WO2015123942A1 - 一种实现波束成型的方法及基站 - Google Patents
一种实现波束成型的方法及基站 Download PDFInfo
- Publication number
- WO2015123942A1 WO2015123942A1 PCT/CN2014/079591 CN2014079591W WO2015123942A1 WO 2015123942 A1 WO2015123942 A1 WO 2015123942A1 CN 2014079591 W CN2014079591 W CN 2014079591W WO 2015123942 A1 WO2015123942 A1 WO 2015123942A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- channel
- base station
- serving cell
- matrix
- model
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 32
- 239000011159 matrix material Substances 0.000 claims abstract description 182
- 238000005457 optimization Methods 0.000 claims abstract description 131
- 238000012549 training Methods 0.000 claims description 46
- 238000004891 communication Methods 0.000 claims description 43
- 238000004422 calculation algorithm Methods 0.000 claims description 30
- 230000005540 biological transmission Effects 0.000 claims description 21
- 230000000670 limiting effect Effects 0.000 claims description 8
- 230000036961 partial effect Effects 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 abstract description 8
- 230000007246 mechanism Effects 0.000 description 16
- 238000010586 diagram Methods 0.000 description 13
- 238000013461 design Methods 0.000 description 10
- 230000002452 interceptive effect Effects 0.000 description 10
- 238000005516 engineering process Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 7
- 230000003993 interaction Effects 0.000 description 5
- 238000011160 research Methods 0.000 description 4
- 239000000654 additive Substances 0.000 description 3
- 230000000996 additive effect Effects 0.000 description 3
- 238000012804 iterative process Methods 0.000 description 3
- 238000010295 mobile communication Methods 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- 230000001629 suppression Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000000052 comparative effect Effects 0.000 description 2
- 230000003111 delayed effect Effects 0.000 description 2
- 230000002829 reductive effect Effects 0.000 description 2
- 230000003595 spectral effect Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000002860 competitive effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- WYROLENTHWJFLR-ACLDMZEESA-N queuine Chemical compound C1=2C(=O)NC(N)=NC=2NC=C1CN[C@H]1C=C[C@H](O)[C@@H]1O WYROLENTHWJFLR-ACLDMZEESA-N 0.000 description 1
- 238000007493 shaping process Methods 0.000 description 1
- 230000011664 signaling Effects 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Classifications
-
- 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
- H04B7/06—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station
- H04B7/0613—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission
- H04B7/0615—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal
- H04B7/0617—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal for beam forming
Definitions
- the present invention relates to a multi-input single-output (MISO) wireless communication technology, and more particularly to a method and a base station for implementing beamforming based on a time division multiplexing (TDD) system.
- MISO multi-input single-output
- TDD time division multiplexing
- the evolutionary system of the third generation mobile communication technology represented by Long Term Evolution (LTE) realized a phased change of mobile communication after 3G.
- LTE Long Term Evolution
- 3GPP adopted LTE enhanced technology (LTE-A, LTE-Advanced) as the fourth generation mobile communication.
- LTE-A LTE enhanced technology
- 4G Multi-cell MIMO is one of the key technologies of LTE-A.
- the LTE-A downlink uses an orthogonal frequency division multiple access (OFDMA) access mode, so that the channel frequency may be the same at the combined edge of the two cells. Therefore, users at the edge of the cell will have strong interference with neighboring cells.
- OFDMA orthogonal frequency division multiple access
- a research hotspot of current multi-cell interference suppression is to adjust the distributed beamforming of the antenna. Its design purpose is to adjust the allocation of the antenna, that is, how to configure the transmitting antenna when each user transmits on each subcarrier, and how the base station should be configured. Configure the receive antenna so that the transmit beam is at most orthogonal to the receive antenna.
- the purpose of beamforming technology is to reduce inter-cell interference by combining multiple base stations.
- the traditional solution aggregates all the information into a central processing unit (CPU) to calculate the optimization problem of the entire system.
- CPU central processing unit
- this solution More difficult to achieve. Therefore, a distributed solution with less information interaction overhead and less computational overhead is needed.
- Each base station obtains other cell information through interaction, and then calculates an optimization problem of the cell according to local information, and obtains a solution of the system in an iterative manner. Therefore, the distributed framework is highly feasible.
- the convergence and convergence of algorithm iterations under the distributed framework Convergence speed is also a problem to be considered. Algorithms that require iterative processes often have obvious defects in time. Therefore, distributed solutions do not have application value under actual conditions.
- CSI Channel State Information
- the industry 3GPP standard specifies that the uplink and downlink of the TDD system communicate using time slots of different frequency bands. Therefore, the base station can estimate the channel according to the training sequence sent by the user uplink, also called the Sounding signal. Therefore, the unique mechanism of the TDD system can be used as a starting point. Design a distributed cooperation mechanism.
- the technical problem to be solved by the embodiments of the present invention is to provide a method and a base station for implementing beamforming, which can effectively suppress channel errors of the TDD system and eliminate complicated iterative calculation steps, thereby effectively suppressing interference between cells.
- a method of implementing beamforming comprising:
- the method further includes: the base station establishing the integrated channel error model by using channel reciprocity of the time division multiplexing TDD system.
- the step of the base station establishing the integrated channel error model by using channel reciprocity of the time division multiplexing TDD system comprises:
- the step of establishing a downlink beam optimization model includes:
- the integrated channel error model is aggregated into a matrix form to obtain a comprehensive channel error model in the form of a matrix;
- the downlink beam optimization model is obtained by performing a summation operation on the indicators indicating the communication quality of all the serving cells, minimizing the sum of the indicators indicating the communication quality, and the transmission power of each base station being less than the predetermined power as a limiting condition.
- the step of acquiring an indicator indicating a communication quality of the serving cell includes: obtaining a serving cell by limiting a receiving coefficient of each serving cell according to the obtained integrated channel error model in a matrix form and the downlink model An indicator of communication quality.
- the steps of separately acquiring a channel matrix of each cell except the serving cell and a non-ideal parameter corresponding to each channel matrix in the cooperative cluster where the base station is located and the base station where the base station is located include:
- the step of acquiring the channel matrix of the serving cell and the corresponding non-ideal parameters according to the received first training sequence signal includes:
- the base station measures the non-ideal parameters corresponding to the channel matrix of the serving cell according to the acquired channel matrix of the serving cell, and obtains a non-ideal parameter corresponding to the channel matrix of the monthly service cell.
- the step of acquiring, according to the received second training sequence signal, a channel matrix of the cell other than the serving cell in the cooperative cluster and corresponding non-ideal parameters includes:
- the step of obtaining an optimal beamforming matrix optimization target according to the established downlink beam optimization model includes:
- the simultaneous equations are established according to the obtained optimal conditions ⁇ and the maximum transmit power as constraints.
- the equations are solved to obtain the optimal beamforming matrix optimization target.
- the indicator indicating communication quality includes an average mean square error AMSE.
- a base station includes a first establishing module, an arithmetic module, an obtaining module, and a processing module, where: the first establishing module is configured to: establish a downlink beam optimization model according to the integrated channel error model and the downlink model;
- the operation module is configured to: obtain an optimal beam shaping matrix optimization target according to the established downlink beam optimization model;
- the obtaining module is configured to: separately obtain a channel matrix of each cell except the serving cell and a non-ideal parameter corresponding to each channel matrix in the cooperative cluster where the base station is located; and the processing module is configured to: obtain And a non-ideal parameter corresponding to each channel matrix and each channel matrix, and the optimal beamforming matrix optimization target, acquiring a beamforming matrix of the base station, The base station transmits downlink data according to the obtained beamforming matrix.
- the base station further includes a second establishing module, where:
- the second establishing module is configured to: establish the integrated channel error model by using channel reciprocity of the time division multiplexing TDD system before establishing the downlink beam optimization model.
- the second establishing module is configured to establish the integrated channel error model by using channel reciprocity of the time division multiplexing TDD system as follows:
- the first establishing module is configured to establish the downlink beam optimization model as follows:
- the integrated channel error model is aggregated into a matrix form to obtain a comprehensive channel error model in the form of a matrix;
- the downlink beam optimization model is obtained by performing a summation operation on the indicators indicating the communication quality of all the serving cells, minimizing the sum of the indicators indicating the communication quality, and using the transmission power of each base station to be less than the predetermined power as a limiting condition.
- the first establishing module is configured to obtain, according to the manner, an indicator indicating a communication quality of the serving cell:
- an indicator indicating the communication quality of the serving cell is obtained.
- the obtaining module includes a first obtaining module and a second obtaining module, where the first obtaining module is configured to: receive a first training sequence signal sent by all terminal devices UE in the serving cell, according to the received Obtaining, by the first training sequence signal, a channel matrix of the serving cell and corresponding non-ideal parameters;
- the second obtaining module is configured to: receive a second training sequence signal sent by all terminal equipment UEs in the cell other than the serving cell in the cooperative cluster, and acquire the cooperative cluster according to the received second training sequence signal A channel matrix of cells other than the serving cell and its corresponding non-ideal parameters.
- the first obtaining module is configured to obtain, according to the received first training sequence signal, a channel matrix of the serving cell and corresponding non-ideal parameters according to the following manner:
- the base station measures a non-ideal parameter corresponding to the channel matrix of the serving cell according to the obtained channel matrix of the serving cell, and obtains a non-ideal parameter corresponding to the channel matrix of the monthly service cell.
- the second obtaining module is configured to acquire, according to the received second training sequence signal, a channel matrix of the cell other than the serving cell in the cooperative cluster and corresponding non-ideal parameters in the cooperative cluster according to the received information:
- the operation module is configured to obtain an optimal beamforming matrix optimization target according to the established downlink beam optimization model as follows:
- the simultaneous equations are established, and the optimal beamforming matrix optimization target is obtained by solving the equations.
- the indicator indicating the communication quality includes an average mean square error AMSE.
- the technical solution of the present application includes: establishing a downlink beam optimization model according to the integrated channel error model and the downlink model; obtaining an optimal beamforming matrix optimization target according to the established downlink beam optimization model; respectively acquiring a cooperative cluster where the serving cell and the base station where the base station is located a channel matrix of each cell except the serving cell and a non-ideal parameter corresponding to each channel matrix; obtaining, according to each obtained channel matrix and its corresponding non-ideal parameters, and the optimal beamforming matrix optimization target a beamforming matrix of the base station, the base station transmitting downlink data according to the obtained beamforming matrix.
- the solution of the beamforming matrix of each base station is in a closed form, and the implementation can be obtained only by one calculation, and the computational complexity is low, and according to the above algorithm process, The secondary scheduling base station only needs to calculate a closed solution, and there is no iterative process, and the project is easy to implement.
- the embodiment of the invention considers that the estimation error and the delay error exist in the TDD system at the same time, and establishes a comprehensive channel error model for the channel of the TDD system, and achieves robustness.
- the solution having the same reception coefficient of each serving cell and the sum of the indicators indicating the communication quality are minimized, and the solutions of the beamforming matrix of each base station are all in a closed form.
- the embodiment of the invention designs a cooperation mechanism between the base stations, and realizes that the solution has a natural distributed property under the TDD system.
- FIG. 1 is a flowchart of a method for implementing beamforming according to an embodiment of the present invention
- FIG. 2 is a schematic structural diagram of a base station according to an embodiment of the present invention.
- FIG. 3 is an application scenario diagram of an embodiment of the present invention.
- FIG. 4 is a schematic diagram of a cell cooperation mechanism in an application scenario diagram according to an embodiment of the present invention.
- FIG. 5 is a comparison diagram of throughput curves of a technical solution and other beamforming schemes according to an embodiment of the present invention
- FIG. 6 is a comparison diagram of an AMSE curve achieved by a technical solution of an embodiment of the present invention and a non-robust conventional MMSE beamforming scheme.
- FIG. 1 is a flowchart of a method for implementing beamforming according to an embodiment of the present invention, including the following steps: Step 101: Establish a downlink beam optimization model according to an integrated channel error model and a downlink model.
- the method further includes: establishing, by the base station, the integrated channel error model by using channel reciprocity of the time division multiplexing TDD system as follows:
- an estimated error model is obtained based on the actual value and the estimated value of the channel state information CSI of the base station.
- the channel reciprocity of the TDD system means that the uplink and downlink air interface radio propagation channels are the same.
- the estimation error is caused by the channel estimation characteristics of the TDD system. Since the TDD system channel has reciprocity, the system estimates the CSI through the uplink sounding signal of the UE. Due to the non-ideality of the Sounding estimation sequence, additive estimation errors are produced.
- an antenna correlation matrix of BS-m; e me for each element of independent and identically distributed error vector, E ⁇ e m3 ⁇ 4 e 3 ⁇ 4 ⁇ I Wr, 3 ⁇ 4 of the power spectral density. Characterizing the CSI error value due to the limited Sounding signal when estimating channel estimation in a TDD system.
- a delay error model is obtained.
- the delay error is caused by the delay between the CSI estimation time and the downlink transmission time of the system.
- the delay is too large, the actual channel used for downlink transmission and the estimated CSI may have a large deviation. If h m3 ⁇ 4 oh M3 ⁇ 4 [U] is used to indicate the kth time (the time when the data is transmitted) and the kd time, respectively.
- n m3 ⁇ 4 [W is the additive part of the delay error, each element obeys the 0 mean, and the Gaussian independent distribution of the power spectral density is 1-.
- n m3 ⁇ 4 [A] characterizes the dissimilarity of CSI at two different times in a TDD system.
- the delay error model is substituted into the estimated error model to obtain the integrated channel error model.
- Equation (1) represents the (k - d)th time (CSI estimation time) The estimated value of CSI and the kth time (time of transmission of data) CSI The relationship between the actual values. If the time interval between the channel estimation and the downlink data transmission is known, the base station estimates the downlink transmission data according to the estimated time to be the true CSI.
- the integrated channel error model is aggregated into a matrix form to obtain an integrated channel error model in the form of a matrix.
- the channels of one cell are aggregated into a matrix
- the integrated channel model in the form of a matrix is:
- an indicator indicating the communication quality of the serving cell is obtained.
- ⁇ uE cell is q
- the channel matrix of all UEs, h M3 ⁇ 4 is a ⁇ ⁇ dimension vector, representing (BS-m) to (CSI between US-; 11 3 ⁇ 4 has rows and columns, is a beamforming matrix of base station q, is an embodiment of the present invention Design target; x : f is used to represent the signal transmitted by BS-q, is a scalar, indicating a useful signal to be transmitted to UE-, without loss of generality, assuming it has unit energy, ie £
- 2 l
- Zg [z 3 ⁇ 4 ,z 3 ⁇ 4 ,...,z 3 ⁇ 4
- the system evaluation index MSE indicates the deviation between the expected signal and the actual signal, which can be expressed as formula (2):
- ? is the receiving coefficient corresponding to the cell q.
- the system overhead is reduced, the MSE model is simplified, and the receiving coefficients of each serving cell are limited. Set to equal.
- the base station needs to obtain the average optimization target AMSE because the channel matrix after introducing the integrated error model contains the part that obeys the probability distribution.
- the accurate value cannot be obtained in real time during the scheduling process.
- the design can only be based on the distribution characteristics.
- the compliance probability part includes the estimation error.
- the average optimization target is expressed as
- the indicator indicating the communication quality includes the average mean square error AMSE.
- a summation operation is performed on the indicators indicating the communication quality of all the serving cells, and the downlink beam optimization model is obtained by minimizing the sum of the indicators indicating the communication quality and the transmission power of each base station being less than the predetermined power.
- the optimization problem of the whole system is constructed by the above modeling and analysis, specifically, the sum of the average MSEs (AMSEs) of all UEs is minimized under the condition that the transmission power of each base station is limited by the upper limit, and the downlink beam optimization model can be Expressed as:
- Step 102 according to the established downlink beam optimization model, obtain an optimal beamforming matrix optimization target.
- the step specifically includes: First, according to the downlink beam optimization model, the Lagrangian function of the downlink beam optimization model is obtained.
- the Lagrangian function of the downlink beam optimization model is subjected to a partial derivative operation with an index indicating the communication quality and a receiving coefficient as optimization variables, and the Lagrang function of the downlink beam optimization model is optimized.
- Condition ⁇ the Lagrang function of the downlink beam optimization model is optimized.
- the following example illustrates this step, using the optimization condition ⁇ algorithm to calculate the downlink beam optimization model.
- the downlink beam optimization model is solved according to the ⁇ condition in the optimization basic theory, and the specific steps are as follows: (1) Write a Lagrangian function of the problem;
- the optimization problem in the embodiment of the present invention is to use the beamforming matrix of each base station ⁇ ⁇ i and the receiving coefficient? as the optimization variables , obtaining a total of Q+1 KKT conditional equations;
- This step includes
- the steps to obtain the non-ideal parameters include:
- the base station measures a non-ideal parameter corresponding to the channel matrix of the serving cell according to the obtained channel matrix of the serving cell, and obtains a non-ideal parameter corresponding to the channel matrix of the monthly service cell.
- the second training sequence signal sent by all the terminal equipments in the cell except the serving cell in the cooperation cluster is received, and the cooperative cluster is acquired according to the received second training sequence signal.
- the channel matrix of the cell outside the serving cell and its corresponding non-ideal parameters. The steps to obtain the non-ideal parameters include:
- the cell cooperation mechanism is specifically: the base station in the cooperative cluster that has been set up shares the UE information, including the cell ID of the UE, the cell group ID, and the like.
- the new UE updates the shared UE information in real time when performing cell search to join a certain cell.
- the Q base stations in the cooperative cluster are synchronized with all UEs in the cluster.
- the base station estimates the downlink channel according to the Sounding signal sent by the UE uplink. When the UE sends the Sounding signal on the uplink, it not only sends it to its own serving base station, but also sends this signal to the interfering base station in the neighboring area.
- a base station After the uplink sounding of all UEs in the cluster ends, a base station obtains CSI information of all the UEs in the cluster that interfere with it.
- the information that the base station needs to exchange is only the cell search and synchronization information of the UE. Specifically include:
- the base station shares the cell search and synchronization information of all UEs. Establishing a temporary synchronization between the base station and all UEs (including the UE in the local cell and other UEs in the cooperative cluster), so that the UE can send a Sounding signal to the interfering base station;
- UE transmits all the base stations (including the serving base station and the interfering base station) signals an uplink Sounding; base station estimates the channel 3 ⁇ 4 cell interference and inter-cell channel measurement corresponding to their non-ideal parameters k ⁇ , Kt according to the uplink Sounding received signal.
- Step 104 Acquire a beamforming matrix of the base station according to the obtained channel matrix and its corresponding non-ideal parameters, and the optimal beamforming matrix optimization target, and the base station sends downlink data according to the obtained beamforming matrix.
- the obtained channel matrices and their corresponding non-ideal parameters are substituted into the optimal beamforming matrix optimization target, that is, the beamforming matrix of the base station can be obtained, so that the base station can send the downlink data according to the obtained beamforming matrix.
- Information embodiment of the present invention is to construct the optimal solution requires include three types: internal (1) cell CSI information, including CSI estimation value 3 ⁇ 4 its associated channel error parameter to BS-q according to the present cell and the UE, UE cell number , base station transmit power and transmit antenna correlation matrix R 1 ; (2) cross-cell CSI information, including BS-q to other cell UE interference channel CSI estimation value ⁇ 0 J and
- the embodiment of the present invention designs the supporting TDD by utilizing the channel reciprocity of the TDD system.
- the system cell cooperation mechanism enables the base station to obtain cross-cell CSI information with as little information interaction as possible.
- the beamforming matrix of each base station is in the closed form as shown below:
- the uplink sounding signal sent by the UE and other cell UEs to the base station of the cell can be estimated, and the CSI is not required to be exchanged between the base stations; therefore, the signaling interaction can be limited. Get the information you need to implement a distributed solution.
- the integrated channel model of the TDD system is used, and the algorithm not only uses the channel estimation value for optimization, but also considers the CSI estimation error R ⁇ E and the delay error and N m3 ⁇ 4 for the target.
- the averaging is optimized for robustness.
- the optimal beamforming matrix optimization target is obtained by solving the downlink beam optimization model, that is, the optimal beamforming matrix optimization target is the optimal solution of the downlink beam optimization model, and the beamforming matrix of each base station is A matrix of all that is obtained by substituting the non-ideal parameters corresponding to each base station into the optimal beamforming matrix optimization target.
- the second is a schematic structural diagram of a base station according to an embodiment of the present invention, including: a first establishing module 201, an operation module 202, an obtaining module 203, and a processing module 204.
- the first establishing module 201 is configured to: establish a downlink beam optimization model according to the integrated channel error model and the downlink model.
- the first establishing module is specifically configured to:
- the integrated channel error model is aggregated into a matrix form to obtain a comprehensive channel error model in a matrix form; Obtaining an indicator indicating the communication quality of the serving cell according to the obtained integrated channel error model and the downlink model in the matrix form; performing summation calculation on the indicators indicating communication quality of all serving cells to minimize the indicator indicating the communication quality And the transmission power of each base station is less than the predetermined power as a limiting condition, and a downlink beam optimization model is obtained.
- the indicator indicating the communication quality includes an average mean square error AMSE.
- the operation module 202 is configured to: obtain an optimal beamforming matrix optimization target according to the established downlink beam optimization model.
- the operation module is specifically configured to: obtain a Lagrangian function of the downlink beam optimization model according to a downlink beam optimization model; and use an indicator indicating a communication quality and a reception coefficient as an optimization variable to pull the downlink beam optimization model
- the Grande function performs a partial derivative operation to obtain an optimization condition KTT of the Lagrangian function of the downlink beam optimization model;
- the obtaining module 203 is configured to: respectively obtain a channel matrix of each cell other than the serving cell and a non-ideal parameter corresponding to each channel matrix in the cooperative cluster where the base station is located and the base station where the base station is located.
- the obtaining module 203 includes a first obtaining module 2031 and a second obtaining module 2032, wherein
- the first obtaining module 2031 is configured to: receive a first training sequence signal sent by all terminal equipment UEs in the serving cell, acquire a channel matrix of the serving cell according to the received first training sequence signal, and corresponding non-ideal a parameter, wherein the first obtaining module 2031 is configured to: obtain, according to the received first training sequence signal, a channel matrix of the serving cell by using an existing channel estimation algorithm;
- the base station measures a non-ideal parameter corresponding to the channel matrix of the serving cell according to the obtained channel matrix of the serving cell, and obtains a non-ideal parameter corresponding to the channel matrix of the monthly service cell.
- the second obtaining module 2032 is configured to: receive a second training sequence signal sent by all the terminal devices UE in the cell other than the serving cell in the cooperation cluster, and acquire the collaboration according to the received second training sequence signal Channel matrix of a cell other than the serving cell in the cluster and its correspondence Non-ideal parameters.
- the second obtaining module 2032 is configured to:
- the processing module 204 is configured to: obtain a beamforming matrix of the base station according to the obtained channel matrix and its corresponding non-ideal parameters, and the optimal beamforming matrix optimization target, and the base station according to the obtained beamforming matrix Send downlink data.
- the base station further includes: the second establishing module 205 is configured to: before the establishing the downlink beam optimization model, the base station establishes the integrated channel error model by using channel reciprocity of the time division multiplexing TDD system.
- the second establishing module 205 is configured to:
- the channel reciprocity of the TDD system means that the uplink and downlink air interface radio propagation channels are the same.
- the second establishing module 205 is further configured to: obtain an indicator indicating the communication quality of the serving cell by limiting the receiving coefficients of the respective serving cells according to the obtained integrated channel error model and the downlink model in the form of a matrix.
- This embodiment considers a cooperative cluster having three cells. Each cell includes a 2-antenna base station and two single-antenna UEs, and one UE is also interfered by the other two cell base stations while receiving the base station service of the own cell.
- the channel in the disclosed embodiment is a randomly generated Gaussian channel having a path loss coefficient of two.
- FIG. 3 is a schematic diagram of an application scenario according to an embodiment of the present invention, including: a base station 1, a base station 2, a base station 3, and users (UE-11, UE-12) served by the base station 1 and users served by the base station 2 (UE- 21. UE-22) and users under the service of base station 3 (UE-31, UE-32).
- FIG. 4 is a schematic diagram of a cell cooperation mechanism in an application scenario diagram according to an embodiment of the present invention.
- FIG. 4 is a user (UE) served by a base station (base station 1) and its neighboring base station (base station 2) in the scenario shown in FIG. -21)
- base station 1 a base station served by a base station (base station 1) and its neighboring base station (base station 2) in the scenario shown in FIG. -21)
- base station 1 base station 1
- base station 2 base station 2
- base station 2 base station 2
- the cooperation mechanism provided by the embodiment of the present invention is specifically described as a base station establishes temporary synchronization with a UE that interferes with itself in a cluster before transmitting data in the downlink, receives an uplink sounding signal across the cell, and designs beamforming by using the estimated cross-cell interference channel state. matrix.
- UE-11 and UE-12 there are two users, UE-11 and UE-12, in cell 1, and two users, UE-21 and UE-22, in cell 2.
- UE-1 and UE-2 are their serving users, and UE-21 is their interfering user.
- the base stations 1 and 2 first share the cell search and synchronization information of the UE-11, the UE-1, the UE-21, and the UE-22 to determine the cell and location information of the user, to ensure that the interfering user UE-21 of the base station 1 can
- a temporary synchronization is established with the base station 1 and an uplink Sounding signal is transmitted.
- the base station 1 estimates the cross-cell interference channel CSI and the non-ideal parameters.
- This embodiment requires a base station to cooperate with the serving base station of each of its interfering UEs.
- This example only demonstrates the process in which a base station cooperates with an interfering user under a neighboring base station.
- each base station establishes synchronization with its own service users due to the cell search and random access mechanism in the 3GPP standard.
- Each base station first receives the Sounding signal of its serving user, base station 1 receives the Sounding signal of UE-11 and UE-12, base station 2 receives the Sounding signal of UE-21 and UE-22, and base station 3 receives UE-31 and UE-32. Sounding signal.
- the base station shares cell search and synchronization information of all UEs. Through the cell search and random access mechanism, temporary synchronization between the base station and its interfering UE is established, so that the interfering UE can send a Sounding signal to its base station. 4. According to the cooperation mechanism, each UE sends an uplink sounding signal to its own interfering base station.
- the base station 1 receives the Sounding signals of the UE-21, the UE-22, the UE-31, and the UE-32; the base station 2 receives the Sounding signals of the UE-11, UE-12, UE-31, and UE-32; 3 Receive Sounding signals of UE-11, UE-12, UE-21, and UE-22.
- All base stations receive the out-of-cell uplink sounding signal according to the Sounding mode defined in step 4, and use the existing channel estimation area interference channel according to the received cross-cell uplink sounding signal.
- ⁇ ⁇ 3 measuring their corresponding non-ideal parameters
- the base station has obtained the information required by the optimal solution, and starts to transmit data according to the calculation of the optimal beamforming matrix ⁇ ⁇ 3 .
- FIG. 5 is a comparison diagram of throughput curves of a technical solution and other beamforming schemes according to an embodiment of the present invention.
- FIG. 5 is a comparison diagram of throughput curves of the present invention and the system capacity upper limit and the non-cooperative scheme, respectively, when the system signal to noise ratio (SNR) is increased from 0 dB to 30 dB.
- SNR system signal to noise ratio
- each cell uses different receiving coefficients to transmit the beamforming matrix u 3 ⁇ 4 and p q to iterate to solve the optimal solution, and the optimal solution is the upper limit of the system capacity under ideal conditions.
- the non-cooperative scheme uses the idea of game theory. Each cell UE treats the received interference as noise, and optimizes only the AMSE of the local cell. The solution only needs to use the intra-cell channel, and there is no information interaction between the cells. Since the base stations in the non-cooperative scheme are mutually competitive, the overall throughput cannot obtain an effective gain from the SNR. As can be seen from FIG. 5, the system throughput achieved by the disclosed embodiments is significantly higher than the comparative game-based non-cooperative solution, and can fully approximate the system's capacity ceiling.
- FIG. 6 is a comparison diagram of an AMSE curve achieved by a technical solution of an embodiment of the present invention and a non-robust conventional MMSE beamforming scheme.
- the ordinate represents the summed AMSE
- the abscissa represents the signal to noise ratio
- the solid line represents the scheme of the embodiment of the present invention
- the broken line represents the non-robust MMSE beam scheme.
- Figure 6 is a comparison of AMSE curves using the embodiment of the present invention and the non-robust MMSE beam scheme, respectively, when the system SNR is increased from OdB to 30 dB.
- the comparative MMSE beam scheme is based on the premise that CSI estimation error and delay error are not present, and the optimal solution obtained by solving MSE and minimizing problems is solved.
- the MMSE beam is equivalent to the non-ideal parameter of the minimum AMSI robust beam in the embodiment of the present invention, which is a degenerate form of the embodiment of the present invention and is not robust.
- the non-robust MMSE beam scheme does not effectively derive the gain from the SNR.
- the AMSE of the system increases at the higher SNR as the SNR increases.
- the disclosed embodiment can effectively solve this problem, and can effectively obtain gain when the SNR is increased, effectively suppress CSI error, and improve throughput.
- the embodiment of the invention considers that the estimation error and the delay error exist in the TDD system at the same time, and establishes a comprehensive channel error model for the channel of the TDD system, and achieves robustness.
- the embodiment of the present invention achieves the solution of the beamforming matrix of each base station in a closed form with the same scheme of the same receiving coefficient of each serving cell and the sum of the indicators indicating the communication quality is minimized.
- the embodiment of the present invention designs a cooperation mechanism between base stations, and realizes that the scheme has a natural distributed property under the TDD system. Therefore, the present invention has strong industrial applicability.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
一种实现波束成型的方法及基站,其中该实现波束成型的方法包括:根据综合信道误差模型和下行链路模型建立下行波束优化模型;根据建立的下行波束优化模型,获取最优波束成型矩阵优化目标;分别获取基站所在服务小区和基站所在协作簇内除服务小区之外的各小区的信道矩阵及各信道矩阵分别对应的非理想参数;根据获得的各信道矩阵及其各自对应的非理想参数,和所述最优波束成型矩阵优化目标,获取所述基站的波束成型矩阵,所述基站根据获得的波束成型矩阵发送下行数据。该波束成型的方法和基站有效抑制了TDD系统信道误差,消除了复杂迭代计算步骤,能够有效的抑制小区间的干扰。
Description
一种实现波束成型的方法及基站
技术领域
本发明涉及多小区多入单出 ( MISO, Multi-Input Single-Output )无线通 信技术, 尤指一种基于时分复用(TDD )系统的实现波束成型的方法及基站。
背景技术
以长期演进( LTE )为代表的第三代移动通信技术( 3G )的演进型系统, 实现了移动通信在 3G之后的一次阶段性变革。 为了进一步满足国际电信联 盟的高级国际通信技术( ITU IMT— Advanced ) 的要求, 同时也作为 LTE技 术的演进, 3GPP通过了 LTE的增强技术(LTE-A, LTE-Advanced )作为第 四代移动通信技术( 4G )标准的一个提案。 多小区 MIMO是 LTE-A关键技 术之一。 同时, 为了满足系统高速率及高频谱利用率的要求, LTE-A下行釆 用正交频分多址(OFDMA )的接入方式, 这样在两个小区的结合边缘处, 信 道频率可能相同, 因而在小区边缘处的用户对邻小区将会产生很强的干扰。 这样导致的结果就是: 为了保证用户的通信量, 需要大幅提升整个小区发送 功率; 而如果保证较低的发送功率, 小区的吞吐量又会大幅下降。 因此小区 间的干扰抑制是 LTE-A系统研究的关键课题之一。
当前多小区干扰抑制的一个研究热点就是调整天线的分配波束成型, 它 的设计目的在于通过调整天线的分配, 即确定每个用户在每个子载波上传输 时应该怎样配置发射天线, 以及基站应该怎样配置接收天线, 来使发送波束 在最大限度上与接收天线相正交。 波束成型技术的目的就是通过多个基站合 作来降低小区间的干扰。
对于多天线系统而言, 传统方案把所有信息汇总到一个中央处理器 ( CPU )来计算整个系统的最优化问题, 但是, 由于 X2接口的带宽和中央处 理器的计算能力受限, 这种方案比较难实现。 所以需要釆用信息交互开销和 计算开销比较少的分布式方案, 每个基站通过交互得到其它小区信息, 再根 据本地的信息, 计算本小区的最优化问题, 以迭代的方式得到系统的解。 所 以分布式框架具有较高可行性。 但是, 分布式框架下算法迭代的收敛性与收
敛速度也是需要考虑的问题, 往往需要迭代过程的算法在时间上有明显的缺 陷, 因此, 分布式方案在实际情况下不具备应用价值。
此外, 传统的波束成型算法大多假设系统进行理想的信道状态信息
( CSI, Channel State Information )估计, 即估计到的 CSI与实际 CSI相比不 存在任何误差。 而在实际的时分复用 (TDD ) 系统中是存在估计误差和延时 误差的, 这样, 造成了传统的波束成型算法在实际系统中会有严重的性能下 降。
为了提高传统的波束成型算法在实习系统中的性能, 相关技术中考虑 CSI误差对系统的影响, B. Dai, W. Xu, and C. Zhao的研究成果 "带有延迟的 码本 CSI 反馈的多小区通信系统中的最优 MMSE 下行波束成型算法" ( " Optimal MMSE beamforming for multiuserdownlink with delayed CSI feedback using codebooks" in Proc. IEEEGlobecom, Houston, USA, Dec. 2011 ) 中, 考虑了延时误差和估计误差同时存在的信道, 对非理想信道进行了建 模。 基于非理想信道模型, 文献设计了单小区下的鲁棒性波束成型问题, 并 且得到了鲁棒性问题的最优解。 通过仿真, 文献证明了提出的算法与传统波 束成型算法相比, 能够有效的克服延时误差和估计误差这两种误差带来的性 能损失。 但是文章考虑的模型仅局限于单小区范围内, 不具有实用价值。
为了提高传统的波束成型算法在实习系统中的性能, 以及为了有更好的 实用价值, 该领域的研究者 Bogale, T.E. , Vandendorpe, L.和 Chalise, B.K.的 研究成果 "Robust Transceiver Optimization for Downlink Coordinated Base Station Systems: Distributed Algorithm (下行合作的基站系统中的鲁棒性发送波 束: 分布式算法 ),, in IEEE Trans, on Sig. Process. Vol. 60, No. 1, January 2012 , 考虑了多小区系统中信道估计误差和延时误差对系统的影响, 并以平 均均方误差 (AMSE ) 为优化目标进行计算。 计算中釆用了发送和接收波束 交替优化的迭代形式的计算方式, 最终得到最优解。 但是, 因为算法需要大 量迭代步骤, 工程实现性不高。
另外, 工业界 3GPP标准规定了 TDD系统的上下行链路釆用同一个频段 不同的时隙进行通信。 所以基站可以根据用户上行发送的训练序列, 也叫做 Sounding信号来估计信道。 所以, 可以以 TDD系统的特有机制作为出发点,
设计分布式的合作机制。
基于以上系统需求和已有的学术成果与标准的分析, 需要设计一种既能 够有效抑制 TDD 系统信道误差, 又不需要复杂迭代步骤的计算, 同时还能 够适应 TDD系统的通信特点的波束成型算法。
发明内容
本发明实施例要解决的技术问题是提供一种实现波束成型的方法及基 站, 以有效的抑制 TDD 系统信道误差, 消除复杂迭代计算步骤, 从而有效 的抑制小区间的干扰。
为解决上述技术问题, 釆用如下技术方案:
一种实现波束成型的方法, 包括:
根据综合信道误差模型和下行链路模型建立下行波束优化模型; 根据建立的所述下行波束优化模型, 获取最优波束成型矩阵优化目标; 分别获取基站所在服务小区和基站所在协作簇内除服务小区之外的各小 区的信道矩阵及各信道矩阵分别对应的非理想参数;
根据获得的各信道矩阵及各信道矩阵对应的非理想参数, 和所述最优波 束成型矩阵优化目标, 获取所述基站的波束成型矩阵, 所述基站根据获得的 所述波束成型矩阵发送下行数据。
可选地, 所述建立下行波束优化模型之前, 该方法还包括: 所述基站利 用时分复用 TDD系统的信道互易性建立所述综合信道误差模型。
可选地,所述基站利用时分复用 TDD系统的信道互易性建立所述综合信 道误差模型的步骤包括:
根据所述基站的信道状态信息 CSI的实际值和估计值, 获得估计误差模 型;
4艮据所述基站的信道状态信息 CSI在估计时刻的估计值和传输时刻的实 际值, 获得延时误差模型;
将所述延时误差模型代入所述估计误差模型, 得到所述综合信道误差模
型。
可选地, 所述建立下行波束优化模型的步骤包括:
将所述综合信道误差模型聚合成矩阵形式, 得到矩阵形式的综合信道误 差模型;
根据得到的矩阵形式的综合信道误差模型和所述下行链路模型, 获取服 务小区的表示通信质量的指标;
对所有服务小区的表示通信质量的指标做求和运算, 将最小化表示通信 质量的指标之和作为目标, 以及将各个基站的发射功率小于预定功率作为限 制条件, 得到所述下行波束优化模型。
可选地, 所述获取服务小区的表示通信质量的指标的步骤包括: 根据得到的矩阵形式的综合信道误差模型和所述下行链路模型, 通过限 制各个服务小区的接收系数相同, 得到服务小区的表示通信质量的指标。
可选地, 所述分别获取基站所在服务小区和基站所在协作簇内除所述服 务小区之外的各小区的信道矩阵及各信道矩阵分别对应的非理想参数的步骤 包括:
接收所述服务小区内的所有终端设备 UE发送的第一训练序列信号, 根 据接收到的第一训练序列信号获取所述服务小区的信道矩阵及其对应的非理 想参数;
接收所述协作簇内除所述服务小区之外的小区内的所有终端设备 UE发 送的第二训练序列信号, 根据接收到的第二训练序列信号获取所述协作簇内 除所述服务小区之外的小区的信道矩阵及其对应的非理想参数。
可选地, 根据接收到的第一训练序列信号获取所述服务小区的信道矩阵 及其对应的非理想参数的步骤包括:
根据接收到的第一训练序列信号, 使用已有的信道估计算法获得所述服 务小区的信道矩阵;
根据获取的所述服务小区的信道矩阵, 所述基站测量所述服务小区的信 道矩阵对应的非理想参数,得到所述月良务小区的信道矩阵对应的非理想参数。
可选地, 根据接收到的第二训练序列信号获取所述协作簇内除所述服务 小区之外的小区的信道矩阵及其对应的非理想参数的步骤包括:
根据接收到的第二训练序列信号 , 使用已有的信道估计算法获得所述协 作簇内除所述服务小区之外的小区的信道矩阵;
根据获取的所述协作簇内除所述服务小区之外的小区的信道矩阵, 所述 基站测量所述协作簇内除所述服务小区之外的小区的信道矩阵对应的非理想 参数, 得到所述协作簇内除所述服务小区之外的小区的信道矩阵对应的非理 想参数。
可选地, 所述根据建立的下行波束优化模型, 获取最优波束成型矩阵优 化目标的步骤包括:
根据下行波束优化模型, 得到所述下行波束优化模型的拉格朗日函数; 以表示通信质量的指标和接收系数为优化变量, 对所述下行波束优化模 型的拉格朗日函数进行求偏导数运算, 得到所述下行波束优化模型的拉格朗 曰函数的最优化条件 ΚΤΤ;
根据得到的最优化条件 ΚΤΤ和发射功率最大作为约束条件建立联立方程 组, 求解方程组获取所述最优波束成型矩阵优化目标。
可选地, 所述表示通信质量的指标包括平均均方误差 AMSE。
一种基站, 包括第一建立模块、 运算模块、 获得模块和处理模块, 其中: 所述第一建立模块设置成: 根据综合信道误差模型和下行链路模型建立 下行波束优化模型;
所述运算模块设置成: 根据建立的所述下行波束优化模型, 获取最优波 束成型矩阵优化目标;
所述获得模块设置成: 分别获取基站所在服务小区和基站所在协作簇内 除服务小区之外的各小区的信道矩阵及各信道矩阵分别对应的非理想参数; 所述处理模块设置成: 根据获得的各信道矩阵及各信道矩阵对应的非理 想参数, 和所述最优波束成型矩阵优化目标, 获取所述基站的波束成型矩阵,
所述基站根据获得的所述波束成型矩阵发送下行数据。
可选地, 所述基站还包括第二建立模块, 其中:
所述第二建立模块设置成: 在所述建立下行波束优化模型之前, 利用时 分复用 TDD系统的信道互易性建立所述综合信道误差模型。
可选地,所述第二建立模块设置成按照如下方式利用时分复用 TDD系统 的信道互易性建立所述综合信道误差模型:
根据所述基站的信道状态信息 CSI的实际值和估计值, 获得估计误差模 型; 4艮据所述基站的信道状态信息 CSI在估计时刻的估计值和传输时刻的实 际值, 获得延时误差模型; 将所述延时误差模型代入所述估计误差模型, 得 到所述综合信道误差模型。
可选地, 所述第一建立模块设置成按照如下方式建立所述下行波束优化 模型:
将所述综合信道误差模型聚合成矩阵形式, 得到矩阵形式的综合信道误 差模型;
根据得到的矩阵形式的综合信道误差模型和所述下行链路模型, 获取服 务小区的表示通信质量的指标;
对所有服务小区的表示通信质量的指标做求和运算, 以最小化表示通信 质量的指标之和为目标, 以及以各个基站的发射功率小于预定功率作为限制 条件, 得到所述下行波束优化模型。
可选地, 所述第一建立模块设置成按照如下方式获取服务小区的表示通 信质量的指标:
根据得到的矩阵形式的综合信道误差模型和所述下行链路模型, 通过限 制各个服务小区的接收系数相同, 得到服务小区的表示通信质量的指标。
可选地, 所述获得模块包括第一获得模块和第二获得模块, 其中 第一获得模块设置成: 接收所述服务小区内的所有终端设备 UE发送的 第一训练序列信号, 根据接收到的第一训练序列信号获取所述服务小区的信 道矩阵及其对应的非理想参数;
第二获得模块设置成: 接收所述协作簇内除所述服务小区之外的小区内 的所有终端设备 UE发送的第二训练序列信号, 根据接收到的第二训练序列 信号获取所述协作簇内除所述服务小区之外的小区的信道矩阵及其对应的非 理想参数。
可选地, 所述第一获得模块设置成按照如下方式根据接收到的第一训练 序列信号获取所述服务小区的信道矩阵及其对应的非理想参数:
根据接收到的第一训练序列信号, 使用已有的信道估计算法获得所述服 务小区的信道矩阵;
根据获取的所述服务小区的信道矩阵, 所述基站测量所述服务小区的信 道矩阵对应的非理想参数,得到所述月良务小区的信道矩阵对应的非理想参数。
可选地, 所述第二获得模块设置成按照如下方式根据接收到的第二训练 序列信号获取所述协作簇内除所述服务小区之外的小区的信道矩阵及其对应 的非理想参数:
根据接收到的第二训练序列信号 , 使用已有的信道估计算法获得所述协 作簇内除所述服务小区之外的小区的信道矩阵;
根据获取的所述协作簇内除所述服务小区之外的小区的信道矩阵, 所述 基站测量所述协作簇内除所述服务小区之外的小区的信道矩阵对应的非理想 参数, 得到所述协作簇内除所述服务小区之外的小区的信道矩阵对应的非理 想参数。
可选地, 所述运算模块是设置成按照如下方式根据建立的下行波束优化 模型, 获取最优波束成型矩阵优化目标:
根据下行波束优化模型, 得到所述下行波束优化模型的拉格朗日函数; 以表示通信质量的指标和接收系数为优化变量, 对所述下行波束优化模 型的拉格朗日函数进行求偏导数运算, 得到所述下行波束优化模型的拉格朗 日函数的最优化条件 KTT;
根据得到的最优化条件 KTT和发射功率最大作为约束条件建立联立方程 组, 求解方程组获取所述最优波束成型矩阵优化目标。
可选地, 所述表示通信质量的指标包括平均均方误差 AMSE。
本申请技术方案包括: 根据综合信道误差模型和下行链路模型建立下行 波束优化模型; 根据建立的下行波束优化模型, 获取最优波束成型矩阵优化 目标; 分别获取基站所在服务小区和基站所在协作簇内除服务小区之外的各 小区的信道矩阵及各信道矩阵分别对应的非理想参数; 根据获得的各信道矩 阵及其各自对应的非理想参数, 和所述最优波束成型矩阵优化目标, 获取所 述基站的波束成型矩阵, 所述基站根据获得的波束成型矩阵发送下行数据。 根据所述问题的最优解的形式, 可知, 每个基站的波束成型矩阵的解都为闭 式形式, 实现只需经过一次计算即可得到, 计算复杂度低, 且根据上述算法 过程, 每次调度基站只须计算一次闭式解, 没有迭代的过程, 工程易实现。
本发明实施例考虑了 TDD 系统同时存在估计误差和延时误差, 并对 TDD系统的信道建立了综合信道误差模型, 实现了鲁棒性。 另外, 本发明实施例釆用各个服务小区的接收系数相同的方案, 以及以 最小化表示通信质量的指标之和为目标, 实现了每个基站的波束成型矩阵的 解都为闭式形式。
最后, 本发明实施例设计了基站之间的协作机制, 实现了本方案在 TDD 系统下具有天然的分布式性质。
附图概述
此处所说明的附图用来提供对本发明的进一步理解, 构成本申请的一部 分, 本发明的示意性实施例及其说明用于解释本发明, 并不构成对本发明的 不当限定。 在附图中:
图 1为本发明实施例实现波束成型的方法的流程图;
图 2为本发明实施例一种基站的结构示意图;
图 3为本发明实施例的应用场景图;
图 4为本发明实施例的应用场景图内小区协作机制示意图;
图 5 为本发明实施例技术方案与其他波束成型方案的吞吐量曲线对比 图;
图 6为本发明实施例技术方案与非鲁棒性的传统 MMSE波束成型方案达 到的 AMSE曲线对比图。 本发明的较佳实施方式
下面结合附图及具体实施例对本发明进行详细说明。 图 1为本发明实施例实现波束成型的方法的流程图, 包括以下步骤: 步骤 101 , 根据综合信道误差模型和下行链路模型建立下行波束优化模 型。
进一步地, 在建立下行波束优化模型之前, 还包括: 按照如下方式所述 基站利用时分复用 TDD系统的信道互易性建立所述综合信道误差模型:
首先, 才艮据所述基站的信道状态信息 CSI的实际值和估计值, 获得估计 误差模型。
TDD 系统的信道互易性指的是上行和下行的空口无线传播信道是相同 的。 举例说明, 估计误差由 TDD系统的信道估计特征造成。 由于 TDD系统 信道具有互易性,系统通过 UE的上行 Sounding信号来估计 CSI。由于 Sounding 估计序列的非理想, 会产生加性的估计误差。 用 hm¾和 ίι∞¾分别表示基站 CSI 的实际值和估计值, 估计误差模型可表示为 h∞¾ = ii∞¾ + 2 e∞¾。 其中; 1 为 BS-m 的天线相关矩阵; eme为各个元素服从独立同分布的误差向量, E{em¾e ¾ } = IWr , ¾为功率谱密度。 表征 TDD系统中信道估计时由 于 Sounding信号有限带来的 CSI误差值。
其次, 4艮据所述基站的信道状态信息 CSI在估计时刻的估计值和传输时 刻的实际值, 获得延时误差模型。
举例说明, 延时误差由系统的 CSI估计时刻和下行传输时刻之间延时造 成, 当延时过大时, 下行传输所用的实际信道和估计到的 CSI会有较大偏差。
如果用 hm¾ ohM¾[U]分别表示第 k时刻 (传输数据的时刻)和第 k-d时刻
(CSI估计时刻) 的 CSI, 那么延时误差模型可表示为两个时刻 CSI的关系: hm¾W = hMJU] + 其中相关系数 = (2 #¾;;)表示两个时刻 CSI 的相关性, J。(*)为 0阶 1型 Bessel函数, /¾为 US- 的最大 Doppler频移, Ts 为一个符号的长度。 nm¾[W为延时误差加性部分, 各个元素服从 0均值, 功率 谱密度为 1- 的高斯独立同分布。 nm¾[A]表征 TDD系统中两个不同时刻 CSI 的相异性。
最后 , 将所述延时误差模型代入所述估计误差模型得到所述综合信道误 差模型。
举例说明 , 将延时误差模型代入估计误差模型得到综合信道误差模型 , 如公式 (1)所示: qiik] = pmqi ( qi[k-d] +
[k-d]) + nH mqi[k] (1) 公式 (1)表示的是第 (k - d)个时刻( CSI估计时刻) CSI的估计值和第 k个 时刻 (传输数据的时刻) CSI 的实际值之间的关系。 如果已知信道估计和下 行数据传输之间的时间间隔, 基站根据估计时刻估计到的 CSI预测下行发送 数据是真实的 CSI。
进一步地, 将所述综合信道误差模型聚合成矩阵形式, 得到矩阵形式的 综合信道误差模型。
举例说明, 将一个小区的信道聚合成矩阵, 矩阵形式的综合信道模型为:
进一步地, 根据得到的矩阵形式的综合信道误差模型和下行链路模型, 获取服务小区的表示通信质量的指标。
其 中 , 下行链路模型 为 已知的模型 , 可以表示为 : y¾ = HwU¾x¾ +∑Hm¾Umxm+z¾, 其中, y b ^,...^^为小区 q uE接收的 信号, _y¾表示小区 q 中编号为 i 的 UE (记作 UE- )接收到的信号; Hm¾ =[hm¾,hm¾,...,hm¾½f为 BS-m到小区 q的所有 UE的信道矩阵, hM¾为 Ντ维 列向量, 表示 (BS- m)到(US- 之间的 CSI; 11¾有 ^行 列, 为基站 q的波 束成型矩阵, 是本发明实施例的设计目标; x : f用来表示 BS-q 传输的信号, 为标量, 表示要传输给 UE- 的有用信号, 不失一般性, 假 设它具有单位能量, 即 £|x¾|2 = l; Zg=[z¾,z¾,...,z¾f为噪声向量, 是功率 为 σ2的力口性高斯白噪声 (AWGN, Additive White Gaussian Noise )
在公式 (2)中, ?为小区 q对应的接收系数, 这里为了能够使构建的最优 化问题得到闭式解, 降低系统开销, 对 MSE模型做了简化, 限制每个服务小 区的接收系数都设为相等。
基站需要得到平均优化目标 AMSE是由于引入综合误差模型后的信道矩 阵 包含服从概率分布的部分,调度过程中无法实时得到其准确值, 只能根 据其分布特性来进行设计 服从概率部分具体包括估计误差中的高斯误差矩 阵 Em¾以及延时误差中用来表征两个不同时刻 CSI相异性的部分 N 对于小 区 q 来 说 平 均 优 化 目 标 表 示 为
AMSE, = {廳3 } = ^+ ― 2 lRe [tr (Ψ )] +
其中
ΚΙ - )¼ + ¾« 1^]为信道 均方矩阵。 其中, 表示通信质量的指标包括平均均方误差 AMSE。 进一步地, 对所有服务小区的表示通信质量的指标做求和运算, 以最小 化表示通信质量的指标之和和各个基站的发射功率小于预定功率作为限制条 件, 得到下行波束优化模型。 举例说明, 整个系统的优化问题^^于以上建模分析构建的, 具体是在 保证各个基站发送功率受上限制约的条件下最小化所有 UE 的平均 MSE ( AMSE )之和, 下行波束优化模型可以表述为:
Q
mm Y AMSE。
^ 6 q 步骤 102, 根据建立的下行波束优化模型, 获取最优波束成型矩阵优化 目标。
本步骤具体包括: 首先, 根据下行波束优化模型, 得到所述下行波束优化模型的拉格朗日 函数。
其次, 以表示通信质量的指标和接收系数为优化变量, 对所述下行波束 优化模型的拉格朗日函数进行求偏导数运算, 得到所述下行波束优化模型的 拉格朗日函数的最优化条件 κττ。
最后,根据得到的最优化条件 ΚΤΤ和发射功率最大作为约束条件建立联 立方程组, 求解方程组获取所述最优波束成型矩阵优化目标。
下面举例说明本步骤,釆用最优化条件 ΚΤΤ算法对下行波束优化模型进 行运算。
所述的下行波束优化模型是根据最优化基础理论中 ΚΚΤ条件求解的,具 体步骤如下:
( 1 )写出问题的拉格朗日函数;
(U, ?)
( 2 )对关于问题的优化变量对拉格朗日函数求偏导数,得到其 KKT条件; 本发明实施例中的优化问题以各个基站的波束成型矩阵 {υί i以及接收系数 ?为优化变量, 得到共 Q+1个 KKT条件组成的方程组;
^^ = -2^ (Ψ„Η„Γ +2 qmVq + 2XqVq = 0, V^
— = Υ-2β-3σ2Κ +2β~2ίΓ Ψ H U ) -2厂?"3V V
与 Q+1个 KKT条件联立, 组成 Q+2个方程的方程组, 求解得到问题的最优 解。
所述下行波束优化模型的最优解的形式为:
Ψ H
可以看到最优解 为闭式形式, 并且满足最优化条件 tr (t;ffU;、 = Pq 。 本步骤中, 用到了最优化问题的数学知识, 下面对本步骤涉及到的部分 进行解释。 假设求解的最优化问题为:
min f(x)
si. g(x)≤0
最优化理论定义: 求解最优化问题有三个步骤:
1、 写出问题的拉格朗日函数, 拉格朗日函数在最优化理论中已有定义, 求解任何一个最优化问题都需要写出拉格朗日函数, 详见《最优化理论与算 法》 (第二版, 清华大学出版社)相关章节, 为: J(x, /l) = /(x) + (x) , 其中 为拉格朗日对偶变量。
2、 对拉格朗日函数求偏导数, 令其为零, 得到问题的 KKT条件(即最优
解需要满足的关系式) :
dL(x, ) _ Q dL(x, ) _ Q
dx ' 3
3、 根据以上 ΚΚΤ条件解方程得到最优解。
综上所述, 根据定义就可以直接写出问题的拉格朗日函数:
0 0
Ii U. ? i = AMSE + Y Λ., I tr ( I: X:' ": ) ~ P \ 步骤 103 , 分别获取基站所在服务小区和基站所在协作簇内除服务小区 之外的各小区的信道矩阵及各信道矩阵分别对应的非理想参数。
该步骤包括,
接收所述服务小区内的所有终端设备 UE发送的第一训练序列信号, 根 据接收到的第一训练序列信号获取所述服务小区的信道矩阵及其对应的非理 想参数。 得到该非理想参数的步骤包括:
根据接收到的第一训练序列信号 , 使用已有的信道估计算法获得所述服 务小区的信道矩阵;
根据获取的所述服务小区的信道矩阵, 所述基站测量所述服务小区的信 道矩阵对应的非理想参数,得到所述月良务小区的信道矩阵对应的非理想参数。
进一步地, 接收所述协作簇内除所述服务小区之外的小区内的所有终端 设备 UE发送的第二训练序列信号, 根据接收到的第二训练序列信号获取所 述协作簇内除所述服务小区之外的小区的信道矩阵及其对应的非理想参数。 得到该非理想参数的步骤包括:
根据接收到的第二训练序列信号 , 使用已有的信道估计算法获得所述协 作簇内除所述服务小区之外的小区的信道矩阵;
根据获取的所述协作簇内除所述服务小区之外的小区的信道矩阵, 所述 基站测量所述协作簇内除所述服务小区之外的小区的信道矩阵对应的非理想 参数, 得到所述协作簇内除所述服务小区之外的小区的信道矩阵对应的非理 想参数。
本步骤中涉及到了小区的协作机制, 小区协作机制具体为, 已经设定好 的协作簇内的基站共享其 UE信息, 包括 UE所属小区 ID, 小区组 ID等, 在
新 UE进行小区搜索加入某个小区时实时更新共享的 UE信息。协作簇内的 Q 个基站与簇内所有 UE同步。 在 TDD系统中, 由于上下行釆用相同的信道, 基站根据 UE 上行发送的 Sounding信号估计下行信道。 UE 在上行发送 Sounding信号时不仅要发给自己的服务基站, 这个信号同时也发给了邻区的 干扰基站。 在簇内所有 UE上行 Sounding结束后, 一个基站就获得了自己到 簇内所有被其干扰的 UE的 CSI信息, 这一过程中基站需要交互的信息仅为 UE的小区搜索和同步信息。 具体包括:
在已经组成的协作簇内, 基站共享所有 UE的小区搜索和同步信息。 建 立基站与所有 UE (包括本小区 UE与协作簇内其他小区 UE )之间的暂时同 步, 使 UE能够给其干扰基站发送 Sounding信号;
UE对所有基站 (包括其服务基站与干扰基站)发送上行 Sounding信号; 基站根据接受到的上行 Sounding信号估计小区内信道 ¾和跨小区干扰 信道 测量他们对应的非理想参数 k∞ , Kt 。
步骤 104 , 根据获得的各信道矩阵及其各自对应的非理想参数, 和所述 最优波束成型矩阵优化目标, 获取所述基站的波束成型矩阵, 所述基站根据 获得的波束成型矩阵发送下行数据。
举例说明, 将获得的各信道矩阵及其各自对应的非理想参数, 代入最优 波束成型矩阵优化目标, 即可以得到基站的波束成型矩阵, 这样, 基站就可 以根据得到的波束成型矩阵发送下行数据。 本发明实施例中, 构建最优解需要的信息包括 3类: (1 )小区内部 CSI 信息, 包括 BS-q到本小区 UE的 CSI估计值 ¾及其相关的信道误差参数 和 , 小区 UE数 , 基站发送功率 和发送天线相关矩阵 R1 ; ( 2 )跨小 区 CSI信息, 包括 BS-q到其他小区 UE的干扰信道 CSI估计值 {0 J 及其
L
相关的信道误差参数 { } , {σ2 \ ( 3 )
L qm J L 」 ; 公用信息, 高斯白噪声功率 。 其中小区内部信息和公用信息在小区内部为已知信息, 为了实现最优解的分 布式特性,本发明实施例利用 TDD系统的信道互易性,设计了其配套的 TDD
系统小区协作机制, 使基站以尽可能少的信息交互获得跨小区 CSI信息。 本发明实施例的最优解形式和相应的协作机制的优点在于:
( 1 )根据最优波束成型矩阵优化目标的形式, 每个基站的波束成型矩阵 都为如下所示的闭式形式:
在计算时只经过一次计算可以得到, 计算复杂度低。 且根据上述算法过 程, 每次调度基站只须计算一次闭式解, 没有迭代的过程, 工程易实现。
( 2 )根据上诉闭式解的形式, 其中只包含本小区 CSI, 包括 w aqq , 在小区 q内为已知信息; 还包括本小区基站到其他小区 UE的跨小区 CSI, 包 括 ή pq 2 mi aq 2 mi , 在 TDD系统中可以根据本小区 UE和其他小区 UE发送给本 小区基站的上行 Sounding信号就可以估计得到, 不需要基站之间交互 CSI; 所以能够以有限的信令交互得到需要的信息, 实现分布式求解。
( 3 )根据本发明实施例釆用的 TDD系统综合信道模型, 算法在计算时 不仅仅使用信道估计值进行优化, 而是考虑了 CSI估计误差 R^E 和延时误 差 和 Nm¾ , 对目标求平均值进行优化, 具有鲁棒性。 本发明实施例中, 最优波束成型矩阵优化目标是通过求解下行波束优化 模型得到, 也就是说最优波束成型矩阵优化目标是下行波束优化模型的最优 解, 而各个基站的波束成型矩阵是将各个基站所对应的非理想参数代入最优 波束成型矩阵优化目标得到的一个全部是常量的矩阵。 图 2为本发明实施例基站的结构示意图, 包括: 第一建立模块 201 , 运 算模块 202 , 获得模块 203 , 处理模块 204。 其中, 第一建立模块 201设置成: 根据综合信道误差模型和下行链路模型建立 下行波束优化模型。 其中, 第一建立模块具体用于:
将所述综合信道误差模型聚合成矩阵形式, 得到矩阵形式的综合信道误 差模型;
根据得到的矩阵形式的综合信道误差模型和下行链路模型, 得到服务小 区的表示通信质量的指标; 对所有服务小区的表示通信质量的指标做求和运算, 以最小化表示通信 质量的指标之和和各个基站的发射功率小于预定功率作为限制条件, 得到下 行波束优化模型。 本发明实施例中, 表示通信质量的指标包括平均均方误差 AMSE。
运算模块 202设置成: 根据建立的下行波束优化模型, 获取最优波束成 型矩阵优化目标。 其中, 运算模块具体用于: 根据下行波束优化模型, 得到所述下行波束优化模型的拉格朗日函数; 以表示通信质量的指标和接收系数为优化变量, 对所述下行波束优化模 型的拉格朗日函数进行求偏导数运算, 得到所述下行波束优化模型的拉格朗 曰函数的最优化条件 KTT;
根据得到的最优化条件 KTT和发射功率最大作为约束条件建立联立方程 组, 求解方程组获取所述最优波束成型矩阵优化目标。 获得模块 203设置成: 分别获取基站所在服务小区和基站所在协作簇内 除服务小区之外的各小区的信道矩阵及各信道矩阵分别对应的非理想参数。
进一步地, 所述获得模块 203 包括第一获得模块 2031 和第二获得模块 2032 , 其中
第一获得模块 2031设置成: 接收所述服务小区内的所有终端设备 UE发 送的第一训练序列信号, 根据接收到的第一训练序列信号获取所述服务小区 的信道矩阵及其对应的非理想参数;其中,所述第一获得模块 2031是设置成: 根据接收到的第一训练序列信号, 使用已有的信道估计算法获得所述服 务小区的信道矩阵;
根据获取的所述服务小区的信道矩阵, 所述基站测量所述服务小区的信 道矩阵对应的非理想参数,得到所述月良务小区的信道矩阵对应的非理想参数。
第二获得模块 2032设置成:接收所述协作簇内除所述服务小区之外的小 区内的所有终端设备 UE发送的第二训练序列信号, 根据接收到的第二训练 序列信号获取所述协作簇内除所述服务小区之外的小区的信道矩阵及其对应
的非理想参数。 其中, 第二获得模块 2032是设置成:
根据接收到的第二训练序列信号 , 使用已有的信道估计算法获得所述协 作簇内除所述服务小区之外的小区的信道矩阵;
根据获取的所述协作簇内除所述服务小区之外的小区的信道矩阵, 所述 基站测量所述协作簇内除所述服务小区之外的小区的信道矩阵对应的非理想 参数, 得到所述协作簇内除所述服务小区之外的小区的信道矩阵对应的非理 想参数。
处理模块 204设置成: 根据获得的各信道矩阵及其各自对应的非理想参 数, 和所述最优波束成型矩阵优化目标, 获取所述基站的波束成型矩阵, 所 述基站根据获得的波束成型矩阵发送下行数据。
所述基站还包括: 第二建立模块 205设置成: 在所述建立下行波束优化 模型之前,所述基站利用时分复用 TDD系统的信道互易性建立所述综合信道 误差模型。 其中, 所述第二建立模块 205是设置成:
根据所述基站的信道状态信息 CSI的实际值和估计值, 获得估计误差模 型;
才艮据所述基站的信道状态信息 CSI在估计时刻的估计值和传输时刻的实 际值, 获得延时误差模型;
将所述延时误差模型代入所述估计误差模型得到所述综合信道误差模 型。
其中, TDD系统的信道互易性指的是上行和下行的空口无线传播信道是 相同的。
进一步地, 第二建立模块 205还设置成: 根据得到的矩阵形式的综合信 道误差模型和下行链路模型, 通过限制各个服务小区的接收系数相同, 得到 服务小区的表示通信质量的指标。 下面结合具体实施例对本发明实施例实现波束成型的方法进行说明。 实施例一
本实施例考虑一个有三个小区构成的协作簇。 每个小区包含一个 2天线 的基站和两个单天线的 UE, —个 UE在接收本小区基站服务的同时也会受到 另两个小区基站的干扰。 公开实施例中的信道为随机生成的高斯信道, 其路 损系数为 2。 典型地, 公开实施例中的误差模型参数设为 = 0.1,/。; = 0.1 。
下面结合附图 3和附图 4详细说明本发明具体实施例。
图 3为本发明实施例的应用场景图, 包括: 基站 1 , 基站 2, 基站 3 , 以 及包括基站 1服务下的用户(UE-11、 UE-12 )、基站 2服务下的用户(UE-21、 UE-22 )和基站 3服务下的用户 (UE-31、 UE-32 ) 。
图 4为本发明实施例的应用场景图内小区协作机制示意图, 图 4以图 3 所示的场景中的一个基站(基站 1 )和其相邻的基站(基站 2 )服务下的用户 ( UE-21 )为示例, 解释了小区协作的具体实施方式。
本发明实施例提供的协作机制具体描述为一个基站在下行发送数据之前 与自己在簇内干扰到的 UE建立暂时同步, 接收跨小区上行 Sounding信号, 用估计到的跨小区干扰信道状态设计波束成型矩阵。
如图 4所示, 小区 1 内存在 UE-11和 UE-12两个用户, 小区 2内存在 UE-21和 UE-22两个用户。 对于基站 1来说, UE-1和 UE-2为其服务用户, UE-21为其干扰用户。 基站协作时, 基站 1和 2先共享 UE-11、 UE-1 , UE-21 和 UE-22的小区搜索和同步信息确定用户所属小区和位置信息, 以确保基站 1的干扰用户 UE-21能够和基站 1建立暂时的同步并发送上行 Sounding信号。 基站 1接收到 UE-21发送来的 Sounding后,估计出跨小区干扰信道 CSI和非 理想参数。
本实施例需要一个基站与每个其干扰到的 UE的服务基站进行协作, 本 示例仅演示了一个基站与一个相邻基站下的一个干扰用户进行协作的过程。
基于提出的协作机制, 下面给出本实施例的具体实施过程:
1、 在已经组成的协作簇(小区 1、 2、 3 ) 内, 由于 3GPP标准中小区搜 索与随机接入机制, 每个基站都与自己的服务用户建立了同步。 每个基站首 先接受其服务用户的 Sounding信号,基站 1接收 UE-11和 UE-12的 Sounding 信号, 基站 2接收 UE-21和 UE-22的 Sounding信号, 基站 3接收 UE-31和 UE-32的 Sounding信号。
2、基站根据步骤 1中定义的 Sounding模式,接收到小区内上行 Sounding 信号, 从而使用已有的信道估计算法估计小区内信道矩阵 {ήΜ^=ι, 测量他们 对应的非理想参数 { }3— ,
3、 参考本实施例所述的协作机制, 基站共享所有 UE的小区搜索和同步 信息。 通过小区搜索和随机接入机制, 建立基站与其干扰 UE之间的暂时同 步, 使干扰 UE能够给其基站发送 Sounding信号。
4、 根据协作机制中所述, 每个 UE给自己的干扰基站发送上行 Sounding 信号。 本实施例中, 基站 1接收 UE-21、 UE-22、 UE-31和 UE-32的 Sounding 信号; 基站 2接收 UE-11、 UE-12、 UE-31和 UE-32的 Sounding信号; 基站 3 接收 UE-11、 UE-12、 UE-21和 UE-22的 Sounding信号。
5、 所有基站根据步骤 4 中定义的 Sounding模式, 接收到小区外上行 Sounding信号, 根据接收到的跨小区上行 Sounding信号使用已有的信道估计 区干扰信道 {ή。∞}3 , 测量他们对应的非理想参数
6、 基站已获得所述最优解所需要的信息, 根据计算最优波束成型矩阵 {υ }3 , 开始下行发送数据。
本实施例并不能限制本发明的方案,协作簇内小区的数量是没有限制的, 根据实际情况而定, 之所以选择三个小区的协作簇是为了便于描述。 图 5 为本发明实施例技术方案与其他波束成型方案的吞吐量曲线对比 图。
如图 5所示, 纵坐标表示吞吐量, 横坐标表示信噪比, 实线代表本发明 实施例方案, 虚线代表系统容量上限, 三角虚线代表不协作方案。 经过仿真, 结果显示本实施例的最优解能够有效逼近图 1所示场景的系统吞吐量上限。 图 5是公开实施例在系统信噪比 ( Signal to Noise Ratio, SNR )从 OdB增加至 30dB 时分别釆用本发明方案和系统容量上限以及不协作方案的吞吐量曲线 对比图。 系统容量上限方案中每个小区釆用不同的接收系数 需要发送波 束成型矩阵 u¾与 pq轮流迭代求解最优解, 达到的最优解在理想条件下为系统 容量上限。 不协作方案釆用博弈论的思想, 每个小区 UE把受到的干扰当作 噪声处理, 只优化本小区部的 AMSE, 求解时仅需要使用小区内部信道, 各 个小区间无任何信息交互。 由于不协作方案中各个基站是相互竟争的关系, 其整体吞吐量无法从 SNR中获得有效增益。 从图 5可以看出, 公开实施例达 到的系统吞吐量要明显高于对比的基于博弈思想的非协作方案, 并且能够充 分逼近系统的容量上限。 图 6为本发明实施例技术方案与非鲁棒性的传统 MMSE波束成型方案达 到的 AMSE曲线对比图。
如图 6所示, 纵坐标表示求和 AMSE, 横坐标表示信噪比, 实线代表本 发明实施例方案, 虚线代表非鲁棒 MMSE波束方案。 在仿真结果中也可以观
察到本实施例对于信道估计误差和延时误差的抑制作用。 图 6是在系统 SNR 从 OdB增加至 30dB时分别釆用本发明实施例方案和非鲁棒 MMSE波束方案 的 AMSE曲线对比图。 对比的 MMSE波束方案以 CSI估计误差和延时误差 都不存在为前提, 求解 MSE和最小化的问题得到的最优解。 MMSE波束相 当于本发明实施例中最小化 AMSE鲁棒性波束的非理想参数 , 是本发明实 施例的一种退化形式, 不具有鲁棒性。 从图 6可以看出, 非鲁棒 MMSE波束 方案并不能有效从 SNR 中获得增益。 相反, 由于过大的发送功率能够放大 CSI误差带来的性能损失, 在较高 SNR下系统的 AMSE反而随着 SNR的增 大而增大。 而公开实施例方案能够有效解决这个问题, 在 SNR增大时能够有 效获得增益, 有效抑制 CSI误差, 提高吞吐量。
以上所述, 仅为本发明的较佳实例而已, 并非用于限定本发明的保护范 围。 凡在本发明的精神和原则之内, 所做的任何修改、 等同替换、 改进等, 均应包含在本发明的保护范围之内。
工业实用性
本发明实施例考虑了 TDD 系统同时存在估计误差和延时误差, 并对 TDD系统的信道建立了综合信道误差模型, 实现了鲁棒性。 另外, 本发明实 施例釆用各个服务小区的接收系数相同的方案, 以及以最小化表示通信质量 的指标之和为目标, 实现了每个基站的波束成型矩阵的解都为闭式形式。 最 后, 本发明实施例设计了基站之间的协作机制, 实现了本方案在 TDD 系统 下具有天然的分布式性质。 因此本发明具有很强的工业实用性。
Claims
1、 一种实现波束成型的方法, 包括:
根据综合信道误差模型和下行链路模型建立下行波束优化模型; 根据建立的所述下行波束优化模型, 获取最优波束成型矩阵优化目标; 分别获取基站所在服务小区和基站所在协作簇内除服务小区之外的各小 区的信道矩阵及各信道矩阵分别对应的非理想参数;
根据获得的各信道矩阵及各信道矩阵对应的非理想参数, 和所述最优波 束成型矩阵优化目标, 获取所述基站的波束成型矩阵, 所述基站根据获得的 所述波束成型矩阵发送下行数据。
2、 根据权利要求 1所述的实现波束成型的方法, 其中, 所述建立下行波 束优化模型之前, 该方法还包括: 所述基站利用时分复用 TDD系统的信道互 易性建立所述综合信道误差模型。
3、 根据权利要求 2所述的实现波束成型的方法, 其中, 所述基站利用时 分复用 TDD系统的信道互易性建立所述综合信道误差模型的步骤包括: 根据所述基站的信道状态信息 CSI的实际值和估计值, 获得估计误差模 型;
才艮据所述基站的信道状态信息 CSI在估计时刻的估计值和传输时刻的实 际值, 获得延时误差模型;
将所述延时误差模型代入所述估计误差模型 , 得到所述综合信道误差模 型。
4、 根据权利要求 1所述的实现波束成型的方法, 其中, 所述建立下行波 束优化模型的步骤包括:
将所述综合信道误差模型聚合成矩阵形式, 得到矩阵形式的综合信道误 差模型;
根据得到的矩阵形式的综合信道误差模型和所述下行链路模型, 获取服 务小区的表示通信质量的指标;
对所有服务小区的表示通信质量的指标做求和运算, 将最小化表示通信
质量的指标之和作为目标, 以及将各个基站的发射功率小于预定功率作为限 制条件 , 得到所述下行波束优化模型。
5、 根据权利要求 4所述的实现波束成型的方法, 其中, 所述获取服务小 区的表示通信质量的指标的步骤包括:
根据得到的矩阵形式的综合信道误差模型和所述下行链路模型, 通过限 制各个服务小区的接收系数相同, 得到服务小区的表示通信质量的指标。
6、 根据权利要求 1所述的实现波束成型的方法, 其中, 所述分别获取基 站所在服务小区和基站所在协作簇内除所述服务小区之外的各小区的信道矩 阵及各信道矩阵分别对应的非理想参数的步骤包括:
接收所述服务小区内的所有终端设备 UE发送的第一训练序列信号, 根 据接收到的第一训练序列信号获取所述服务小区的信道矩阵及其对应的非理 想参数;
接收所述协作簇内除所述服务小区之外的小区内的所有终端设备 UE发 送的第二训练序列信号 , 根据接收到的第二训练序列信号获取所述协作簇内 除所述服务小区之外的小区的信道矩阵及其对应的非理想参数。
7、 根据权利要求 6所述的实现波束成型的方法, 其中, 根据接收到的第 一训练序列信号获取所述服务小区的信道矩阵及其对应的非理想参数的步骤 包括:
根据接收到的第一训练序列信号, 使用已有的信道估计算法获得所述服 务小区的信道矩阵;
根据获取的所述服务小区的信道矩阵, 所述基站测量所述服务小区的信 道矩阵对应的非理想参数,得到所述月良务小区的信道矩阵对应的非理想参数。
8、 根据权利要求 6所述的实现波束成型的方法, 其中, 根据接收到的第 二训练序列信号获取所述协作簇内除所述服务小区之外的小区的信道矩阵及 其对应的非理想参数的步骤包括:
根据接收到的第二训练序列信号 , 使用已有的信道估计算法获得所述协 作簇内除所述服务小区之外的小区的信道矩阵;
根据获取的所述协作簇内除所述服务小区之外的小区的信道矩阵, 所述
基站测量所述协作簇内除所述服务小区之外的小区的信道矩阵对应的非理想 参数, 得到所述协作簇内除所述服务小区之外的小区的信道矩阵对应的非理 想参数。
9、 根据权利要求 1所述的实现波束成型的方法, 其中, 所述根据建立的 下行波束优化模型, 获取最优波束成型矩阵优化目标的步骤包括:
根据下行波束优化模型, 得到所述下行波束优化模型的拉格朗日函数; 以表示通信质量的指标和接收系数为优化变量, 对所述下行波束优化模 型的拉格朗日函数进行求偏导数运算, 得到所述下行波束优化模型的拉格朗 曰函数的最优化条件 KTT;
根据得到的最优化条件 KTT和发射功率最大作为约束条件建立联立方程 组, 求解方程组获取所述最优波束成型矩阵优化目标。
10、 根据权利要求 4所述的实现波束成型的方法, 其中, 所述表示通信 质量的指标包括平均均方误差 AMSE。
11、 一种基站, 包括第一建立模块、 运算模块、 获得模块和处理模块, 其中:
所述第一建立模块设置成: 根据综合信道误差模型和下行链路模型建立 下行波束优化模型;
所述运算模块设置成: 根据建立的所述下行波束优化模型, 获取最优波 束成型矩阵优化目标;
所述获得模块设置成: 分别获取基站所在服务小区和基站所在协作簇内 除服务小区之外的各小区的信道矩阵及各信道矩阵分别对应的非理想参数; 所述处理模块设置成: 根据获得的各信道矩阵及各信道矩阵对应的非理 想参数, 和所述最优波束成型矩阵优化目标, 获取所述基站的波束成型矩阵, 所述基站根据获得的所述波束成型矩阵发送下行数据。
12、根据权利要求 11所述的基站,所述基站还包括第二建立模块,其中: 所述第二建立模块设置成: 在所述建立下行波束优化模型之前, 利用时 分复用 TDD系统的信道互易性建立所述综合信道误差模型。
13、 根据权利要求 12所述的基站, 其中, 所述第二建立模块设置成按照 如下方式利用时分复用 TDD系统的信道互易性建立所述综合信道误差模型: 根据所述基站的信道状态信息 CSI的实际值和估计值, 获得估计误差模 型; 4艮据所述基站的信道状态信息 CSI在估计时刻的估计值和传输时刻的实 际值, 获得延时误差模型; 将所述延时误差模型代入所述估计误差模型, 得 到所述综合信道误差模型。
14、 根据权利要求 11所述的基站, 其中, 所述第一建立模块设置成按照 如下方式建立所述下行波束优化模型:
将所述综合信道误差模型聚合成矩阵形式, 得到矩阵形式的综合信道误 差模型;
根据得到的矩阵形式的综合信道误差模型和所述下行链路模型, 获取服 务小区的表示通信质量的指标;
对所有服务小区的表示通信质量的指标做求和运算, 以最小化表示通信 质量的指标之和为目标, 以及以各个基站的发射功率小于预定功率作为限制 条件, 得到所述下行波束优化模型。
15、 根据权利要求 14所述的基站, 其中, 所述第一建立模块设置成按照 如下方式获取服务小区的表示通信质量的指标:
根据得到的矩阵形式的综合信道误差模型和所述下行链路模型, 通过限 制各个服务小区的接收系数相同, 得到服务小区的表示通信质量的指标。
16、 根据权利要求 11所述的基站, 其中, 所述获得模块包括第一获得模 块和第二获得模块, 其中
第一获得模块设置成: 接收所述服务小区内的所有终端设备 UE发送的 第一训练序列信号, 根据接收到的第一训练序列信号获取所述服务小区的信 道矩阵及其对应的非理想参数;
第二获得模块设置成: 接收所述协作簇内除所述服务小区之外的小区内 的所有终端设备 UE发送的第二训练序列信号, 根据接收到的第二训练序列 信号获取所述协作簇内除所述服务小区之外的小区的信道矩阵及其对应的非 理想参数。
17、 根据权利要求 16所述的基站, 其中, 所述第一获得模块设置成按照 如下方式根据接收到的第一训练序列信号获取所述服务小区的信道矩阵及其 对应的非理想参数:
根据接收到的第一训练序列信号, 使用已有的信道估计算法获得所述服 务小区的信道矩阵;
根据获取的所述服务小区的信道矩阵, 所述基站测量所述服务小区的信 道矩阵对应的非理想参数,得到所述月良务小区的信道矩阵对应的非理想参数。
18、 根据权利要求 16所述的基站, 其中, 所述第二获得模块设置成按照 如下方式根据接收到的第二训练序列信号获取所述协作簇内除所述服务小区 之外的小区的信道矩阵及其对应的非理想参数:
根据接收到的第二训练序列信号 , 使用已有的信道估计算法获得所述协 作簇内除所述服务小区之外的小区的信道矩阵;
根据获取的所述协作簇内除所述服务小区之外的小区的信道矩阵, 所述 基站测量所述协作簇内除所述服务小区之外的小区的信道矩阵对应的非理想 参数, 得到所述协作簇内除所述服务小区之外的小区的信道矩阵对应的非理 想参数。
19、 根据权利要求 11所述的基站, 其中, 所述运算模块是设置成按照如 下方式根据建立的下行波束优化模型, 获取最优波束成型矩阵优化目标: 根据下行波束优化模型, 得到所述下行波束优化模型的拉格朗日函数; 以表示通信质量的指标和接收系数为优化变量, 对所述下行波束优化模 型的拉格朗日函数进行求偏导数运算, 得到所述下行波束优化模型的拉格朗 曰函数的最优化条件 KTT;
根据得到的最优化条件 KTT和发射功率最大作为约束条件建立联立方程 组, 求解方程组获取所述最优波束成型矩阵优化目标。
20、 根据权利要求 14所述的基站, 其中, 所述表示通信质量的指标包括 平均均方误差 AMSE。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP14883399.9A EP3096466A4 (en) | 2014-02-20 | 2014-06-10 | Method and base station for enabling beamforming |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410058819.3 | 2014-02-20 | ||
CN201410058819.3A CN104868947A (zh) | 2014-02-20 | 2014-02-20 | 一种实现波束成型的方法及基站 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2015123942A1 true WO2015123942A1 (zh) | 2015-08-27 |
Family
ID=53877588
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2014/079591 WO2015123942A1 (zh) | 2014-02-20 | 2014-06-10 | 一种实现波束成型的方法及基站 |
Country Status (3)
Country | Link |
---|---|
EP (1) | EP3096466A4 (zh) |
CN (1) | CN104868947A (zh) |
WO (1) | WO2015123942A1 (zh) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110650479A (zh) * | 2019-09-12 | 2020-01-03 | 中国人民解放军战略支援部队信息工程大学 | 异构携能通信网络中鲁棒的物理层安全传输方法及装置 |
CN111416352A (zh) * | 2020-04-29 | 2020-07-14 | 华北电力大学 | 一种多能源耦合系统的调度方法及计算设备 |
CN113543068A (zh) * | 2021-06-07 | 2021-10-22 | 北京邮电大学 | 一种基于层次化分簇的林区无人机网络部署方法与系统 |
CN113644946A (zh) * | 2021-08-11 | 2021-11-12 | 北京邮电大学 | 一种面向用户为中心网络的深度学习动态波束赋形方法 |
CN113825236A (zh) * | 2021-09-03 | 2021-12-21 | 浙江大学 | 一种无线网络中感知、计算和通信的融合方法 |
Families Citing this family (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107222288B (zh) * | 2016-03-22 | 2020-12-15 | 中国移动通信集团公司 | 一种无线传输方法及装置 |
CN105721028B (zh) * | 2016-04-18 | 2018-08-21 | 东南大学 | Vlc mu-miso系统下行波束成型方法、装置及vlc mu-miso系统 |
US10218785B2 (en) * | 2016-12-02 | 2019-02-26 | Qualcomm Incorporated | Opportunistic sync block transmission for mm wave NR-SS |
CN109150271B (zh) * | 2017-06-15 | 2022-07-19 | 大唐移动通信设备有限公司 | 一种最优波束确定方法、用户终端和网络侧设备 |
CN109831823B (zh) * | 2017-11-23 | 2021-01-12 | 上海诺基亚贝尔股份有限公司 | 用于通信的方法、终端设备和网络设备 |
CN107994933B (zh) * | 2017-11-30 | 2021-01-29 | 重庆邮电大学 | 认知mimo网络中次用户系统容量的优化方法 |
CN117715216A (zh) | 2018-02-14 | 2024-03-15 | 华为技术有限公司 | 一种数据传输的方法及相关装置 |
CN113141228B (zh) * | 2021-04-15 | 2022-09-27 | 青岛鼎信通讯股份有限公司 | 一种基于传输延时的台区时钟自校正方法 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1533065A (zh) * | 2003-03-18 | 2004-09-29 | 中兴通讯股份有限公司 | 智能天线上行波束形成方法及其装置 |
CN102142876A (zh) * | 2011-03-04 | 2011-08-03 | 西安交通大学 | Td-lte系统中时频域联合处理的波束形成方法 |
CN103245956A (zh) * | 2013-04-19 | 2013-08-14 | 哈尔滨工程大学 | 一种基于稳健波束形成算法的gps抗多径方法 |
CN103595453A (zh) * | 2013-10-29 | 2014-02-19 | 上海交通大学 | 多小区时分复用无线系统波束成型方法 |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101415229B (zh) * | 2008-11-20 | 2010-06-02 | 上海交通大学 | 空分多址系统基于有限反馈限制的方法及装置 |
KR101452732B1 (ko) * | 2009-03-10 | 2014-10-21 | 삼성전자주식회사 | 기지국 협력 시스템 및 방법 |
CN101986574B (zh) * | 2010-10-02 | 2013-05-01 | 上海交通大学 | 用于多基站协作系统的波束成型矩阵的优化方法 |
-
2014
- 2014-02-20 CN CN201410058819.3A patent/CN104868947A/zh active Pending
- 2014-06-10 EP EP14883399.9A patent/EP3096466A4/en not_active Withdrawn
- 2014-06-10 WO PCT/CN2014/079591 patent/WO2015123942A1/zh active Application Filing
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1533065A (zh) * | 2003-03-18 | 2004-09-29 | 中兴通讯股份有限公司 | 智能天线上行波束形成方法及其装置 |
CN102142876A (zh) * | 2011-03-04 | 2011-08-03 | 西安交通大学 | Td-lte系统中时频域联合处理的波束形成方法 |
CN103245956A (zh) * | 2013-04-19 | 2013-08-14 | 哈尔滨工程大学 | 一种基于稳健波束形成算法的gps抗多径方法 |
CN103595453A (zh) * | 2013-10-29 | 2014-02-19 | 上海交通大学 | 多小区时分复用无线系统波束成型方法 |
Non-Patent Citations (1)
Title |
---|
See also references of EP3096466A4 * |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110650479A (zh) * | 2019-09-12 | 2020-01-03 | 中国人民解放军战略支援部队信息工程大学 | 异构携能通信网络中鲁棒的物理层安全传输方法及装置 |
CN110650479B (zh) * | 2019-09-12 | 2023-02-10 | 中国人民解放军战略支援部队信息工程大学 | 异构携能通信网络中鲁棒的物理层安全传输方法及装置 |
CN111416352A (zh) * | 2020-04-29 | 2020-07-14 | 华北电力大学 | 一种多能源耦合系统的调度方法及计算设备 |
CN111416352B (zh) * | 2020-04-29 | 2023-08-15 | 华北电力大学 | 一种多能源耦合系统的调度方法及计算设备 |
CN113543068A (zh) * | 2021-06-07 | 2021-10-22 | 北京邮电大学 | 一种基于层次化分簇的林区无人机网络部署方法与系统 |
CN113543068B (zh) * | 2021-06-07 | 2024-02-02 | 北京邮电大学 | 一种基于层次化分簇的林区无人机网络部署方法与系统 |
CN113644946A (zh) * | 2021-08-11 | 2021-11-12 | 北京邮电大学 | 一种面向用户为中心网络的深度学习动态波束赋形方法 |
CN113644946B (zh) * | 2021-08-11 | 2023-12-01 | 北京邮电大学 | 一种面向用户为中心网络的深度学习动态波束赋形方法 |
CN113825236A (zh) * | 2021-09-03 | 2021-12-21 | 浙江大学 | 一种无线网络中感知、计算和通信的融合方法 |
CN113825236B (zh) * | 2021-09-03 | 2023-08-01 | 浙江大学 | 一种无线网络中感知、计算和通信的融合方法 |
Also Published As
Publication number | Publication date |
---|---|
EP3096466A1 (en) | 2016-11-23 |
EP3096466A4 (en) | 2017-01-11 |
CN104868947A (zh) | 2015-08-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2015123942A1 (zh) | 一种实现波束成型的方法及基站 | |
CN114208051B (zh) | 使用互易几何预编码的无线通信方法和装置 | |
US9854589B2 (en) | Interference management and interference alignment in wireless networks | |
CN112042147B (zh) | 具有多个假设的信道状态信息(csi)反馈 | |
KR20210109598A (ko) | 희소 채널 표현에 기초한 무선 셀들의 분산형 협력 동작 | |
WO2017133434A1 (zh) | 无线通信方法和无线通信装置 | |
US9923685B2 (en) | Method and transmitter apparatus for generating and transmitting channel feedback and method and receiver apparatus for receiving and retrieving channel feedback | |
TWI846900B (zh) | 時分雙工多輸入多輸出校準 | |
JP5497966B2 (ja) | 無線ネットワークの基地局内でアンテナ相互関係を較正する方法およびそのデバイス | |
KR101414665B1 (ko) | 부분 채널 상태 정보를 이용한 다층 빔포밍 | |
WO2014000449A1 (zh) | 传输模式选择方法、天线收发组合确定方法、装置及系统 | |
AU2021316062B2 (en) | Localization and auto-calibration in a wireless network | |
JP5384754B2 (ja) | 相互関係誤差を校正するデバイスおよび方法 | |
Zhao et al. | Massive MIMO in 5G networks: selected applications | |
US10979119B2 (en) | Beamforming in cellular systems using the same feedback information for different physical channels | |
EP3868033B1 (en) | Multi-user pairing and sinr calculation based on relative beam power for codebook-based dl mu-mimo | |
WO2021092746A1 (zh) | 一种信道信息获取的方法 | |
CN105450274B (zh) | 基于能效最优的大规模多天线中继系统用户数优化方法 | |
TWI586129B (zh) | 用於下行鏈路中協調式多點傳輸之預編碼權重之分散式計算 | |
WO2010124488A1 (zh) | 多小区预编码权值的配置方法和装置 | |
WO2013131339A1 (zh) | 一种干扰消除的方法及装置 | |
WO2020061781A1 (zh) | 通信方法、装置及系统 | |
WO2019166995A1 (en) | Method to estimate snr for mu-mimo based on channel orthogonality | |
CN103595453B (zh) | 多小区时分复用无线系统波束成型方法 | |
US20140205040A1 (en) | System and Method for Digital Communications Using Channel Statistics |
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: 14883399 Country of ref document: EP Kind code of ref document: A1 |
|
REEP | Request for entry into the european phase |
Ref document number: 2014883399 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2014883399 Country of ref document: EP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |