CN104158773A - Channel estimation method and channel estimation device of multi-antenna system - Google Patents

Abstract

An embodiment of the invention provides a channel estimation method and a channel estimation device of a multi-antenna system. The channel estimation method comprises the following steps of: determining first channel estimation of an angle delay domain of an expected channel of a base station of lth cell of the current training period at each training period according to a pilot frequency signal; determining a multipath component of the expected channel of the current training period according to a preset threshold and channel estimation of the angle delay domain of the expected channel of the former training period; and determining second channel estimation of the angle delay domain of the expected channel of the current training period according to the multipath component of the expected channel of the current training period. According to the channel estimation method and the channel estimation device, which are provided by the embodiment of the invention, the multipath component corresponding to the expected channel is extracted by utilizing the preset threshold and the channel estimation of the angle delay domain of the expected channel of each training period, so that accurate channel estimation capable of eliminating pilot frequency pollution can be implemented.

Description

The channel estimation methods of multiaerial system and channel estimating apparatus

Technical field

The application relates to multiaerial system technical field, relates in particular to a kind of multiaerial system channel estimation methods and channel estimating apparatus.

Background technology

In recent years, multiaerial system forwards the large-scale application modern wireless cellular system to from theoretical research stage.At current wireless communication field, no matter be in real system or in theoretical research, the interference of minizone becomes a bottleneck already, and this bottleneck has seriously limited channel capacity and the throughput of communication system.In order to reduce presence of intercell interference, current study hotspot is that alignment techniques is disturbed in multi-user's multiple-input, multiple-output (MIMO, Multiple-Input Multiple-Output), and collaborative network.These technology have been proved to be and can have effectively reduced presence of intercell interference.Yet, in the face of being the data volume of exponential increase, comprise radiotelephone service, and ever-increasing wireless data demand, these technology can not fundamentally be brought the leap of power system capacity, also cannot meet user's demand.

Extensive MIMO (Massive MIMO) installs a large amount of antennas to improve the new technique of spatial resolution and antenna array gain in base station, this technology provides higher data transfer rate with lower power consumption, make the availability of frequency spectrum reach unprecedented level, therefore become the very potential candidate technologies of the 5th generation cellular network.

The subject matter of extensive MIMO is obtaining of instantaneous channel condition information (CSI, Channel State Information).In this wireless transmission scheme, base station end has the low-power miniature antenna of enormous amount, and number of antennas is considerably beyond the single antenna number of users of dispatching simultaneously.Extensive MIMO can make wireless communication system reach very high throughput, and when the antenna number of base station levels off to infinitely, channel capacity should be infinitely great.But in the scene of practical application, this is also false.Unique limiting factor is exactly pilot pollution (Pilot Contamination).In thering is the time division multiplexing of desirable channel reciprocity (TDD, Time-Division Duplex) system, can obtain CSI by ascending pilot frequency supplemental training.Due to the coherence time of finite length, the quantity of available orthogonal pilot frequency sequence is limited.Therefore, between several communities by multiplexed pilot sequence, it is not the channel between desired user and base station by the result that causes base station to carry out channel estimating that the user of neighbor cell sends identical pilot frequency sequence, but the training sequence being sent by the user of other communities pollutes estimation afterwards.Because the quantity of antenna levels off to infinitely, pilot pollution becomes the bottleneck of systematic function (downlink transfer performance).

Summary of the invention

The application's object is to provide a kind of multiaerial system channel estimation scheme.

According to the application's first aspect, a kind of channel estimation methods of multiaerial system is provided, described system comprises L community, and each community comprises base station and K the single antenna user with M root antenna, and described method comprises:

In each cycle of training, according to pilot signal, determine that the base station expectation channel angle of current cycle of training of l community postpones first channel estimating in territory;

According to a default thresholding and expect that before channel angle postpones the channel estimating in territory cycle of training, determine the multipath component of expecting channel described in described current cycle of training;

According to the multipath component of expecting channel described in described current cycle of training, determine and expect that the second channel that channel angle postpones territory estimates described in described current cycle of training;

Wherein, L, M and K are positive integer, l=(0 ..., L-1).

According to the application's second aspect, a kind of channel estimating apparatus of multiaerial system is provided, described system comprises L community, and each community comprises base station and K the single antenna user with M root antenna, and described device comprises:

One first determination module, in each cycle of training, determines that according to pilot signal the base station expectation channel angle of current cycle of training of l community postpones first channel estimating in territory;

One second determination module, for according to a default thresholding and expect that before channel angle postpones the channel estimating in territory cycle of training, determines the multipath component of expecting channel described in described current cycle of training;

According to the multipath component of expecting channel described in described current cycle of training, determine and expect that the second channel that channel angle postpones territory estimates described in described current cycle of training;

Wherein, L, M and l are positive integer, l=(0 ..., L-1).

The method of the embodiment of the present application and device are preset thresholding and are expected that channel angle postpones the channel estimating extraction in territory corresponding to the multipath component of expectation channel each cycle of training by utilizing, and then can realize the channel estimating accurately of eliminating pilot pollution.

Accompanying drawing explanation

Fig. 1 (a) and Fig. 1 (b) are the multiaerial system channel estimation methods of the embodiment of the present application and principle signal and the chart signal that the angle in device postpones territory;

Fig. 2 is the flow chart of the multiaerial system channel estimation methods of the embodiment of the present application;

Fig. 3 is the block diagram of the first implementation of the multiaerial system channel estimating apparatus of the embodiment of the present application;

Fig. 4 is a kind of block diagram of implementation of the first determination module of the multiaerial system channel estimating apparatus of the embodiment of the present application;

Fig. 5 is a kind of block diagram of implementation of the second determination module of the multiaerial system channel estimating apparatus of the embodiment of the present application;

The block diagram of the second implementation of the multiaerial system channel estimating apparatus of Fig. 6 the embodiment of the present application;

Fig. 7 is the block diagram of the third implementation of the multiaerial system channel estimating apparatus of the embodiment of the present application;

Fig. 8 is a kind of application scenarios schematic diagram of each embodiment of the application;

Fig. 9 is pilot frequency distribution schematic diagram in the example shown in Fig. 8;

Figure 10 (a) is respectively three angles of gathering multipath component cycles of training interior first in Fig. 8 example illustrated to Figure 10 (c) and postpones schematic diagram;

Figure 11 is the block diagram of the 4th kind of implementation of the multiaerial system channel estimating apparatus of the embodiment of the present application.

Embodiment

Below in conjunction with accompanying drawing (in some accompanying drawings, identical label represents identical element) and embodiment, the application's embodiment is described in further detail.Following examples are used for illustrating the application, but are not used for limiting the application's scope.

It will be understood by those skilled in the art that the terms such as " first " in the application, " second ", only for distinguishing different step, equipment or module etc., neither represent any particular technology implication, also do not represent the inevitable logical order between them.

Each embodiment of the application provides a kind of new channel estimation scheme of multiaerial system, multiaerial system refers to have the ofdm system of the antenna of infinite many or extensive quantity, this system has L community, each community is comprised of base station and an individual single antenna user of K (K≤M) with M root antenna, and L, M and K are positive integer.For the base station of arbitrary community, the user in its coverage is called desired user, and the channel between desired user and this base station is called expectation channel; Be called interference user with other community users of desired user multiplexed pilot, the channel between interference user and this base station becomes interference channel.In the multiaerial system of each embodiment of the application, expectation channel and interference channel are the multipath channel with sparse characteristic.Because each multipath component of expectation channel and interference channel is difficult to the same receiver side (base station) that delays to reach in condition of sparse channel, the expectation multipath component of channel and the multipath component of interference channel can divide in time domain, in addition, if Fig. 1 (a) is due to the dispersion of channel in angle, signal is also different by the angle of every footpath arrival antenna of channel, therefore, the expectation multipath component of channel and the multipath component of interference channel also can divide in angle domain, in the description of each embodiment of the application, signal is arrived to antenna by every footpath with different angles and time delay and be called " angle delay ", as shown in Fig. 1 (b), the non-zero position (square of filling) that angle postpones on territory represents that signal postpones to reach antenna with corresponding angle.If it is known that the angle in every footpath of channel postpones, can from contaminated channel estimating, extract the multipath component of expectation channel, can realize the channel estimating accurately of eliminating pilot pollution.

As shown in Figure 2, the multiaerial system channel estimation methods of the embodiment of the present application comprises:

S220. in each cycle of training, according to pilot signal, determine that the base station expectation channel angle of current cycle of training of l community postpones first channel estimating in territory, wherein, l=(0 ..., L-1).

In each embodiment of the application, " cycle of training " refers to that user is undertaken the cycle of channel training by ascending pilot frequency, is also the time interval of every uplink pilot signal of user.In step S220, the first channel estimating that postpones territory according to the determined angle of pilot signal is "ball-park" estimate, has wherein both comprised the estimation of the multipath component of expecting channel, also comprises the estimation of the multipath component of interference channel.

S240. according to a default thresholding and expect that before channel angle postpones the channel estimating in territory cycle of training, determine the multipath component of expecting channel described in described current cycle of training.

Because multipath channel presents sparse characteristic, multipath transmisstion consists of the stronger main footpath of energy (dominant paths) conventionally, therefore, most components in first channel estimating in angle delay territory are zero, and the part in nonzero component in corresponding footpath belongs to expectation channel, by belonging to, expect that the corresponding nonzero component in footpath of channel and the corresponding nonzero component in footpath that belongs to interference channel distinguish, and can eliminate pilot pollution.In step S240, according to default thresholding and before each cycle of training channel the angle channel estimating that postpones territory angle can be postponed to belong to and expect that the multipath component of channel extracts in the first corresponding footpath of channel estimating nonzero component in territory.

S260. determine in step S240 the multipath component of expecting channel described current cycle of training of determining, determine and expect that the second channel that channel angle postpones territory estimates described in described current cycle of training.The second channel of expecting channel angle delay territory described in described current cycle of training is estimated as accurate channel estimating, only comprises the estimation that belongs to the multipath component of expecting channel.

Determined in the first corresponding footpath of channel estimating nonzero component of expecting channel angle delay territory current cycle of training and belonged to after the multipath component of expecting channel, can determine the channel estimating of the multipath component of expectation channel.

To sum up, the method for the embodiment of the present application can be preset thresholding and expect that before channel angle postpones the channel estimating extraction in territory corresponding to the multipath component of expectation channel each cycle of training by utilizing, and then has realized the channel estimating accurately of eliminating pilot pollution.

In a kind of possible implementation, S220 further comprises:

S222. according to pilot signal, determine that each antenna space of base station of current cycle of training of l community postpones the channel estimating in territory:

${\hat{h}}_{l_m{l}_k}^b=h_{l_m{l}_k}^b+\underset{j=0,j\≠l}{\overset{L-1}{\mathrm{\Σ}}}{h}_{l_m{j}_{k_j}}^b+v_{l_m}$

Wherein, be illustrated in k user (being desired user) in b the cycle of training of l community to the multipath channel of the m root antenna of l cell base station, N _hthe length that represents channel impulse response (CIR, Channel Impulse Response), be illustrated in the k in b the cycle of training of j community _jindividual user (be interference user, determined by pilot frequency distribution result) is to the multipath channel of the m root antenna of l cell base station, b=(0 ..., B-1), k=(0 ..., K-1), j=(0 ... L-1) and j ≠ l, m=(0 ..., M-1) represent additive Gaussian noise.

Estimation of multipath channel from desired user on each antenna is expressed as expect that channel space postpones the channel estimating on territory

${\hat{H}}_{{\mathrm{ll}}_k}=\left[\begin{array}{c}{\hat{h}}_{l_0{l}_k}^T\\ {\hat{h}}_{l_1{l}_k}^T\\ .\\ .\\ .\\ {\hat{h}}_{l_{M-1}{l}_k}^T\end{array}\right]\∈C^{{M\×N}_h}$

Wherein, () ^trepresent transposition.

S224. according to the antenna array structure of the base station of described l community, determine the orthogonal basis vector of an angle domain:

For example, when aerial array is spacing while being the even linear array (ULA) of half-wavelength, this group orthogonal basis vector is { e ₀, e ₁..., e _m-1, wherein:

$e_m=\frac{1}{\sqrt{M}}\left[\begin{array}{c}1\\ \exp (-j2\mathrm{\π}\frac{m}{M})\\ .\\ .\\ .\\ \exp (-j2\mathrm{\π}(M-1)\frac{m}{M})\end{array}\right],m\∈\{\mathrm{0,1},...,M-1\}$

S226. according to described orthogonal basis vector, the channel estimating in delay territory, described space is converted to the channel estimating that angle postpones territory, obtains expecting that channel angle postpones first channel estimating in territory.Particularly:

This group orthogonal basis is formed to transformation matrix will being transformed into angle delay domain representation is: represent that angle postpones the channel estimating in territory, () ^hrepresent conjugate transpose.As above, because multipath channel presents sparse characteristic, multipath channel only comprises the main footpath that a small amount of energy is stronger, therefore matrix in only have oligo-element non-zero, and each nonzero element represents the multipath component of special angle and delay. in other a great number of elements value be approximately zero, the part in the corresponding footpath of nonzero component belongs to expectation channel, should will belong to expectation channel the corresponding nonzero component in footpath with belong to interference channel the corresponding nonzero component in footpath distinguish, to eliminate pilot pollution.In the method for the embodiment of the present application, step S240 further comprises:

S242. according to described default thresholding and described expectation channel angle, postpone first channel estimating in territory, determine that the angle of the first set multipath component postpones.This first set multipath component comprises the multipath component of expecting channel, also comprises the multipath component of interference channel.

Also, default thresholding T being set filters obtain some nonzero elements of middle mould value maximum, and record each nonzero element and exist in position (x, y), the angle that obtains the first set multipath component postpones S _b.S _bphysical significance be b cycle of training, in the corresponding angle of each multipath component postpone the set forming.Default thresholding can be set to relative threshold, gets front T multipath of energy maximum, (now T is integer, represents quantity); Default thresholding also can be set to absolute thresholding, gets all multipaths that energy is greater than T, (now T is real number, represents that angle postpones size).In addition, can default thresholding T be set by several modes.For example, according to statistical information setting before, etc.

S244. according to the channel estimating of expecting before the angle delay territory of channel cycle of training, (can be rough estimate and also can be accurate estimation, also be, can be the first channel estimating and also can be second channel estimation) the corresponding angle delay of multipath component and the common factor of the angle delay of described the first set multipath component, determine that the angle of the second set multipath component postpones: S ₀∩ S ₁∩ ..., ∩ S _b, the second set multipath component only comprises the multipath component of described expectation channel.

In step S260, directly extract the multipath component of described expectation channel, form the channel estimating that new angle postpones territory

${\hat{G}}_{l_m{l}_k}^a[x,y]=\left\{\begin{array}{cc}{\hat{H}}_{l_m{l}_k}^a[x,y],& (x,y)\∈S_0\∩S_1\∩,...,\∩S_b\\ 0,& ,(x,y)\∉S_0\∩S_1\∩,...,\∩S_b\end{array}\right.$

In a kind of possible implementation, can be according to above-mentioned orthogonal basis vector, the channel estimating that this angle is postponed to territory is transformed into delay territory, space.In this implementation, the method for the embodiment of the present application also comprises step:

S280. the second channel of expecting channel angle delay territory described in described current cycle of training is estimated to be converted to the channel estimating in delay territory, space multipath channel is projected on above-mentioned orthogonal basis vector, thereby make channel be transformed into angle from delay territory, space, postpones territory; It is its inverse process that channel is transformed into delay territory, space from angle delay territory.

It should be noted that, the method for the embodiment of the present application can be carried out by each base station, or is carried out by the device that is independent of base station, needs to dispatch according to the pilot frequency distribution scheme of system the up training of each community before carrying out channel estimating.The method of the embodiment of the present application also comprises step:

S200. determine the pilot frequency distribution scheme of described system.

Wherein, pilot frequency distribution scheme can be formulated by the device of carrying out the method for the embodiment of the present application, also can obtain from the outside of this device.

S210. according to described allocative decision, dispatch the up training of this community.

In a kind of possible implementation, in step S210 determined pilot frequency distribution scheme can be can randomization the pilot frequency distribution scheme of pollution in several continuous cycles of training.Tool is sayed it, and in a kind of possible implementation, in multiaerial system Ge community, number of users comprises K in maximum community _maxindividual user, represents pilot number to be allocated with K ', with B, represents quantity cycle of training, and K ' is for being not less than K _maxprime number, B≤K _max.Preferably, if K _maxfor prime number, get K '=K _maxif, K _maxbe not prime number, K ' gets and is slightly larger than K _maxprime number.For can randomization pollution in several continuous cycles of training, described pilot schemes comprises: in each cycle of training, the pilot frequency distribution of at least described desired user is given to the different user of other communities in described system.Can realize by the mode of cyclic shift above-mentioned distribution, for example, be the user assignment pilot tone of each community according to the following formula:

$A_l=\left(\begin{array}{cccccc}0& 1& 2& ...& K-2& K-1\\ l\mathrm{mod}K& (l+1)\mathrm{mod}K& (l+2)\mathrm{mod}K& ...& (l-2)\mathrm{mod}K& (l-1)\mathrm{mod}K\\ .& .& .& & .& .\\ .& .& .& ...& .& .\\ .& .& .& & .& .\\ \left[(B-1)l\right]\mathrm{mod}K& [(B-1)l+1]\mathrm{mod}K& [(B-1)l+2]\mathrm{mod}K& ...& [(B-1)l-2]\mathrm{mod}K& [(B-1)l-1]\mathrm{mod}K\end{array}\right)$

Wherein, A _lthe pilot frequency distribution matrix that represents l community, mod represents complementation computing, is that two numerical expressions are made the remainder after division arithmetic.A _lindex corresponding to pilot tone that distribute within b cycle of training of k user of l community of [b, k] individual element representation, B represents quantity cycle of training, and B≤K _max, K _maxthe number of users that represents the community that in described system, user is maximum, b=(0 ..., B-1).In above-mentioned matrix, secondary series is symmetric to the constant " C " in (B-1) l+C of the dividend on last row " mod " left side, and " K " in matrix gets K conventionally _max.In each embodiment of the application, pilot tone to be allocated is pre-designed, for sake of convenience, is pilot settings to be allocated index number separately.

Also it should be noted that, mode by above-mentioned cyclic shift is carried out pilot frequency distribution, all users that can make all communities in each cycle of training of suffered interference from different users, this is a kind of preferred implementation, also can be as required, only make individual user in each cycle of training of suffered interference from different users, correspondingly, the method for salary distribution of pilot tone can be the mode of any appropriate outside cyclic shift.

By the pollution in above-mentioned pilot frequency distribution scheme randomization B continuous cycle of training, by utilizing all B to there is the angle delay distribution that randomization is polluted, can obtain more accurately the delay accurately of extraction, the final multipath component postponing corresponding to accurately that extracts, and then obtain channel estimating more accurately, and needed prior information is few, do not need minizone cooperation, complexity is low.

It will be appreciated by those skilled in the art that, in the said method of the application's embodiment, the sequence number size of each step does not also mean that the priority of execution sequence, the execution sequence of each step should be definite with its function and internal logic, and should not form any restriction to the implementation process of the application's embodiment.

In addition, the embodiment of the present application also provides a kind of computer-readable medium, is included in the computer-readable instruction that carries out following operation while being performed: carry out the operation of each step of the method in above-mentioned Fig. 1 illustrated embodiment.

The application also provides a kind of multiaerial system channel estimating apparatus.This device can belong to or be independent of any base station in multiaerial system.In each embodiment of the application, base station refers to that " base station " broadly refers to the random node with the network terminal of terminal communication, such as Node B, eNode B, base station, AP (access point) etc.As shown in Figure 3, the channel estimating apparatus 300 of the embodiment of the present application comprises:

As shown in Figure 3, the multiaerial system channel estimation methods of the embodiment of the present application comprises:

The first determination module 320, in each cycle of training, determines that according to pilot signal the base station expectation channel angle of current cycle of training of l community postpones first channel estimating in territory, wherein, l=(0 ..., L-1).The first channel estimating that the first determination module 320 postpones territory according to the determined angle of pilot signal is "ball-park" estimate, has wherein both comprised the estimation of the multipath component of expecting channel, also comprises the estimation of the multipath component of interference channel.

The second determination module 340, for according to a default thresholding and expect that before channel angle postpones the channel estimating in territory cycle of training, determines the multipath component of expecting channel described in described current cycle of training.

Because multipath channel presents sparse characteristic, multipath transmisstion consists of the stronger main footpath of energy (dominant paths) conventionally, therefore, most components in first channel estimating in angle delay territory are zero, and the part in nonzero component in corresponding footpath belongs to expectation channel, by belonging to, expect that the corresponding nonzero component in footpath of channel and the corresponding nonzero component in footpath that belongs to interference channel distinguish, and can eliminate pilot pollution.The second determination module 340 according to default thresholding and before each cycle of training channel the part multipath component angle channel estimating that postpones territory angle can be postponed to belong to and expect that the multipath component of channel extracts in the first corresponding footpath of channel estimating nonzero component in territory.

The 3rd determination module 360, for determining the determined described multipath component of expecting channel current cycle of training of the second determination module 340, determines and expects that the second channel that channel angle postpones territory estimates described in described current cycle of training.The second channel that angle postpones territory is estimated as accurate channel estimating, only comprises the estimation that belongs to the multipath component of expecting channel.

Determined in the first corresponding footpath of channel estimating nonzero component of expecting channel angle delay territory current cycle of training and belonged to after the multipath component of expecting channel, can determine the channel estimating of the multipath component of expectation channel.

To sum up, the device of the embodiment of the present application can obtain the multipath component corresponding to expectation channel by the channel estimating of utilizing the angle of presetting thresholding and expecting before channel each cycle of training to postpone territory, and then has realized the channel estimating accurately of eliminating pilot pollution.

As shown in Figure 4, in a kind of possible implementation, the first determination module 320 can further comprise:

The first determining unit 322, for determining the base station of current cycle of training of l community according to pilot signal each antenna space postpones the channel estimating in territory:

${\hat{h}}_{l_m{l}_k}^b=h_{l_m{l}_k}^b+\underset{j=0,j\≠l}{\overset{L-1}{\mathrm{\Σ}}}{h}_{l_m{j}_{k_j}}^b+v_{l_m}$

Wherein, be illustrated in k user (being desired user) in b the cycle of training of l community to the multipath channel of the m root antenna of l cell base station, N _hthe length that represents channel impulse response, be illustrated in the k in b the cycle of training of j community _jindividual user (be interference user, determined by pilot frequency distribution result) is to the multipath channel of the m root antenna of l cell base station, b=(0 ..., B-1), k=(0 ..., K-1), j=(0 ... L-1) and j ≠ l, m=(0 ..., M-1) represent additive Gaussian noise.

Estimation of multipath channel from desired user on each antenna is expressed as expect that channel space postpones the channel estimating on territory

${\hat{H}}_{{\mathrm{ll}}_k}=\left[\begin{array}{c}{\hat{h}}_{l_0{l}_k}^T\\ {\hat{h}}_{l_1{l}_k}^T\\ .\\ .\\ .\\ {\hat{h}}_{l_{M-1}{l}_k}^T\end{array}\right]\∈C^{{M\×N}_h}$

Wherein, () ^trepresent transposition.

The second determining unit 324, for determine the orthogonal basis vector of an angle domain according to the antenna array structure of the base station of described l community:

For example, when aerial array is spacing while being the even linear array (ULA) of half-wavelength, this group orthogonal basis vector is { e ₀, e ₁..., e _m-1, wherein:

$e_m=\frac{1}{\sqrt{M}}\left[\begin{array}{c}1\\ \exp (-j2\mathrm{\π}\frac{m}{M})\\ .\\ .\\ .\\ \exp (-j2\mathrm{\π}(M-1)\frac{m}{M})\end{array}\right],m\∈\{\mathrm{0,1},...,M-1\}$

Converting unit 326, for the channel estimating in delay territory, described space being converted to the channel estimating that angle postpones territory according to described orthogonal basis vector, obtains expecting that channel angle postpones first channel estimating in territory.Particularly,

this group orthogonal basis is formed to transformation matrix, will be transformed into angle and postpone domain representation and be: represent that angle postpones the channel estimating in territory, () ^hrepresent conjugate transpose.As above, because multipath channel presents sparse characteristic, multipath channel only comprises the main footpath that a small amount of energy is stronger, therefore matrix in only have oligo-element non-zero, and each nonzero element represents the multipath component of special angle and delay. in other a great number of elements value be approximately zero, the part in the corresponding footpath of nonzero component belongs to expectation channel, should will belong to expectation channel the corresponding nonzero component in footpath with belong to interference channel the corresponding nonzero component in footpath distinguish, to eliminate pilot pollution.As shown in Figure 5, in the device of the embodiment of the present application, the second determination module 340 can further comprise:

The 3rd determining unit 342, for postponing first channel estimating in territory according to described default thresholding and described expectation channel angle, determines that the angle of the first set multipath component postpones.This first set multipath component comprises the multipath component of expecting channel, also comprises the multipath component of interference channel.

Also, default thresholding T being set filters obtain some nonzero elements of middle mould value maximum, and record each nonzero element and exist in position (x, y), the angle that obtains the first set multipath component postpones S _b.S _bphysical significance be b cycle of training, in the corresponding angle of each multipath component postpone the set forming.Default thresholding can be set to relative threshold, gets front T multipath of energy maximum, (now T is integer, represents quantity); Default thresholding also can be set to absolute thresholding, gets all multipaths that energy is greater than T, (now T is real number, represents that angle postpones size).In addition, can default thresholding T be set by several modes.For example, according to statistical information setting before, etc.

The 4th determining unit 344, for (can be rough estimate and also can be accurate estimation according to the channel estimating of expecting before channel angle delay territory cycle of training, also be, can be the first channel estimating also can be second channel and estimates) angle of corresponding multipath component postpones to postpone to occur simultaneously with the angle of described the first set multipath component, determines that the angle of the second set multipath component of described expectation channel postpones: S ₀∩ S ₁∩ ..., ∩ S _b, the second set multipath component only comprises the multipath component of described expectation channel.

The 3rd determination module 360 can directly extract the multipath component of described expectation channel, forms the channel estimating that new angle postpones territory

${\hat{G}}_{l_m{l}_k}^a[x,y]=\left\{\begin{array}{cc}{\hat{H}}_{l_m{l}_k}^a[x,y],& (x,y)\∈S_0\∩S_1\∩,...,\∩S_b\\ 0,& ,(x,y)\∉S_0\∩S_1\∩,...,\∩S_b\end{array}\right.$

In a kind of possible implementation, can be according to above-mentioned orthogonal basis vector, the channel estimating that this angle is postponed to territory is transformed into delay territory, space.In this implementation, as shown in Figure 6, the device 300 of the embodiment of the present application also can comprise:

Modular converter 380, for estimating that the second channel of expecting the angle delay territory of channel described in described current cycle of training be converted to the channel estimating in delay territory, space multipath channel is projected on above-mentioned orthogonal basis vector, thereby make channel be transformed into angle from delay territory, space, postpones territory; It is its inverse process that channel is transformed into delay territory, space from angle delay territory.

It should be noted that, the device of the embodiment of the present application needed to dispatch according to the pilot frequency distribution scheme of system the up training of each community before carrying out channel estimating.As shown in Figure 7, the device 300 of the embodiment of the present application also comprises:

The 4th determination module 311, for determining the pilot frequency distribution scheme of described system.

Wherein, pilot frequency distribution scheme can be formulated by the device of carrying out the method for the embodiment of the present application, also can obtain from the outside of this device.

Scheduler module 312, for dispatching the up training of this community according to described allocative decision.

In a kind of possible implementation, the determined pilot frequency distribution scheme of the 4th determination module 311 can be can randomization the pilot frequency distribution scheme of pollution in several continuous cycles of training.Tool is sayed it, and in a kind of possible implementation, in multiaerial system Ge community, number of users comprises K in maximum community _maxindividual user, represents pilot number to be allocated with K ', with B, represents quantity cycle of training, and K ' is for being not less than K _maxprime number, B≤K _max.Preferably, if K _maxfor prime number, get K '=K _maxif, K _maxbe not prime number, K ' gets and is slightly larger than K _maxprime number.For can randomization pollution in several continuous cycles of training, described pilot schemes comprises: in each cycle of training, the pilot frequency distribution of at least described desired user is given to the different user of other communities in described system.Can realize by the mode of cyclic shift above-mentioned distribution, for example, be the user assignment pilot tone of each community according to the following formula:

$A_l=\left(\begin{array}{cccccc}0& 1& 2& ...& K-2& K-1\\ l\mathrm{mod}K& (l+1)\mathrm{mod}K& (l+2)\mathrm{mod}K& ...& (l-2)\mathrm{mod}K& (l-1)\mathrm{mod}K\\ .& .& .& & .& .\\ .& .& .& ...& .& .\\ .& .& .& & .& .\\ \left[(B-1)l\right]\mathrm{mod}K& [(B-1)l+1]\mathrm{mod}K& [(B-1)l+2]\mathrm{mod}K& ...& [(B-1)l-2]\mathrm{mod}K& [(B-1)l-1]\mathrm{mod}K\end{array}\right)$

Wherein, A _lthe pilot frequency distribution matrix that represents l community, mod represents complementation computing, is that two numerical expressions are made the remainder after division arithmetic.A _lindex corresponding to pilot tone that distribute within b cycle of training of k user of l community of [b, k] individual element representation, B represents quantity cycle of training, and B≤K _max, K _maxthe number of users that represents the community that in described system, user is maximum, b=(0 ..., B-1).In above-mentioned matrix, secondary series is symmetric to the constant " C " in (B-1) l+C of the dividend on last row " mod " left side, and " K " in matrix gets K conventionally _max.In each embodiment of the application, pilot tone to be allocated is pre-designed, for sake of convenience, is pilot settings to be allocated index number separately.

Also it should be noted that, mode by above-mentioned cyclic shift is carried out pilot frequency distribution, all users that can make all communities in each cycle of training of suffered interference from different users, this is a kind of preferred implementation, also can be as required, only make individual user in each cycle of training of suffered interference from different users, correspondingly, the method for salary distribution of pilot tone can be the mode of any appropriate outside cyclic shift.

By the pollution in above-mentioned pilot frequency distribution scheme randomization B continuous cycle of training, by utilizing all B to there is the angle delay distribution that randomization is polluted, can obtain more accurately the delay accurately of extraction, the final multipath component postponing corresponding to accurately that extracts, and then obtain channel estimating more accurately, and needed prior information is few, do not need minizone cooperation, complexity is low.

With instantiation, further illustrate method and the device of each embodiment of the application below.

The ofdm system shown in Fig. 8 with N subcarrier of take is example, and this system comprises San Ge community, Cell 0, Cell 1 and Cell 2, also be l ∈ { 0,1,2}, in each community, comprise respectively there is the base station of M root antenna and three single antenna user UE 0, UE 1, UE2, be also K _max=3, k ∈ { 0,1,2}.The up training stage, pilots insertion is individual to the Np in N subcarrier, to survey each user's CSI.In order to simplify, suppose Np=N, also, all subcarriers are for channel detection.In order to eliminate the interference of multi-user in community, the pilot tone quadrature of the user in same community.Adopt time division mechanism to realize quadrature: each user monopolizes an OFDM symbol for training, and each cycle of training is K '=K altogether _max=3 OFDM symbols are used for the training of each community.Suppose the full time division multiplexing in minizone, these 3 OFDM symbols are shared by all communities, and then cause serious pilot pollution.Take community Cell 0 as example, in the up training stage, the pilot signal of desired user UE 0 in community Cell 0 (the thickest solid line) is subject to the interference (dotted line) of the interference user UE 0 in Cell 1 and Cell 2, cause the base station of Cell 0 inaccurate to the channel estimating of desired user UE 0, also have pilot pollution.In order to obtain CSI accurately, by the base station of Cell 0 or by being independent of the device of this base station according to the method for the embodiment of the present application, first determine the pilot frequency distribution scheme of each community:

According to number of users, definite pilot number K ' is 3, total B=K _max=3 cycles of training.3 pilot tones take respectively front 3 OFDM symbols of each cycle of training.Adopt the mode of cyclic shift to generate 3 users' of 3 communities in 3 cycles of training pilot frequency distribution scheme, make pilot pollution that all users are subject in each training from different user.Allocation matrix A with 3 * 3 _lthe index of k user shared OFDM symbol within b cycle of training of l community of [b, k] individual element representation.A _lthe 1st row element be [012], r row element is that r-1 row element is carried out to a ring shift left l position.The allocation matrix A obtaining ₀, A ₁, A ₂be respectively:

$\left(\begin{array}{ccc}0& 1& 2\\ 0& 1& 2\\ 0& 1& 2\end{array}\right),\left(\begin{array}{ccc}0& 1& 2\\ 1& 2& 0\\ 2& 0& 1\end{array}\right),\left(\begin{array}{ccc}0& 1& 2\\ 2& 0& 1\\ 1& 2& 0\end{array}\right),$

Element value in matrix is the shared notation index of user's pilot tone, and its corresponding pilot frequency distribution result as shown in Figure 9.

Within different cycles of training, the range value of multipath channel may be different (times change), yet the multidiameter of channel and multipath arrive angle substantially constant (with respect to amplitude, it is slow that angle postpones temporal evolution).

According to above-mentioned pilot frequency distribution scheme, carry out, after the up training scheduling of this community, according to following step, carrying out channel estimating:

In cycle of training 0, according to pilot signal, determine the channel estimating of current cycle of training of Cell each antenna of base station of 0:

${\hat{h}}_{l_m{l}_k}^0=h_{l_m{l}_k}^0+\underset{j=0,j\≠l}{\overset{L-1}{\mathrm{\Σ}}}{h}_{l_m{j}_{k_j}}^0+v_{l_m}$

Estimation of multipath channel from desired user on each antenna is expressed as

${\hat{H}}_{{\mathrm{ll}}_k}=\left[\begin{array}{c}{\hat{h}}_{l_0{l}_k}^T\\ {\hat{h}}_{l_1{l}_k}^T\\ .\\ .\\ .\\ {\hat{h}}_{l_{M-1}{l}_k}^T\end{array}\right]\∈C^{{M\×N}_h}$

Wherein, () ^trepresent transposition.

According to aerial array, be that spacing is the even linear array (ULA) of half-wavelength, determine that one group of orthogonal basis vector is { e ₀, e ₁..., e _m-1, wherein:

$e_m=\frac{1}{\sqrt{M}}\left[\begin{array}{c}1\\ \exp (-j2\mathrm{\π}\frac{m}{M})\\ .\\ .\\ .\\ \exp (-j2\mathrm{\π}(M-1)\frac{m}{M})\end{array}\right],m\∈\{\mathrm{0,1},...,M-1\}$

This group orthogonal basis is formed to transformation matrix will being transformed into angle delay domain representation is: represent that angle postpones the channel estimating in territory, () ^hrepresent conjugate transpose.

Be first cycle of training cycle of training 0, therefore do not have can be for reference before cycle of training, therefore, directly according to default thresholding T, determine some nonzero elements of middle mould value maximum, and record each nonzero element and exist in position (x, y), the angle that obtains the first set multipath component postpones S ₀, as shown in Figure 10 (a), wherein, the angle delay that the color lump that black is filled is the multipath component of expectation channel, the angle of the multipath component that the color lump that left oblique line is filled is interference channel postpones.

Directly extract this part multipath component, form the channel estimating that new angle postpones territory:

${\hat{G}}_{l_m{l}_k}^a[x,y]=\left\{\begin{array}{cc}{\hat{H}}_{l_m{l}_k}^a[x,y],& (x,y)\∈S_0\\ 0,& ,(x,y)\∉S_0\end{array}\right.$

The channel estimating that again this angle is postponed to territory is transformed into delay territory, space, obtains the channel estimating in delay territory, space:

In cycle of training 1, according to pilot signal, determine that current cycle of training, Cell 0 base station expectation channel angle postponed first channel estimating in territory and determine according to default thresholding T some nonzero elements of middle mould value maximum, and record each nonzero element in position (x, y), the angle that obtains the first set multipath component postpones S ₁, as shown in Figure 10 (b), wherein, the angle delay that the color lump that black is filled is the multipath component of expectation channel, the angle of the multipath component that the color lump that right oblique line is filled is interference channel postpones.According to pilot frequency distribution result, the 0th, the interference in 1 cycle of training is different, so S ₁with S ₀different.

According to expecting that the angle of channel postpones the corresponding angle delay of multipath component and the common factor of the angle delay of described the first set multipath component of channel estimating in territory, determines the angle delay of the second set multipath component of described expectation channel: S cycle of training 0 ₀∩ S ₁, the second set multipath component only comprises the multipath component of described expectation channel.

Directly extract the multipath component of described expectation channel, form the channel estimating that new angle postpones territory:

${\hat{G}}_{l_m{l}_k}^a[x,y]=\left\{\begin{array}{cc}{\hat{H}}_{l_m{l}_k}^a[x,y],& (x,y)\∈S_0\∩S_1\\ 0,& ,(x,y)\∉S_0\∩S_1\end{array}\right.$

The channel estimating that again this angle is postponed to territory is transformed into delay territory, space, obtains the channel estimating in delay territory, space:

In cycle of training 2, according to pilot signal, determine that current cycle of training, Cell 0 base station expectation channel angle postponed first channel estimating in territory and determine according to default thresholding T some nonzero elements of middle mould value maximum, and record each nonzero element in position (x, y), the angle that obtains the first set multipath component postpones S ₂, as shown in Figure 10 (c), wherein, the angle delay that the color lump that black is filled is the multipath component of expectation channel, the angle of the multipath component that the color lump that nicking is filled is interference channel postpones.According to pilot frequency distribution result, the 0th, the interference in 1,2 cycle of training is different, so S ₂with S ₁, S ₀different.

According to expecting that the angle of channel postpones the corresponding angle delay of multipath component and the common factor of the angle delay of described the first set multipath component of channel estimating in territory, determines the angle delay of the second set multipath component of described expectation channel: S cycle of training 0 ₀∩ S ₁∩ S ₂, the second set multipath component only comprises the multipath component of described expectation channel.

Directly extract the multipath component of described expectation channel, form the channel estimating that new angle postpones territory:

${\hat{G}}_{l_m{l}_k}^a[x,y]=\left\{\begin{array}{cc}{\hat{H}}_{l_m{l}_k}^a[x,y],& (x,y)\∈S_0\∩S_1\∩S_2\\ 0,& ,(x,y)\∉S_0\∩S_1\∩S_2\end{array}\right.$

The channel estimating that again this angle is postponed to territory is transformed into delay territory, space, obtains the channel estimating in delay territory, space:

In the present embodiment, through the up training of B=3 cycle of training, enough carry out accurately channel estimating.

The structural representation of the multiaerial system channel estimating apparatus 1100 that Figure 11 provides for the embodiment of the present application, the application's specific embodiment does not limit the specific implementation of multiaerial system channel estimating apparatus 1100.As shown in figure 11, this multiaerial system channel estimating apparatus 1100 can comprise:

Processor (processor) 1110, communication interface (Communications Interface) 1120, memory (memory) 1130 and communication bus 1140.Wherein:

Processor 1110, communication interface 1120 and memory 1130 complete mutual communication by communication bus 940.

Communication interface 1120, for the net element communication with such as client etc.

Processor 1110, for executive program 1132, specifically can realize the correlation function of multiaerial system channel estimating apparatus in the device embodiment of above-mentioned Fig. 3.

Particularly, program 1132 can comprise program code, and described program code comprises computer-managed instruction.

Processor 1110 may be a central processor CPU, or specific integrated circuit ASIC (Application Specific Integrated Circuit), or is configured to implement one or more integrated circuits of the embodiment of the present application.Program 1132 specifically can be for making described multiaerial system channel estimating apparatus 1100 carry out following steps:

In each cycle of training, according to pilot signal, determine that the base station expectation channel angle of current cycle of training of l community postpones first channel estimating in territory;

According to a default thresholding and expect that before channel angle postpones the channel estimating in territory cycle of training, determine the multipath component of expecting channel described in described current cycle of training;

According to the multipath component of expecting channel described in described current cycle of training, determine and expect that the second channel that channel angle postpones territory estimates described in described current cycle of training;

Wherein, L, M and K are positive integer, l=(0 ..., L-1).

In program 1132, the specific implementation of each step can, referring to description corresponding in the corresponding steps in above-described embodiment and unit, be not repeated herein.Those skilled in the art can be well understood to, and for convenience and simplicity of description, the specific works process of the equipment of foregoing description and module, can describe with reference to the corresponding process in preceding method embodiment, does not repeat them here.

Those skilled in the art can be well understood to, and for convenience and simplicity of description, the specific works process of the equipment of foregoing description and module, can describe with reference to the correspondence in aforementioned means embodiment, does not repeat them here.

Although theme described herein is the execution in computer system and providing in the general context carried out in binding operation system and application program, but also one skilled in the art will recognize that and can carry out other realizations in conjunction with the program module of other types.Generally speaking, program module comprises the structure of carrying out particular task or realizing routine, program, assembly, data structure and the other types of particular abstract data type.It will be appreciated by those skilled in the art that, this theme described herein can be put into practice by other computer system configurations, comprise portable equipment, multicomputer system, based on microprocessor or programmable consumer electronics, minicom, mainframe computer etc., also can use in the distributed computing environment (DCE) that task is carried out by the teleprocessing equipment connecting by communication network therein.In distributed computing environment (DCE), program module can be arranged in both of local and remote memory storage device.

Those of ordinary skills can recognize, unit and the method step of each example of describing in conjunction with embodiment disclosed herein, can realize with the combination of electronic hardware or computer software and electronic hardware.These functions are carried out with hardware or software mode actually, depend on application-specific and the design constraint of technical scheme.Professional and technical personnel can specifically should be used for realizing described function with distinct methods to each, but this realization should not thought and exceeds the application's scope.

If described function usings that the form of SFU software functional unit realizes and during as production marketing independently or use, can be stored in a computer read/write memory medium.Understanding based on such, the part that the application's technical scheme contributes to original technology in essence in other words or the part of this technical scheme can embody with the form of software product, this computer software product is stored in a storage medium, comprise that some instructions are with so that a computer equipment (can be personal computer, server, or the network equipment etc.) carry out all or part of step of method described in each embodiment of the application.And aforesaid computer read/write memory medium comprises storing as any mode of the information such as computer-readable instruction, data structure, program module or other data or physics volatibility that technology realizes and non-volatile, removable and can not be because of eastern medium.Computer read/write memory medium specifically comprises, but be not limited to, USB flash disk, portable hard drive, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), Erasable Programmable Read Only Memory EPROM (EPROM), EEPROM (Electrically Erasable Programmable Read Only Memo) (EEPROM), flash memory or other solid-state memory technology, CD-ROM, digital versatile disc (DVD), HD-DVD, blue light (Blue-Ray) or other light storage devices, tape, disk storage or other magnetic storage apparatus, maybe can be for storing information needed and can be by any other medium of computer access.

Above execution mode is only for illustrating the application; and the not restriction to the application; the those of ordinary skill in relevant technologies field; in the situation that do not depart from the application's spirit and scope; can also make a variety of changes and modification; therefore all technical schemes that are equal to also belong to the application's category, and the application's scope of patent protection should be defined by the claims.

Claims (18)

1. a channel estimation methods for multiaerial system, is characterized in that, described system comprises L community, and each community comprises base station and K the single antenna user with M root antenna, and described method comprises:

In each cycle of training, according to pilot signal, determine that the base station expectation channel angle of current cycle of training of l community postpones first channel estimating in territory;

According to a default thresholding and expect that before channel angle postpones the channel estimating in territory cycle of training, determine the multipath component of expecting channel described in described current cycle of training;

According to the multipath component of expecting channel described in described current cycle of training, determine and expect that the second channel that channel angle postpones territory estimates described in described current cycle of training;

Wherein, L, M and K are positive integer, l=(0 ..., L-1).

2. method according to claim 1, is characterized in that, described method also comprises:

Determine the pilot frequency distribution scheme of described system;

According to described allocative decision, dispatch the up training of this community.

3. method according to claim 2, is characterized in that, described expectation channel is that desired user is to the channel between base station;

Described pilot schemes comprises: in each cycle of training, the pilot frequency distribution of at least described desired user is given to the different user of other communities in described system.

4. method according to claim 3, is characterized in that, in the described pilot frequency distribution scheme of determining described system, and the user assignment pilot tone that the mode by cyclic shift is each community.

5. method according to claim 4, is characterized in that, is that the pilot frequency distribution matrix of user assignment pilot tone of each community is as follows in described pilot schemes:

$A_l=\left(\begin{array}{cccccc}0& 1& 2& ...& K-2& K-1\\ l\mathrm{mod}K& (l+1)\mathrm{mod}K& (l+2)\mathrm{mod}K& ...& (l-2)\mathrm{mod}K& (l-1)\mathrm{mod}K\\ .& .& .& & .& .\\ .& .& .& ...& .& .\\ .& .& .& & .& .\\ \left[(B-1)l\right]\mathrm{mod}K& [(B-1)l+1]\mathrm{mod}K& [(B-1)l+2]\mathrm{mod}K& ...& [(B-1)l-2]\mathrm{mod}K& [(B-1)l-1]\mathrm{mod}K\end{array}\right)$

Wherein, A _lthe pilot frequency distribution matrix that represents l community, A _lindex corresponding to pilot tone that distribute within b cycle of training of k user of l community of [b, k] individual element representation, B represents quantity cycle of training, and B≤K _max, K _maxthe number of users that represents the community that in described system, user is maximum, b=(0 ..., B-1).

6. method according to claim 3, is characterized in that, described pilot schemes also comprises: the pilot tone quadrature of distributing to the user in identical community; User's multiplexed pilot in different communities.

7. according to the method described in any one in claim 1 to 6, it is characterized in that, describedly according to pilot signal, determine that the first channel estimating that the base station expectation channel angle of current cycle of training of l community postpones territory comprises:

The base station expectation channel space of determining current cycle of training of l community according to pilot signal postpones the channel estimating in territory;

According to the antenna array structure of the base station of described l community, determine the orthogonal basis vector of an angle domain;

According to described orthogonal basis vector, the channel estimating in delay territory, described space is converted to the channel estimating that angle postpones territory.

8. according to the method described in any one in claim 1 to 7, it is characterized in that, described according to a default thresholding and expect that before channel angle postpones the channel estimating in territory cycle of training, determine described in described current cycle of training and expect that the multipath component of channel comprises:

The first channel estimating that postpones territory according to the described expectation channel angle of described default thresholding and described current cycle of training, determines that the angle of the first set multipath component postpones;

According to expecting that before channel angle postpones the corresponding angle delay of multipath component and the common factor of the angle delay of described the first set multipath component of channel estimating in territory, determines the second set multipath component cycle of training.

9. according to the method described in any one in claim 1 to 8, it is characterized in that, described method also comprises:

The second channel of expecting channel angle delay territory described in described current cycle of training is estimated to be converted to the channel estimating in delay territory, space.

10. a channel estimating apparatus for multiaerial system, is characterized in that, described system comprises L community, and each community comprises base station and K the single antenna user with M root antenna, and described device comprises:

One first determination module, in each cycle of training, determines that according to pilot signal the base station expectation channel angle of current cycle of training of l community postpones first channel estimating in territory;

One second determination module, for according to a default thresholding and expect that before the angle of channel postpones the channel estimating in territory cycle of training, determines the multipath component of expecting channel described in described current cycle of training;

One the 3rd determination module, for according to the multipath component of expecting channel described in described current cycle of training, determines that the second channel that the angle of expecting channel described in described current cycle of training postpones territory estimates;

Wherein, L, M and K are positive integer, l=(0 ..., L-1).

11. devices according to claim 10, is characterized in that, described device also comprises:

One the 4th determination module, for determining the pilot frequency distribution scheme of described system;

One scheduler module, for dispatching the up training of this community according to described allocative decision.

12. devices according to claim 11, is characterized in that, described expectation channel is that desired user is to the channel between base station;

The described pilot schemes that described the 4th determination module is determined comprises: in each cycle of training, the pilot frequency distribution of at least described desired user is given to the different user of other communities in described system.

13. devices according to claim 12, is characterized in that, described the 4th determination module is given the pilot frequency distribution of at least described desired user the different user of other communities in described system by the mode of cyclic shift.

14. devices according to claim 13, is characterized in that, the user assignment pilot tone that described the 4th determination module is each community according to pilot frequency distribution matrix below:

$A_l=\left(\begin{array}{cccccc}0& 1& 2& ...& K-2& K-1\\ l\mathrm{mod}K& (l+1)\mathrm{mod}K& (l+2)\mathrm{mod}K& ...& (l-2)\mathrm{mod}K& (l-1)\mathrm{mod}K\\ .& .& .& & .& .\\ .& .& .& ...& .& .\\ .& .& .& & .& .\\ \left[(B-1)l\right]\mathrm{mod}K& [(B-1)l+1]\mathrm{mod}K& [(B-1)l+2]\mathrm{mod}K& ...& [(B-1)l-2]\mathrm{mod}K& [(B-1)l-1]\mathrm{mod}K\end{array}\right)$

Wherein, A _lthe pilot frequency distribution matrix that represents l community, A _lindex corresponding to pilot tone that distribute within b cycle of training of k user of l community of [b, k] individual element representation, B represents quantity cycle of training, and B≤K _max, K _maxthe number of users that represents the community that in described system, user is maximum, b=(0 ..., B-1).

15. devices according to claim 12, is characterized in that, the described pilot schemes that described the 4th determination module is determined also comprises: the pilot tone quadrature of distributing to the user in identical community; User's multiplexed pilot in different communities.

16. according to claim 10 to the device described in any one in 15, it is characterized in that, described the first determination module comprises:

One first determining unit, for determining that according to pilot signal the base station expectation channel space of current cycle of training of l community postpones the channel estimating in territory;

One second determining unit, for determining the orthogonal basis vector of an angle domain according to the antenna array structure of the base station of described l community;

One converting unit, for being converted to by the channel estimating in delay territory, described space the channel estimating that angle postpones territory according to described orthogonal basis vector.

17. according to claim 10 to the device described in any one in 16, it is characterized in that, described the second determination module comprises:

One the 3rd determining unit, for postpone first channel estimating in territory according to the described expectation channel angle of described default thresholding and described current cycle of training, determines that the angle of the first set multipath component postpones;

One the 4th determining unit, for the common factor with the angle delay of described the first set multipath component according to the angle delay of multipath component corresponding to the channel estimating of expecting before channel angle delay territory cycle of training, determines the second set multipath component.

18. according to claim 10 to the method described in any one in 17, it is characterized in that, described method also comprises:

One modular converter, for estimating that the second channel of expecting channel angle delay territory described in described current cycle of training be converted to the channel estimating in delay territory, space.