CN101232358A - Method and system of channel blind equalization in MIMO system - Google Patents

Abstract

The invention discloses a channel blind equalization method in a multi-input multi-output system, the method comprises that: the transmitting ends respectively add prefixes in the data signals on all the transmitting antennas, so as to obtain the transmitting signals of a self-related function which meet the self-related independent conditions and transmit; the receiving ends of all the receiving antennas receive the transmitting signals and carry out the channel blind equalization which is based on the self-related matching principles. The invention further discloses a system for realizing the channel blind equalization. The system complexity for realizing the channel blind equalization can be reduced and the error bit rate performances of the system can be improved according to the method.

Description

A kind of in multi-input multi-output system channel blind balance method and system

Technical field

The present invention relates to balancing technique, be meant a kind of channel blind balance method and system in multiple-input and multiple-output (MIMO) system especially.

Background technology

In mimo system, when the data that transmitting terminal sends were transferred to receiving terminal by wireless channel, receiving terminal need be eliminated the interference that wireless channel brings by equilibrium in order to obtain sending the estimated value of signal.

Balancing technique in the wireless communication system mainly is divided into the equilibrium of adopting training sequence and the blind equalization that need not training sequence.The former need use the coefficient of all known training sequence estimation balancing device of transmitting terminal and receiving terminal, and the latter does not need to use training sequence, directly the coefficient of equalizer is estimated.Blind equalization does not use training sequence, and this is different from the equilibrium of adopting training sequence, and the band efficiency of system can not be affected when using blind equalization, and therefore, Blind Equalization Technique is one of industry research focus always.

Blind Equalization Technique based on auto-correlation coupling (AM, Autocorrelation Matching) principle is a kind of balancing technique commonly used at present.Blind Equalization Technique based on the auto-correlation matching principle need satisfy following condition: the transmission signal { s on each transmitting antenna _i(n) | i=1 ... N _tUncorrelated mutually, and respectively send signal s _i(n) auto-correlation function $R_{s_i}\left(\mathrm{\τ}\right)=E\left[s_i\left(n\right)s_i^{*}(n-\mathrm{\τ})\right]$ Satisfy the linear displacement independent condition, i.e. the set of auto-correlation function shown in the following formula is linear uncorrelated.

$\left\{R_{s_i}(\mathrm{\τ}-l)\right|i=\mathrm{1,2}\·\·\·,N_t;l=0,\±1,\±2,\·\·\·,\±L_G,\mathrm{\τ}\≥\mathrm{SF}\}---\left(1\right)$

Wherein, SF is a positive constant.

Because under normal conditions, time-domain signal on each transmitting antenna in the mimo system is a white signal, the auto-correlation function that is the time-domain signal on each transmitting antenna does not satisfy the auto-correlation independent condition, therefore as if the channel blind equalization that will use, then need to allow the transmission signal satisfy the auto-correlation independent condition based on the auto-correlation matching principle.Two kinds of methods that propose below all utilize the auto-correlation matching principle to realize the channel blind equalization.

In in May, 2002 IEEE International Symposium on Circuits and Systems, in disclosed " Autocorrelation-a new differentiating domain for multiple accesswireless communications " document one, provide a kind of mimo channel blind balance method based on the auto-correlation matching principle.This method produces corresponding transmitted signals by different prefilters respectively with the white data signal on each transmitting antenna, the auto-correlation function of the transmission signal that obtains like this satisfies the auto-correlation independent condition, thereby, receiving terminal is complementary with the auto-correlation function that sends signal by the auto-correlation function with equalizer output signal, realizes the blind equalization of channel.In the method, transmitting terminal need use prefilter, and corresponding, receiving terminal also needs to use inverse filter, so the complexity that system realizes increases.

In in May, 2003 disclosed in IEEE International Conference on Communications " A new multiplex-access scheme based on the diversity of autocorrelation " document two, provide a kind of mimo channel blind balance method based on the auto-correlation matching principle.In this method, white data signal on each transmitting antenna different coloured random signal that superposes respectively, make the auto-correlation function that sends signal satisfy the auto-correlation independent condition, at this moment, receiving terminal also can be realized the channel blind equalization based on the auto-correlation matching principle.In this method, when receiving terminal recovers to send data-signal, data-signal has been caused interference, made the performance of BER of system descend to some extent in coloured random signal that transmitting terminal superposeed.

Summary of the invention

In view of this, the embodiment of the invention provide a kind of in mimo system the method and system of channel blind equalization, reduce the system complexity of realizing the channel blind equalization, improve system's performance of BER.

The embodiment of the invention provide a kind of in mimo system the channel blind balance method, the wireless channel between the transmitting antenna of transmitting terminal and the reception antenna of receiving terminal is separate, this method comprises:

Add prefix respectively in the data-signal of transmitting terminal on each transmitting antenna, obtain the transmission signal that auto-correlation function satisfies the auto-correlation independent condition, and send; Each reception antenna of receiving terminal receives above-mentioned transmission signal, and carries out the channel blind equalization based on the auto-correlation matching principle to received signal.

The embodiment of the invention provide a kind of in mimo system the system of channel blind equalization, comprise transmitting terminal and receiving terminal; Wherein, transmitting terminal is used for adding prefix at data-signal and obtains sending signal and transmission, and this prefix makes the auto-correlation function that respectively sends signal satisfy the auto-correlation independent condition; Receiving terminal is used for the signal that receiving end/sending end sends, and carries out to received signal obtaining sending the estimated value of signal based on the channel blind equalization of auto-correlation matching principle, and removes the estimated value that prefix obtains data-signal from the estimated value of above-mentioned transmission signal.

The channel blind balance method and the system that provide according to the embodiment of the invention, adding in the transmission signal on each transmitting antenna makes the auto-correlation function that respectively sends signal satisfy the prefix of auto-correlation independent condition, method by this interpolation prefix, reduction system effectively realizes the ground complexity, can improve the performance of BER of system significantly.

Description of drawings

Figure 1 shows that the schematic flow sheet of realizing the channel blind equalization in the embodiment of the invention;

Figure 2 shows that MIMO-OFDM system block diagram in the embodiment of the invention;

Figure 3 shows that the schematic diagram that adds prefix in the embodiment of the invention;

Figure 4 shows that and send signal s in the embodiment of the invention ₁The schematic diagram of auto-correlation function (n);

Figure 5 shows that and send signal s in the embodiment of the invention ₂The schematic diagram of auto-correlation function (n);

Figure 6 shows that the graph of relation of the embodiment of the invention and bit error rate of the prior art and signal to noise ratio;

Figure 7 shows that the graph of relation of the embodiment of the invention and bit error rate of the prior art and signal to noise ratio;

Figure 8 shows that the system block diagram of realizing the channel blind equalization in the embodiment of the invention.

Embodiment

For making the purpose, technical solutions and advantages of the present invention clearer, lift specific embodiment below, the present invention is further detailed explanation.

Fig. 1 is a schematic flow sheet of realizing blind equalization in the embodiment of the invention, comprises following step.

Step 101: add prefix respectively in the data-signal of transmitting terminal on each transmitting antenna and obtain sending signal, this prefix makes the auto-correlation function of the transmission signal on each transmitting antenna satisfy the auto-correlation independent condition.

At this, comprise a part of data-signal in the former data-signal in the prefix, be called Cyclic Prefix, the use of Cyclic Prefix guarantees to add the auto-correlation function that respectively sends signal that obtains after the prefix also peak value can occur in the position that is not equal to zero.In addition, each sends the prefix of adding in the signal and has different structure, can guarantee to add the auto-correlation function that respectively sends signal that obtains after the prefix like this and peak value occur in different positions respectively, satisfies the auto-correlation independent condition.Described different structure is meant the composition structure difference of the each several part in the prefix, promptly sends in the prefix of signal at each, and the position of Cyclic Prefix has nothing in common with each other.In the prefix, the part except Cyclic Prefix can adopt null sequence, and like this, the length of the null sequence of front and back that can be by adjusting Cyclic Prefix obtains the prefix of different structure.When each sends in the prefix of signal, the difference between the length of the null sequence of inserting between Cyclic Prefix and the data-signal is during greater than the twice of the exponent number of equalizer and channel exponent number sum, and each auto-correlation function that sends signal satisfies the linear displacement independent condition.

At this, the interpolation of prefix is to be unit with data transmission block in the mimo system, promptly adds above-mentioned prefix in each data transmission block.The length of prefix can be according to the channel circumstance setting, and the length of the each several part in the prefix also can specifically be provided with according to channel circumstance.

In sending signal, add prefix, not only can eliminate the influence of multidiameter delay, send the satisfied needed specified conditions of channel blind equalization of signal based on the auto-correlation matching principle but also can make.

Step 102: transmitting terminal sends signal arrives receiving terminal by wireless channel each reception antenna by each transmission antennas transmit corresponding transmitted signals.

Wherein, the wireless channel between transmitting terminal and the receiving terminal need satisfy: the wireless channel between each transmitting antenna and the reception antenna is separate.

Step 103: receiving terminal carries out the channel blind equalization based on the auto-correlation matching principle to the received signal on each reception antenna.

The concrete steps of channel blind equalization comprise following two steps:

Step 103-1: calculate the auto-correlation function and the auto-correlation function that sends signal of the signal that received signal obtains by equilibrium, and, obtain equalizing coefficient by auto-correlation function that mates the signal that equilibrium obtains and the auto-correlation function that sends signal.

At this, can utilize iterative algorithm, minimize the auto-correlation function of the signal that equilibrium obtains and send the difference between the auto-correlation function of signal absolute value square, obtain equalizing coefficient.

Step 103-2: utilize resulting equalizing coefficient, each received signal is carried out the estimated value that equilibrium is respectively sent signal.

Respectively sent the estimated value of signal by above step after, remove the estimated value that prefix can obtain data-signal.

Below, be given in detail in multi-I/O OFDM (MIMO-OFDM) system, realize channel blind balance method based on the auto-correlation matching principle.

Figure 2 shows that the MIMO-OFDM system block diagram.In this MIMO-OFDM system, transmitting terminal has N _tIndividual transmitting antenna, receiving terminal has N _rIndividual reception antenna.U among Fig. 2 _i(k), u _i(n) and s _i(n) be respectively i (i=1,2 ..., N _t) on the individual transmitting antenna frequency domain data signal, time domain data signal and send signal.x _j(n) be j (j=1,2 ..., N _r) received signal on the individual reception antenna,

With

Be respectively that receiving terminal is to sending the estimated value of signal, time domain data signal and frequency domain data signal.

In MIMO+OFDM system shown in Figure 2,, realize by following concrete steps based on the channel blind equalization of auto-correlation matching principle.

Step 1: at the time domain data signal u of each transmitting antenna _i(n) preceding interpolation prefix obtains sending signal s _i(n).

In the MIMO+OFDM system, the time domain data signal u of transmitting terminal _i(n) be meant the data that obtain by the OFDM modulation.Generally speaking, time domain data signal u _i(n) be white, send signal s in order to make _i(n) satisfy the required specified conditions of auto-correlation matching principle, the transmission signal s of each transmitting antenna _i(n) be at time domain data signal u _i(n) front is added different prefixes and is obtained, as shown in Figure 3.

As shown in Figure 3, send signal s _i(n) comprise that length is N _gPrefix and length be N _SubcData two parts.Wherein, send signal s for i _i(n), length is N _gPrefix be (N by length _g-N _Cp-N _i) null sequence, length be N _CpCyclic Prefix (CP) and length be N _iNull sequence form.Wherein, CP can be time domain data signal u _i(n) any N _CpIndividual continuous data-signal for example, can be time domain data signal u _i(n) back N _CpIndividual data-signal also can be the N of mid portion _CpIndividual data-signal.

Following surface analysis adds the autocorrelation performance of the transmission signal that obtains after the prefix in the time domain data signal according to structure as shown in Figure 3.

On i transmitting antenna transmission signal s _i(n) auto-correlation function can be expressed as:

$R_{s_i}\left(m\right)=\frac{1}{N}\underset{n=0}{\overset{(N_{\mathrm{subc}}+N_g)-1-\left|m\right|}{\mathrm{\Σ}}}{s}_i\left(n\right)s_i^{*}(n+m)$

Because time domain data signal u _i(n) be white, thus following formula (2) but approximate representation be:

In the following formula (3), $A=\frac{1}{N}\underset{n=0}{\overset{(N_{\mathrm{subc}}+N_g)-1}{\mathrm{\Σ}}}{\left|s_i\left(n\right)\right|}^2.$ By (3) formula as can be seen, for different transmission signal s _i(n), its auto-correlation function has nonzero value at different m value places, i.e. s except m=0 _i(n) auto-correlation function has following form:

In the following formula (4), for convenience's sake, the nonzero value in (3) formula is made as respectively, a=A and $b=A\frac{N_{\mathrm{cp}}}{N_{\mathrm{subc}}+N_{\mathrm{cp}}}.$ Obviously, when each auto-correlation function that sends signal has the form shown in (4) formula, and respectively send the N of the prefix of adding on the signal _iDifference between the length is greater than 2L _GThe time, the auto-correlation function collection $\left\{R_{s_i}(\mathrm{\τ}-l)\right|i=\mathrm{1,2},\·\·\·,N_t;l=0,\±1,\±2,\·\·\·,\±L_G\}$ It is linear independence.Wherein, L _G=L _w+ L, L _wBe the exponent number of equalizer, L is the exponent number of channel, and L _wSatisfy $L_w\≥\frac{N_{t}L}{N_r-N_t}-1.$

Step 2: transmitting terminal respectively send signal s _i(n) arrive receiving terminal by wireless channel, the received signal on each reception antenna is respectively x _i(n).

By Fig. 2, the received signal x on j reception antenna _j(n) can be expressed as:

$x_j\left(n\right)=\underset{i=1}{\overset{N_t}{\mathrm{\Σ}}}{h}_{\mathrm{ij}}{\left(n\right)*s}_i\left(n\right)+v_j\left(n\right)$

$=\underset{i=1}{\overset{N_t}{\mathrm{\Σ}}}\underset{m=0}{\overset{L-1}{\mathrm{\Σ}}}{h}_{\mathrm{ij}}\left(m\right)\×{s^{*}}_i(n-m)+v_j\left(n\right)$ 0?≤n≤(N _subc+N _g)-1 (5)

In the formula (5), h _Ij(n), n=0,1 ..., L-1 is the channel impulse response between i transmitting antenna and j the reception antenna, L is the exponent number of channel, v _j(n) be that average is zero on j the reception antenna, variance is σ _v ²Additive white Gaussian noise, N _SubcBe the sub-carrier number of OFDM, N _gBe the length of prefix.

Step 3: receiving terminal is x to received signal _j(n) carry out channel blind equalization, obtain sending signal s based on the auto-correlation matching principle _i(n) estimated value

Sending signal s _i(n) have under the situation of prefix as shown in Figure 3, hypothesis sends signal s simultaneously _i(n) and noise v _j(n) be incoherent mutually, and wireless channel h _Ij(n), n=0,1 ..., each the antithetical phrase interchannel among the L-1 does not have common factor, then can be according to the channel blind equalization of following method realization based on the auto-correlation matching principle.

Consider i (i=1,2 ..., N _t) individual transmitting antenna s _i(n).By Fig. 2, s _i(n) estimated value

Can be expressed as:

${\hat{s}}_i\left(n\right)=\underset{j=1}{\overset{N_r}{\mathrm{\Σ}}}{w}_{\mathrm{ji}}\left(n\right)*x_j\left(n\right)---\left(6\right)$

In the formula (6), w _Ji(n), n=0,1 ..., L _w-1, j=1 ..., N _rIt is the tap coefficient between j receiving terminal and i output of equalizer.

The tap coefficient w of equalizer _Ji(n) be to obtain, be specially: calculate received signal x by the iterative algorithm recursion _j(n) signal of exporting by equalizer

Auto-correlation function $R_{{\hat{s}}_i}\left(m\right)=E\left[{\hat{s}}_i\left(n\right){\hat{s}}_i^{*}(n-m)\right]$ With transmission signal s _i(n) auto-correlation function $R_{s_i}\left(m\right)=E\left[s_i\left(n\right)s_i^{*}(n-m)\right],$ Utilize iterative algorithm then, with auto-correlation function

With

Mate, obtain the tap coefficient w of equalizer _Ji(n), n=0,1 ..., L _w-1, j=1 ..., N _r

Can prove, work as equalizer output signal

Auto-correlation function equal to send signal s _iDuring (n) auto-correlation function, can realize the zero forcing equalization of channel.Therefore, can be by minimizing equalizer output signal

Auto-correlation function with send signal s _iThe absolute value of the difference between the auto-correlation function (n) square, utilize iterative algorithm to obtain the tap coefficient of equalizer.That is, can try to achieve the tap coefficient w of equalizer by the cost function in the formula of minimizing (7) _Ji(n).

$J=\underset{m}{\mathrm{\Σ}}{|R_{{\hat{s}}_i}\left(m\right)-R_{s_i}\left(m\right)|}^2---\left(7\right)$

After the tap coefficient of equalizer is obtained, can obtain sending signal s according to formula (6) _i(n) estimated value

Obtain sending the estimated value of signal

After, from

N before middle the removal _gIndividual prefix can obtain the estimated value of data-signal

Be given in below in the MIMO-OFDM system as shown in Figure 2, as described above the equalization performance that obtains of method emulation.

At first suppose simulated conditions: transmitting antenna N _t=2, reception antenna N _r=4, system bandwidth is 10MHz, the number N of OFDM subcarrier _Subc=1024, the prefix length N of each signal frame _g=1/4 * N _Subc, the length N of Cyclic Prefix _Cp=1/4 * N _Subc-50.In the prefix of being added in the transmission signal on the transmission signal on first transmitting antenna and second transmitting antenna, the length N of null sequence ₁And N ₂Be respectively N ₁=0 and N ₂=20.The exponent number of used channel is respectively L=2 and L=4 in the emulation, and the time delay in each footpath evenly distributes between 0～2us and 0～4us respectively, and it is zero that average is obeyed in the decline in each footpath, and variance is 0.8 multiple Gaussian Profile.The exponent number of equalizer is L _w=5.

Fig. 4 and Fig. 5 have provided transmission signal s respectively ₁(n) and s ₂(n) auto-correlation function.As can be seen, consistent with situation about being analyzed in (3) formula, s ₁(n) and s ₂(n) auto-correlation function is except the m=0 place is non-vanishing, also respectively at m=± N _SubcAnd m=± (N _Subc+ 20) locate to have nonzero value.Obviously, s ₁(n) and s ₂(n) auto-correlation function satisfies the auto-correlation independent condition.

Fig. 6 has provided when the frequency domain data signal and has been respectively BPSK and QPSK modulation, and during the exponent number L=2 of channel, (Bit Error Rate is BER) with signal to noise ratio (Signal to Noise Ratio, the SNR) relation curve between for the bit error rate of system.Here signal to noise ratio is defined as the mean value of signal to noise ratio on each reception antenna:

$\mathrm{SNR}=\frac{1}{N_r}\underset{j=1}{\overset{N_r}{\mathrm{\Σ}}}{\left(\mathrm{SNR}\right)}_j---\left(7\right)$

Provided the systematic function curve that the method mentioned in employing " the A new multiplex-access scheme based on thediversity of autocorrelation " document two obtains among Fig. 6 simultaneously, the energy of the pilot signal of stack is 1/5 of a data-signal energy.As can be seen, owing to there be not the interference of pilot signal to data-signal in the method that the embodiment of the invention proposed, the performance of BER of system has had obvious improvement.It may be noted that when the method in the employing document two is carried out blind equalization to channel, for fear of because channel delay is expanded the interference between the adjacent data blocks of being brought, also is to need to add to protect at interval between the adjacent two data blocks.

Fig. 7 has provided when the frequency domain data signal and has been respectively BPSK and QPSK modulation, and during the exponent number L=4 of channel, (Bit Error Rate is BER) with signal to noise ratio (Signal to Noise Ratio, the SNR) relation curve between for the bit error rate of system.The conclusion that draws from Fig. 7 is similar with the conclusion that draws from Fig. 6.

Simulation result according to top can be clearly seen that, the channel blind balance method based on the auto-correlation matching principle that the embodiment of the invention proposed under the situation with low system implementation complexity, can effectively improve the performance of BER of system.

Below, be given in the system that realizes the channel blind equalization in the mimo system, as shown in Figure 8, in realizing the channel blind equalization systems, comprise transmitting terminal and receiving terminal.Wherein, transmitting terminal adds prefix and obtains sending signal and transmission in data-signal, and this prefix makes the auto-correlation function that respectively sends signal satisfy the auto-correlation independent condition; The signal that transmitting terminal sends arrives receiving terminal by channel; Behind the signal that the receiving terminal receiving end/sending end sends, carry out to received signal obtaining sending the estimated value of signal based on the channel blind equalization of auto-correlation matching principle, and from the estimated value that sends signal, remove the estimated value that prefix obtains data-signal.Concrete, transmitting terminal comprises prefix adding device and transmitting element, receiving terminal comprises receiving element, equalizer and prefix removal unit.Described prefix adding device for example receives data-signal the modulating unit from the previous stage processing unit, add in data-signal and export to described transmitting element after prefix obtains sending signal, this prefix makes the auto-correlation function that respectively sends signal satisfy the auto-correlation independent condition; The transmission signal that described transmitting element is used for being received sends; Described receiving element receives the signal that described transmitting element sends, and exports to described equalizer; Described equalizer carries out obtaining sending the estimated value of signal based on the channel blind equalization of auto-correlation matching principle to received signal, and passes to described prefix and remove the unit; Described prefix is removed the unit and is removed prefix obtain the estimated value of data-signal from the estimated value that sends signal, and the estimated value that will remove the data-signal that obtains after the prefix is exported to for example demodulating unit of next stage processing unit.In this system, the data-signal that is input to the prefix adding device is the signal that obtains by modulation, and the signal that prefix is removed unit output recovers initial data by demodulation.

Wherein, equalizer specifically comprises: equalizer output signal auto-correlation function computing unit, transmission signal auto-correlation function computing unit, tap coefficient computing unit, transmission signal estimation unit.Described transmission signal estimation unit obtains the tap coefficient of equalizer from described tap coefficient computing unit, and utilizes tap coefficient, carries out the estimated value that equilibrium obtains sending signal to received signal, and exports to described prefix and remove the unit; Described equalizer output signal auto-correlation function computing unit receives the signal of described transmission signal estimation unit output, and calculates the auto-correlation function of this signal, passes to described tap coefficient computing unit; Described transmission signal auto-correlation function computing unit calculates the auto-correlation function that sends signal, passes to described tap coefficient computing unit; Described tap coefficient computing unit is used for by coupling from the auto-correlation function of described equalizer output signal auto-correlation function computing unit reception and the auto-correlation function that receives from described transmission signal auto-correlation function computing unit, obtain the tap coefficient of equalizer, and pass to described transmission signal estimation unit.

The above only is preferred embodiment of the present invention, and is in order to restriction the present invention, within the spirit and principles in the present invention not all, any modification of being done, is equal to replacement, improvement etc., all should be included within protection scope of the present invention.

Claims (14)

1. channel blind balance method in multi-input multi-output system is characterized in that described method comprises:

Add prefix respectively in the data-signal of transmitting terminal on each transmitting antenna, obtain the transmission signal that auto-correlation function satisfies the auto-correlation independent condition, and send;

Each reception antenna of receiving terminal receives described transmission signal, and described received signal is carried out channel blind equalization based on the auto-correlation matching principle.

2. method according to claim 1 is characterized in that, comprises Cyclic Prefix in the described prefix, and described Cyclic Prefix is to extract to obtain from described data-signal.

3. method according to claim 2 is characterized in that, described Cyclic Prefix, and the position in the described prefix that respectively sends signal has nothing in common with each other.

4. method according to claim 3 is characterized in that, described prefix further comprises: first null sequence and second null sequence;

Described first null sequence is positioned at before the described Cyclic Prefix, and described second null sequence is between described Cyclic Prefix and described data-signal.

5. method according to claim 4 is characterized in that, described Cyclic Prefix each send position in the prefix of signal have nothing in common with each other for: in the described prefix that respectively sends signal, the length ratio of described first null sequence and second null sequence has nothing in common with each other.

6. method according to claim 4, it is characterized in that, the position of described Cyclic Prefix in the described prefix that respectively sends signal have nothing in common with each other for: in the described prefix that respectively sends signal, the difference between the described second null sequence length is greater than 2 times of equalizer exponent number and channel exponent number sum.

7. method according to claim 1 is characterized in that, described channel blind equalization based on the auto-correlation matching principle may further comprise the steps:

Calculate the auto-correlation function and the auto-correlation function that sends signal of the signal that received signal obtains by equilibrium, and, obtain equalizing coefficient by auto-correlation function that mates the signal that equilibrium obtains and the auto-correlation function that sends signal;

Utilize described equalizing coefficient to carry out equilibrium to received signal, respectively sent the estimated value of signal.

8. method according to claim 7, it is characterized in that, the described concrete steps that obtain equalizing coefficient are: utilize iterative algorithm, minimize the auto-correlation function of the signal that equilibrium obtains and send the difference between the auto-correlation function of signal absolute value square, obtain equalizing coefficient.

9. method according to claim 7 is characterized in that, after the step of the described estimated value that is respectively sent signal, described method further comprises: remove prefix from the described estimated value that respectively sends signal, obtain the estimated value of data-signal.

10. method according to claim 1 is characterized in that, in multi-input multi-output orthogonal frequency division multiplexing system, described data-signal is: modulate the data that obtain by OFDM.

11. the system of a channel blind equalization in multi-input multi-output system is characterized in that, comprises transmitting terminal and receiving terminal; Wherein,

Described transmitting terminal is used for adding prefix at data-signal and obtains sending signal and transmission, and described prefix makes the auto-correlation function that respectively sends signal satisfy the auto-correlation independent condition;

Described receiving terminal, be used for the signal that receiving end/sending end sends, and carry out to received signal obtaining sending the estimated value of signal based on the channel blind equalization of auto-correlation matching principle, and from the estimated value of described transmission signal, remove the estimated value that described prefix obtains data-signal.

12. system according to claim 11 is characterized in that, described transmitting terminal comprises: prefix adding device and transmitting element, wherein,

Described prefix adding device is used for adding prefix to obtain sending signal at data-signal;

Described transmitting element is used for described transmission signal is sent.

13. system according to claim 12 is characterized in that, described receiving terminal comprises: receiving element, equalizer and prefix are removed the unit, wherein,

Described receiving element is used to receive the signal that described transmitting element sends;

Described equalizer is used for described received signal is carried out obtaining sending based on the channel blind equalization of auto-correlation matching principle the estimated value of signal;

Described prefix is removed the unit, is used for removing the estimated value that prefix obtains data-signal from the estimated value of described transmission signal.

14. system according to claim 13 is characterized in that, described equalizer comprises: equalizer output signal auto-correlation function computing unit, transmission signal auto-correlation function computing unit, tap coefficient computing unit, transmission signal estimation unit, wherein,

Described transmission signal estimation unit is used for obtaining the tap coefficient of equalizer from described tap coefficient computing unit, and utilizes tap coefficient, carries out the estimated value that equilibrium obtains sending signal to received signal;

Described equalizer output signal auto-correlation function computing unit is used to calculate the auto-correlation function of the signal of described transmission signal estimation unit output;

Described transmission signal auto-correlation function computing unit is used to calculate the auto-correlation function of described transmission signal;

Described tap coefficient computing unit is used for obtaining the tap coefficient of equalizer by mating the auto-correlation function that auto-correlation function that described equalizer output signal auto-correlation function computing unit obtains and described transmission signal auto-correlation function computing unit obtain.

Images