A method for calibrating antenna reciprocity in a base station
of wireless network and a device thereof
Field of the Invention
The present invention relates to a wireless network, particularly to a method for calibrating the antenna of a CoMP(cooperative multi-point processing) base station in a base station of wireless network and a device thereof the present invention Base is configured to over-the-air calibration CoMP Base antennas.
Background of the Invention
Channel reciprocity, i.e. characteristics of frequency symmetry
for uplink/downlink, have a bright prospect of application in TDD
(Time Division Duplex) system for LTE-A( Advanced- Long Term Evolution ) .
The hypothesis of reciprocity of uplink/downlink is widely recognized
in the industry, and is utilized to conduct channel estimation
effectively. However, in practical application, due to the difference
of RF Circuit between receiver and transmitter, especially the
difference of RF Circuit between receiver and transmitter on base
station side, hence, assurance of reciprocity between uplink and
downlink in TDD system is difficult to be kept. For a system with
a reciprocity between uplink and downlink based on TDD, system
performance is sensitive to the tolerance between uplink and downlink
channel, and even a little tolerance between uplink and downlink channel
might result in a most serious performance degradation. Thus,
reciprocity calibration of the TDD system draws attention of the
industry.
Without the requirement for extra hardware equipment, the
calibration of an OTA (over-the-air) interface of the TDD system with

a reciprocity between downlink and uplink turns out to be an effective approach to guarantee reciprocity of uplink and downlink channel of the TDD system.
Rl-0 9 46 23 proposed an OTA calibrating method for TDD (Time Division Duplexing) CoMP systems, which can eliminate the requirement of the resource-consuming reciprocity calibration for each transceiver of a CoMP eNB. But analysis in Rl-101016 shows that the quality of OTA calibration is limited by the user equipment's (UE) channel quality. In CoMP, the UE needs to measure all its channels from/to all the CoMP eNBs relevant thereto. With merely a portion of the above channels, if not all, calibration effect would be poor, thus directly resulting in degradation of CoMP performance.
Additionally, for current CoMP, the conventional over-the-air calibration requires the participating UE (s) to be able to accurately measure the downlink and uplink channels to all eNBs in the CoMP cluster to which the UE belongs. This requirement proves to be difficult in CoMP practice because the UE could be at the edge of at least one CoMP cell if not at the edge of all the CoMP cells. That may mean poor calibration performance of the air interface channels due to the inaccurate measurement.
Therefore, there still exist some issues in the prior art, such as how to preferably select UE (s) , how to configure the selected UE (s) to calibrate the antenna(s) configured on the CoMP base station. Especially for a CoMP TDD system, there is still needed a solution about UE-selecting and calibration for reciprocity of the matrix channel between the selected UE(s) and the CoMP base station related thereto.
Summary of the Invention

In view of the deficiency in the background, the present invention provides a method for calibrating antenna reciprocity in a base station of wireless network. The method may be applicable to selection of calibrating UE(s) for calibrating antennas configured for the CoMP base stations in a CoMP scenario and to configuration of antennas, of the CoMP base station, for the selected UE(s) to calibrate, thus enabling CoMP base stations to calibrate, based on the selected UE, reciprocity of the channel matrix.
According to one embodiment of the present invention, there is provided a method of calibrating antenna reciprocity in a base station of wireless network, wherein said base station is one of a plurality of base stations configured to conduct a cooperative multi-point processing, the method comprising : determining, based on a predefined rule, a plurality of calibrating user equipments out of a plurality of user equipment s, andantennas, to be calibrated by said plurality of calibrating user equipments, of said plurality of base stations, wherein, said plurality of calibrating user equipments are configured to calibrate antennas of said plurality of base stations and said predefined rule is that the number L of antennas of said plurality of calibrating user equipments, the number K of antennas configured for said plurality of base stations and to be calibrated by the antenna of each of said plurality of calibrating user equipments, and the number M of the antennas of said plurality of base stations satisfy the relation of KL >M + L-\ ; and the criteria for channel quality between said plurality of calibrating user equipments and the antennas configured for said plurality of base stations is above a threshold; calibrating, according to the determined calibration relationship, the antennas

reciprocity of said plurality of base stations based on said plurality of calibrating user equipments, wherein, said calibration relationship is the correspondence between said plurality of calibrating user equipments and the antennas, to be calibrated respectively by said plurality of calibrating user equipments, of said plurality of base stations.
In another embodiment of the present invention, the above calibrating step further comprises: determining the ratio of the reciprocity error of said plurality of base stations relative to that of a first one among said plurality of calibrating user equipments, according to the ratio of the reciprocity error of said plurality of base stations to that of said plurality of calibrating user equipments; generating a calibration weight matrix for the uplink channel on said base station side, according to the ratio of the reciprocity error of said plurality of base stations relative to that of the first one among said plurality of calibrating user equipments; determining a calibrated transposed matrix for the uplink channel, according to the calibration weight matrix for the uplink channel on said base station side and a transposed matrix for the uplink channel.
In another embodiment of the present invention, the above calibrating step further comprises: determining the ratio of reciprocity error of said plurality of base stations relative to that of a first one among the antennas configured for said plurality of base stations, according to the ratio of the reciprocity error of said plurality of base stations to that of said plurality of calibrating user equipments; generating a calibration weight matrix for the uplink channel on said base station side, according to the ratio of the reciprocity error of said plurality of base

stations relative to that of the first one among the antennas configured for said plurality of base stations; determining a calibrated transposed matrix for the uplink channel, according to the calibration weight matrix for the uplink channel on said base station side and a transposed matrix for the uplink channel. According to yet another embodiment of the present invention, there is provided a device for calibrating antenna reciprocity in a base station of wireless network, wherein said base station is one of a plurality of base stations configured to conduct a cooperative multi-point processing, the device comprising: a calibrating-UE determining unit, configured to determine, based on a first predefined rule, antennas of a plurality of calibrating user equipments out of a plurality of user equipments, wherein the antennas of said plurality of calibrating user equipments are configured to calibrate antennas of said plurality of base stations and said first predefined rule is that the criteria for channel quality between said plurality of calibrating user equipments and the antennas configured for said plurality of base stations is above a threshold; a calibration relationship determining unit, configured to determine, based on a secondpredef ined rule, antennas, to be calibrated respectively by the antennas of said plurality of calibrating user equipments, of said plurality of base stations, wherein said second predefined rule is that the number L of the antennas of said plurality of calibrating user equipments, the number K of antennas configured for said plurality of base stations and to be calibrated by the antenna of each of said plurality of calibrating user equipments, and the number M of the antennas of said plurality of base stations satisfy the relation of KL>M + L-1 ; a calibration unit, configured to calibrate, according to the

determined calibration relationship, the antennas reciprocity of said plurality of base stations based on said plurality of calibrating user equipments, wherein said calibration relationship is the correspondence between said plurality of calibrating user equipments and the antennas, to be calibrated respectively by said plurality of calibrating user equipments, of said plurality of base stations.
In still another embodiment of the present invention, the above calibration unit further comprises: a first reciprocity-error ratio determining unit, configured to determine the ratio of reciprocity error of said plurality of base stations relative to that of a first one among said plurality of calibrating user equipments, according to the ratio of the reciprocity error of said plurality of base stations to that of said plurality of calibrating user equipments; a first BS-side uplink-channel calibration-weight-matrix generating unit, configured to generate a calibration weight matrix for the uplink channel on said base station side, according to the ratio of the reciprocity error of said plurality of base stations relative to that of the first one among said plurality of calibrating user equipments; a first uplink-channel transposed matrix generating unit, configure to determine a calibrated transposed matrix for the uplink channel, according to the calibration weight matrix for the uplink channel on said base station side and a transposed matrix for the uplink channel.
In still another embodiment of the present invention, the above calibration unit further comprises: a second reciprocity-error ratio determining unit, configured to determine the ratio of reciprocity error of said plurality of base stations relative to

that of a first antenna among the antennas configured for said plurality of base stations, according to the ratio of the reciprocity error of said plurality of base stations to that of said plurality of calibrating user equipments; a second BS-side uplink-channel calibration-weight-matrix generating unit, configured to generate a calibration weight matrix for the uplink channel on said base station side, according to the ratio of the reciprocity error of said plurality of base stations relative to that of the first antenna among the antennas configured for said plurality of base stations; a second uplink-channel transposed matrix generating unit, configured to determine a calibrated transposed matrix for the uplink channel, according to the calibration weight matrix for the uplink channel on said base station side and a transposed matrix for the uplink channel.
In the method of the present invention, a plurality of calibrating UEs are determined according to channel quality criteria, and the antennas reciprocity of a plurality of base stations is calibrated based on the determined plurality of calibrating UEs and in accordance to the calibration relationship between the antennas of the plurality of UEs and that of the plurality of CoMP base stations , thus rendering a brand new reciprocity calibration solution for TDD CoMP scenario so as to enhance the flexibility for selection of the calibrating UEs and also improve the accuracy of channel measurement and reciprocity calibration as compared to the reciprocity calibration conducted with only one calibrating UE based thereon. It could be achieved by experiments that: as compared with the conventional OTA calibration, by means of the calibrating method of some embodiments of the present invention, the reciprocity performance gain may be accomplished

by averagely over 21% for cells, and by over 23% for cell edge. The above-mentioned experiments are made under experimental conditions in accordance to LTE-R10 standard, with experimental parameters listed in Table 1:
The results for the above test are reported as below:

Additionally, it is not needed, for calibration method of the present invention, to add extra hardware equipment to base station or UE, and it can independently fulfill the reciprocity calibration of the CoMP base station with the self-calibration in conventional base stations replaced to some extent.
Moreover, according to a calibration method of another embodiment of the present invention, the calibration method in the embodiments of the present invention may be used, in combination with the self-calibration in traditional base stations, to achieve better calibration performance(as shown in Table 2).
Brief Description of the Drawings
Features, aspects and advantages of the present invention will become more obvious by reading the following description of non-limiting embodiments with the aid of appended drawings.
Figurel illustrates the block diagram of a system model capable of reciprocity according to one embodiment of the present invention;
Figure2 illustrates the topology schematic diagram between a plurality of UEs and CoMP base stations according to one embodiment of the present invention;
Figure3 illustrates the flow chart of a method of calibrating

antennas reciprocity according to one embodiment of the present invention;
Figure4 illustrates the flow chart of the calibrating step for a method of calibrating antennas reciprocity according to one embodiment of the present invention;
Figure5 illustrates the block diagram of a device for calibrating antennas reciprocity according to one embodiment of the present invention;
Wherein, same or similar reference numerals refer to the same or similar device(module) or step of method.
Detailed description of embodiments
Reciprocity based MIMO is of most interest for TDD downlink
transmission because an eNB has higher computation power than a
UE . Moreover, it has access to all users' channel information and
thus can perform globally optimum precoding with the full channel
matrix taken into account. Therefore, focus will be applied mainly
on a SU/MU MIMO downlink channel in some of the embodiments of
the present invention as follows.
Figure 1 illustrates the block diagram of a system model capable of reciprocity according to one embodiment of the present invention, the uplink channel model 13 and downlink channel model 14, as shown in Fig.l, represent the matrix channel among at least one UE 11 and at least one base station 12 ( schematically illustrated by only one channel in the Figure) , wherein uplink channel model 13 may also be the object to which at least portion of embodiments of method according to the present invention may mainly apply.
As shown in Figure 1, various reciprocity errors may take effect in the system. In the model of the system, H denotes matrix channel response, subscript 'b' denotes base station or eNB, Jm' denotes

user equipment (e.g. UE), ;AI' denotes air interface, 'UL' and ;DL' denote uplink and downlink respectively, ' t' and ' r ' denote transmitter and receiver, respectively. The interference asymmetry can be compensated via whitening filtering at both sides. In the following, description will be concentrated on the estimation and calibration of the RF mismatch type of reciprocity errors according to some embodiments of the present invention hereinafter. In the system shown in Figure 1, generally, a wireless transmitter may be very different, in terms of RF circuitry, from a receiver (with H^r and H^r to be channel response model thereof) . By including both the eNB and the UE transmitter/receiver circuitry responses, the baseband to baseband channel responses can be formulated as:
IT —WW W
X1DL r±mrr±AI,DLr±bt
W =W W W \*-J
aUL r±brr±AI,ULr±mt
The reciprocity error can be formulated as follows:
Eb =HtoH^ = &mg(ebl9eb2^ebN)
where Em is reciprocity error at the mobile station side, and *^b is reciprocity error at the eNB side, (•)~1 means matrix inverse operation. Apply the above formulations into above equations (1), respectively. The effective downlink and uplink channel can be derived as follows:
HDL =E«HjLEj, (2)
Such equations(2) models the reciprocity of the baseband to baseband uplink and downlink channels with RF mismatches.
The target of over-the-air reciprocity error estimation and calibration is to estimate eNB side reciprocity error E^ and UE side reciprocity error *^m from over-the-air baseband to baseband

measurements of HDL and "-UL * anci compensate the above error back to the user signals so as to guarantee the MIMO performance.
Of course, those skilled in the art could readily appreciate, for cooperated multipoint joint processing (CoMP), a plurality of CoMP base stations can be regarded as a super eNB, and a plurality of CoMP UEs can be regarded as a super UE. This will not impact the above reciprocity error modeling.
For practical systems, HDL and ■H[/L are measurable via pilot signaling, for example, downlink CSI-RS and uplink SRS in 3GPP LTE-A systems. Thus, the following equation (3) is left with two unknown matrices Efe and *^m • And the following embodiments of the present invention will be described with i^b and *^m remaining.
It should be noted that: *^h and *^m are both diagonal matrices . Without loss of generality, assume that all elements of H^L are non-zero, then Equation (3) can be formulated as:
diag-1(E;1)[diag-1(Ei)]r =HDL./H^L (3)
wherein, where diag" (•) denotes the operation to transform a diagonal matrix to a column vector, and ./ denotes matrix dot-division.
Take logarithm on both sides and move the right side to the left, then the linear equation can be formulated into the form ofAX=0. The solvability of this matrix equation is determined by the rank of its coefficient matrix,
1 0 ••• 0 I
x7Vxl ^Nxl ^Nxl *NxN
0 1 ••• 0 I
A _ ^Nxl *Nxl ^Nxl *NxN
^(MxN)x(M+N) — • • • • (4)
* * * *
0 0 ••• 1 I
_™Nxl ^Nxl *Nxl *NxN _
where l^xl or "A^XI is a column vector with N all one or zero elements, respectively, and *NXN is an entity matrix with dimension NxN . By using mathematical induction method, it can

be proved that:
rank(A) = M + N-1 (5)
By selecting Cml as reference, the eNB side reciprocity errors can be defined as,
eb,\ ~ emXlul,\YldlM eb,2 ~ em,\*lul,\2*ldl,\2
(6)
fb,N ~ em,\*lul,\N*ldl,\N
m 1 . . .
By also selecting ' as reference, the UE side reciprocity errors can be defined as,
e~\ = e~\ e~x = e~x h~x h h~x h
(7) e~x = e~x h~x h h~x h
~m,M ^myldlMnulMnulMV dl,MWith the eNB/UE side reciprocity errors both expressed in eml-> the eNB calibration weight can be calculated according to following formulation,
The calibration weight of the calibrating UE is: And the calibration operation can be expressed as,
KL^ = WmHjLW, (8)
wherein:
Wm =dmg(wm^wma,...,wmM) Wfc=diag(ww,ww,...,w^) •
It can be noted from Equation (7) that, only one UE antenna is needed for performing relative eNB antenna calibration. For

performance enhancement , anaverage, from a plurality of UE antennas , can be taken on the estimation of Efc . But the condition is that a common eNB antenna should be used as reference in the estimation of Efc with different UE antennas. Similar average operation can be taken in the calibration of UE side antennas.
Figure2 illustrates the topology schematic diagram between a plurality of UEs and CoMP base stations according to one embodiment of the present invention. Therein is included a base station 121, a base station 122, a base station 123, in the figure, and the above three base stations are configured to perform intra-site CoMP, and participate in reciprocity calibration. Wherein, the base station 121 is equipped with antennas Tl, T2, T3, T4, the base station 122 is equipped with antennas T5, T6, T7, T8, and the base station 123 is equipped with antennas T9, T10, Til, T12. The base station 121, the base station 122 and the base station 123 have three UEs in their belonging coverage: UE111, UE112, UE113. And the above three UEs are capable of providing OTA calibration measurement. Without loss of generality, herein, supposing that each UE is equipped with one antenna, then the number of antennas is actually the number of UEs.
Figure3 illustrates the flow chart of a method of calibrating antennas reciprocity according to one embodiment of the present invention. The method as shown in the figure comprises a step of S301 for determining calibrating UE and calibration relationship between the calibrating UE and CoMP base station, a step of S302 for calibrating antennas reciprocity.
In step S301, the base station 121, the base station 122, the base station 123 or the combination thereof determine, based on a predefined rule, a plurality of calibrating UEs out of a plurality of UEs, and the plurality of base stations ' antennas to be calibrated

by the plurality of calibrating UEs, wherein, the plurality of calibrating UEs are configured to calibrate the antennas of the plurality of base stations, and the predefined rule is: the number L of the plurality of calibrating UEs, the number K of antennas, configured for the plurality of base stations, calibrated by each calibrating UE of the plurality of calibrating UEs, and the number M of the antennas of the plurality of base stations satisfy the relation of KL> M + L — \; and channel quality criteria between the plurality of calibrating UEs and the antennas configured for the plurality of base stations is above a threshold.
For instance, the base station 121, the base station 122, the base station 123 or the combination thereof may select two or three UEs as calibrating UEs which are used during calibration, based on a predefined rule in this embodiment.
Meanwhile, according to the predefined rule given foregoing, those skilled in the art should appreciated, by means of enumeration, calibrating UE (s) satisfying the predefined rule could be selected and the number of antennas, of the base station 12 1, the base station 122, the base station 123, to be calibrated by the calibrating UE(s) could be determined.
The conventional calibration method requires all the channels from certain UE ' s antenna to the twelve eNB antennas, of the three base stations as shown in Fig.2, are sufficiently good in quality. But obviously, a UE, for example UE1, is out of the coverage of base station 3, thus the channel quality from the UE1 to the base station 3 could be poor.
Herein, in the above embodiment, the coefficient matrix (4) of equation (3) needs to be revised. Assume all UE antennas calibrate the same number of eNB antennas while all eNB antennas are calibrated by the same number of UE antennas. Let K denote the number of eNB

antennas calibrated by each UE, and L denote the number of UE antennas for calibrating each eNB antenna. To guarantee the relative calibration is still feasible for the M eNB antennas and the N UE antennas, the rank of the coefficient matrix (4) needs to be maintained, e.g. , the following relations should satisfy:
KL>M+L-l (9)
Thus K and L should be natural number and satisfy:
K>^^^ (10)
Or equivalently,
L^ (11)
Equation (10) gives the minimum number of eNB antennas to be calibrated by each UE antenna when the number of calibrating UE antennas is given, or the minimum number of calibrating UE antennas required when the number of eNB antennas to be calibrated by each calibrating UE is constrained. For instance, the relationship among the number of eNB antennas to be calibrated and the number of antenna of the calibrating UE, the minimum number of eNB antennas to be calibrated by antenna of each calibrating UE may be given in the form listed by the table 3. Of course, more combinations with equation (9) met could be derived by means of exhaustion method.
a) Given the number of UE antennas involved in the calibration

I :_= I - I - I
b) Given the number of eNB antennas to be calibrated for each UE antenna Table 3 Illustration of minimum number of eNB or UE antennas for network calibration when a
plurality of calibrating UEs calibrate CoMP base station
As for the above table 3, it should be notedthat: when a plurality of UE antennas are involved in the OTA calibration, an eNB antenna does not need to be calibrated by all UE antennas in the case that the predefined rule in equation (9) is met.
Additionally, the criteria for channel quality involved in the predefined rule of step S301 comprises, but not limited to at least one of the following: signal-to-noise ratio, signal to interference noise ratio, the number of notification of ACK/NACK of Automatic Feedback Repeat. For example, the criteria for channel quality may be that signal to interference noise ratio of channel is greater than lOdB.
In every possible scheme candidate achieved via equation (9) by means of enumeration in table (3), it is assumed that the last scheme in table 3(b) is adopted for the above embodiment, namely, "the minimum number K of eNB antennas calibrated by each UE antenna" is 8, while "the number L of calibrating UE antennas" is 2.
Furthermore, it is assumed that UE111 and UE112 are determined to be calibrating UEs according to the criteria for channel quality that signal to interference noise ratio of channel is greater than lOdB, yet UE113 does not participate in calibration due to the deficiency of the criteria for channel quality.
Meanwhile, with "minimum number K of eNB antennas calibrated

by each UE Antenna" being 8, there might exist various calibration relationship between UE111, UE112 and antennas T1-T12 configured for the base station 121, the base station 122 and the base station 123. Herein, with the factor of channel quality taken into consideration, without loss of generality, the calibration relationship could be determined to be that: UE111 calibrates antennas T1-T8, and UE112 calibrates antennas T5-T12.
In step S302, based on the plurality of calibrating UEs, the base station 121, the base station 122 and the base station 123 or the combination thereof calibrates, according to the determined calibration relationship, the antennas reciprocity of the plurality of base stations. For instance, first, the base station 121, the base station 122 and the base station 123 or the combination thereof, determines the ratio of the reciprocity error of the twelve antennas T1-T12 of the base station 121, the base station 122 and the base station 123 relative to that of any antenna, wherein the any antenna may be any one antenna configured for the base station or UE within such calibration system. Subsequently, in accordance with such ratio, to generate a calibration weight matrix for the uplink channel on the base station side, and determine the calibrated transposed matrix for the uplink channel according to the generated calibration weight matrix for the uplink channel on the base station side and the transposed matrix for the uplink channel, such that reciprocity error of antennas of the base station system is calibrated for the base station 121, the base station 122 and the base station 123 or the combination thereof with the above the transposed matrix for the uplink channel calibrated, thereby the performance of subsequent precoding, MIMO processing or beamforming is enhanced.
Figure 4 illustrates the flow chart of the calibrating step

for a method of calibrating antennas reciprocity according to one embodiment of the present invention. As shown in the Figure, the calibrating step S302 specifically comprises the step S3021 of ratio-determining of the reciprocity-error of base station to that of calibrating UE, the step S3022 of generating calibration weight matrix, the step S3033 of determining the transposed matrix for the uplink channel.
In step S3021, according to the ratio of the reciprocity error of a plurality of base stations to that of a plurality of calibrating UEs, to determine the ratio of the reciprocity error of the plurality of base stations respectively to that of a first calibrating UE among the plurality of calibrating UEs.
In such embodiment, according to the ratio of the reciprocity error of antennas configured for the base station 121, the base station 122 and the base station 123 to the reciprocity error of the two calibrating UEs UE 111 and UE112, the base station 121, the base station 122 and the base station 123 or the combination thereof collaboratively determines, respectively or collaboratively, the ratio of the reciprocity error of antennas of the base station 121, the base station 122 and the base station 123 relative to reciprocity error of antenna of UE111. Certainly, the base station 121, the base station 122 and the base station 123 or the combination thereof collaboratively may alternatively determine the ratio of the reciprocity error of antennas of the base station 121, the base station 122 and the base station 123 relative to reciprocity error of antenna of UE112.
By selection of the reciprocity error ^mj of UE111 and the reciprocity error £m2 of UE112 as reference, the reciprocity errors of the antennas of the base station 121, the base station 122 and the base station 123 may be expressed respectively as

following equation:
eb,\ = em,Ai,nhdi,n eb5 = e^h^K,-*
eb,2 = em,Ai,nhdi,n eb6 = em2hj26hdl26
• • •
* * *
eb,s ~ em,\Ki,\5^di,\5 eb9 = em2hul29hdl29
• • • ...
From the two set in equation (12), it could be derived of:
_ J em,Ylul,\j'fldl,\j'> J ~ 1>"°
wherein,
1 8
V = T 2^1 ^ul,l,j^dl,lJ^dl,2J^ul,2J . 4 j=5
Thus, in such embodiment, the reciprocity of the base station
121, thebase station 122 and the base station 123 could be calibrated
with calibration weights relative to reciprocity error of UE111:
wbj =< _{
\P^ul,2j^dl,2j » J = "v"4^
In step S3022, to generate a calibration weight matrix for the uplink channel on the base station side according to the ratio of the reciprocity error of a plurality of base stations relative to that of the first calibrating UE among a plurality of calibrating UEs.
In the above embodiment, according to the ratio of the reciprocity error of antennas employed by the base station 121, the base station 12 2 and the base station 123 relative to reciprocity error of calibrating UE 111, the base station 121, the base station
122, thebase station 123 or the combination thereof collaboratively
generate a calibration weight matrix for the uplink channel on

base station side by utilization of the above weight, shown as following equation:
W, =dmg(wbl,wb2,...,wbl2) (14)
Next, instepS3023, to determine a calibrated transposed matrix for the uplink channel according to the calibration weight matrix for the uplink channel on the base station side and the transposed matrix for the uplink channel.
For instance, the calibrated transposed matrix for the uplink channel may be determined according to the following equation:
Hr =Hr W
wherein , ^uLcai denotes the calibrated transposed matrix for the uplink channel which is determined as above, HUL denotes the transposed matrix for the uplink channel before calibration, W^ denotes the calibration weight matrix for the uplink channel on base station side.
In another embodiment of the present invention, prior to step S3022, such method further comprises a step that a plurality of base stations conduct self-calibration respectively, and the step S3022 further comprises a step of averaging respectively uplink channel's calibration weight corresponding to antennas respectively configured for each base station among the plurality of base stations, with the above calibration weight to be averaged being an element of the calibration weight matrix for the uplink channel on the base station side, and substitute the resulting average value for the uplink channel's calibration weight corresponding to antennas respectively configured for each base station among the plurality of base stations so as to generate a average calibration weight matrix for the uplink channel on the base station side. Subsequently, step S3023 further comprises a step of determining the calibrated transposed matrix for the uplink

channel according to the average calibration weight matrix and the transposed matrix for the uplink channel on the base station
side .
Prior to step S3022 of such embodiment, the base station 121, the base station 122, the base station 123 may respectively conduct self-calibration. In response to the self-calibration of above base stations, instep S3022, the base station 121, the base station 122, the base station 123 or the combination thereof average the weight value of the three vectors of ( W&f;, W^J W^j, W^ ) , ( W^j, Wa,tf, W*f7, Ww) and ( Wbi9, Wbiio, W6i7/, WM2) in equation (14), and substitute the three average values of bj234? 6,5678 ? 6,9^12 > for corresponding weight respectively so as to generate the average calibration weight matrix for the uplink channel on the base station side, shown as follows:
W4 = dmg(wb —, wb—-%,wb-%,wb-%,wb — ,...,wb — , w^2, w& —, w^—, w^—
subsequently, instepS3023, to determine the calibrated transposed matrix for the uplink channel according to the calibration weight matrix W& for the uplink channel on the base station side and the transposed matrix for the uplink channel.
For instance, the calibrated transposed matrix for the uplink channel may be determined according to the following equation:
XXt/L,ca/ "-UL VT b ■>
wherein, H01ca/ denotes the calibrated transposed matrix determined for the uplink chajinel, H^ denotes the transposed
matrix for the uplink channel before calibration.
Optionally, the step S302 further comprises a step of generating a calibration weight matrix for the uplink channel on the UE side according to the ratio of the reciprocity error of the first calibrating UE among the plurality of calibrating UEs relative to that of the plurality of base stations . For example, by selection

of the reciprocity error ebyi~ebyn of antennas T1-T12 employed by
the base station 121, the base station 122, the base station 123 as reference, the reciprocity error of antennas equipped in calibrating UE 111 and UE112, may be respectively represented, for example:
em,l = ebAlMh'ldl,U
■ ■ ■
em,2=eb,sK&h~*& ( 15)
■ ■ ■
Subsequently, formula derivation in the course of generation of calibration weight matrix Wm for the uplink channel on UE side
is similar to the formula derivation from equation 13 to equation
14, so no redundant description is made herein.
Next, in step S3023, to determine the calibrated transposed
matrix for the uplink channel according to the calibration weight matrix Wb, the transposed matrix for the uplink channel on the
base station side, and the calibration weight matrix Wm for the
uplink channel on the UE side. For example, the calibrated transposed matrix for the uplink channel may be further determined according to the following equation:
H'T = W Hr W
H
'T .._„„_, uL,cai denotes the calibrated transposed matrix for
H
T ^,w ^r^^iAJ. w^^.^w^. ...^w^ ^^ ^^^,w^ ^w^w^^.w^, UL denotes the
transposed matrix for the uplink channel before calibration, Wm denotes the calibration weight matrix for the uplink channel on the UE side, W^ denotes the calibration weight matrix for the uplink channel on the base station side.
In such embodiment, for step S3021, the ratio of the reciprocity error of a plurality of base stations relative to that of a plurality

of calibrating UEs may be the ratio of the downlink channel response
relative to uplink channel response between corresponding base
station and calibrating UE, as shown in equation 12, optionally,
prior to the step S3021, such method further comprises the steps
that, the base station 121, the base station 122, the base station
123 respectively or collaboratively transmits a signaling to the
plurality of calibrating UEs, for example UE111 and UE112, wherein
the signaling is used for instructing UE111 and UE112 to send
measurement values of the downlink channel response, for example n (15)
By introducing the above transformed equation 15 into equation 13, to determine the ratio of the reciprocity error of the twelve antennas T1-T12 of the base station 121, the base station 122 and the base station 123 relative to that of antennas Tl out of all the antennas employed by the base station 121, the base station 122 and the base station 123, and no redundant description will be made herein.
Furthermore, the base station 121, the base station 122 and the base station 123 may generate a calibration weight matrix for the uplink channel on the base station side according to the ratio of the reciprocity error of antennas T1-T12 configured for the above three base stations relative to that of a first antenna of all the antennas configured for the three base stations in such embodiment, for example the first antenna may be antennas Tl in this embodiment. And such step is similar to the step S3022 in the embodiment shown in Fig.4, and no redundant description needs

to be made herein.
Next, the base station 121, the base station 122 and the base station 123 determines the calibrated transposed matrix for the uplink channel according to the calibration weight matrix for the uplink channel on the base station side and the transposed matrix for the uplink channel. And no redundant description needs to be made herein.
Herein, those skilled in the art could readily appreciate: reciprocity error of above base station(s) may actually be the reciprocity error of antennas configured for base station(s), likewise, reciprocity error of UE may actually be the reciprocity error of antennas employed by UE.
Figure5 illustrates the block diagram of a device 200 for calibrating antennas reciprocity according to one embodiment of the present invention. Device 200 comprises a calibrating-UE determining unit 201, a calibration relationship determining unit 202 and a calibration unit 203.
Wherein, a plurality of base stations are configured to conduct CoMP. The above calibrating-UE determining unit 201 is configured to determine, based on a first predefined rule, antennas of a plurality of calibrating user equipments out of a plurality of user equipments, wherein the antennas of the plurality of calibrating user equipments are configured to calibrate antennas of the plurality of base stations and the first predefined rule is that the criteria for channel quality between the plurality of calibrating user equipments and the antennas configured for said plurality of base stations is above a threshold.
The calibration relationship determining unit 202 is configured to determine, based on a second predefined rule, antennas, to be

calibrated respectively by the antennas of the plurality of calibrating user equipments, of the plurality of base stations, wherein the second predefined rule is that the number L of the antennas of the plurality of calibrating user equipments, the number K of antennas configured for the plurality of base stations and to be calibrated by the antenna of each of the plurality of calibrating user equipments, and the number M of the antennas of the plurality of base stations satisfy the relation of KL>M + L-l .
The calibration unit 203 is configured to calibrate, according to the determined calibration relationship, the antennas reciprocity of the plurality of base stations based on the plurality of calibrating user equipments.
In such embodiment, the calibrating-UE determining unit 201 may select two or three UEs as calibrating UEs used on occurrence of calibration based on a predefined rule in such an embodiment.
Meanwhile, according to the predefined rule given foregoing, those skilled in the art should appreciated, according to formula KL>M + L-\f by means of enumeration, calibrating UE satisfying
the predefined rule could be selected and the number of antennas, of the base station 121, the base station 122 and the base station 123, for the calibrating UE to calibrate could be determined.
Additionally, as a basis of calibrating-UE determining unit 201, the criteria of channel quality involved in the predefined rule of step S301 comprises, but not limited to at least one of the following: signal-to-noise ratio, signal to interference noise ratio, the number of notification of ACK/NACK of Automatic Feedback Repeat. For example, the criteria for channel quality may be that signal to interference noise ratio of channel is greater than lOdB.
In another embodiment of the present invention, the calibration

unit 2 03 further comprises a first reciprocity-error ratio determining unit, a first BS-side uplink-channel calibration-weight-matrix generating unit, and a first uplink-channel transposed matrix generating unit.
The first reciprocity-error ratio determining unit is configured to determine the ratio of reciprocity error of the plurality of base stations relative to that of a first one among the plurality of calibrating user equipments, according to the ratio of the reciprocity error of the plurality of base stations to that of the plurality of calibrating user equipments.
The first BS-side uplink-channel calibration-weight-matrix generating unit is configured to generate a calibration weight matrix for the uplink channel on the base station side, according to the ratio of the reciprocity error of the plurality of base stations relative to that of the first one among the plurality of calibrating user equipments;
The first uplink-channel transposed matrix generating unit is configure to determine a calibrated transposed matrix for the uplink channel, according to the calibration weight matrix for the uplink channel on the base station side and a transposed matrix for the uplink channel.
In such embodiment, according to the ratio of the reciprocity error of antennas configured for the base station 121, the base station 122 and the base station 123 to the reciprocity error of the two calibrating UEs of calibrating UE 111 and UE112, the first reciprocity-error ratio determining unit determines the ratio of the reciprocity error of antennas of the base station 121, the base station 122 and the base station 123 relative to reciprocity error of antenna employed by UE111. Certainly, the base station

121, the base station 122, the base station 123 or the combination
thereof collaboratively may alternatively determine the ratio of
the reciprocity error of antennas of the base station 121, the
base station 122 and the base station 123 relative to reciprocity
error of antenna employed by UE112.
By selection of the reciprocity error ^m\ of UE111 and the
reciprocity error £m2 of UE112 as reference, the reciprocity
errors of the antennas employed by the base station 121, the base station 122 and the base station 123 may be expressed respectively as following equation:
eb,i = em,Ai,nK,n eb5 = e^XlisK^
eb,2 = em,Ai,i2hdi,i2 eb6 = em2hj26hdl26
• • •
* * *
eb,5 - em,iKi,15^1,15 eb9 = em2hul29hdl29 (142')
• • • ...
eb$ ~~ em,\'lul,\%'ldl,\% [^,12 ~~ em,2 ul,2l2^dl,2l2
From the two set in equation (12), it could be derived of:
wherein,
1 8
4 j=5
Thus, in such embodiment, the reciprocity of the base station 121, thebase station 122 and the base station 123 could be calibrated with calibration weights relative to reciprocity error of UE111:
wbj =< _x
\Jl^ul,2j^dl,2j » J ="v?AZ
Subsequently, according to the ratio of the reciprocity error

of antennas employed by the base station 121, the base station 122 and the base station 123 relative to reciprocity error of calibrating UE 111, the first BS-side uplink-channel calibration-weight-matrix generating unit generate a calibration weight matrix for the uplink channel on the Base station side by utilization of the above weight, shown as following equation:
Wb = dmg(wbl,wb2,...,wbl2) (14')
Next, the first uplink-channel transposed matrix generating unit may determine a calibrated transposed matrix for the uplink channel, for example according to the following equation:
Hr =Hr W
wherein, HULcal denotes the calibrated transposed matrix determined for the uplink channel, HUL denotes the transposed
matrix for the uplink channel before calibration, W^ denotes the calibration weight matrix for the uplink channel on the base station side .
Optionally, device 200 further comprises a UE-side uplink-channel calibration-weight-matrix generating unit.
The UE-side uplink-channel calibration-weight-matrix generating unit is configured to generate a calibration weight matrix for the uplink channel on the user equipment side, according to the ratio of the reciprocity error of the first one among the plurality of calibrating user equipments relative to that of the plurality of base stations.
Meanwhile, The first uplink-channel transposed matrix generating unit is further configure to determine the calibrated transposed matrix for the uplink channel, according to the calibration weight matrix for the uplink channel on the base station side, the transposed matrix for the uplink channel, and the calibration weight

matrix for the uplink channel on the user equipment side.
For example, by selection of the reciprocity error ebyi~ebyn of
antennas T1-T12 employed by the base station 121, the base station 122 and the base station 123 as reference, the reciprocity error of antennas equipped in calibrating UE 111 and UE112, may be respectively represented, for example:
a a a
em,2=ebJhul,25h~l*>B (15'}
■ ■ ■
Subsequently, formula derivation in the course of generation of calibration weight matrix Wm for the uplink channel on UE side
is similar to the formula derivation from equation 13' to equation
14', so no more redundant description is presented herein.
Next, the first uplink-channel transposed matrix generating
unit determines the calibrated transposed matrix for the uplink channel according to the calibration weight matrix Wb , the
transposed matrix for the uplink channel on the base station side, and the calibration weight matrix Wm for the uplink channel on
the UE side. For example, the calibrated transposed matrix for the uplink channel may be further determined according to the following equation:
H'T = W Hr W
H
'T .._„ „_, uL,cai denotes the further calibrated transposed
H
T
denotes the transposed matrix for the uplink channel before calibration, Wm denotes the calibration weight matrix for the
uplink channel on the UE side, W^ denotes the calibration weight

matrix for the uplink channel on the base station side.
In still another embodiment of the present invention, the calibration unit 203 further comprises a second reciprocity-error ratio determining unit, a second BS-side uplink-channel calibration-weight-matrix generating unit, and a second uplink-channel transposed matrix generating unit.
The second reciprocity-error ratio determining unit is configured to determine the ratio of reciprocity error of the plurality of base stations relative to that of a first antenna among the antennas configured for the plurality of base stations, according to the ratio of the reciprocity error of the plurality of base stations to that of the plurality of calibrating user equipments.
The second BS-side uplink-channel calibration-weight-matrix generating unit is configured to generate a calibration weight matrix for the uplink channel on the base station side, according to the ratio of the reciprocity error of the plurality of base stations relative to that of the first antenna among the antennas configured for the plurality of base stations.
The second uplink-channel transposed matrix generating unit is configured to determine a calibrated transposed matrix for the uplink channel, according to the calibration weight matrix for the uplink channel on the base station side and a transposed matrix for the uplink channel.
For example, the second reciprocity-error ratio determining unit may determine the ratio of the reciprocity error of antennas T1-T12 relative to the reciprocity error of certain antenna of all antennas T1-T12, by means of equation 12', namely, the ratio of the reciprocity error of antennas T1-T12 of the base station 121, the base station 122 and the base station 123 relative to

the reciprocity error of antennas configured for UE111, UE112 which are used for calibration.
For example, the first part of equation 12' can be transformed as follows:
eb,l=em,lhiuhdlM "^ em,l=ebAl,nh~ldl>n (15')
By introducing the above transformed equation 15 ' into equation 13', the ratio of the reciprocity error of the twelve antennas T1-T12 of the base station 121, the base station 122 and the base station 123 relative to that of antennas Tl out of all the antennas employed by the base station 121, the base station 122 and the base station 123 can be determined and no redundant description will be made herein. Of course, herein, those skilled in the art could readily appreciate: T2, Til or any one of other antennas can be selected as reference to represent the reciprocity error of antennas T1-T12 of the base station 121, the base station 122 and the base station 123, as similar to equation 13'.
Furthermore, the second BS-side uplink-channel calibration-weight-matrix generating unit may generate a calibration weight matrix for the uplink channel on the base station side according to the ratio of the reciprocity error of antennas T1-T12 configured for the above three base stations relative to that of a first antenna of all the antennas configured for the three base stations in such embodiment, for example the first antenna may be antennas Tl in this embodiment.
Subsequently, the second uplink-channel transposed matrix generating unit may determine the calibrated transposed matrix for the uplink channel according to the calibration weight matrix for the uplink channel on the base station side and the transposed matrix for the uplink channel. And no redundant description is

needed herein.
Those skilled in the art could readily appreciate, the device
in the present invention can be implemented by means of either
hardware module, functional module of software, or the hardware
module with functional module of software incorporated therein. Those skilled in the art could readily appreciate, the above
embodiment are intended to be exemplary but not limited thereto.
Different technical feature in various embodiments can be combined,
so as to gain some technical effect. After examining the
accompanying figures, specification and claims, those skilled in
the art should readily appreciate and carry out some variation
for the above disclosed embodiments. In the claims, the term of
"comprise" does not exclude the existence of other device(s) or
step(s); indefinite article "a/an" does not exclude the
"multiple" case; term of "a first", "a second" is just used for
give a name rather than for representing some particular order.
Figure number (s) in the claims, if any, should not be construed
to be any restriction to protection scope. Multiple component
functionalities in the claim(s) may be implemented by means of
an singular hardware or software. The occurrence of some technical
features in different dependent claims does not eliminate the
possibility of the technical effect resulting from the combination
of these technical features.

What is claimed is:
1. Amethod of calibrating antenna reciprocity in a base station
of wireless network, wherein a plurality of base stations are
configured to conduct a cooperative multi-point processing, the
method comprising:
- determining, based on a predefined rule, a plurality of calibrating user equipments out of a plurality of user equipments, and antennas, to be calibrated by said plurality of calibrating user equipments, of said plurality of base stations, wherein, said plurality of calibrating user equipments are configured to calibrate antennas of said plurality of base stations and said predefined rule is that the number L of antennas of said plurality of calibrating user equipments, the number K of antennas configured for said plurality of base stations and to be calibrated by the antenna of each of said plurality of calibrating user equipments, and the number M of the antennas of said plurality of base stations satisfy the relation of KL> M + L-l; and the criteria for channel quality between said plurality of calibrating user equipments and the antennas configured for said plurality of base stations is above a threshold;
calibrating, according to the determined calibration relationship, the antennas reciprocity of said plurality of base stations based on said plurality of calibrating user equipments, wherein, said calibration relationship is the correspondence between said plurality of calibrating user equipments and the antennas, to be calibrated respectively by said plurality of calibrating user equipments, of said plurality of base stations.
2. The method of claim 1, wherein said calibrating step further

comprises:
A. determining the ratio of the reciprocity error of said
plurality of base stations relative to that of a first one among
said plurality of calibrating user equipments, according to the
ratio of the reciprocity error of said plurality of base stations
to that of said plurality of calibrating user equipments;
B. generating a calibration weight matrix for the uplink channel
on said base station side, according to the ratio of the reciprocity error of said plurality of base stations relative to that of said first one among said plurality of calibrating user equipments;
C. determining a calibrated transposed matrix for the uplink
channel, according to said calibration weight matrix for the uplink
channel on said base station side and a transposed matrix for the
uplink channel.
3. The method of claim 1, wherein said calibrating step further comprises:
A', determining the ratio of the reciprocity error of said plurality of base stations relative to that of a first one among the antennas configured for said plurality of base stations, according to the ratio of the reciprocity error of said plurality of base stations to that of said plurality of calibrating user equipments;
B '. generating a calibration weight matrix for the uplink channel on said base station side, according to the ratio of the reciprocity error of said plurality of base stations relative to that of said first one among the antennas configured for said plurality of base stations;
C. determining a calibrated transposed matrix for the uplink channel, according to said calibration weight matrix for the uplink

channel on said base station side and a transposed matrix for the uplink channel.
4. The method of claim 2, wherein the method further comprises:
- generating a calibration weight matrix for the uplink channel
on said user equipment side, according to the ratio of the
reciprocity error of said first one among said plurality of
calibrating user equipments relative to that of said plurality
of base stations;
wherein said step C further comprises:
determining said calibrated transposed matrix for the uplink channel, according to said calibration weight matrix for the uplink channel on said base station side, said transposed matrix for the uplink channel, and said calibration weight matrix for the uplink channel on said user equipment side.
5. The method of claim 2, further comprising, prior to said
step B, the steps of:
- said plurality of base stations respectively conducting
self-calibration; and said step B further comprising:
- respectively averaging uplink channel's calibration weight,
corresponding to antennas respectively configured for each base
station among said plurality of base stations, in said calibration
weight matrix for the uplink channel on said base station side,
and substituting the average value for the uplink channel's
calibration weight corresponding to antennas respectively
configured for each base station among said plurality of base
stations so as to generate an average calibration weight matrix
for the uplink channel on said base station side;
said step C further comprising:
- determining said calibrated transposed matrix for the uplink

channel, according to said average calibration weight matrix on said base station side and said transposed matrix for the uplink channel.
6. The method of claim 2, further comprising, prior to said
step A, the steps of:
- receiving a signaling of request for calibration from a user equipment; or
- receiving a signaling of cooperative request for calibration from said plurality of user equipments.
7. The method of claim 2, wherein the ratio, in said step A,
of the reciprocity error of said plurality of base stations relative
to that of said plurality of calibrating user equipments is the
ratio of the downlink channel response relative to the uplink
channel response between corresponding base stations and
calibrating user equipments, and said method further comprises,
prior to said step A, the steps of:
- transmitting a signaling to said plurality of calibrating
user equipments, wherein the signaling is used for instructing
said plurality of calibrating user equipments to respectively send
measurement values of the downlink channel response to said
plurality of base stations;
- receiving a signaling from said plurality of calibrating user
equipments, said signaling comprising the measurement values of the downlink channel response measured by said plurality of calibrating user equipments.
8. The method of claim 7, further comprising, prior to said
step A, the step of:
- acquiring a corresponding measurement value of the uplink
channel response, in response to a reference signal from said

plurality of calibrating user equipments.
9. The method of claim 1, wherein the criteria for the channel
quality comprises at least one of the following:
- signal-to-noise ratio
- signal to interference noise ratio
- the number of notification of ACK/NACK of automatic feedback
repeat.
10. The method of claim 2, wherein said step C further comprises :
determining said calibrated transposed matrix for the uplink
channel according to the following formula,
^UUcal nUL ™b ?
with HULcal representing the determined calibrated transposed matrix for the uplink channel, HUL representing said transposed matrix for the uplink channel before calibration, W^representing said calibration weight matrix for the uplink channel on said base station side.
11 . The method of claim 4, wherein said step C further comprises : further determining a calibrated transposed matrix for the uplink
channel according to the following formula,
UT =W HT W
llUL,cal ™mllUL "b
with HULcal representing the further determined calibrated transposed matrix for the uplink channel, HUL representing said
transposed matrix for the uplink channel before calibration, Wm representing said calibration weight matrix for the uplink channel on said user equipment side, W^ representing said calibration weight matrix for the uplink channel on said base station side. 12. A device for calibrating antenna reciprocity in a base station of wireless network, wherein a plurality of base stations are configured to conduct a cooperative multi-point processing,

the device comprising:
- a calibrating-UE determining unit, configured to determine, based on a first predefined rule, antennas of a plurality of calibrating user equipments out of a plurality of user equipments, wherein the antennas of said plurality of calibrating user equipments are configured to calibrate antennas of said plurality of base stations and said first predefined rule is that the criteria for channel quality between said plurality of calibrating user equipments and the antennas configured for said plurality of base stations is above a threshold;
- a calibration relationship determining unit, configured to determine, based on a second predefined rule, antennas, to be calibrated respectively by the antennas of said plurality of calibrating user equipments, of said plurality of base stations, wherein said second predefined rule is that the number L of the antennas of said plurality of calibrating user equipments, the number K of antennas configured for said plurality of base stations and to be calibrated by the antenna of each of said plurality of calibrating user equipments, and the number M of the antennas of said plurality of base stations satisfy the relation of KL>M +L-1 ;
- a calibration unit, configured to calibrate, according to the determined calibration relationship, the antennas reciprocity of said plurality of base stations based on said plurality of calibrating user equipments, wherein said calibration relationship is the correspondence between said plurality of calibrating user equipments and the antennas, to be calibrated respectively by said plurality of calibrating user equipments, of said plurality of base stations.
13. The device of claim 12, wherein said calibration unit further

comprises:
- a first reciprocity-error ratio determining unit, configured
to determine the ratio of the reciprocity error of said plurality
of base stations relative to that of a first one among said plurality
of calibrating user equipments, according to the ratio of the
reciprocity error of said plurality of base stations to that of
said plurality of calibrating user equipments;
- a first BS-side uplink-channel calibration-weight-matrix generating unit, configured to generate a calibration weight matrix for the uplink channel on said base station side, according to the ratio of the reciprocity error of said plurality of base stations relative to that of said first one among said plurality of calibrating user equipments;
- a first uplink-channel transposed matrix generating unit, configure to determine a calibrated transposed matrix for the uplink channel, according to said calibration weight matrix for the uplink channel on said base station side and a transposed matrix for the uplink channel.
14. The device of claim 12, wherein said calibration unit further comprises:
- a second reciprocity-error ratio determining unit, configured
to determine the ratio of the reciprocity error of said plurality
of base stations relative to that of a first antenna among the
antennas configured for said plurality of base stations, according
to the ratio of the reciprocity error of said plurality of base
stations to that of said plurality of calibrating user equipments;
- a second BS-side uplink-channel calibration-weight-matrix
generating unit, configured to generate a calibration weight matrix
for the uplink channel on said base station side, according to

the ratio of the reciprocity error of said plurality of base stations relative to that of said first antenna among the antennas configured for said plurality of base stations;
- a second uplink-channel transposed matrix generating unit, configured to determine a calibrated transposed matrix for the uplink channel, according to said calibration weight matrix for the uplink channel on said base station side and a transposed matrix for the uplink channel.
15. The device of claim 13, further comprising:
a UE-side uplink-channel calibration-weight-matrix generating unit, configured to generate a calibration weight matrix for the uplink channel on said user equipment side, according to the ratio of the reciprocity error of said first one among said plurality of calibrating user equipments relative to that of said plurality of base stations;
wherein said first uplink-channel transposed matrix generating unit is further configured to determine said calibrated transposed matrix for the uplink channel, according to said calibration weight matrix for the uplink channel on said base station side, said transposed matrix for the uplink channel, and said calibration weight matrix for the uplink channel on said user equipment side.