CN101557377A - Method, device and system for calculation of pre-filtering coefficient and interference suppression - Google Patents

Abstract

The embodiment of the invention relates to a method, a device and a system for calculation of pre-filtering coefficient and interference suppression, wherein the device comprises a pre-filtering coefficient calculation device and a transmitter or a receiver; the interference suppression method includes: obtaining partial channel information that comprises partial channel response according to the feedback of a receiving end; calculating the pre-filtering coefficient according to the obtained partial channel information; and sending a time domain signal after the pre-filtering treatment is conducted according to the obtained pre-filtering coefficient. By conducting pre-filtering treatment similar to channel shortening equalization on the time domain signal to be sent at a transmission end, the embodiment causes that when the receiving end receives the time domain signal, the energy of equivalent channel response is concentrated in the range shorter than cyclic prefix length as far as possible, thus being capable of remarkably reducing the implementation complexity needed by the receiving end to suppress interference and leading to good interference suppression effect.

Description

Pre-filtering coefficient calculating and disturbance restraining method, device and system

Technical field

The present invention and embodiment relate to the wireless broadband communication field, particularly relate to a kind of pre-filtering coefficient calculating and disturbance restraining method, device and system.

Background technology

OFDM (Orthogonal Frequency Division Multiplexing, abbreviation OFDM) single-carrier system (the single carrier systemwith frequency domain equalization of system or employing frequency-domain equalization technology, be called for short SC-FDE), usually by inserting Cyclic Prefix (Cyclic Prefix, abbreviation CP) eliminates the intersymbol interference (Inter Symbol Interference is called for short ISI) that the wireless channel multipath transmisstion brings.When the length of CP during more than or equal to the channel delay extension length, can eliminate fully in theory because intersymbol interference (the Inter Symbol Interference that wireless channel multipath transmisstion effect produces, be called for short ISI), keep the orthogonality between subcarrier in frequency domain, this means that will there be inter-block-interference (Inter Block Interference in receiving terminal under the prerequisite of ideal synchronisation, be called for short IBI) and inter-carrier interference (Inter CarrierInterference is called for short ICI).But, when the channel delay extension length greater than CP length, when causing alleged usually " CP deficiency ", even if having taked corresponding sign synchronization technology disturbs to reduce, but IBI and ICI still are difficult to avoid, generally can reduce the influence of IBI and ICI, guarantee receptivity by in receiver, adopting interference mitigation technology.

The IBI inhibition technology of an existing quasi-representative is that (Residual ISI cancellation is eliminated in so-called residual intersymbol interference, be called for short RISIC) and similar approach, its basic ideas are by feeding back based on demodulation block decision-feedback or decoding, come the previous symbolic blocks of reconstruct to the I B I of current sign piece and the CP part of current sign piece disappearance, before FFT, the two is compensated then in time domain.This Technology Need carries out interference eliminated based on iteration, and well-known, iteration receives the extra process time delay and the amount of calculation that cause will increase the complexity of receiver greatly.

The IBI inhibition technology of an existing other quasi-representative is at signal receiving end the time domain received signal to be carried out channel to shorten (Channel Shorten) equilibrium, make the concentration of energy of equivalent channel (Equivalent Channel) response arrive in the CP length range, thereby minimize IBI and ICI influence, guarantee the orthogonality of subcarrier in frequency domain; Equivalent channel is the cascade system response that actual channel and channel shorten equalizer.

The inventor finds in realizing embodiment of the invention process, because the time domain channel response tap number is a lot of in the broadband system, correspondingly will cause time domain channel to shorten equalizing coefficient needs in asking for large matrix to invert or decompose, computational complexity is very high, based on current application receiver existing force comparatively widely, the realization difficulty is big, is unfavorable for promoting the use of.

Summary of the invention

Embodiment of the invention first aspect provides a kind of disturbance restraining method, device and system, in order to the length of Cyclic Prefix less than channel response total length situation under, reduce receiving terminal and suppress the computational complexity that disturbs.

Embodiment of the invention second aspect provides a kind of pre-filtering coefficient computational methods and device, in order to obtain pre-filtering coefficient, adopt this pre-filtering coefficient that signal is carried out pre-filtering and carry out interference eliminated, help reducing the computational complexity of transmitting terminal or receiving terminal inhibition interference.

The embodiment of the invention provides a kind of disturbance restraining method, comprising:

Feedback according to receiving terminal obtains partial channel knowledge, and described partial channel knowledge comprises the local channel response;

According to the partial channel knowledge that obtains, calculate pre-filtering coefficient;

According to the described pre-filtering coefficient that obtains, time-domain signal is carried out pre-filtering handle the back transmission.

The embodiment of the invention also provides another kind of disturbance restraining method, comprising:

Obtain time domain channel response by channel estimating;

According to described time domain channel response, extract partial channel knowledge and, calculate pre-filtering coefficient according to described partial channel knowledge, and according to described pre-filtering coefficient time-domain signal to be sent is carried out pre-filtering and handle for described transmitting terminal to the transmitting terminal feedback; Described partial channel knowledge comprises the local channel response.

The embodiment of the invention also provides a kind of transmitter, comprising:

The channel information acquisition module is used for obtaining partial channel knowledge according to the feedback of receiving terminal, and described partial channel knowledge comprises the local channel response;

The pre-filtering coefficient computing module is used for calculating pre-filtering coefficient according to the partial channel knowledge that obtains;

The pre-filtering processing module is used for according to the described pre-filtering coefficient that obtains, and time-domain signal is carried out pre-filtering handle the back transmission.

The embodiment of the invention also provides a kind of receiver, comprising:

Channel estimation module is used for obtaining time domain channel response by channel estimating;

Feedback module, be used for according to described time domain channel response, extract partial channel knowledge and, calculate pre-filtering coefficient according to described partial channel knowledge, and according to described pre-filtering coefficient time-domain signal to be sent is carried out pre-filtering and handle for described transmitting terminal to the transmitting terminal feedback; Described partial channel knowledge comprises the local channel response.

The embodiment of the invention also provides a kind of Interference Suppression System, comprises transmitter and receiver;

Described transmitter is used for obtaining partial channel knowledge according to the feedback of described receiver, calculate pre-filtering coefficient according to the partial channel knowledge that obtains, according to the described pre-filtering coefficient that obtains time-domain signal is carried out pre-filtering and handle the back transmission, described partial channel knowledge comprises the local channel response;

Described receiver is used for obtaining time domain channel response by channel estimating; Extract described partial channel knowledge and to described transmitter feedback from described time domain channel response.

Disturbance restraining method that the embodiment of the invention provides and system, transmitter and receiver, carrying out the pre-filtering processing that similar channel shortens equilibrium to sent time-domain signal by transmitting terminal, make receiving terminal receive in the time-domain signal, the energy of equivalent channel response concentrates in the scope that is shorter than circulating prefix-length as far as possible, for the existing interference mitigation technology in the background technology, can significantly reduce receiving terminal and disturb required implementation complexity, and have good interference suppressioning effect in order to suppress.Realize that based on the receiver of the embodiment of the invention principle is very simple, signal receiving performance also reaches or is better than prior art.

The embodiment of the invention also provides a kind of pre-filtering coefficient computational methods, comprising:

According to the channel response total length that obtains in advance, signal calculated power calculation matrix D _sWith interference power compute matrix D _I

According to the signal power compute matrix D that obtains _sWith interference power compute matrix D _IAnd the local channel that obtains in advance response, calculate constraint matrix and objective matrix;

According to constraint matrix that obtains and objective matrix, calculate the pre-filtering coefficient of correspondence when Signal to Interference plus Noise Ratio is maximum, described Signal to Interference plus Noise Ratio equals the ratio of signal power and interference noise power.

The embodiment of the invention also provides a kind of pre-filtering coefficient calculation element, comprising:

Preset matrix calculation unit, be used for according to the channel response total length that obtains in advance, signal calculated power calculation matrix and interference power compute matrix;

Objective matrix and constraint matrix computing unit are used for calculating constraint matrix and objective matrix according to the signal power compute matrix that obtains and interference power compute matrix and the local channel response of obtaining in advance;

Pre-filtering coefficient is asked for the unit, is used for calculating the pre-filtering coefficient of correspondence when Signal to Interference plus Noise Ratio is maximum according to the constraint matrix and the objective matrix that obtain, and described Signal to Interference plus Noise Ratio equals the ratio of signal power and interference noise power.

The computational methods of the pre-filtering coefficient that the embodiment of the invention provides and device, can use but be not limited in OFDM or the SC-FDE system, in order to obtain pre-filtering coefficient, so that transmitting terminal or receiving terminal carry out interference eliminated according to pre-filtering coefficient to the signal that receives or send; Carry out interference eliminated based on this pre-filtering coefficient, help reducing the computational complexity of transmitting terminal or receiving terminal inhibition interference.

Description of drawings

In order to be illustrated more clearly in the embodiment of the invention or technical scheme of the prior art, to do to introduce simply to the accompanying drawing of required use in embodiment or the description of the Prior Art below, apparently, accompanying drawing in describing below only is some embodiments of the present invention, for those of ordinary skills, under the prerequisite of not paying creative work, can also obtain other accompanying drawing according to these accompanying drawings.

The disturbance restraining method flow chart that Fig. 1 provides for first embodiment of the invention;

The disturbance restraining method flow chart that Fig. 2 provides for second embodiment of the invention;

The disturbance restraining method that Fig. 3 provides for third embodiment of the invention is at the realization block diagram of ofdm system application scenarios;

A kind of ofdm system frame structure schematic diagram that Fig. 4 provides for fourth embodiment of the invention;

The pre-filtering coefficient computational methods flow chart that Fig. 5 provides for fifth embodiment of the invention;

Fig. 6 is that the performance of embodiment of the invention disturbance restraining method and existing disturbance restraining method compares schematic diagram one;

Fig. 7 is that the performance of embodiment of the invention disturbance restraining method and other disturbance restraining methods compares schematic diagram two;

The transmitter architecture figure that Fig. 8 provides for sixth embodiment of the invention;

The receiver structure figure that Fig. 9 provides for seventh embodiment of the invention;

The Interference Suppression System structure chart that Figure 10 provides for eighth embodiment of the invention;

The pre-filtering coefficient computing device structure figure that Figure 11 provides for ninth embodiment of the invention.

Embodiment

Below in conjunction with the accompanying drawing in the embodiment of the invention, the technical scheme in the embodiment of the invention is clearly and completely described, obviously, described embodiment only is the present invention's part embodiment, rather than whole embodiment.Based on the embodiment among the present invention, those of ordinary skills belong to the scope of protection of the invention not making the every other embodiment that is obtained under the creative work prerequisite.

The embodiment of the invention can be applicable to ofdm system, adopt the single-carrier system of frequency domain equalization or symbolization piece to carry out disturbing inhibition in other broadband system that signal sends.For the channel response total length of actual channel situation greater than circulating prefix-length, shorten balanced pre-filtering processing by carrying out similar channel to sent time-domain signal at transmitting terminal, make receiving terminal receive in the time-domain signal, the energy of equivalent channel response concentrates in the scope that is shorter than circulating prefix-length as far as possible, thereby minimise interference, the orthogonality of assurance subcarrier in frequency domain.The embodiment of the invention makes receiving terminal need not to carry out complicated interference inhibition and handles, only need to adopt common simple single tap equalization to finish and receive processing, thereby can significantly reduce receiving terminal and disturb required implementation complexity, and has good interference suppressioning effect in order to suppress.

The disturbance restraining method flow chart that Fig. 1 provides for first embodiment of the invention.Present embodiment describes from sending distolateral technical scheme to the embodiment of the invention.As shown in Figure 1, present embodiment comprises:

Step 11, obtain partial channel knowledge according to the feedback of receiving terminal, described partial channel knowledge comprises: the local channel response.

The local channel response can include but not limited to that actual channel exceeds the channel response of Cyclic Prefix (CP) length part.

In the present embodiment, the channel response total length can be greater than circulating prefix-length.The mode that transmitting terminal obtains the channel response total length can comprise: transmitting terminal is estimated acquisition channel response total length in advance; Perhaps, receiving terminal obtains the channel response total length, and the channel response total length is fed back to transmitting terminal, and under this situation, receiving terminal also can comprise the channel response total length in the partial channel knowledge of transmitting terminal feedback.

Because present embodiment is handled in the pre-filtering (Pre-Processing Filter) that transmitting terminal carries out similar channel isostatic compression (Channel ShorteningEqualization is called for short CSE), therefore needs receiving terminal to transmitting terminal feedback fraction channel response.Receiving terminal can exceed actual channel the response of CP length part, and/or actual channel do not exceed the response of CP length part, as the response of the local channel in the embodiment of the invention, feeds back to transmitter.

Step 12, according to the partial channel knowledge that obtains, calculate pre-filtering coefficient.

The pre-filtering coefficient that step 13, basis obtain carries out pre-filtering to time-domain signal (as: time domain OFDM modulation signal) and handles the back transmission.

Described time-domain signal is for sending symbol through ovennodulation (as: OFDM modulation) and added the time-domain signal behind the CP.

Present embodiment is at the situation of channel response total length greater than Cyclic Prefix (CP) length, carrying out the pre-filtering processing that similar channel shortens equilibrium to sent the OFDM time-domain signal by transmitting terminal, make receiving terminal receive in the time-domain signal, the energy of equivalent channel response concentrates in the scope that is shorter than CP length as far as possible, thereby guarantee that suffered IBI of received signal and ICI are effectively suppressed, for the existing interference mitigation technology in the background technology, present embodiment has reduced receiving terminal in order to suppress to disturb required computational complexity and processing delay, has obviously reduced the difficulty that realizes.

The embodiment of the invention can be applicable to ofdm system or receiving terminal and carries out in other broadband system that signal sends based on the single-carrier system of frequency-domain equalization technology or symbolization piece, particularly be applied in the downlink transfer of said system, can obviously reduce the implementation complexity of receiving terminal when considering that interference that the CP curtailment causes suppresses.

The disturbance restraining method flow chart that Fig. 2 provides for second embodiment of the invention.Present embodiment is that with the difference of the corresponding embodiment of Fig. 1 present embodiment describes from receiving distolateral technical scheme to the embodiment of the invention.As shown in Figure 2, present embodiment comprises:

Step 21, carry out channel estimating to obtain time domain channel response.

Receiving terminal can be known channel response and channel response total length by channel estimating.

Step 22, according to described time domain channel response, obtain partial channel knowledge and feed back to transmitting terminal, calculate pre-filtering coefficient for described transmitting terminal according to described partial channel knowledge, and carry out pre-filtering according to described pre-filtering coefficient to sent time-domain signal and handle.

Described partial channel knowledge comprises the local channel response.

The local channel response can include but not limited to that actual channel exceeds the channel response of Cyclic Prefix (CP) length part.

In the present embodiment, the channel response total length can be greater than circulating prefix-length.Receiving terminal can obtain the channel response total length, and the channel response total length is fed back to transmitting terminal, and under this situation, receiving terminal also can comprise the channel response total length in the partial channel knowledge of transmitting terminal feedback.

Present embodiment is at the channel response total length of the actual channel situation greater than Cyclic Prefix (CP) length, and what be different from prior art is: shorten balanced pre-filtering and handle carrying out channel to sent time-domain signal by transmitting terminal.Because transmitting terminal can't be informed in time-domain signal channel transmitted information in the communication process, also can't estimate channel response, therefore, need receiving terminal with partial channel knowledge, feed back to transmitting terminal, carry out channel for transmitting terminal according to the partial channel knowledge that obtains and shorten balanced pre-filtering processing, make receiving terminal receive in the time-domain signal, the energy of equivalent channel response concentrates in the scope that is shorter than CP length as far as possible, for the existing interference mitigation technology in the background technology, present embodiment has reduced receiving terminal in order to suppress to disturb required computational complexity and processing delay, has obviously reduced the difficulty that realizes.

The disturbance restraining method that Fig. 3 provides for third embodiment of the invention is at the realization block diagram of ofdm system application scenarios.Present embodiment is from transmitting terminal and receiving terminal two sides, is embodied as example with the embodiment of the invention under the ofdm system application scenarios, and the technical scheme of the embodiment of the invention is described.It will be understood by those skilled in the art that following technical scheme is equally applicable to adopt the single-carrier system of frequency domain equalization, after module of illustrating among Fig. 3 or device are done to change slightly, also can be used for adopting the single-carrier system of frequency domain equalization fully.

With reference to the realization block diagram of Fig. 3, present embodiment disturbs the implementation method of inhibition to comprise:

The channel estimation module 91 of step 301, receiving terminal carries out channel estimating to obtain time domain channel response.

Because the present embodiment pre-filtering is handled at transmitting terminal and is realized that therefore, receiving terminal need be to transmitting terminal feedback fraction channel information.Receiving terminal carries out channel estimating according to the signal that receives, obtain time domain channel response, and, carry out the calculating of signal pre-filtering coefficient in the subsequent data frame (as: next Frame etc.) according to this partial channel knowledge for transmitting terminal to transmitting terminal feedback fraction channel information.

Feedback channel response extracting unit 921 in step 302, the receiving terminal extracts partial information from the time domain channel response of described acquisition.

The partial information of the 922 pairs of extractions of feedback channel response quantifying unit in step 303, the receiving terminal is carried out quantification treatment, and gives transmitting terminal with the feedback information after the quantification treatment.

Information after the quantification treatment can comprise: local channel response and channel response total length.

Quantification treatment can comprise: the floating number in the partial information that extracts is changed into fixed-point number, feed back to transmitting terminal to make things convenient for receiving terminal.

31 pairs of symbols to be sent of OFDM modulation module in step 304, the transmitting terminal carry out the OFDM modulation, obtain time domain OFDM signal.

The time domain OFDM signal that this step obtains is for sending symbol { s _n[k] _K=0 ^K-1} _N=-∞ ^{+ ∞}Time-domain signal through the 31 modulation back acquisitions of OFDM modulation module.Wherein, { s _n[k] _K=0 ^K-1} _N=-∞ ^{+ ∞}In n be OFDM symbolic blocks sequence number, k represents OFDM symbolic blocks sub-carrier sequence number, K represents total sub-carrier number.To symbol { s to be sent _n[k] _K=0 ^K-1} _N=-∞ ^{+ ∞}Carry out the OFDM modulation, mainly comprise common serial to parallel conversion, IFFT (anti-Fourier) conversion, parallel serial conversion, insert CP, obtain time domain OFDM signal x[n].X[n] represent that the n here is the express time meaning only, with { s through the time-domain signal of OFDM modulation and insertion CP _n[k] _K=0 ^K-1} _N=-∞ ^{+ ∞}In n do not have direct corresponding relation.

Pre-filtering module 32 in step 305, the transmitting terminal is calculated pre-filtering coefficient based on the partial channel knowledge of feedback, carries out pre-filtering according to pre-filtering coefficient to sent time domain OFDM signal.

Pre-filtering coefficient can calculate based on maximum Signal to Interference plus Noise Ratio criterion (Maximum Signal to Interferenceplus Noise Ratio is called for short MSINR), and the specific implementation method sees the record of the corresponding embodiment of Fig. 5 for details.

Signal after transmitting terminal is handled pre-filtering sends the signal y[n that receiving terminal receives to receiving terminal by channel (channel)] in be superimposed with noise n[m usually].

Present embodiment is adoptable a kind ofly to be convenient to the method that receiving terminal obtains real channel response and to be: transmitting terminal and receiving terminal communicate based on the transmission signal structure of certain particular design, and receiving terminal carries out channel estimating according to the preface information of inserting in the signal structure.

A kind of ofdm system frame structure schematic diagram that Fig. 4 provides for fourth embodiment of the invention.In the ofdm system frame structure as shown in Figure 4, transmitting terminal inserted leading 321 before OFDM symbol 322 to be sent.This leading 321 can adopt PN (pseudorandom) sequence, and this leading 321 can carry out pre-filtering to the OFDM time-domain signal in the pre-filtering processing module and handle the back and insert.Transmitting terminal adopts frame structure as shown in Figure 4 to send time-domain signal, makes receiving terminal to carry out channel estimating based on these the leading 321 employing industries method in common knowledge in the signal, obtains real channel response.

Below just provide a kind of actual channel and estimated acquisition methods, it will be understood by those skilled in the art that and take other method to carry out channel estimating, do not influenced the feature that present embodiment need obtain the actual channel time-domain response.

Next the preferred acquiring method to the pre-filtering coefficient that relates in the pre-filtering link of the present invention describes.For ease of essence and the reasonability thereof of understanding this pre-filtering coefficient acquiring method, provide the detailed derivation that pre-filtering coefficient is asked for below:

The embodiment of the invention can be according to general maximum Signal to Interference plus Noise Ratio criterion, and received signal model and feedback fraction channel response during based on the CP deficiency come concrete derivation pre-filtering coefficient w[n]:

At first the embodiment that provides with reference to Fig. 3 realizes block diagram, and specifically defining real channel response is h[n], do not lose and suppose that prevailingly real channel response is a cause and effect, that is: the original position of real channel response is 0, final position is v, when corresponding n＜0 and n＞v, and h[n]=0.Then the length of channel response (that is: channel delay extension length) is v+1.

Equivalent channel response in the received signal equals the convolution of pre-filtering coefficient and real channel response:

g[n]＝w[n]*h[n] (1)

In the formula (1), w[n] be pre-filtering coefficient, g[n] be equivalent channel response.If the initial response position of pre-filtering is-M ₁, the termination response position of pre-filtering is M ₂Might as well make n＜-M ₁And n＞M ₂The time, pre-filtering coefficient is 0, i.e. w[n]=0, therefore, pre-filtering coefficient length is M ₁+ M ₂+ 1.M ₁And M ₂Value can preestablish according to actual needs.

Formula (1) is expressed as matrix form, can gets:

Matrix form for further simplified style (2), can carry out as giving a definition:

$g=\left(\begin{array}{c}g[-M_1]\\ .\\ .\\ .\\ g\left[0\right]\\ .\\ .\\ .\\ g[M_2+\mathrm{\υ}]\end{array}\right),$ $w=\left(\begin{array}{c}w[-M_1]\\ .\\ .\\ .\\ w\left[0\right]\\ .\\ .\\ .\\ w\left[M_2\right]\end{array}\right),$

Can get thus:

g＝Hw (4)

Wherein, w is the pre-filtering coefficient vector, is (M ₁+ M ₂+ 1) Hang column vector, wherein element is from w[-M ₁] to w[M ₂]; H is total time domain channel response matrix, is (M ₁+ M ₂+ v+1) row * (M ₁+ M ₂+ 1) Lie matrix; G is the equivalent channel response vector, is (M ₁+ M ₂+ v+1) the column vector of row, its element is determined by formula (2).

Because the embodiment of the invention is the not IBI and ICI the inhibitions technology of foot of the CP that causes at the long delay channel, so supposes the CP length N in the following analysis _gLess than channel length, i.e. N _g＜v+1.The local channel response of supposing the receiving terminal feedback is described for convenient for surpassing the channel response part of CP length, channel response is split as two parts, wherein h ₁[n]=(h[0], h[1] ..., h[N _g]) the expression actual channel does not exceed CP length part (being called first's channel response), and actual channel exceeds the channel response h of CP length part ₂[n]=(h[N _g+ 1] ..., h[v]) (being called the second portion channel response), just need the part fed back, correspondingly define channel matrix corresponding to first's channel response and second portion channel response:

Wherein, H ₁Be the matrix form of first's channel response, be called first's channel response matrix, H ₂Be the matrix form of second portion channel response, be called the second portion channel response matrix.H ₁And H ₂All be (M ₁+ M ₂+ v+1) row * (M ₁+ M ₂+ 1) Lie matrix, wherein, H ₁By the channel response h of first ₁[n]=(h[0], h[1] ..., h[N _g]) middle element constitutes, and specifically is exactly with all second portion channel response h that comprise in the total channel matrix H ₂[n]=(h[N _g+ 1] ..., h[v]) in element put 0; H ₂By second portion channel response h ₂[n]=(h[N _g+ 1] ..., h[v]) middle element constitutes, and specifically is exactly with all channel response h of first that comprise in the total channel matrix H ₁[n]=(h[0], h[1] ..., h[N _g]) in element put 0; Therefore, in the formula (4), the equivalent channel response after the pre-filtering can be expressed as:

g＝(H ₁+H ₂)W

Received signal equals the convolution of equivalent channel response and time-domain signal and the stack of noise:

$y\left[m\right]=g\left[m\right]*x\left[m\right]+n\left[m\right]=g\left[m\right]*\underset{l=-\∞}{\overset{+\∞}{\mathrm{\Σ}}}{x}_{n+l}[m-\mathrm{lN}]+n\left[m\right]---\left(6\right)$

Wherein, n is the ordinal number of OFDM symbolic blocks, and m is the ordinal number of sampling point, x _n[m] is m sampling point of n OFDM symbolic blocks, and every OFDM symbol total sample is N=N _g+ K, 0≤m≤N, K are the integration lengths of fast Fourier transform (FFT),

For n OFDM symbolic blocks after the OFDM modulation sends time-domain signal, n (m) is a noise superimposed.

Because inter-block-interference (IBI) is usually from adjacent two the OFDM symbolic blocks of current OFDM symbolic blocks, therefore, only consider the inter-block-interference of adjacent two OFDM symbolic blocks (n-1 OFDM symbolic blocks and n+1 OFDM symbolic blocks) to current OFDM symbolic blocks (n OFDM symbolic blocks), the model of received signal (formula (6)) can further be expressed as:

$y_n\left[m\right]=g\left[m\right]*x_n\left[m\right]+g\left[m\right]*x_{n-1}[m+N]+g\left[m\right]*x_{n+1}[m-N]+n\left[m\right]$

$=\underset{l=-M_1}{\overset{M_2+v}{\mathrm{\Σ}}}g\left[l\right]x_n[m-l]+\underset{l=N_g}{\overset{M_2+v}{\mathrm{\Σ}}}g\left[l\right]x_{n-1}[m+N-l]$

$+\underset{l=-{\mathrm{<figure-callout\; id="1"\; label="M"\; filenames="A20091007865000351.png"\; state="\{\{state\}\}">M</figure-callout>}}_1}{\overset{0}{\mathrm{\&Sigma;}}}g\left[l\right]x_{n+1}[m-N-l]+n\left[m\right]---\left(7\right)$

The time domain received signal of formula (7) expression is carried out discrete Fourier transform (being called for short DFT), obtains the subcarrier in frequency domain symbol:

${\hat{s}}_n\left[k\right]=\frac{1}{K}\underset{m=0}{\overset{K-1}{\mathrm{\&Sigma;}}}{y}_n\left[m\right]e^{-j\frac{2\mathrm{\&pi;km}}{K}}$

Wherein, 0≤m＜K, K are the integration lengths of discrete Fourier transform (DFT).

With formula (7) substitution following formula, can get:

${\hat{s}}_n\left[k\right]=\frac{1}{K}\underset{m=0}{\overset{K-1}{\mathrm{\&Sigma;}}}\left(\underset{l=-M_1}{\overset{M_2+v}{\mathrm{\&Sigma;}}}g\right[l\left]x_n\right[m-l]+\underset{l=1}{\overset{M_2+v}{\mathrm{\&Sigma;}}}g[l\left]x_{n-1}\right[m-l+N\left]\right)e^{-j\frac{2\mathrm{\&pi;km}}{K}}$

$+\frac{1}{K}\underset{m=0}{\overset{K-1}{\mathrm{\&Sigma;}}}\underset{l=-M_1}{\overset{-(N_G+1)}{\mathrm{\&Sigma;}}}g\left[l\right]x_{n+1}[m-l-N]e^{-j\frac{2\mathrm{\&pi;km}}{K}}+\stackrel{\&OverBar;}{n}\left[k\right]---\left(8\right)$

$={\stackrel{\&OverBar;}{s}}_n\left[k\right]+{\stackrel{\&OverBar;}{e}}_n\left[k\right]+\stackrel{\&OverBar;}{n}\left[k\right]$

In the formula (8), noise section is:

$\stackrel{\&OverBar;}{n}\left[k\right]=\frac{1}{K}\underset{m=0}{\overset{K-1}{\mathrm{\&Sigma;}}}n\left[m\right]e^{-j\frac{2\mathrm{\&pi;km}}{K}}---\left(9\right)$

Inter-block-interference (IBI) part is:

${\stackrel{\&OverBar;}{e}}_n\left[k\right]=\frac{1}{K}\underset{m=0}{\overset{K-1}{\mathrm{\&Sigma;}}}\left(\underset{l=N_g}{\overset{M_2+v}{\mathrm{\&Sigma;}}}g\right[l\left]x_{n-1}\right[m-l+N]+\underset{l=-{\mathrm{<figure-callout\; id="1"\; label="M"\; filenames="A20091007865000351.png"\; state="\{\{state\}\}">M</figure-callout>}}_1}{\overset{0}{\mathrm{\&Sigma;}}}g[l\left]x_{n+1}\right[m-l-N\left]\right)e^{-j\frac{2\mathrm{\&pi;km}}{K}}---\left(10\right)$

K sub-carrier and its suffered inter-carrier interference (ICI) part are:

${\stackrel{\&OverBar;}{s}}_n\left[k\right]=\underset{l=-M_1}{\overset{M_2+v}{\mathrm{\&Sigma;}}}g\left[l\right]\frac{1}{K}\underset{m=0}{\overset{K-1}{\mathrm{\&Sigma;}}}{x}_n[m-l]e^{-j\frac{2\mathrm{\&pi;km}}{K}}$

$=\underset{l=-M_1}{\overset{M_2+v}{\mathrm{\&Sigma;}}}g\left[l\right]{\stackrel{\&OverBar;}{s}}_n[k,l]e^{-j\frac{2\mathrm{\&pi;kl}}{K}}---\left(11A\right)$

In the formula (11A):

${\stackrel{\&OverBar;}{s}}_n[k,l]=\frac{1}{K}\underset{m=0}{\overset{K-1}{\mathrm{\&Sigma;}}}{x}_n[m-l]e^{-j\frac{2\mathrm{\&pi;k}(m-l)}{K}}=\underset{m=0}{\overset{K-1}{\mathrm{\&Sigma;}}}{s}_n\left[m\right]{\mathrm{\&alpha;}}_{k-m}\left[l\right]---\left(12\right)$

Wherein, l is the sequence number of the tap signal of equivalent channel, and m and k are the sequence number of subcarrier, and k is the target sub-carriers sequence number, and m is other subcarrier sequence number that target sub-carriers is caused ICI.α _K-mIn the frequency domain of the received signal correspondence of [l] expression l tap correspondence, m subcarrier is to inter-carrier interference (ICI) coefficient of k subcarrier.

According to formula (11A) and (12), can be further with s _n[k] splits into subcarrier in frequency domain symbolic component (being useful signal part (the desired signal)) and inter-carrier interference (ICI) part:

${\stackrel{\&OverBar;}{s}}_n\left[k\right]=\underset{l=-M_1}{\overset{M_2+v}{\mathrm{\&Sigma;}}}g\left[l\right]\left({\mathrm{\&alpha;}}_0\right[l\left]s_n\right[k]+\underset{k\&NotEqual;m}{\mathrm{\&Sigma;}}{\mathrm{\&alpha;}}_{k-m}[l\left]s_n\right[m\left]\right)e^{-j\frac{2\mathrm{\&pi;kl}}{K}}$

What the ICI that causes with common Doppler frequency deviation or expansion was different is present embodiment ICI factor alpha _K-m[l] is relevant with the tap signal l in the equivalent channel, the ICI factor alpha of different tap signal correspondences _K-mThe value difference of [l], concrete, the ICI coefficient can be expressed as:

${\mathrm{\&alpha;}}_m\left[l\right]=\frac{1}{K}\underset{\mathrm{\&lambda;}=\max (-l,0)}{\overset{\min (K-l,N)-1}{\mathrm{\&Sigma;}}}{e}^{-j\frac{2\mathrm{\&pi;m\&lambda;}}{K}}---\left(13\right)$

According to the tap sequence number l of equivalent channel at valid interval-M ₁≤ l≤M ₂Diverse location among the+v can obtain corresponding ICI factor alpha respectively _m[l]:

${\mathrm{\&alpha;}}_m\left[l\right]=\left\{\begin{array}{cc}\frac{N+l}{K},& \mathrm{ifm}=0\mathrm{and}-{\mathrm{<figure-callout\; id="1"\; label="M"\; filenames="A20091007865000351.png"\; state="\{\{state\}\}">M</figure-callout>}}_1\&le;l<-N_g,\\ -\frac{1}{K}\frac{\sin \frac{\mathrm{\&pi;}(-l-N_g+1)m}{K}}{\sin \frac{\mathrm{\&pi;m}}{K}}{e}^{j\frac{\mathrm{\&pi;}(-l-N_g+1)m}{K}},& \mathrm{ifm}\&NotEqual;0\mathrm{and}-{\mathrm{<figure-callout\; id="1"\; label="M"\; filenames="A20091007865000351.png"\; state="\{\{state\}\}">M</figure-callout>}}_1\&le;l<-N_g,\\ 1,& \mathrm{ifm}=0\mathrm{and}-N_g\&le;l\&le;0,\\ 0,& \mathrm{ifm}\&NotEqual;0\mathrm{and}-N_g\&le;l\&le;0,\\ 1-\frac{l}{K},& \mathrm{ifm}=0\mathrm{and}0<l\&le;M_2+v,\\ -\frac{1}{K}\frac{\sin \frac{\mathrm{\&pi;lm}}{K}}{\sin \frac{\mathrm{\&pi;m}}{K}}{e}^{j\frac{\mathrm{\&pi;}(l+1)m}{K}},& \mathrm{ifm}\&NotEqual;0\mathrm{and}0<l\&le;M_2+v.\end{array}\right.---\left(14\right)$

According to formula (14) as can be known, to the footpath of delay positions in the CP scope, promptly-N _g≤ l≤0 o'clock, will there be ICI in signal after the FFT conversion.

Describe carrying out the pre-filtering coefficient calculating principle below based on maximum Signal to Interference plus Noise Ratio criterion.

Total Signal to Interference plus Noise Ratio is the ratio of the interference noise gross power on all sub-carrier gross powers and all sub-carrier, can adopt following formula to represent:

$\mathrm{SINR}\left(w\right)=\frac{P_S\left(w\right)}{P_{\mathrm{ICI}}\left(w\right)+P_{\mathrm{IBI}}\left(w\right)+P_{\mathrm{AWGN}}}---\left(15\right)$

In the formula (15), SINR (w) is total Signal to Interference plus Noise Ratio; P _s(w) be total power signal, equal all sub-carrier gross powers of current OFDM symbol (down together); P _AWGN(w) be noise average power, equal noise gross power on all subcarriers; P _ICI(w) be the suffered ICI gross powers of all subcarriers; P _IBI(w) be the IBI gross power.Obviously, total Signal to Interference plus Noise Ratio is all relevant with pre-filtering coefficient w with each power.

Next, based on the signal section in the above-mentioned derivation, the signal model of ICI part, IBI part and noise section is asked for the each several part power that relates in the formula (15) respectively.

1, all sub-carrier signal gross power P _sCalculating

According to formula (8), (11B) and (14), total power signal P _sCan be expressed as:

$P_S\left(w\right)=E\left\{\underset{k=0}{\overset{K-1}{\mathrm{\&Sigma;}}}{\left|s_n\right[k\left]\underset{l=-M_1}{\overset{M_2+v}{\mathrm{\&Sigma;}}}g\right[l\left]{\mathrm{\&alpha;}}_0\right[l\left]e^{-j\frac{2\mathrm{\&pi;kl}}{K}}\right|}^2\right\}$

$={\mathrm{\&sigma;}}_s^2\underset{k=0}{\overset{K-1}{\mathrm{\&Sigma;}}}{\left|\underset{l=-M_1}{\overset{M_2+v}{\mathrm{\&Sigma;}}}g\right[l\left]{\mathrm{\&alpha;}}_0\right[l\left]e^{-j\frac{2\mathrm{\&pi;kl}}{K}}\right|}^2---\left(16A\right)$

According to Wei Na-Xin Qin principle, can further obtain:

$P_S\left(w\right)=K{\mathrm{\&sigma;}}_s^2\underset{l=-M_1}{\overset{M_2+v}{\mathrm{\&Sigma;}}}{\left|g\right[l\left]\right|}^2{\left|{\mathrm{\&alpha;}}_0\right[l\left]\right|}^2=K{\mathrm{\&sigma;}}_s^2{w}^H{H}^H{D}_S\mathrm{Hw}---\left(16B\right)$

$=K{\mathrm{\&sigma;}}_s^2(w^H{H}_1^H{D}_S{H}_{1}w+w^H{H}_2^H{D}_S{H}_{2}w)$

σ wherein _s ²For sending sub-carrier power, subscript H represents the conjugate transpose of corresponding matrix.D _s(might as well be called the signal power compute matrix) is to define for further formula (16B) being expressed as more succinct mode:

Can get by formula (14)

D _sBe (M ₁+ M ₂+ v) row * (M ₁+ M ₂+ v) row diagonal matrix, diagonal element can be expressed as:

Wherein, i is for satisfying inequality 1≤i≤M ₁+ M ₂The integer of+v ,-M ₁Be the initial response position of default pre-filtering, M ₂Termination response position for default pre-filtering; The original position of default real channel response is 0, and final position v is the final position of default real channel response; K is default FFT (fast Fourier) transform length.Through type (17B) as can be known, signal power compute matrix D _sOnly comprising a known variables v, (v+1) is the channel response total length, therefore, is obtaining channel length information, can calculate D according to formula (17) _s

2, the suffered ICI gross power of all subcarriers P _ICI(w) calculating

At first can obtain other subcarrier according to formula (8), (11B) and (14) to the ICI of k subcarrier is:

$I_n\left[k\right]=\underset{m\&NotEqual;k}{\mathrm{\&Sigma;}}\left\{\underset{l}{\mathrm{\&Sigma;}}g\right[l\left]{\mathrm{\&alpha;}}_{k-m}\right[l\left]e^{-j\frac{2\mathrm{\&pi;kl}}{K}}\right\}{s}_n\left[m\right]---\left(19\right)$

The corresponding suffered ICI power of k subcarrier is:

$P_{\mathrm{ICI}}\left[k\right]==E{\left|I_n\right[k\left]\right|}^2={\mathrm{\&sigma;}}_s^2\underset{l}{\mathrm{\&Sigma;}}\underset{m\&NotEqual;0}{\mathrm{\&Sigma;}}{\left|g\right[l\left]{\mathrm{\&alpha;}}_m\right[l\left]\right|}^2---\left(20\right)$

For formula (20) is expressed as more succinct mode so that calculate definable inter-carrier interference power calculation matrix (that is: ICI power calculation matrix) D ₁:

Wherein

$\mathrm{\&Lambda;}\left[l\right]=\left\{\begin{array}{cc}\frac{-l-N_g}{K}(1-\frac{-l-N_g}{K}),& -{\mathrm{<figure-callout\; id="1"\; label="M"\; filenames="A20091007865000351.png"\; state="\{\{state\}\}">M</figure-callout>}}_1\&le;l<-{\mathrm{<figure-callout\; id="0"\; label="N\; g"\; filenames="A20091007865000351.png"\; state="\{\{state\}\}">N</figure-callout>}}_g\\ 0,& -N_g\&le;l\&le;0\\ \frac{1}{K}(1-\frac{1}{K}),& 0<l\&le;M_2+v\end{array}\right.$

ICI power calculation matrix D ₁Be (M ₁+ M ₂+ v) row * (M ₁+ M ₂+ v) row diagonal matrix.L is for satisfying inequality-M ₁≤ l≤M ₂The integer of+v ,-M ₁Be the initial response position of default pre-filtering, M ₂Termination response position for default pre-filtering; Default channel response total length is v+1; K is default fast Fourier transform length; N _gLength for Cyclic Prefix.Through type (21) as can be known, the CP length N _gK is system's known parameters with the FFT transform length, having obtained channel response length information, can calculate D ₁

From formula (20) as can be known, suffered ICI power is identical on the different sub carrier.Further, can obtain the suffered ICI gross power of all subcarriers P according to formula (20) and (21) _ICI(w) adopt ICI power calculation matrix D ₁The computing formula of expression is:

$P_{\mathrm{ICI}}\left(w\right)=K{\mathrm{\&sigma;}}_s^2{w}^H{\mathrm{<figure-callout\; id="1"\; label="H\; H\; D"\; filenames="A20091007865000351.png"\; state="\{\{state\}\}">H</figure-callout>}}^H{D}_{}{}_1\mathrm{Hw}---\left(22\right)$

3, inter-block-interference gross power P _IBI(w) calculating

According to formula (8) and (10), inter-block-interference gross power P _IBI(w) be:

$P_{\mathrm{IBI}}\left(w\right)=\underset{k=0}{\overset{K-1}{\mathrm{\&Sigma;}}}E{\left|e_n\right[k\left]\right|}^2=K{\mathrm{\&sigma;}}_s^2{w}^H{H}^H{D}_2\mathrm{Hw}---\left(23\right)$

Inter-block-interference power calculation matrix D wherein ₂(that is: IBI power calculation matrix) is for further with P _IBI(w) being expressed as more succinct mode defines:

Wherein

${\mathrm{\&Lambda;}}_1\left[l\right]=\left\{\begin{array}{cc}\frac{-l-N_g}{K},& -{\mathrm{<figure-callout\; id="1"\; label="M"\; filenames="A20091007865000351.png"\; state="\{\{state\}\}">M</figure-callout>}}_1\&le;l<-{\mathrm{<figure-callout\; id="0"\; label="N\; g"\; filenames="A20091007865000351.png"\; state="\{\{state\}\}">N</figure-callout>}}_g\\ 0,& -N_g\&le;l\&le;0\\ \frac{1}{K},& 0<l\&le;M_2+v\end{array}\right.$

IBI power calculation matrix D ₂Be (M ₁+ M ₂+ v) row * (M ₁+ M ₂+ v) row diagonal matrix.L is for satisfying inequality-M ₁≤ l≤M ₂The integer of+v ,-M ₁Be the initial response position of default pre-filtering, M ₂Termination response position for default pre-filtering; Preset channel response total length is v+1; K is default fast Fourier transform length; N _gLength for Cyclic Prefix.Through type (23) as can be known, the CP length N _gK is system's known parameters with the FFT transform length, having obtained channel length information, can calculate D ₂

4, noise gross power P on all subcarriers _AWGN(w) calculating:

Can get according to formula (9):

$P_{\mathrm{AWGN}}=\underset{k=0}{\overset{K-1}{\mathrm{\&Sigma;}}}E{\left|\stackrel{\&OverBar;}{n}\right[k\left]\right|}^2=K{\mathrm{\&sigma;}}_n^2---\left(25\right)$

σ wherein _n ²Be the noise power on the sub-carrier.

5, the expression formula of the total Signal to Interference plus Noise Ratio SINR of all sub-carrier (w)

Can get according to formula (16B), (22), (23) and (25):

$\mathrm{SINR}\left(w\right)=\frac{P_S\left(w\right)}{P_{\mathrm{ICI}}\left(w\right)+P_{\mathrm{IBI}}\left(w\right)+P_{\mathrm{AWGN}}}$

$=\frac{K{\mathrm{\&sigma;}}_s^2(w^H{H}_1^H{D}_S{H}_{1}w+w^H{H}_2^H{D}_S{H}_{2}w)}{K{\mathrm{\&sigma;}}_s^2(w^H{\mathrm{<figure-callout\; id="1"\; label="H\; H\; D"\; filenames="A20091007865000351.png"\; state="\{\{state\}\}">H</figure-callout>}}^H{D}_{}{}_1\mathrm{Hw}+w^H{H}^H{D}_2\mathrm{Hw})+K{\mathrm{\&sigma;}}_n^2}---\left(27\right)$

$=\frac{w^H{H}_1^H{D}_S{H}_{1}w+w^H{H}_2^H{D}_S{H}_{2}w}{w^H{H}_1^H{D}_I{H}_{1}w+w^H{H}_2^H{D}_I{H}_{2}w)+\frac{1}{\mathrm{\&rho;}}I}=\frac{w^H\mathrm{Bw}}{w^H\mathrm{Aw}+\frac{1}{\mathrm{\&rho;}}I}$

In the formula (27),

$\mathrm{\&rho;}=\frac{{\mathrm{\&sigma;}}_s^2}{{\mathrm{\&sigma;}}_n^2}$ Be sub-carrier signal to noise ratio, D _I=D ₁+ D ₂(might as well claim D _IBe the interference power compute matrix).

$B=H_1^H{\mathrm{<figure-callout\; id="1"\; label="D\; S\; H"\; filenames="A20091007865000351.png"\; state="\{\{state\}\}">D</figure-callout>}}_S{H}_{}{}_1+H_2^H{D}_S{H}_2---\left(28\right)$

$A=H_1^H{\mathrm{<figure-callout\; id="1"\; label="D\; I\; H"\; filenames="A20091007865000351.png"\; state="\{\{state\}\}">D</figure-callout>}}_I{H}_{}{}_1+H_2^H{D}_I{H}_2---\left(29\right)$

By maximizing total Signal to Interference plus Noise Ratio SINR (w), can obtain pre-filtering coefficient w under the maximum Signal to Interference plus Noise Ratio criterion.By formula (27) as can be known, total Signal to Interference plus Noise Ratio SINR (w) gets the peaked condition of equivalence and is: at w ^HMinimize its denominator under the constraints of Bw=1

Further be equivalent to again and minimize w ^HAw.Based on this equivalence solving model, might as well claim that B is a constraint matrix, title A is an objective matrix.Obviously, A and B matrix will directly influence and find the solution the pre-filtering coefficient that obtains.

Because receiving terminal exceeds CP length channel response partly to the local channel response of transmitting terminal feedback for actual channel in the present embodiment, so the channel response matrix H of first ₁Be unknown for transmitting terminal, second portion channel response matrix H ₂Then be as can be known.Therefore, B, the A matrix of (28) and (29) definition all are unavailable.To the section H that can not ask in B, the A matrix ₁ ^HD _sH ₁And H ₁ ^HD _IH ₁Expect E{H with it respectively ₁ ^HD _sH ₁And E{H ₁ ^HD _IH ₁(to some channel scene, can according to standard channel model by calculating this two parts in advance) replace, then can address the above problem.

Know based on experiment and theory analysis, maximize the demodulation performance that proper total Signal to Interference plus Noise Ratio might not strengthen OFDM, because demodulation performance also is decided by signal and the distribution character of interference power on each subcarrier.Therefore, present embodiment has further provided about the following alternative method of A and B matrix and (for the purpose of difference, has used respectively

With

Represent, still be called objective matrix and constraint matrix):

$\left(1\right).\hat{A}=E\left\{H_1^H{D}_I{H}_1\right\}+H_2^H{D}_I{H}_2$

$\left(2\right).\hat{A}=H_2^H{D}_I{H}_2---\left(30\right)$

$\left(3\right).\hat{A}=E\left\{H_1^H{D}_I{H}_1\right\}$

$\left(a\right).\hat{B}=E\left\{H_1^H{D}_S{H}_1\right\}+H_2^H{D}_S{H}_2$

$\left(b\right).\hat{B}=H_2^H{D}_S{H}_2---\left(31\right)$

$\left(c\right).\hat{B}=E\left\{H_1^H{D}_S{H}_1\right\}$

The objective matrix that formula (30) illustrates

Three kinds of computational methods (1～3), the constraint matrix that illustrates with formula (31)

Three kinds of computational methods (a～c), can carry out combination in any in twos, therefore, combination can comprise:

$\begin{array}{cc}\hat{A}=E\left\{H_1^H{D}_I{H}_1\right\}+H_2^H{D}_I{H}_2& \hat{B}=E\left\{H_1^H{D}_S{H}_1\right\}+H_2^H{D}_S{H}_2\end{array}---(1-a)$

$\begin{array}{cc}\hat{A}=E\left\{H_1^H{D}_I{H}_1\right\}+H_2^H{D}_I{H}_2& \hat{B}=H_2^H{D}_S{H}_2\end{array}---(1-b)$

$\begin{array}{cc}\hat{A}=E\left\{H_1^H{D}_I{H}_1\right\}+H_2^H{D}_I{H}_2& \hat{B}=E\left\{H_1^H{D}_S{H}_1\right\}\end{array}---(1-c)$

$\begin{array}{cc}\hat{A}=H_2^H{D}_I{H}_2& \hat{B}=E\left\{H_1^H{D}_S{H}_1\right\}+H_2^H{D}_S{H}_2\end{array}---(2-a)$

$\begin{array}{cc}\hat{A}=H_2^H{D}_I{H}_2& \hat{B}=H_2^H{D}_S{H}_2\end{array}---(2-b)$

$\begin{array}{cc}\hat{A}=H_2^H{D}_I{H}_2& \hat{B}=E\left\{H_1^H{D}_S{H}_1\right\}\end{array}---(2-c)$

$\begin{array}{cc}\hat{A}=E\left\{H_1^H{D}_I{H}_1\right\}& \hat{B}=E\left\{H_1^H{D}_S{H}_1\right\}+H_2^H{D}_S{H}_2\end{array}---(3-a)$

$\begin{array}{cc}\hat{A}=E\left\{H_1^H{D}_I{H}_1\right\}& \hat{B}=H_2^H{D}_S{H}_2\end{array}---(3-b)$

$\begin{array}{cc}\hat{A}=E\left\{H_1^H{D}_I{H}_1\right\}& \hat{B}=E\left\{H_1^H{D}_S{H}_1\right\}\end{array}---(3-c)$

Based on above-mentioned analytic process as can be known, objective matrix And constraint matrix

Calculating, the signal power compute matrix D that relates to _sWith interference power compute matrix D _I, all can calculate in advance according to the channel length of receiving terminal feedback, because the wireless channel length variations is slow relatively, therefore feeding back the update cycle can be longer, so D _sAnd D _ICan calculate in advance.Based on D _sAnd D _IThis characteristic, with D _sAnd D _IAlso can be called preset signal power calculation matrix and preset the interference power compute matrix.

By following one section the finding the solution and the analysis of corresponding equivalent solving model of formula (29), can further adopt existing matrix mathematical method to find the solution and obtain pre-filtering coefficient w based on pre-filtering coefficient w under the maximum Signal to Interference plus Noise Ratio criterion:

$w={\sqrt{\hat{B}}}^{-H}{v}_{\min }---\left(30\right)$

Wherein,

For

Qiao Lisiji (Cholesky) decompose,

With

Satisfy relation:

$\hat{B}=\sqrt{\hat{B}}{\sqrt{\hat{B}}}^H---\left(31\right)$

v _MinBe matrix

Minimal eigenvalue characteristic of correspondence vector.

Pre-filtering in this example under the not enough situation of the above-mentioned CP that provides (can think also that in essence channel shortens equalizer) coefficient acquiring method, describe at the pre-filtering of carrying out at transmitter although be, but one of ordinary skill in the art will appreciate that: described prefilter or signal channel shortening equalization coefficient acquiring method, change slightly, the channel that can be used for implementing at receiver fully shortens the coefficient of equalizer to be asked for, and described change does not need to pay extra novelty work, does not make the spirit and scope of the essence disengaging embodiment of the invention technical scheme of appropriate technical solution yet.

Ask for the analysis of principle by above-mentioned prefilter coefficient, the pre-filtering coefficient computational methods can be according to the channel response total length that obtains in advance as can be known, signal calculated power calculation matrix D _sWith interference power compute matrix D _IAccording to the signal power compute matrix D that obtains _sWith interference power compute matrix D _IAnd the local channel that obtains in advance response, calculate constraint matrix and objective matrix; According to constraint matrix that obtains and objective matrix, calculate the pre-filtering coefficient of correspondence when Signal to Interference plus Noise Ratio is maximum, described Signal to Interference plus Noise Ratio equals the ratio of signal power and interference noise power.

Below the explanation present embodiment carries out the realization flow that pre-filtering coefficient calculates based on maximum Signal to Interference plus Noise Ratio criterion.

The pre-filtering coefficient computational methods flow chart that Fig. 5 provides for fifth embodiment of the invention.Flow process shown in Figure 5 also can be used as a realization flow of step 305 among Fig. 3.As shown in Figure 5, the pre-filtering coefficient computational methods comprise:

Step 51, according to the channel response total length, signal calculated power calculation matrix D respectively _sWith interference power compute matrix D _I

The channel response total length is (v+1), calculates signal power compute matrix D according to formula (17B), (21) and (24) respectively _s, ICI power calculation matrix D ₁With IBI power calculation matrix D ₂, and further by D _I=D ₁+ D ₂Calculate interference power compute matrix D _I

Step 52, surpass the local channel response h of CP length according to actual channel ₂[n]=(h[N _g+ 1] ..., h[v]), structure second portion channel response matrix H ₂

H ₂In the 1st the row to N _gEach element of row is 0, and H ₂In N _gRow is to M ₁+ M ₂N in each element that+v+1 is capable and the total channel matrix H _gRow is to M ₁+ M ₂Each element that+v+1 is capable is identical.Relevant parameter N _g, M ₁, M ₂All are system intialization parameters, concrete meaning sees above, and (v+1) is the channel response total length.

Step 53, the signal power compute matrix D that obtains according to step 51 _sWith interference power compute matrix D _IAnd the second portion channel response matrix H that obtains of step 52 ₂, calculate objective matrix respectively And constraint matrix

Concrete calculating can be adopted formula (1-a), (1-b), (1-c), (2-a), (2-b), (2-c), (3-a), (3-b) or (3-c).

Step 54, the objective matrix that obtains according to step 53

And constraint matrix

Calculate pre-filtering coefficient w by formula (30).

Computational methods flow process by pre-filtering coefficient shown in Figure 5 obtains after the pre-filtering coefficient, just can carry out pre-filtering to signal and handle.Based on the pre-filtering of above-mentioned pre-filtering coefficient, can implement at transmitting terminal, also can implement at receiving terminal.Implement at transmitting terminal, need receiving terminal feedback channel response total length and local channel response, described local channel response is for surpassing the local channel response of CP length.Implement at receiving terminal, receiving terminal self just can get access to channel response total length and local channel response message.

In order to verify the effect that adopts the embodiment of the invention to disturb inhibition, present embodiment is to adopting objective matrix

And constraint matrix

The pre-filtering coefficient that obtains of different values carry out pre-filtering and handle making a start, the IBI that is obtained suppresses effect, with typical IBI inhibition technology in the ofdm system of being introduced in the background technology, be that the IBI inhibition effect of (RISIC algorithm) is eliminated in residual intersymbol interference, compare based on emulation.

Fig. 6 is that the performance of embodiment of the invention disturbance restraining method and existing disturbance restraining method compares schematic diagram one; Fig. 7 is that the performance of embodiment of the invention disturbance restraining method and other disturbance restraining methods compares schematic diagram two.The emulation of Fig. 6 and Fig. 7 correspondence is based on micro-wave access global inter communication technology (WorldwideInteroperability for Microwave Access, abbreviation WIMAX) ofdm system in the standard, each OFDM symbolic blocks sub-carrier number is 914, total bandwidth is MHz, and CP length is 1/8 symbol period length.The scene of emulation is a kind of typical mountain area channel model (HT channel), and sending the symbol-modulated mode is 4QAM (quadrature amplitude modulation).Receiver adopts ideal symbol synchronous.Abscissa is the signal to noise ratio (snr) of received signal among Fig. 6, and unit is dB; Ordinate is error sign ratio (SER).Abscissa is the signal to noise ratio (snr) of received signal among Fig. 7, and unit is dB; Ordinate is mistake symbolic blocks rate (WER).Among Fig. 6 and Fig. 7, be the performance curve after carrying out the received signal pre-filtering at the existing RISIC algorithm of receiving terminal employing than curve; Optimal curve 1 is the objective matrix based on formula (2-b)

And constraint matrix

The pre-filtering coefficient of trying to achieve sends the performance curve after the signal pre-filtering; In like manner, optimal curve 2-5 is respectively based on formula (3-b), (1-b), (1-a) and objective matrix (3-c)

And constraint matrix The pre-filtering coefficient of trying to achieve sends the performance curve after the signal pre-filtering.Each performance curve by comparison diagram 6 and Fig. 7 as can be known, the best performance of optimal curve 1, and will obviously being better than by than curve is being example in 10% working point, optimal curve 1 has the performance gain about 5dB and 10dB respectively on SER and WER performance index meaning.After this means that transmitting terminal adopts the pre-filtering coefficient of optimal curve 1 correspondence that the transmission signal is carried out the pre-filtering processing, suppress to handle even receiving terminal no longer carries out extra interference, also can obtain complicated IBI suitable even that be better than implementing at receiving terminal at present and disturb the performance that suppresses processing.

To sum up analyze, present embodiment transmitting terminal in multicarrier system obtains partial channel knowledge according to the feedback of receiving terminal, calculate pre-filtering coefficient based on given method among the embodiment 5, and described pre-filtering coefficient, time-domain signal is carried out pre-filtering handle the back transmission, can fully reach the purpose that similar channel shortens, make receiving terminal receive in the time-domain signal, the energy of equivalent channel response concentrates in the scope that is shorter than circulating prefix-length as far as possible, thereby, avoided the implementation complexity of receiver simultaneously for the required remarkable increase of IBI to because the I BI that the channel delay expansion causes greater than CP length has effective inhibition effect.Therefore, with respect to prior art, present embodiment has obviously reduced receiver to be suppressed to disturb required computational complexity, has realization and advantage such as simply, easily applies.

The transmitter architecture figure that Fig. 8 provides for sixth embodiment of the invention.As shown in Figure 8, the present embodiment transmitter comprises: channel information acquisition module 81, pre-filtering coefficient computing module 82 and pre-filtering processing module 83.

Channel information acquisition module 81 is used for obtaining partial channel knowledge according to the feedback of receiving terminal, and described partial channel knowledge can comprise the local channel response.

Pre-filtering coefficient computing module 82 is used for calculating pre-filtering coefficient according to the partial channel knowledge that obtains.

Pre-filtering processing module 83 is used for according to the described pre-filtering coefficient that obtains, and time-domain signal is carried out pre-filtering handle the back transmission.

In on the basis of technique scheme, described partial channel knowledge also can comprise the channel response total length.Pre-filtering coefficient computing module 82 can further comprise: preset matrix calculation unit 821, objective matrix and constraint matrix computing unit 822 and pre-filtering coefficient and ask for unit 823.

Preset matrix calculation unit 821 and be used for according to described channel response total length signal calculated power calculation matrix and interference power compute matrix.

Objective matrix and constraint matrix computing unit 822 are used for the local channel response of obtaining according to the signal power compute matrix that obtains and interference power compute matrix and channel information acquisition module 81, calculate constraint matrix and objective matrix.

The local channel response can comprise: actual channel exceeds the channel response (that is: second portion channel response) of circulating prefix-length part.The second portion channel response matrix that the local channel response that objective matrix and constraint matrix computing unit 822 can select for use channel information acquisition module 81 to obtain is constructed calculates constraint matrix and objective matrix according to signal power compute matrix, interference power compute matrix and second portion channel response matrix.

Pre-filtering coefficient is asked for unit 823 and is used for according to the constraint matrix and the objective matrix that obtain, based on maximum Signal to Interference plus Noise Ratio criterion, and the pre-filtering coefficient of correspondence when calculating the maximization Signal to Interference plus Noise Ratio.The function of present embodiment channel information acquisition module 81, pre-filtering coefficient computing module 82 and pre-filtering processing module 83 can be integrated into a functional module, as is integrated into the pre-filtering module 32 as shown in the corresponding embodiment of Fig. 3.In the application scenarios of channel response total length, realize that at transmitting terminal the realization principle of the pre-filtering processing of signal sees the record of the corresponding embodiment of Fig. 3-Fig. 5 for details, repeats no more based on the present embodiment transmitter greater than Cyclic Prefix (CP) length.

Present embodiment is at the situation of channel response total length greater than Cyclic Prefix (CP) length, carry out similar channel by transmitter to sent the OFDM time-domain signal and shorten balanced pre-filtering processing, make in the receiving terminal received signal, the energy of equivalent channel response concentrates in the scope that is shorter than CP length as far as possible, thereby guarantees that suffered IBI of received signal and ICI obtain effectively pre-the inhibition.Present embodiment helps reducing receiving terminal in order to suppress to disturb required computational complexity and processing delay, has obviously reduced the difficulty that realizes.

The receiver structure figure that Fig. 9 provides for seventh embodiment of the invention.As shown in Figure 9, the present embodiment receiver comprises: channel estimation module 91 and feedback module 92.

Channel estimation module 91 is used to carry out channel estimating to obtain time domain channel response.

Feedback module 92 is used for extracting partial channel knowledge and to the transmitting terminal feedback, calculating pre-filtering coefficient for described transmitting terminal according to described partial channel knowledge from described time domain channel response, and according to described pre-filtering coefficient time-domain signal is carried out pre-filtering and handle; Described partial channel knowledge comprises the local channel response.

On the basis of technique scheme, described partial channel knowledge also can comprise the channel response total length.Feedback module 92 can further comprise: feedback channel response extracting unit 921 and feedback channel response quantifying unit 922.

Feedback channel response extracting unit 921 is used for extracting partial information at described time domain channel response.

Feedback channel response quantifying unit 922 is used for that the described partial information that extracts is carried out quantification treatment and obtains described local channel response and channel response total length, and to described transmitting terminal feedback.

Based on present embodiment receiver channel estimation module 91 and feedback module 92 (comprising feedback channel response extracting unit 921 and feedback channel response quantifying unit 922), in the application scenarios of channel response total length greater than Cyclic Prefix (CP) length, realize that at transmitting terminal the realization principle of the pre-filtering processing of signal sees the record of the corresponding embodiment of Fig. 3-Fig. 5 for details, repeats no more.

Present embodiment is in the application scenarios of channel response total length greater than Cyclic Prefix (CP) length, receiver is with channel informations such as local channel responses, feed back to transmitting terminal, carrying out pre-filtering for transmitting terminal according to the partial channel knowledge that obtains handles, make receiver receive in the time-domain signal, the energy of equivalent channel response concentrates in the scope that is shorter than CP length as far as possible, present embodiment has reduced receiving terminal in order to suppress to disturb required computational complexity and processing delay, has obviously reduced the difficulty that realizes.

The Interference Suppression System structure chart that Figure 10 provides for eighth embodiment of the invention.As shown in figure 10, the present embodiment receiver comprises: transmitter 101 and receiver 102.

Transmitter 101 is used for obtaining partial channel knowledge according to the feedback of receiver 102, calculate pre-filtering coefficient according to the partial channel knowledge that obtains, according to the described pre-filtering coefficient that obtains time-domain signal is carried out pre-filtering and handle the back transmission, described partial channel knowledge comprises the local channel response.

Receiver 102 is used for obtaining time domain channel response by channel estimating; Extract described partial channel knowledge and to transmitter 101 feedback from described time domain channel response.

The refined structure of present embodiment transmitter can be referring to the record of the corresponding embodiment of Fig. 8, the refined structure of receiver can be referring to the record of the corresponding embodiment of Fig. 9, based on the present embodiment Interference Suppression System in the application scenarios of channel response total length greater than Cyclic Prefix (CP) length, disturb the realization principle of inhibition, see the record of the corresponding embodiment of Fig. 1-Fig. 7 for details, repeat no more.

Embodiment of the invention Interference Suppression System can be applicable to ofdm system or adopts in the single-carrier system of frequency domain equalization and disturb inhibition, for the channel response total length of actual channel situation greater than circulating prefix-length, shorten balanced pre-filtering processing by carrying out similar channel to sent time-domain signal at transmitter, make receiver receive in the time-domain signal, the energy of equivalent channel response concentrates in the scope that is shorter than circulating prefix-length as far as possible, thereby can significantly reduce receiving terminal and disturb required implementation complexity, and has good interference suppressioning effect in order to suppress.

The pre-filtering coefficient computing device structure figure that Figure 11 provides for ninth embodiment of the invention.As shown in figure 11, present embodiment pre-filtering coefficient calculation element comprises: preset matrix calculation unit 111, objective matrix and constraint matrix computing unit 112 and pre-filtering coefficient and ask for unit 113.

Presetting matrix calculation unit 111 is used for according to the channel response total length, signal calculated power calculation matrix and the interference power compute matrix that obtain in advance.

Objective matrix and constraint matrix computing unit 112 are used for calculating constraint matrix and objective matrix according to the signal power compute matrix that obtains and interference power compute matrix and the local channel response of obtaining in advance.

Pre-filtering coefficient is asked for unit 113 according to the constraint matrix and the objective matrix that obtain, calculates the pre-filtering coefficient of correspondence when Signal to Interference plus Noise Ratio is maximum, and described Signal to Interference plus Noise Ratio equals the ratio of signal power and interference noise power.

The pre-filtering coefficient calculation element that present embodiment provides can be used but is not limited in ofdm system or the SC-FDE system, in order to obtain pre-filtering coefficient, so that transmitting terminal or receiving terminal carry out interference eliminated according to pre-filtering coefficient to the signal that receives or send; Carry out interference eliminated based on this pre-filtering coefficient, help reducing the computational complexity of transmitting terminal or receiving terminal inhibition interference.

In addition, present embodiment also can be used as a functional module that is used to calculate pre-filtering coefficient, is integrated in transmitting terminal or the receiving terminal.Fig. 8 shows present embodiment and is integrated in a application scenarios in the transmitting terminal, under this sight, needs receiving terminal feedback fraction channel response, and described local channel response can comprise the local channel response that surpasses CP length.In addition, present embodiment also can be integrated in the receiving device (as: receiver), and under this situation, receiving terminal self just can get access to channel response total length and local channel response message, therefore, can carry out the calculating of pre-filtering coefficient according to the information of obtaining.The operation principle of the calculation element of relevant present embodiment pre-filtering coefficient can repeat no more referring to the record of Fig. 3 and the corresponding embodiment of Fig. 5.

One of ordinary skill in the art will appreciate that: accompanying drawing is the schematic diagram of a preferred embodiment, and module in the accompanying drawing or flow process might not be that enforcement the present invention is necessary.

One of ordinary skill in the art will appreciate that: the module in the device among the embodiment can be described according to embodiment and be distributed in the device of embodiment, also can carry out respective change and be arranged in the one or more devices that are different from present embodiment.The module of the foregoing description can be merged into a module, also can further split into a plurality of submodules.

The invention described above embodiment sequence number is not represented the quality of embodiment just to description.

One of ordinary skill in the art will appreciate that: all or part of step that realizes said method embodiment can be finished by the relevant hardware of program command, aforesaid program can be stored in the computer read/write memory medium, this program is carried out the step that comprises said method embodiment when carrying out; And aforesaid storage medium comprises: various media that can be program code stored such as ROM, RAM, magnetic disc or CD.

It should be noted that at last: above embodiment only in order to technical scheme of the present invention to be described, is not intended to limit; Although with reference to previous embodiment the present invention is had been described in detail, those of ordinary skill in the art is to be understood that: it still can be made amendment to the technical scheme that previous embodiment is put down in writing, and perhaps part technical characterictic wherein is equal to replacement; And these modifications or replacement do not make the essence of appropriate technical solution break away from the spirit and scope of embodiment of the invention technical scheme.

Claims (18)

1, a kind of disturbance restraining method is characterized in that, comprising:

Feedback according to receiving terminal obtains partial channel knowledge, and described partial channel knowledge comprises the local channel response;

According to the partial channel knowledge that obtains, calculate pre-filtering coefficient;

According to the described pre-filtering coefficient that obtains, time-domain signal is carried out pre-filtering handle the back transmission.

2, disturbance restraining method according to claim 1 is characterized in that, described partial channel knowledge also comprises the channel response total length; Described calculating pre-filtering coefficient comprises:

According to described channel response total length, signal calculated power calculation matrix D _sWith interference power compute matrix D _I

According to the signal power compute matrix D that obtains _sWith interference power compute matrix D _IAnd the described local channel response of obtaining, calculate constraint matrix and objective matrix;

According to constraint matrix that obtains and objective matrix, calculate the pre-filtering coefficient of correspondence when Signal to Interference plus Noise Ratio is maximum, described Signal to Interference plus Noise Ratio equals the ratio of signal power and interference noise power.

3, a kind of disturbance restraining method is characterized in that, comprising:

Obtain time domain channel response by channel estimating;

According to described time domain channel response, extract partial channel knowledge and, calculate pre-filtering coefficient according to described partial channel knowledge, and according to described pre-filtering coefficient time-domain signal to be sent is carried out pre-filtering and handle for described transmitting terminal to the transmitting terminal feedback; Described partial channel knowledge comprises the local channel response.

4, disturbance restraining method according to claim 3 is characterized in that, described partial channel knowledge also comprises the channel response total length; Describedly obtain partial channel knowledge and, comprising to transmitting terminal feedback:

In described time domain channel response, extract partial information;

The described partial information that extracts is carried out quantification treatment obtain described local channel response and channel response total length, and to described transmitting terminal feedback.

5, a kind of pre-filtering coefficient computational methods is characterized in that, comprising:

According to the channel response total length that obtains in advance, signal calculated power calculation matrix D _sWith interference power compute matrix D _I

According to the signal power compute matrix D that obtains _sWith interference power compute matrix D _IAnd the local channel that obtains in advance response, calculate constraint matrix and objective matrix;

According to constraint matrix that obtains and objective matrix, calculate the pre-filtering coefficient of correspondence when Signal to Interference plus Noise Ratio is maximum, described Signal to Interference plus Noise Ratio equals the ratio of signal power and interference noise power.

6, pre-filtering coefficient computational methods according to claim 5 is characterized in that, described signal power compute matrix D _sFor:

Wherein, D _sBe (M ₁+ M ₂+ v) row * (M ₁+ M ₂+ v) row diagonal matrix, diagonal element is:

I is for satisfying inequality 1≤i≤M ₁+ M ₂The integer of+v ,-M ₁Be the initial response position of default pre-filtering, M ₂Termination response position for default pre-filtering; (v+1) be the channel response total length; K is default fast Fourier transform length.

7, pre-filtering coefficient computational methods according to claim 6 is characterized in that, described interference power compute matrix D _IFor: inter-carrier interference power calculation matrix D ₁With inter-block-interference power calculation matrix D ₂Stack.

8, pre-filtering coefficient computational methods according to claim 7 is characterized in that, described inter-carrier interference power calculation matrix D ₁For:

Wherein, D ₁Be (M ₁+ M ₂+ v) row * (M ₁+ M ₂+ v) row diagonal matrix, diagonal element is:

$\mathrm{\&Lambda;}\left[l\right]=\left\{\begin{array}{cc}\frac{-l-N_g}{K}(1-\frac{-l-N_g}{K}),& -M_1\&le;l<-N_g\\ 0,& -N_g\&le;l\&le;0\\ \frac{l}{K}(1-\frac{l}{K}),& 0<l\&le;M_2+v\end{array}\right.$

L is for satisfying inequality-M ₁≤ l≤M ₂The integer of+v ,-M ₁Be the initial response position of default pre-filtering, M ₂Termination response position for default pre-filtering; (v+1) be the channel response total length; K is default fast Fourier transform length; N _gLength for Cyclic Prefix.

9, pre-filtering coefficient computational methods according to claim 7 is characterized in that, described inter-block-interference power calculation matrix D ₂For:

Wherein, D ₂Be (M ₁+ M ₂+ v) row * (M ₁+ M ₂+ v) row diagonal matrix, diagonal element is:

${\mathrm{\&Lambda;}}_1\left[l\right]=\left\{\begin{array}{cc}\frac{-l-N_g}{K},& -M_1\&le;l<-N_g\\ 0,& -N_g\&le;l\&le;0\\ \frac{l}{K},& 0<l\&le;M_2+v\end{array}\right.$

L is for satisfying inequality-M ₁≤ l≤M ₂The integer of+v ,-M ₁Be the initial response position of default pre-filtering, M ₂Termination response position for default pre-filtering; (v+1) be the channel response total length; K is default fast Fourier transform length; N _gLength for Cyclic Prefix.

According to the described arbitrary pre-filtering coefficient computational methods of claim 5～9, it is characterized in that 10, described local channel response comprises: first's channel response and/or second portion channel response; Described first channel response is the channel response that actual channel does not exceed the circulating prefix-length part, and described second portion channel response is the channel response that actual channel exceeds the circulating prefix-length part.

11, pre-filtering coefficient computational methods according to claim 10 is characterized in that, described objective matrix and constraint matrix are respectively:

$\hat{A}=H_2^H{D}_I{H}_2$

$\hat{B}=H_2^H{D}_s{H}_2$

Wherein, The expression objective matrix;

The expression constraint matrix, H ₂Be second portion channel response matrix, H ₂ ^HConjugate transpose for the second portion channel response matrix.

12, pre-filtering coefficient computational methods according to claim 11 is characterized in that, described second portion channel H ₂For:

H ₂Be (M ₁+ M ₂+ v+1) row * (M ₁+ M ₂+ 1) Lie matrix is made of element in the second portion channel response, (v+1) is the channel response total length; N _gLength for Cyclic Prefix.

13, a kind of transmitter is characterized in that, comprising:

The channel information acquisition module is used for obtaining partial channel knowledge according to the feedback of receiving terminal, and described partial channel knowledge comprises the local channel response;

The pre-filtering coefficient computing module is used for calculating pre-filtering coefficient according to the partial channel knowledge that obtains;

The pre-filtering processing module is used for according to the described pre-filtering coefficient that obtains, and time-domain signal is carried out pre-filtering handle the back transmission.

14, transmitter according to claim 13 is characterized in that, described partial channel knowledge also comprises the channel response total length; Described pre-filtering coefficient computing module comprises:

Preset matrix calculation unit, be used for according to described channel response total length signal calculated power calculation matrix and interference power compute matrix;

Objective matrix and constraint matrix computing unit are used for calculating constraint matrix and objective matrix according to the signal power compute matrix that obtains and interference power compute matrix and the described local channel response of obtaining;

Pre-filtering coefficient is asked for the unit, is used for calculating the pre-filtering coefficient of correspondence when Signal to Interference plus Noise Ratio is maximum according to the constraint matrix and the objective matrix that obtain, and described Signal to Interference plus Noise Ratio equals the ratio of signal power and interference noise power.

15, a kind of receiver is characterized in that, comprising:

Channel estimation module is used for obtaining time domain channel response by channel estimating;

Feedback module, be used for according to described time domain channel response, extract partial channel knowledge and, calculate pre-filtering coefficient according to described partial channel knowledge, and according to described pre-filtering coefficient time-domain signal to be sent is carried out pre-filtering and handle for described transmitting terminal to the transmitting terminal feedback; Described partial channel knowledge comprises the local channel response.

16, receiver according to claim 15 is characterized in that, described partial channel knowledge also comprises the channel response total length; Described feedback module comprises:

Feedback channel response abstraction module is used for extracting partial information at described time domain channel response;

Feedback channel response quantization modules is used for that the described partial information that extracts is carried out quantification treatment and obtains described local channel response and channel response total length, and to described transmitting terminal feedback.

17, a kind of pre-filtering coefficient calculation element is characterized in that, comprising:

Preset matrix calculation unit, be used for according to the channel response total length that obtains in advance, signal calculated power calculation matrix and interference power compute matrix;

Objective matrix and constraint matrix computing unit are used for calculating constraint matrix and objective matrix according to the signal power compute matrix that obtains and interference power compute matrix and the local channel response of obtaining in advance;

Pre-filtering coefficient is asked for the unit, is used for calculating the pre-filtering coefficient of correspondence when Signal to Interference plus Noise Ratio is maximum according to the constraint matrix and the objective matrix that obtain, and described Signal to Interference plus Noise Ratio equals the ratio of signal power and interference noise power.

18, a kind of Interference Suppression System comprises transmitter and receiver, it is characterized in that,

Described transmitter is used for obtaining partial channel knowledge according to the feedback of described receiver, calculate pre-filtering coefficient according to the partial channel knowledge that obtains, according to the described pre-filtering coefficient that obtains time-domain signal is carried out pre-filtering and handle the back transmission, described partial channel knowledge comprises the local channel response;

Described receiver is used for obtaining time domain channel response by channel estimating; Extract described partial channel knowledge and to described transmitter feedback from described time domain channel response.

Images