码本）。
303、 接入设备将各待调度 UE分别划分到候选用户队列中；
其中， 接入设备根据各待调度 UE分别对应的信道量化码本， 将各待调 度 UE分别划分到与其对应的信道量化码本相对应的候选用户队列中。
在实际应用中， 每个候选用户队列中的 UE, 例如按照调度优先级顺序 排列或者任意排列或按照其它方式排列。 如此， 每个候选用户队列中的所 有 UE对应的信道量化码本相同，每个候选用户队列中的每个 UE对应的信道 量化码本即为该候选用户队列所对应的信道量化码本。
这样通过信道量化码本对各待调度 UE进行信道量化， 实现基于信道量 化码本对各待调度 UE进行分组。
304、 接入设备从候选用户队列中选取一个 UE作为待配对 UE;
确定出该 M个信道量化码本中与该待配对 UE对应的信道量化码本相关 性最小的另一信道量化码本， 将该另一信道量化码本所对应的候选用户队 列中的一个 UE和该待配对 UE配对；
在实际应用中， 接入设备例如可先从候选用户队列中选取调度优先级 最高的 UE (或调度优先级次高的 UE或其它 UE或任意 UE )作为待配对 UE, 为便于引述， 可将该选取出的待配对 UE称之为 UE-1 , 将 UE-1所对应的信道 量化码本称之为第一信道量化码本， 将 UE-1所属的候选用户队列称之为第 一候选用户队列。
其中， 接入设备可根据各个信道量化码本之间的相关性， 确定出该 M 个信道量化码本中与 UE-1对应的第一信道量化码本相关性最小的另一信道 量化码本， 为便于引述， 可将确定出的该 M个信道量化码本中与 UE-1对应 的第一信道量化码本相关性最小的另一信道量化码本称之为第二信道量化 码本， 第二信道量化码本所对应的候选用户队列称之为第二候选用户队列。
可参考信道量化码本的星座图， 确定相对于第一信道量化码本， 欧式 距离最大或者相关性最小的星座点所代表的另一信道量化码本（即第二信 道量化码本）， 将第二信道量化码本对应的第二候选用户队列中调度优先级 最高的一个 UE (或调度优先级次高的 UE或其它 UE或任意 UE )和 UE-1进行 配对。
其中， 信道量化码本映射的星座图， 由于最大欧式距离或者最小相关 性的两个星座点所表示的两个信道量化码本之间的相关性最小， 该两个信 道量化码本所对应的候选用户队列中的 UE之间的干扰也相对最小， 因此将 满足该条件的 UE直接配对具有最小的 UE间干扰损失 t
配对公式可如公式（ 8 ):
F, F e {Fk \ k = l...n} ^ j = l...n
Figure imgf000018_0001
for j = 1 to n
) 其中
Figure imgf000018_0002
表示 最小相关性。
为便于快速完成配对， 还可事先建立一张配对队列映射表， 在配对队 列映射表中记录各候选用户队列之间的优选配对关系， 在配对时则根据配 对队列映射表中记录的优选配对关系进行配对。
可以理解， 釆用上述 UE配对机制， 由于是先将待调度 UE按照信道相关 性大小进行分组， 将信道相关性最小的候选用户队列中的 UE进行配对， 可 有效解决目前基于 DS域的贪婪算法造成的调度时延波动大的问题，提升 UE 配对成功概率， 简化调度计算量的同时提升 MU-MIMO系统吞吐量。
305、 接入设备分别根据 UE-1及与 UE-1成功配对的 UE所组成的配对 UE 组中的每个 UE的信道量化码本计算该每个 UE配对后的预编码权值，以便后 续消除多 UE间的干扰；
306、 接入设备基于配对 UE组中的每个 UE计算出配对后的预编码权值 和该每个 UE配对前的预编码权值的差值， 得到该每个 UE的 CQI调整量， 并 基于该每个 UE的 CQI调整量调整该每个 UE的 CQI;
其中， 在配对前， 每个待调度 UE对应一个预编码权值和 CQI (可以是 UE自行上报的）， 配对后， 引入了配对 UE之间的干扰， 为后续消除多 UE间 的干扰， 可对配对后的每个待调度 UE配对前的 CQI进行调整，调整后的 CQI 能更好的适应配对传输场景。
307、 接入设备根据配对 UE组中的每个 UE对应的 CQI确定该每个 UE对 应的调制编码方式（ MCS , Modulation and Coding Set );
308、 接入设备根据配对 UE组中的每个 UE对应的调制编码方式为配对 UE组分配传输资源。
在另一种应用场景下， 接入设备也可分别根据 UE-1及与 UE-1成功配对 的 UE所组成的配对 UE组中的每个 UE的信道量化码本之间的相关性 , 得到 每个 UE的 CQI调整量（信道量化码本之间的相关性确定后， 对应个码本的 CQI调整量也就可随之确定）， 并基于该每个 UE的 CQI调整量调整该每个 UE 的 CQI; 根据配对 UE组中的每个 UE对应的 CQI确定该每个 UE对应的调制编 码方式； 根据该每个 UE对应的调制编码方式为配对 UE组分配传输资源。
配对 UE组中包括 UE k和 UE i, 则 UE k的 CQI可通过公式（ 9 )表示：
CQL oc SINR' = S1NR、 + A L SINR、
ASINR κ = ( \F kk, ' F i. ) / H IFI
公式（ 9 ) 其中， Cg/ 表示 的 CQI
同理， 则 UE i的 CQI可表示为公式（ 10 ):
CQL oc S直 = SINRk + ASINR k
ASINR = {Fk, F ) HkFk
公式 ( 10 ) 其中， ^表示 的 CQi
由上可见， 本实施例通过预先构建的 M个信道量化码本对各待调度 UE 的信道进行量化得到各待调度 UE的信道量化码本， 将各待调度 UE中的某 UE作为待配对 UE后， 将该待配对 UE和与该 UE对应信道量化码本相关性最 小的信道量化码本对应 UE中的一个进行配对。 由于是根据各待调度 UE对应 的信道量化码本间的相关性来配对待调度 UE, 无需使用 Greedy算法对所有 待调度 UE按照配对准则进行搜索计算，故而有利于极大降低两个 UE进行配 对的计算复杂度和时延， 简化两个 UE配对的机制， 增强可实施性。 为便于理解和实施本发明实施例的上述方案，下面以 MU-MIMO系统支 持 N+1个 UE配对的一个具体应用场景为例进行介绍。
其中， N大于 1且小于 M
参见图 4、 本发明实施例提供的另一种资源调度方法可包括： 401、 接入设备创建与预先构建的 M个信道量化码本分别对应的候选用 户队列；
其中， 为便于后续管理各待调度 UE, 接入设备可创建与预先构建的 M 个信道量化码本分别对应的候选用户队列， 即创建 M个候选用户队列，每个 信道量化码本对应不同的候选用户队列。 接入设备还可为创建的每个候选 用户队列分配不同的队列索引， 例如可直接将每个候选用户队列对应的信 道量化码本作为该候选用户队列的队列索引。
在实际应用中，预先构建的 M个信道量化码本可尽量均勾分布， 该 M个 信道量化码本之间的相关性变化量也尽量均勾 ,该 M个信道量化码本尽量充 分遍历当前小区 UE可能遍历的所有信道， 使得该 M个信道量化码本能够尽 可能准确的对 UE的信道进行量化， 尽可能减少量化误差。
402、 接入设备获取到各待调度 UE分别对应的信道量化码本； 在实际应用中， 接入设备可能通过多种方式来获取各待调度 UE分别对 应的信道量化码本。
举例来说， 例如可在 UE中预置 M个信道量化码本， 或者接入设备可向 UE下发预先构建的 M个信道量化码本； UE则通过该 M个信道量化码本对其 信道进行量化以得到该 UE对应的信道量化码本；各待调度 UE可将获得的其 对应的信道量化码本（和 /或能够指示出该信道量化码本的码本指示信息） 上报给接入设备； 接入设备则可接收各待调度 UE中的每个 UE上报的该 UE 对应的信道量化码本（或该码本对应的指示信息）， 根据接收到的各待调度 UE中的每个 UE上报的该每个 UE对应的信道量化码本， 得到该各待调度 UE 分别对应的信道量化码本， 即， UE根据信道测量结果对其信道进行量化， 接入设备直接从 UE获取该 UE对应的信道量化码本。 或， 待调度 UE可向接 入设备上报信道状态信息 （CSI ); 接入设备可按照预先构建的 M个信道量 化码本，对各待调度 UE的信道进行量化以得到各待调度 UE分别对应的信道 量化码本（即， 接入设备根据 UE上报的 CSI对该 UE的信道进行量化）。 或， 由第三方设备按照预先构建的 M个信道量化码本对各待调度 UE的信道进行 量化， 以得到各待调度 UE分别对应的信道量化码本， 第三方设备向接入设 备上报得到的各待调度 UE对应的信道量化码本（和 /或能够指示出该信道量 化码本的码本指示信息），接入设备则据此获取各待调度 UE分别对应的信道 量化码本（即， 接入设备从第三方设备获取各待调度 UE分别对应的信道量 化码本）。
403、 接入设备将各待调度 UE分别划分到候选用户队列中；
其中， 接入设备根据各待调度 UE分别对应的信道量化码本， 将各待调 度 UE分别划分到与其对应的信道量化码本相对应的候选用户队列中。
在实际应用中， 每个候选用户队列中的 UE, 例如按照调度优先级顺序 排列或者任意排列或按照其它方式排列。 如此， 每个候选用户队列中的所 有 UE对应的信道量化码本相同，每个候选用户队列中的每个 UE对应的信道 量化码本即为该候选用户队列所对应的信道量化码本。
这样通过信道量化码本对各待调度 UE进行信道量化， 实现基于信道量 化码本对各待调度 UE进行分组。
404、 接入设备从候选用户队列中选取一个 UE作为待配对 UE (为便于 引述， 可将该选取出的待配对 UE称之为 UE-1 , 将 UE-1所对应的信道量化码 本称之为第一信道量化码本， 将 UE-1所属的候选用户队列称之为第一候选 用户队列）；
确定出该 M个信道量化码本中， 相对于 UE-1所对应的第一信道量化码 本满足组合最小相关性准则的第一信道量化码本集合， 该第一信道量化码 本集合包含 N个信道量化码本 (即第一信道量化码本集合包含 M个信道量化 码本中的 N个， 但第一信道量化码本集合不包含第一信道量化码本， N+1为 接入设备支持的最大配对 UE数 );分别选出第一信道量化码本集合中每个信 道量化码本对应的候选用户队列中一个 UE (例如可分别选出第一信道量化 码本集合中每个信道量化码本对应的对应的候选用户队列中调度优先级最 高的 UE、调度优先级次高的 UE或其它 UE或任意一个 UE ),得到 N个 UE, 将 UE-1和该得到的 N个 UE进行配对；
其中， 第一信道量化码本集合和第一信道量化码本所组成的第二信道 量化码本集合（其中， 第二信道量化码本集合包括 N+1个信道量化码本）中 的信道量化码本两两之间相关性的和值，小于或者等于 M个信道量化码本中 任何包含第一信道量化码的 N+ 1个信道量化码本两两之间相关性的和值。
其中， 则当选出 UE-1作为待配对 UE后， 从 M-1个信道量化码本选出 N ( -l)
个信道量化码本的选择方式共有 ^— w— 种可能的组合。 举例来说， 假设 M=16, N=3 , 则当选出 UE-1作为待配对 UE后， 从 M-1 个信道量化码本选出 N个信道量化码本的选择方式共有
-1)!
Figure imgf000022_0001
( 15*14*13) ) I ( 3*2*1 ) = 455种组合可能。
该 455种不包括第一信道量化码本的可能组合（第一信道量化码本集合 为其中的一种组合）加上第一信道量化码本也是 455种可能组合（第二信道 量化码本集合为其中的一种组合；)； 从中选取信道量化码本两两之间相关性 的和值最小的一种组合作为第二信道量化码本集合。
可以理解， 釆用上述 UE配对机制， 由于是先将待调度 UE按照信道相关 性大小进行分组， 将信道相关性最小的候选用户队列中的 UE进行配对， 可 有效解决目前基于 DS域的贪婪算法造成的调度时延波动大的问题，提升 UE 配对成功概率， 简化调度计算量的同时提升 MU-MIMO系统吞吐量。
405、 接入设备分别根据 UE-1及与 UE-1成功配对的 UE所组成的配对 UE 组中的每个 UE的信道量化码本计算该每个 UE配对后的预编码权值，以便后 续消除多 UE间的干扰；
406、 接入设备基于配对 UE组（共 N+1个 UE ) 中的每个 UE计算出配对 后的预编码权值和该每个 UE配对前的预编码权值的差值，得到该每个 UE的 CQI调整量， 并基于该每个 UE的 CQI调整量调整该每个 UE的 CQI;
其中， 在配对前， 每个待调度 UE对应一个预编码权值和 CQI (可以是 UE自行上报的）， 配对后， 引入了配对 UE之间的干扰， 为后续消除多 UE间 的干扰， 可对配对后的每个待调度 UE配对前的 CQI进行调整，调整后的 CQI 能更好的适应配对传输场景。
407、 接入设备根据配对 UE组中的每个 UE对应的 CQI确定该每个 UE对 应的调制编码方式（MCS );
408、 接入设备根据配对 UE组中的每个 UE对应的调制编码方式为配对 UE组分配传输资源。
在另一种应用场景下， 接入设备也可分别根据 UE-1及与 UE-1成功配对 的 UE所组成的配对 UE组中的每个 UE的信道量化码本之间的相关性， 得到 每个 UE的 CQI调整量（信道量化码本之间的相关性确定后， 对应个码本的 CQI调整量也就可随之确定 ), 并基于该每个 UE对应的 CQI调整量调整该每 个 UE对应的 CQI;根据配对 UE组中的每个 UE对应的 CQI确定该每个 UE对应 的调制编码方式；根据配对 UE组中的每个 UE对应的调制编码方式为该配对 UE组分配传输资源。
由上可见， 本实施例通过预先构建的 M个信道量化码本对各待调度 UE 的信道进行量化得到各待调度 UE的信道量化码本， 将各待调度 UE中的某 UE作为待配对 UE后， 将相对于该待配对 UE所对应的信道量化码本， 满足 组合最小相关性准则的信道量化码本集合中的 N个信道量化码本分别对应 UE中的一个 UE, 共 N个 UE和该待配对 UE进行配对。 由于是根据各待调度 UE对应的信道量化码本间的相关性来配对待调度 UE,无需使用 Greedy算法 对所有待调度 UE按照配对准则进行搜索计算，故而有利于极大降低多 UE配 对的计算复杂度和时延， 简化多 UE配对机制， 增强可实施性。
为便于更好的实施本发明实施例的上述方案， 下面还提供可用于实施 上述方法的相关装置。
参见图 5-a, 本发明实施例提供一种接入设备 500, 可包括：
获取模块 501、 配对模块 502和资源分配模块 503。
获取模块 501 , 用于获取各待调度 UE分别对应的信道量化码本， 其中， 该各待调度 UE中的每个 UE对应的信道量化码本通过预先构建的 M个信道 量化码本对该每个 UE的信道进行量化得到；
配对模块 502, 用于确定出上述 M个信道量化码本中， 与上述各待调度 UE中待配对的第一 UE所对应的第一信道量化码本相关性最小的第二信道 量化码本， 将第二信道量化码本所对应 UE中的一个和第一 UE进行配对； 或 者， 确定出上述 M个信道量化码本中， 相对于第一 UE所对应的第一信道量 化码本满足组合最小相关性准则的第一信道量化码本集合， 其中， 第一信 道量化码本集合包含 N个信道量化码本，分别选出第一信道量化码本集合中 每个信道量化码本对应 UE中的一个， 得到 N个 UE, 将第一 UE和该得到的 N 个 UE进行配对， 其中， 第一信道量化码本集合和第一信道量化码本所组成 的第二信道量化码本集合中的信道量化码本两两之间相关性的和值， 小于 或等于上述 M个信道量化码本中任何包含第一信道量化码本的 N+l个信道 量化码本两两之间相关性的和值， 其中， N大于 1且小于 M;
资源分配模块 503 , 用于为第一 UE及与第一 UE成功配对的 UE分配传输 资源。
在一种应用场景下， 获取模块 501可具体用于， 按照预先构建的 M个信 道量化码本对各待调度 UE的信道进行量化以得到上述各待调度 UE分别对 应的信道量化码本；或，接收到的各待调度 UE中的每个 UE上报的该每个 UE 对应的信道量化码本， 得到该各待调度 UE分别对应的信道量化码本。
在实际应用中，获取模块 501可通过多种方式来获取各待调度 UE分别对 应的信道量化码本。 举例来说， 获取模块 501可按照预先构建的 M个信道量 化码本对各待调度 UE的信道进行量化，以得到该各待调度 UE分别对应的信 道量化码本（即接入设备 500根据 UE的信道测量信息进行信道量化）； 或者， 获取模块 501可接收各待调度 UE中的每个 UE上报的该 UE对应的信道量化 码本（即 UE根据信道测量信息进行信道量化， 获取模块 501直接从 UE获取 该 UE对应的信道量化码本 ), 根据接收到的各待调度 UE中的每个 UE上报的 该每个 UE对应的信道量化码本，得到该各待调度 UE分别对应的信道量化码 本，或者，也可由第三方设备按照预先构建的 M个信道量化码本对各待调度 UE的信道进行量化， 以得到该各待调度 UE分别对应的信道量化码本， 第三 方设备向接入设备 500上报得到的各待调度 UE对应的信道量化码本，而获取 模块 501则据此获取各待调度 UE分别对应的信道量化码本。
参见图 5-b, 在一种应用场景下， 接入设备 500还可包括：
队列模块 504, 用于在获取模块 501获取各待调度 UE分别对应的信道量 化码本之后， 根据上述各待调度 UE分别对应的信道量化码本， 将该各待调 度 UE分别划分到与其信道量化码本相对应的候选用户队列中， 其中， 上述 M个信道量化码本分别对应创建有候选用户队列。
在一种应用场景下， 配对模块 502可确定出该 M个信道量化码本中， 与 各待调度 UE中待配对的第一 UE (其中， 第一 UE可能是当前各待调度 UE中 调度优先级最高的 UE、 调度优先级次高的 UE或其它 UE或任意一个 UE )所 对应的第一信道量化码本相关性最小的第二信道量化码本， 将该第二信道 量化码本所对应 UE中一个 （该一个 UE例如为第二信道量化码本所对应 UE 中调度优先级最高的 UE、调度优先级次高的 UE或其它 UE或者任意一个 UE ) 和第一 UE配对。 其中， 若队列模块 504创建了 M个候选用户队列来分别管理 对应不同信道量化码本的 UE,则配对模块 502可将第二信道量化码本所对应 的候选用户队列中的一个 UE (该 UE例如为第二信道量化码本所对应的候选 用户队列中调度优先级最高的 UE、 调度优先级次高的 UE或其它 UE或者任 意一个 UE )和第一 UE进行配对（其中， 此方式主要针对 MU-MIMO系统只 支持两个 UE配对的场景）；
或， 配对模块 502可确定出该 M个信道量化码本中， 相对于各待调度 UE 中待配对的第一 UE所对应的第一信道量化码本满足组合最小相关性准则的 第一信道量化码本集合， 该第一信道量化码本集合包含 N个信道量化码本 (即第一信道量化码本集合包含 M个信道量化码本中的 N个，但第一信道量 化码本集合不包含第一信道量化码本， N+1为接入设备支持的最大配对 UE 数）；分别选出第一信道量化码本集合中每个信道量化码本对应的 UE中一个 (如可分别选出第一信道量化码本集合中每个信道量化码本对应的 UE中调 度优先级最高的 UE、调度优先级次高的 UE或其它 UE或任意一个 UE ),得到 N个 UE (其中， 若队列模块 504创建 M个候选用户队列来分别管理对应不同 信道量化码本的 UE ,则配对模块 502可分别选出第一信道量化码本集合中每 个信道量化码本所对应的候选用户队列中的一个 UE (其中， 该 UE例如为对 应的候选用户队列中调度优先级最高的 UE、 调度优先级次高的 UE或其它 UE或任意一个 UE ), 以得到 N个 UE ), 将第一 UE和该得到的 N个 UE进行配 对； 其中， 第一信道量化码本集合和第一信道量化码本所组成的第二信道 量化码本集合（其中， 第二信道量化码本集合包括 N+1个信道量化码本）中 的信道量化码本两两之间相关性的和值，小于或者等于 M个信道量化码本中 任何包含第一信道量化码本的 N+1个信道量化码本两两之间相关性的和值 , 其中， N大于 1且小于 M (其中， 此方式主要针对 MU-MIMO系统支持 N+1 个 UE配对的场景）。
参见图 5-c , 在一种应用场景下， 资源分配模块 503可包括：
预编码权值计算子单元 5031 , 用于分别利用上述第一 UE及与上述第一 UE成功配对的 UE所组成的配对 UE组中的每个 UE的信道量化码本， 确定上 述每个 UE配对后的预编码权值；
第一 CQI调整单元 5032, 用于基于上述每个 UE对应的预编码权值差值 得到上述每个 UE对应的 CQI调整量， 基于上述每个 UE对应的 CQI调整量调 整上述每个 UE对应的 CQI, 其中， 上述每个 UE对应的预编码权值差值为上 述每个 UE配对后的预编码权值和配对前的预编码权值的差值；
调制编码确定单元 5033 ,用于根据第一 CQI调整单元 5032调整后的每个 UE对应的 CQI确定上述每个 UE对应的调制编码方式；
资源分配单元 5034, 用于根据调制编码确定单元 5033确定出的每个 UE 参见图 5-d, 在另一种应用场景下， 资源分配模块 503可包括： 第二 CQI调整单元 5035 , 用于分别利用第一 UE以及与第一 UE成功配对 的 UE所组成的配对 UE组中的每个 UE的信道量化码本之间的相关性 , 得到 每个 UE的 CQI调整量， 并基于该 UE的 CQI调整量调整该 UE的 CQI;
调制编码确定单元 5033 ,用于根据第二 CQI调整单元 5035调整后的每个 UE对应的 CQI , 确定上述每个 UE对应的调制编码方式
资源分配单元 5034, 用于根据调制编码确定单元 5033确定出的每个 UE 需要说明的是， 本实施例接入设备 500例如可以是基站、 接入点或网络 中其它具有 UE接入功能的实体。
可以理解的是， 本实施例中的接入设备 500可如上述方法实施例中的接 入设备中， 其各个功能模块的功能可根据上述方法实施例中的方法具体实 现， 其具体实现过程可以参照上述方法实施例的相关描述， 此处不再赘述。
由上可见， 本实施例通过预先构建的 M个信道量化码本对各待调度 UE 的信道进行量化得到各待调度 UE的信道量化码本，资源调度装置 500将各待 调度 UE中的某 UE作为待配对 UE后，将该待配对 UE和与该 UE对应信道量化 码本相关性最小的信道量化码本对应 UE中的一个进行配对， 或， 将相对于 该待配对 UE所对应的信道量化码本， 满足组合最小相关性准则的信道量化 码本集合中的 N个信道量化码本分别对应 UE中的一个 UE, 共 N个 UE和该待 配对 UE进行配对。 由于是根据各待调度 UE对应的信道量化码本间的相关性 来配对待调度 UE,无需使用 Greedy算法对所有待调度 UE按照配对准则进行 搜索计算， 故而有利于极大降低 UE配对的计算复杂度和时延， 简化 UE配对 机制， 增强可实施性。
本发明实施例还提供一种通信系统， 可包括： 接入设备 500。
需要说明的是， 对于前述的各方法实施例， 为了简单描述， 故将其都 表述为一系列的动作组合， 但是本领域技术人员应该知悉， 本发明并不受 所描述的动作顺序的限制， 因为依据本发明， 某些步骤可以釆用其他顺 序或者同时进行。 其次， 本领域技术人员也应该知悉， 说明书中所描述 的实施例均属于优选实施例， 所涉及的动作和模块并不一定是本发明所 必须的。
在上述实施例中， 对各个实施例的描述都各有侧重， 某个实施例中 没有详述的部分， 可以参见其他实施例的相关描述。
综上， 本发明实施例通过预先构建的 M个信道量化码本对各待调度 UE 的信道进行量化得到各待调度 UE的信道量化码本， 将各待调度 UE中的某 UE作为待配对 UE后， 将该待配对 UE和与该 UE对应信道量化码本相关性最 小的信道量化码本对应 UE中的一个进行配对， 或， 将相对于该待配对 UE所 对应的信道量化码本，满足组合最小相关性准则的信道量化码本集合中的 N 个信道量化码本分别对应 UE中的一个 UE, 共 N个 UE和该待配对 UE进行配 对。 由于是根据各待调度 UE对应的信道量化码本间的相关性来配对待调度 UE, 无需使用 Greedy算法对所有待调度 UE按照配对准则进行搜索计算， 故 而有利于极大降低 UE配对的计算复杂度和时延， 简化 UE配对机制， 增强可 实施性。
本领域普通技术人员可以理解上述实施例的各种方法中的全部或部分 步骤是可以通过程序来指令相关的硬件来完成， 该程序可以存储于一计算 机可读存储介质中， 存储介质可以包括： 只读存储器、 随机存储器、 磁盘 或光盘等。、 ， 、、 I 、 、； ；承 ^
述， 以上实施例的说明只是用于帮助理解本发明的方法及其核心思想； 同 时， 对于本领域的一般技术人员， 依据本发明的思想， 在具体实施方式及 应用范围上均会有改变之处， 综上， 本说明书内容不应理解为对本发明的 限制。

Claims

权利要求
1、 一种资源调度方法， 其特征在于， 包括：
获取各待调度用户设备 UE分别对应的信道量化码本， 其中， 所述各待 调度 UE中的每个 UE对应的信道量化码本通过预先构建的 M个信道量化码 本对所述每个 UE的信道进行量化得到；
确定出所述 M个信道量化码本中， 与所述各待调度 UE中待配对的第一 UE所对应的第一信道量化码本相关性最小的第二信道量化码本， 将所述第 二信道量化码本所对应 UE中的一个和所述第一 UE进行配对； 或者， 确定出 所述 M个信道量化码本中， 相对于所述第一 UE所对应的所述第一信道量化 码本满足组合最小相关性准则的第一信道量化码本集合， 其中， 所述第一 信道量化码本集合包含 N个信道量化码本，分别选出所述第一信道量化码本 集合中每个信道量化码本对应 UE中的一个， 得到 N个 UE, 将所述第一 UE 和该得到的 N个 UE进行配对， 其中， 所述第一信道量化码本集合和所述第 一信道量化码本所组成的第二信道量化码本集合中的信道量化码本两两之 间相关性的和值，小于或等于所述 M个信道量化码本中任何包含所述第一信 道量化码本的 N+1个信道量化码本两两之间相关性的和值， 其中， N大于 1 且小于 M;
为所述第一 UE及与所述第一 UE成功配对的 UE分配传输资源。
2、 根据权利要求 1所述的方法， 其特征在于，
所述方法还包括：
所述获取各待调度 UE分别对应的信道量化码本之后， 根据所述各待调 度 UE分别对应的信道量化码本，将该各待调度 UE分别划分到与其信道量化 码本相对应的候选用户队列中，其中，所述 M个信道量化码本分别对应创建 有候选用户队列。
3、 根据权利要求 2所述的方法， 其特征在于，
所述将所述第二信道量化码本所对应 UE中一个和所述第一 UE进行配 对包括： 将所述第二信道量化码本所对应的候选用户队列中的一个 UE和所 述第一 UE进行配对； 和 /或，
所述分别选出所述第一信道量化码本集合中每个信道量化码本对应的
UE中一个， 得到 N个 UE, 包括： 分别从所述第一信道量化码本集合中的每 个信道量化码本对应候选用户队列中选出一个 UE, 得到 N个 UE。
4、 根据权利要求 1至 3任一项所述的方法， 其特征在于，
所述为所述第一 UE及与所述第一 UE成功配对的 UE分配传输资源， 包 括：
分别利用所述第一 UE及与所述第一 UE成功配对的 UE所组成的配对 UE 组中的每个 UE的信道量化码本， 确定所述每个 UE配对后的预编码权值； 基 于所述每个 UE对应的预编码权值差值得到所述每个 UE对应的信道质量指 示 CQI调整量， 基于所述每个 UE对应的 CQI调整量调整所述每个 UE对应的 CQI, 其中， 所述每个 UE对应的预编码权值差值为所述每个 UE配对后的预 编码权值和配对前的预编码权值的差值； 根据所述每个 UE对应的 CQI确定 所述每个 UE对应的调制编码方式；根据所述每个 UE对应调制编码方式为所 述配对 UE组分配传输资源；
或者，
分别利用所述第一 UE及与所述第一 UE成功配对的 UE所组成的配对 UE 组中的每个 UE对应的信道量化码本之间的相关性，得到所述每个 UE对应的 CQI调整量， 基于所述每个 UE对应的 CQI调整量调整所述每个 UE对应的 CQI; 根据所述每个 UE对应的 CQI确定所述每个 UE对应的调制编码方式；
5、 根据权利要求 1至 3任一项所述的方法， 其特征在于，
所述第一 UE为所述各待调度 UE中调度优先级最高的 UE或任意 UE; 和 /或，
所述将所述第二信道量化码本所对应 UE中一个和所述第一 UE配对得 到配对 UE组， 包括： 将所述第二信道量化码本所对应 UE中调度优先级最高 的 UE和所述第一 UE配对；
和 /或，
所述分别选出所述第一信道量化码本集合中每个信道量化码本对应 UE 中的一个， 得到 N个 UE, 包括： 分别选出所述第一信道量化码本集合中每 个信道量化码本对应的 UE中调度优先级最高的 UE, 得到 N个 UE。
6、 一种接入设备， 其特征在于， 包括：
获取模块， 用于获取各待调度用户设备 UE分别对应的信道量化码本， 其中， 所述各待调度 UE中的每个 UE对应的信道量化码本通过预先构建的 M 个信道量化码本对所述每个 UE的信道进行量化得到；
配对模块， 用于确定出所述 M个信道量化码本中， 与所述各待调度 UE 中待配对的第一 UE所对应的第一信道量化码本相关性最小的第二信道量化 码本， 将所述第二信道量化码本所对应 UE中的一个和所述第一 UE进行配 对； 或者， 确定出所述 M个信道量化码本中， 相对于所述第一 UE所对应的 所述第一信道量化码本满足组合最小相关性准则的第一信道量化码本集 合， 其中， 所述第一信道量化码本集合包含 N个信道量化码本， 分别选出所 述第一信道量化码本集合中每个信道量化码本对应 UE中的一个， 得到 N个 UE, 将所述第一 UE和该得到的 N个 UE进行配对， 其中， 所述第一信道量化 码本集合和所述第一信道量化码本所组成的第二信道量化码本集合中的信 道量化码本两两之间相关性的和值，小于或等于所述 M个信道量化码本中任 何包含所述第一信道量化码本的 N+1个信道量化码本两两之间相关性的和 值， 其中， N大于 1且小于 M;
资源分配模块， 用于为所述第一 UE及与所述第一 UE成功配对的 UE分 配传输资源。
7、 根据权利要求 6所述的接入设备， 其特征在于， 还包括：
队列模块， 用于在所述获取模块获取各待调度 UE分别对应的信道量化 码本之后， 根据所述各待调度 UE分别对应的信道量化码本， 将该各待调度 UE分别划分到与其信道量化码本相对应的候选用户队列中， 其中， 所述 M 个信道量化码本分别对应创建有候选用户队列。
8、 根据权利要求 6或 7所述的接入设备， 其特征在于，
所述资源分配模块包括：
预编码权值计算子单元，用于分别利用所述第一 UE及与所述第一 UE成 功配对的 UE所组成的配对 UE组中的每个 UE的信道量化码本， 确定所述每 个 UE配对后的预编码权值；
第一 CQI调整单元， 用于基于所述每个 UE对应的预编码权值差值得到 所述每个 UE对应的 CQI调整量， 基于所述每个 UE对应的 CQI调整量调整所 述每个 UE对应的 CQI, 其中， 所述每个 UE对应的预编码权值差值为所述每 个 UE配对后的预编码权值和配对前的预编码权值的差值；
调制编码确定单元，用于根据所述第一 CQI调整单元调整后的所述每个 UE对应的 CQI , 确定所述每个 UE对应的调制编码方式；
资源分配单元， 用于根据所述调制编码确定单元确定出的每个 UE对应 调制编码方式为所述配对 UE组分配传输资源。
9、 根据权利要求 6或 7所述的接入设备， 其特征在于，
所述资源分配模块包括：
第二 CQI调整单元， 分别利用所述第一 UE及与所述第一 UE成功配对的 UE所组成的配对 UE组中的每个 UE对应的信道量化码本之间的相关性 , 得 到所述每个 UE对应的 CQI调整量， 基于所述每个 UE对应的 CQI调整量调整 所述每个 UE对应的 CQI;
调制编码确定单元， 用于根据所述第二 CQI调整单元调整后的每个 UE 对应的 CQI , 确定所述每个 UE对应的调制编码方式；
资源分配单元， 用于根据所述调制编码确定单元确定出的每个 UE对应 调制编码方式为所述配对 UE组分配传输资源。
10、 根据权利要求 6或 7所述的接入设备， 其特征在于，
所述接入设备为基站或接入点。
PCT/CN2012/086742 2011-12-16 2012-12-17 资源调度方法及设备 WO2013087036A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201110424343.7A CN102438322B (zh) 2011-12-16 2011-12-16 资源调度方法及相关设备
CN201110424343.7 2011-12-16

Publications (1)

Publication Number Publication Date
WO2013087036A1 true WO2013087036A1 (zh) 2013-06-20

Family

ID=45986136

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2012/086742 WO2013087036A1 (zh) 2011-12-16 2012-12-17 资源调度方法及设备

Country Status (2)

Country Link
CN (1) CN102438322B (zh)
WO (1) WO2013087036A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11323167B2 (en) * 2020-04-13 2022-05-03 National Tsing Hua University Communication time allocation method using reinforcement learning for wireless powered communication network and base station

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102438322B (zh) * 2011-12-16 2014-03-26 华为技术有限公司 资源调度方法及相关设备
CN103748943A (zh) 2012-08-17 2014-04-23 华为技术有限公司 用户设备配对处理方法、网络侧设备和用户设备
CN105472754A (zh) * 2014-09-02 2016-04-06 中兴通讯股份有限公司 资源分配方法及装置
CN106413110B (zh) * 2015-07-31 2019-09-17 电信科学技术研究院 一种调度方法、装置及网络节点
CN106230560A (zh) * 2016-09-28 2016-12-14 西北工业大学 无线网络中协作多媒体多播数据传输方法
CN108810980B (zh) * 2017-04-28 2020-10-23 华为技术有限公司 一种数据传输的方法、装置及设备
CN111757330B (zh) * 2019-03-27 2021-12-28 华为技术有限公司 一种用户配对方法及相关设备

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101383645A (zh) * 2007-09-07 2009-03-11 中兴通讯股份有限公司 一种上行多用户设备虚拟多输入多输出的配对方法
CN101465684A (zh) * 2009-01-08 2009-06-24 上海交通大学 Mimo系统用户端码本快速匹配方法
CN101577574A (zh) * 2008-05-09 2009-11-11 中兴通讯股份有限公司 控制信令发送方法
CN102438322A (zh) * 2011-12-16 2012-05-02 华为技术有限公司 资源调度方法及相关设备

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7961775B2 (en) * 2007-01-09 2011-06-14 Broadcom Corporation Method and system for a delta quantizer for MIMO pre-coders with finite rate channel state information feedback
CN101277523A (zh) * 2007-03-30 2008-10-01 松下电器产业株式会社 减少预编码和信道质量信息反馈的方法和装置
CN101615935B (zh) * 2009-07-29 2011-01-12 普天信息技术研究院有限公司 一种用户选择方法、装置和系统

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101383645A (zh) * 2007-09-07 2009-03-11 中兴通讯股份有限公司 一种上行多用户设备虚拟多输入多输出的配对方法
CN101577574A (zh) * 2008-05-09 2009-11-11 中兴通讯股份有限公司 控制信令发送方法
CN101465684A (zh) * 2009-01-08 2009-06-24 上海交通大学 Mimo系统用户端码本快速匹配方法
CN102438322A (zh) * 2011-12-16 2012-05-02 华为技术有限公司 资源调度方法及相关设备

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11323167B2 (en) * 2020-04-13 2022-05-03 National Tsing Hua University Communication time allocation method using reinforcement learning for wireless powered communication network and base station

Also Published As

Publication number Publication date
CN102438322A (zh) 2012-05-02
CN102438322B (zh) 2014-03-26

Similar Documents

Publication Publication Date Title
CN109952716B (zh) 用于高级csi反馈开销减少的可配置码本
KR102270375B1 (ko) 개선된 무선 통신 시스템에서의 선형 조합 pmi 코드북 기반 csi 보고
WO2013087036A1 (zh) 资源调度方法及设备
US8923427B2 (en) Codebook sub-sampling for frequency-selective precoding feedback
KR101506922B1 (ko) 정보 피드백 및 프리코딩을 위한 방법 및 장치
JP6148325B2 (ja) 進歩したアンテナシステムにおけるチャンネル品質情報の決定
JP6118423B2 (ja) チャネル状態情報をフィードバックする方法、ユーザ装置及び基地局
CN109951217B (zh) 一种数据传输方法、用户设备和基站
WO2011134322A1 (zh) 信道状态信息的反馈方法及终端
WO2013159517A1 (zh) 一种实现用户调度的用户配对方法、装置和系统
WO2011020428A1 (zh) 用于实现下行多输入多输出传输的方法和装置
WO2011160451A1 (zh) 信道状态信息的反馈方法及终端
WO2012068880A1 (zh) 一种信道状态信息反馈方法及装置
WO2015109796A1 (zh) 信道状态信息处理方法、装置、终端及基站
JP2009542052A (ja) プリコード化されたｍｉｍｏ−ｏｆｄｍシステムにおいてフィードバック情報のオーバヘッドを低減させるための方法
WO2011140062A1 (en) Method and precoder information feedback in multi-antenna wireless communication systems
WO2012041107A1 (zh) 信道状态信息反馈方法及终端
JP2016509405A (ja) 垂直及び全次元ビーム形成のためのチャネルフィードバック
WO2012041102A1 (zh) 反馈信道信息的方法及终端
JP2011505744A (ja) 空間多重化のためのプリコーダ、多重アンテナ送信機
TWI538428B (zh) 在網路中通信的方法、副站台及主站台
JP6595694B2 (ja) 下りプリコーディング方法および基地局
US9686005B2 (en) Method and apparatus for feeding back channel estimation in multi-input multi-output system
KR102050745B1 (ko) 무선통신 시스템에서 기지국과 단말 사이의 정보 송수신 방법 및 장치
WO2014180319A1 (zh) 一种基于码本的信息反馈方法和装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 12858449

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 12858449

Country of ref document: EP

Kind code of ref document: A1