CN101982953A - Frequency domain multi-dimensional parameterized model of broadband wireless communication channel and modeling method - Google Patents

Abstract

The invention discloses a frequency domain multi-dimensional parameterized model of a broadband wireless communication channel and a modeling method. The method comprises the following steps: (1) computing and acquiring a real-time frequency domain impulse response of the channel by utilizing transmission signal frequency domain data measured by the wireless communication channel and receiving signal frequency domain acquisition data measured by the wireless communication channel; (2) constructing a channel estimation matching function and a frequency domain parameterized model according to the needed multi-dimensional parameters; (3) searching multiple paths from the real-time frequency domain impulse response of the channel, and initializing multi-dimensional parameters of each of the multiple paths; (4) computing maximum likelihood estimation, and acquiring multi-dimensional parameter values of each of the multiple paths through iterative estimation; and (5) if the multi-dimensional parameter values of each of the multiple paths meet the condition of convergence, stopping the iterative estimation and outputting the finally estimated multi-dimensional parameter values of each of the multiple paths, and if the multi-dimensional parameter values of each of the multiple paths do not meet the condition of convergence, returning to the step (4). The invention meets the requirement of high-precision multi-dimensional parameterized modeling of the broadband wireless communication channel, and has the advantages of low realization complexity and high speed of convergence.

Description

Broadband wireless communications channel frequency domain multidimensional parameterized model and modeling method

Technical field

The invention belongs to wireless communication technology field, relate to a kind of broadband wireless communications channel frequency domain multidimensional parameterized model and modeling method.

Background technology

In recent years, the 4th third-generation mobile communication system is towards the technique direction development in broadband.The IMT-Advanced system bandwidth of International Telecommunication Union definition can reach 100MHz, and can support city microcellulor, city macrocellular, macrocellular high-speed mobile, multiple wireless scene such as indoor.Simultaneously, international WRC 07-11 conference has been a global evolution wireless mobile communications service negotiation abundant frequency spectrum resource is had laid a good foundation for forming the scale industry.But, because the residing frequency range of the various air interface physical layer of wireless communication system, application scenarios difference, the decline feature of wireless channel is also far from each other, especially show that time delay, Doppler frequency shift, spatial coherence etc. are for cell mobile communication systems of new generation, for example: key technology Effect on Performance such as LTE, B3G, 4G are example with the MIMO-OFDM system, research and experiment proves that high-precision wireless channel model is very important for the true capacity of accurately estimating the actual environment lower channel, directly influences the performance optimization of such communication equipment in the actual environment; In addition, utilize decline the also many testers of emulation or emulation instrument of high-precision wireless channel model to press for.

The signal processing that obtains the parameterized channel model of high accuracy multidimensional based on actual measurement is very complicated.The present invention be intended to by at a high speed, in real time, the continuous measurement magnanimity dissemination channel data of gathering, estimate Space Time high accuracy channel parameter quickly and accurately, set up succinct, effective, as to accord with true environment broadband wireless communications channel model, make it be suitable for application such as communication equipment optimization, channel test equipment, emulator, for improving systematic function, improving the efficient of the network optimization, bring huge facility.

Carrying out the nature of radio propagation measurement in actual application environment is the optimal path of understanding true radio channel characteristic.Common method of measurement can be divided into time-domain sampling and measure and the frequency domain sample measurement.Practical application request concentrates near the certain bandwidth of a certain frequency usually, wishes that the measuring-signal energy in the band is high as far as possible and smooth, is suppressed as far as possible outside the band.Time-domain sampling is measured the method that the PN sequence is periodically launched in common employing, yet the signal energy that this method of measurement obtains has fluctuating in band, it is serious to reveal outside the frequency domain band; Frequency domain sample is measured and is adopted a plurality of sub-carrier signals of emission in measuring frequency band usually, and the smooth and abundant inhibition zone of band build-in test signal energy is revealed outward.

Obtaining to need data are analyzed extraction actual measurement channel parameter behind the actual measurement channel data.Present actual measurement channel parameter estimation method can be three classes generally: the space is estimated, is estimated, determines parameter Estimation based on the parametrization of subspace.The realization of these algorithms generally all realizes on time domain at the time-domain sampling data.If what obtain is the frequency domain sample data, need in advance time domain to be arrived in the frequency domain sample data conversion.Algorithm has increased the step and the complexity of data processing in this, and the time domain precision of artificial quantized channel parameter.

Summary of the invention

Technical problem to be solved by this invention is: a kind of broadband wireless communications channel frequency domain multidimensional parametric modeling method is provided, this method can estimate the radio channel characteristic parameter, describe wireless channel comprehensively from the actual measurement channel data, satisfy the parameter demand to Channel Modeling;

In addition, the present invention also provides a kind of broadband wireless communications channel frequency domain multidimensional parameterized model of setting up according to said method.

For solving the problems of the technologies described above, the present invention adopts following technical scheme.

A kind of broadband wireless communications channel frequency domain multidimensional parametric modeling method is characterized in that described modeling method comprises:

Step 1 utilizes the received signal frequency domain image data of the transmit frequency domain data and the radio communication channel measurement of radio communication channel measurement to calculate the real-time frequency domain impulse response that obtains channel;

Step 2, as required the time-frequently-the multidimensional parameter in spatial domain makes up channel estimating adaptation function and frequency domain parameter model;

Step 3 is utilized described channel estimating adaptation function and frequency domain parameter model, searches out multipath from the real-time frequency domain impulse response of described channel, and the multidimensional parameter of every multipath of initialization;

Step 4 for every multipath, is utilized the real-time frequency domain impulse response of described channel and channel estimating adaptation function to calculate maximal possibility estimation, and is estimated to obtain the multidimensional parameter value of every multipath by iteration;

Step 5 if the multidimensional parameter value of every multipath satisfies the condition of convergence, then stops iteration and estimates, exports the final multidimensional parameter value of estimating of every multipath; Otherwise return step 4.

As a preferred embodiment of the present invention, the multidimensional parameter of described every multipath is that delay parameter, Doppler frequency shift parameter, wave beam arrive the angle of pitch, wave beam arrival bearing angle, wave beam and leave the angle of pitch, wave beam and leave partial parameters or whole parameters in azimuth, the complex magnitude parameter.

The implementation method of described step 1 is: the received signal frequency domain image data Y (f) of transmit frequency domain data X (f) and radio communication channel measurement that radio communication channel is measured is by the real-time frequency domain impulse response H (f) of matrix division arithmetic H (f)=Y (f)/X (f) acquisition channel, and wherein f is the interior frequency sampling point of measurement bandwidth of described broadband wireless communications channel.

The implementation method of described step 2 is: choose the channel estimating adaptation function that delay parameter τ and complex magnitude parameter alpha make up every multipath l

Wherein f is the interior frequency sampling point of measurement bandwidth of described broadband wireless communications channel, τ _lBe the delay parameter of multipath l, α _lComplex magnitude parameter for multipath l; If the always total L bar of multipath, then the frequency domain parameter model of radio communication channel is

$H\left(f\right|\mathrm{\τ},\mathrm{\α})=\underset{l=1}{\overset{L}{\mathrm{\Σ}}}{\mathrm{\α}}_l\exp \{-j2\mathrm{\πf}{\mathrm{\π}}_l\}.$

Described step 3 comprises following detailed step: steps A 1 according to the frequency domain parameter model of described radio communication channel, makes the complex magnitude parameter alpha of multipath l _lNormalizing, delay parameter τ _lBe the maximal possibility estimation variable, utilize the real-time frequency domain impulse response of described channel to make maximal possibility estimation, obtain the delay parameter initialization value of multipath l; Wherein the maximal possibility estimation function is

F wherein ₁..., f _MBe M sample point at frequency domain; Steps A 2 according to the frequency domain parameter model of described radio communication channel, makes the complex magnitude parameter alpha of multipath l _lNormalizing adopts the delay parameter initialization value reconstruct channel frequency domain impulse response of the multipath l that steps A 1 estimates, and with the channel frequency domain impulse response of the channel frequency domain impulse response after the reconstruct before divided by reconstruct, obtains the complex magnitude parameter initialization value of multipath l; Steps A 3 judges whether many delay parameter initialization value and complex magnitude parameter initialization values through l satisfy the loop ends condition, if satisfy condition then export the initiation parameter result, if do not satisfy condition then proceed steps A 4; The loop ends condition is: whether the number of times that judging circulates carries out need equals the multipath number of estimation; Steps A 4, frequency domain parameter model according to described radio communication channel, the delay parameter initialization value and the complex magnitude parameter initialization value reconstruct channel frequency domain impulse response that adopt steps A 1, step 62 to estimate, and with the real-time frequency domain impulse response of the channel frequency domain impulse response negative feedback after the reconstruct to channel, obtain the real-time frequency domain impulse response H (f) of the measured channel of renewal, return steps A 1.

Described step 4 comprises following detailed step: step B1, frequency domain parameter model according to described radio communication channel, order treats that the multidimensional parameter value of all multipaths that iteration is estimated is last iteration results estimated value, if be iteration first, make then that described each estimated parameter value is the corresponding initialization value of step 3 output, reconstruct channel frequency domain impulse response, and will give the real-time frequency domain impulse response of channel except that the reconstruct channel frequency domain impulse response negative feedback of other multipaths of this multipath that needs iteration to estimate, obtain the real-time frequency domain impulse response H of the channel of renewal _Update(f); Step B2, according to the frequency domain parameter model of described radio communication channel, the complex magnitude parameter alpha of the multipath l that iteration is estimated is treated in order _lNormalizing, the delay parameter τ of multipath l _lAs the variable of maximal possibility estimation, utilize the real-time frequency domain impulse response H of the measured channel of the described renewal of step B1 _Update(f) make maximal possibility estimation, obtain the delay parameter value of the multipath l of this iteration estimation; Wherein the maximal possibility estimation function is

f ₁..., f _MBe M sample point at frequency domain; Step B3, according to the frequency domain parameter model of described radio communication channel, the complex magnitude parameter alpha of the multipath l that iteration is estimated is treated in order ₁Normalizing adopts the delay parameter value reconstruct channel frequency domain impulse response of this iteration of the multipath l that step B2 obtains, and with the real-time frequency domain impulse response H of the channel frequency domain impulse response after the reconstruct divided by the measured channel of steps A 4 described renewals _Update(f), obtain the complex magnitude parameter value of the multipath l of this iteration estimation; Whether the delay parameter value of the multipath l that step B4, determining step B2 obtain satisfies the condition of convergence, if satisfy the condition of convergence then export delay parameter and complex magnitude parameter Estimation result; If do not satisfy the condition of convergence then return execution in step B1; The condition of convergence is: whether the iterations of judging delay parameter equals the multipath number that needs are estimated.

As another kind of preferred version of the present invention, the implementation method of described step 2 is: choose delay parameter τ, complex magnitude parameter alpha, Doppler frequency shift parameter v, wave beam arrival pitching angle theta ₂, wave beam arrival bearing angle φ ₂, wave beam leaves pitching angle theta ₁, wave beam leaves azimuth φ ₁, make up the channel estimating adaptation function of every multipath l

${\mathrm{\θ}}_l(f,t,{\stackrel{\→}{r}}_1,{\stackrel{\→}{r}}_2|{\mathrm{\τ}}_l,{\mathrm{\α}}_l,v_l,{\stackrel{\→}{\mathrm{\Ω}}}_{1,l},{\stackrel{\→}{\mathrm{\Ω}}}_{2,l})={\mathrm{\α}}_l\exp \left\{j2\mathrm{\πf}{\mathrm{\τ}}_l\right\}\exp \left\{{j2\mathrm{\πv}}_{l}t\right\}$

$\exp \left\{j\frac{2\mathrm{\π}}{\mathrm{\λ}}({\stackrel{\→}{\mathrm{\Ω}}}_{1,l}\·{\stackrel{\→}{r}}_1)\right\}\exp \left\{j\frac{2\mathrm{\π}}{\mathrm{\λ}}({\stackrel{\→}{\mathrm{\Ω}}}_{2,l}\·{\stackrel{\→}{r}}_2)\right\},$

Wherein f is the interior frequency sampling point of measurement bandwidth of described broadband wireless communications channel, and t is the time sampling point in the Measuring Time,

The vector position of sampled point between the transmitting terminal multi-antenna space during for many antenna measurements, The vector position of sampled point between the receiving terminal multi-antenna space during for many antenna measurements, symbol tau _lBe the delay parameter of multipath l, α _lBe the complex magnitude parameter of multipath l, v _lDoppler frequency shift parameter for multipath l;

For the velocity of wave of multipath l leaves direction vector, comprise that velocity of wave leaves pitching angle theta ₁Leave azimuth φ with velocity of wave ₁Two parameters;

Velocity of wave arrival direction vector for multipath l comprises that velocity of wave arrives pitching angle theta ₂With velocity of wave arrival bearing angle φ ₂Two parameters; If the always total L bar of the multipath that receives, then the frequency domain parameter model of radio communication channel is

$H(f,t,{\stackrel{\→}{r}}_1,{\stackrel{\→}{r}}_2|\mathrm{\τ},\mathrm{\α},v,{\stackrel{\→}{\mathrm{\Ω}}}_1,{\stackrel{\→}{\mathrm{\Ω}}}_2)=\underset{l=1}{\overset{L}{\mathrm{\Σ}}}{\mathrm{\α}}_l\exp \{-j2\mathrm{\πf}{\mathrm{\τ}}_l\}\exp \{-{j2\mathrm{\πv}}_{l}t\}$

$\exp \{-j\frac{2\mathrm{\π}}{\mathrm{\λ}}({\stackrel{\→}{\mathrm{\Ω}}}_{1,l}\·{\stackrel{\→}{r}}_1)\}\exp \{-j\frac{2\mathrm{\π}}{\mathrm{\λ}}({\stackrel{\→}{\mathrm{\Ω}}}_{2,l}\·{\stackrel{\→}{r}}_2)\}.$

Described step 3 comprises following detailed step: steps A 1 ' according to the frequency domain parameter model of described radio communication channel, makes the complex magnitude parameter alpha of multipath l _lNormalizing makes multipath l remove delay parameter τ _lWith the complex magnitude parameter alpha _lOutside other parameters be 0, with delay parameter τ _lBe the maximal possibility estimation variable, utilize the real-time frequency domain impulse sound of described channel to make maximal possibility estimation, obtain the delay parameter initialization value of multipath l; Wherein the maximal possibility estimation function is that the real-time frequency domain impulse of described channel rings and the corresponding corresponding frequency point data of the channel estimating adaptation function delivery square of afterwards suing for peace that multiplies each other; Steps A 2 ' according to the frequency domain parameter model of described radio communication channel, makes the complex magnitude parameter alpha of multipath l _lNormalizing makes delay parameter τ _lDelay parameter initialization value for steps A 1 ' estimation, make multipath l that other parameters except that complex magnitude parameter, delay parameter and Doppler frequency shift parameter are 0, Doppler frequency shift parameter with multipath l is the maximal possibility estimation variable, utilize the real-time frequency domain impulse response of described channel to make maximal possibility estimation, obtain the Doppler frequency shift parameter initialization value of multipath; Wherein the maximal possibility estimation function is that the real-time frequency domain impulse of described channel rings and the corresponding corresponding frequency point data of the channel estimating adaptation function delivery square of afterwards suing for peace that multiplies each other; Steps A 3 '; Frequency domain parameter model according to described radio communication channel, the delay parameter that makes multipath l is the delay parameter value that steps A 1 ' is estimated, the Doppler frequency shift parameter is the Doppler frequency shift parameter value that steps A 2 ' is estimated, complex magnitude parameter normalizing, make multipath l remove delay parameter, the Doppler frequency shift parameter, other parameters outside complex magnitude parameter and the wave beam arrival direction vector parameters are 0, wave beam arrival direction vector parameters with multipath l is the maximal possibility estimation variable, utilize the real-time frequency domain impulse response of described channel to make maximal possibility estimation, obtain the wave beam arrival direction vector parameters initialization value of multipath l; Wherein the maximal possibility estimation function is that the real-time frequency domain impulse of described channel rings and the corresponding corresponding frequency point data of the channel estimating adaptation function delivery square of afterwards suing for peace that multiplies each other; Steps A 4 ', frequency domain parameter model according to described radio communication channel, the delay parameter that makes multipath l is the delay parameter value that steps A 1 ' is estimated, the Doppler frequency shift parameter that makes multipath l is the Doppler frequency shift parameter value that steps A 2 ' is estimated, the wave beam arrival direction vector that makes multipath l is the wave beam arrival direction vector parameters value that steps A 3 ' is estimated, make the complex magnitude parameter normalizing of multipath l, with wave beam departure direction vector parameters is the maximal possibility estimation variable, utilize the real-time frequency domain impulse response of described channel to make maximal possibility estimation, obtain the wave beam departure direction vector parameters initialization value of multipath l; Wherein the maximal possibility estimation function is that the real-time frequency domain impulse of described channel rings and the corresponding corresponding frequency point data of the channel estimating adaptation function delivery square of afterwards suing for peace that multiplies each other; Steps A 5 ', frequency domain parameter model according to described radio communication channel, make the complex magnitude parameter normalizing of multipath l, the delay parameter value, Doppler frequency shift parameter value, wave beam arrival direction vector parameters value, the wave beam departure direction vector parameters value that adopt steps A 1 ' to estimate to A4 ', reconstruct channel frequency domain impulse response, and with the channel frequency domain impulse response of the channel frequency domain impulse response after the reconstruct before, obtain the complex magnitude parameter initialization value of multipath l divided by reconstruct; Steps A 6 ' judges whether each parameter value satisfies the loop ends condition, if satisfy condition then export the initiation parameter result, if do not satisfy condition then proceed steps A 7 '; The loop ends condition is: whether the number of times that judging circulates carries out need equals the multipath number of estimation; Steps A 7 ', frequency domain parameter model according to described radio communication channel, the delay parameter value, Doppler frequency shift parameter value, wave beam arrival direction vector parameters value, the wave beam departure direction vector parameters value that adopt steps A 1 ' to estimate to A5 ', complex magnitude parameter value reconstruct channel frequency domain impulse response, and with the real-time frequency domain impulse response of the channel frequency domain impulse response negative feedback after the reconstruct to channel, obtain the real-time frequency domain impulse response H (f) of the channel of renewal, return steps A 1 '.

Described step 4 comprises following detailed step: step B1 ', frequency domain parameter model according to described radio communication channel, order treats that the multidimensional parameter value of all multipaths that iteration is estimated is last iteration results estimated value, if be iteration first, make then that described each estimated parameter value is the corresponding initialization value of step 3 output, reconstruct channel frequency domain impulse response, and will give the real-time frequency domain impulse response of channel except that the reconstruct channel frequency domain impulse response negative feedback of other multipaths of this multipath that needs iteration to estimate, obtain the real-time frequency domain impulse response H of the channel of renewal _Update(f); Step B2 ', frequency domain parameter model according to described radio communication channel, order treats that multipath l each estimated parameter value except that delay parameter that iteration is estimated is last iteration results estimated value, if be iteration first, make then that described each estimated parameter value is the corresponding initialization value of step 3 output, complex magnitude parameter normalizing is the maximal possibility estimation variable with the delay parameter, utilizes the real-time frequency domain impulse response H of the channel of described renewal _Update(f) make maximal possibility estimation, obtain the delay parameter value of the multipath l of this iteration estimation; Wherein the maximal possibility estimation function is that the real-time frequency domain impulse of described channel rings and the corresponding corresponding frequency point data of the channel estimating adaptation function delivery square of afterwards suing for peace that multiplies each other; Step B3 ', frequency domain parameter model according to described radio communication channel, order treats that multipath l each estimated parameter value except that the Doppler frequency shift parameter that iteration is estimated is last iteration results estimated value, if be iteration first, make then that described each estimated parameter value is the corresponding initialization value of step 3 output, complex magnitude parameter normalizing is the maximal possibility estimation variable with the Doppler frequency shift parameter, utilizes the real-time frequency domain impulse response H of the channel of described renewal _Update(f) make maximal possibility estimation, obtain the Doppler frequency shift parameter value of the multipath l of this iteration estimation; Wherein the maximal possibility estimation function is that the real-time frequency domain impulse of described channel rings and the corresponding corresponding frequency point data of the channel estimating adaptation function delivery square of afterwards suing for peace that multiplies each other; Step B4 ', frequency domain parameter model according to described radio communication channel, order treats that multipath l each estimated parameter value except that wave beam arrival direction vector parameters that iteration is estimated is last iteration results estimated value, if be iteration first, make then that described each estimated parameter value is the corresponding initialization value of step 3 output, complex magnitude parameter normalizing is the maximal possibility estimation variable with wave beam arrival direction vector parameters, utilizes the real-time frequency domain impulse response H of the channel of described renewal _Update(f) make maximal possibility estimation, obtain the wave beam arrival direction vector parameters value of the multipath l of this iteration estimation; Wherein the maximal possibility estimation function is that the real-time frequency domain impulse of described channel rings and the corresponding corresponding frequency point data of the channel estimating adaptation function delivery square of afterwards suing for peace that multiplies each other; Step B5 ', frequency domain parameter model according to described radio communication channel, order treats that multipath l each estimated parameter value except that wave beam departure direction vector parameters that iteration is estimated is last iteration results estimated value, if be iteration first, make then that described each estimated parameter value is the corresponding initialization value of step 3 output, complex magnitude parameter normalizing is the maximal possibility estimation variable with wave beam departure direction vector parameters, utilizes the real-time frequency domain impulse response H of the channel of described renewal _Update(f) make maximal possibility estimation, obtain the wave beam departure direction vector parameters value of the multipath l of this iteration estimation; Wherein the maximal possibility estimation function is that the real-time frequency domain impulse of described channel rings and the corresponding corresponding frequency point data of the channel estimating adaptation function delivery square of afterwards suing for peace that multiplies each other; Step B6 ', frequency domain parameter model according to described radio communication channel, the delay parameter value, Doppler frequency shift parameter value, wave beam arrival direction vector parameters value, the wave beam departure direction vector parameters value that adopt step B2 ' to estimate to B5 ', make complex magnitude parameter normalizing, reconstruct channel frequency domain impulse response, and with the real-time frequency domain impulse response H of the channel of channel frequency domain impulse response after the reconstruct and described renewal _Update(f) be divided by, obtain the complex magnitude parameter value of the multipath l of this iteration estimation; Step B7 ' judges whether to satisfy the condition of convergence, if satisfy the condition of convergence then export the parameter Estimation result of this iteration of multipath l; If do not satisfy condition then return step B2 '; The described condition of convergence is whether the iterations of judging each parameter equals the multipath number that needs are estimated.

A kind of broadband wireless communications channel frequency domain multidimensional parameterized model comprises:

The received signal frequency domain image data that transmit frequency domain data and the radio communication channel that utilizes that radio communication channel measures measured is calculated the actual measurement channel impulse response generation module of the real-time frequency domain impulse response that obtains channel;

Set up the frequency domain parameter model of radio communication channel and the frequency domain coupling MBM of channel estimating adaptation function according to the model parameter demand;

Utilize described frequency domain parameter model and channel estimating adaptation function from the real-time frequency domain impulse response of described channel, to search out multipath, and the frequency domain Multipath searching module of the parameter of every multipath of initialization;

For every multipath, utilize described actual measurement channel frequency domain impulse response and channel estimating adaptation function to calculate maximal possibility estimation, and obtain the frequency-domain model parameter iteration estimation module of the multidimensional parameter value of every multipath by the iteration estimation;

Judge whether to circulate and carry out the estimated parameter determination of stability module that iteration is estimated.

Beneficial effect of the present invention is: modeling method of the present invention can directly be handled the wide-band width measurement frequency domain sample data of measuring instrument output, has reduced implementation complexity;

Another beneficial effect of the present invention is: designed multidimensional and estimated channel parameter, can describe wireless channel comprehensively, not only satisfied present parameter demand to Channel Modeling, can also not be subjected to the restriction of time-domain sampling rate, improved the delay parameter estimated accuracy.

Description of drawings

Fig. 1 is a broadband wireless communications channel frequency domain multidimensional parametric modeling method flow chart of the present invention;

Fig. 2 is the initial method flow chart of two channel parameters among the embodiment one;

Fig. 3 is the iteration method of estimation flow chart of two channel parameters among the embodiment one;

Fig. 4 is the initial method flow chart of seven channel parameters among the embodiment two;

Fig. 5 is the iteration method of estimation flow chart of seven channel parameters among the embodiment two.

Embodiment

The invention discloses a kind of broadband wireless communications channel frequency domain multidimensional parameterized model and modeling method.This model comprises actual measurement channel impulse response generation module, frequency domain coupling MBM, frequency domain Multipath searching module, frequency-domain model parameter iteration estimation module, estimated parameter determination of stability module.The present invention can be based on channel wireless radio multi communication channel frequency domain measurement data, the multidimensional parameterized model of rapid extraction broadband wireless communications channel, keeping reducing the signal processing complexity effectively, the very suitable future wireless systems such as LTE, 4G that are applied under the high-precision situation.

Below in conjunction with accompanying drawing the specific embodiment of the present invention is described in further detail.

Embodiment one

Present embodiment provides a kind of radio communication channel frequency domain modeling method based on the actual measurement broadband, referring to Fig. 1, this method can be used for directly handling the broadband frequency domain sampled data of measuring instrument output, carries out high precision wireless communication channel multidimensional parameter Estimation, specifically may further comprise the steps:

Step 1: the received signal frequency domain image data Y (f) of transmit frequency domain data X (f) and radio communication channel measurement that the input radio communication channel is measured, by the real-time frequency domain impulse response H (f) of matrix division arithmetic H (f)=Y (f)/X (f) acquisition measured channel, wherein f is for measuring the frequency sampling point in the bandwidth.For example: the frequency domain data X (f) that transmits of measurement is identical with received signal frequency domain image data Y (f) bandwidth, all is the matrix of M * 1, and H (f)=Y (f)/X (f) is calculated as Y (f) is divided by with X (f) corresponding element, and obtaining H (f) is the matrix of M * 1.

Step 2: choose certain parameter according to demand, set up the frequency domain parameter model of actual measurement radio communication channel impulse response.

Present embodiment is chosen delay parameter τ, the complex magnitude parameter alpha makes up every channel estimating adaptation function of propagating multipath l

Wherein sample variation symbol f is for measuring the frequency sampling point in the bandwidth, τ _lBe the delay parameter of multipath l, α _lComplex magnitude parameter for multipath l.Suppose the always total L bar of the multipath that receives, the frequency domain parameter model of then surveying the radio communication channel impulse response is:

$H\left(f\right|\mathrm{\τ},\mathrm{\α})=\underset{l=1}{\overset{L}{\mathrm{\Σ}}}{\mathrm{\α}}_l\exp \{-j2\mathrm{\πf}{\mathrm{\π}}_l\}.$

Step 3: the frequency domain parameter model of the radio communication channel of setting up according to step 2, utilize the real-time frequency domain impulse response of the channel that step 1 obtains, each parameter value of initialization frequency domain parameter model.

The initial method flow process of two channel parameters as shown in Figure 2 in the present embodiment.

Step 301, the frequency domain parameter model of the radio communication channel of setting up according to step 2 makes complex magnitude parameter normalizing, is the maximal possibility estimation variable with the delay parameter, utilize the real-time frequency domain impulse response of channel to make maximal possibility estimation, obtain the delay parameter initialization value of each bar multipath.Wherein the maximal possibility estimation function is: earlier the measured channel frequency domain impulse response data on each frequency be multiply by the channel estimating adaptation function of the corresponding frequency of this multipath, again to sue for peace back delivery square of above result:

F wherein ₁..., f _MBe M sample point at frequency domain.

Step 302, the frequency domain parameter model of the radio communication channel of setting up according to step 2, the delay parameter initialization value that adopts step 301 to estimate, make complex magnitude parameter normalizing, reconstruct channel frequency domain impulse response, and with the real-time frequency domain impulse response of the channel frequency domain impulse response after the reconstruct, obtain the complex magnitude value of every multipath divided by channel.

Step 303 judges whether each parameter value of every many warps satisfies the loop ends condition, if satisfy condition then export the initiation parameter result, if do not satisfy condition then proceed step 304.The condition of convergence can be: whether the number of times that judging circulates carries out need equals the multipath number of estimation.

Step 304, the frequency domain parameter model of the radio communication channel of setting up according to step 2, the delay parameter value, the complex magnitude parameter value that adopt step 301, step 302 to estimate, reconstruct channel frequency domain impulse response, and with the real-time frequency domain impulse response of the channel frequency domain impulse response negative feedback after the reconstruct to channel, obtain the real-time frequency domain impulse response of the measured channel of renewal, estimate in order to the next round initialization.

Above step is carried out in circulation, judges until step 303 and finishes initialization.

Step 4: the frequency domain parameter model of the radio communication channel of setting up according to step 2, utilize the real-time frequency domain impulse response of the channel that step 1 obtains and each initiation parameter value of every multipath that step 3 obtains, iteration is estimated each parameter value of frequency-domain model, satisfies the condition of convergence until each parameter value.

Two channel parameter iteration method of estimation flow processs as shown in Figure 3 in the present embodiment.

Step 401, frequency domain parameter model according to described radio communication channel, order treats that the multidimensional parameter value of all multipaths that iteration is estimated is last iteration results estimated value, if be iteration first, make then that described each estimated parameter value is the corresponding initialization value of step 3 output, reconstruct channel frequency domain impulse response, and will give the real-time frequency domain impulse response of channel except that the reconstruct channel frequency domain impulse response negative feedback of other multipaths of this multipath that needs iteration to estimate, obtain the real-time frequency domain impulse response H of the channel of renewal _Update(f);

Step 402, the frequency domain parameter model of the radio communication channel of setting up according to step 2 makes complex magnitude parameter normalizing, and delay parameter is as the variable of maximal possibility estimation, utilizes the real-time frequency domain impulse response H of the measured channel of the renewal that step 3 obtains _Update(f) make maximal possibility estimation, obtain the delay parameter value of this iteration of every multipath.Wherein the maximal possibility estimation function is

f ₁..., f _MBe M sample point at frequency domain.

Step 403, the frequency domain parameter model of the radio communication channel of setting up according to step 2, the delay parameter value that adopts step 402 to estimate makes complex magnitude parameter normalizing, reconstruct channel frequency domain impulse response, and with the real-time frequency domain impulse response H of the channel frequency domain impulse response after the reconstruct divided by measured channel _Update(f) (step 402 output result) obtains the complex magnitude parameter value of every multipath.

Step 404, whether the delay parameter value that determining step 402 obtains satisfies the condition of convergence, if satisfy condition then the output parameter estimated result, if do not satisfy condition then proceed step 401.The condition of convergence can be: whether the iterations of judging delay parameter equals the multipath number that needs are estimated.

Step 5: whether the parameter Estimation result of determining step 4 outputs satisfies stable iterated conditional, if satisfy stable iterated conditional, then exports the estimated channel model parameter value; If do not satisfy stable iterated conditional, then return execution in step 4.

Embodiment two

The difference of present embodiment and embodiment one is to need to consider that time delay, complex magnitude, Doppler frequency shift, wave beam leave the angle of pitch, wave beam leaves azimuth, the velocity of wave arrival angle of pitch, seven parameters in wave beam arrival bearing angle.Below in conjunction with Fig. 1 present embodiment is described in further detail.

The concrete steps of present embodiment comprise:

Step 1 ': the frequency domain parameter model of setting up actual measurement radio communication channel impulse response according to demand.

According to the actual requirements, choose certain parameter, set up channel impulse response frequency domain parameter model.Present embodiment is chosen time delay, complex magnitude, Doppler frequency shift, wave beam and is left the angle of pitch, wave beam and leave that azimuth, velocity of wave arrive the angle of pitch, wave beam arrival bearing angle makes up every channel estimating adaptation function of propagating multipath l, and the expression formula of this function is:

${\mathrm{\θ}}_l(f,t,{\stackrel{\→}{r}}_1,{\stackrel{\→}{r}}_2|{\mathrm{\τ}}_l,{\mathrm{\α}}_l,v_l,{\stackrel{\→}{\mathrm{\Ω}}}_{1,l},{\stackrel{\→}{\mathrm{\Ω}}}_{2,l})={\mathrm{\α}}_l\exp \left\{j2\mathrm{\πf}{\mathrm{\τ}}_l\right\}\exp \left\{{j2\mathrm{\πv}}_{l}t\right\}$

$\exp \left\{j\frac{2\mathrm{\π}}{\mathrm{\λ}}({\stackrel{\→}{\mathrm{\Ω}}}_{1,l}\·{\stackrel{\→}{r}}_1)\right\}\exp \left\{j\frac{2\mathrm{\π}}{\mathrm{\λ}}({\stackrel{\→}{\mathrm{\Ω}}}_{2,l}\·{\stackrel{\→}{r}}_2)\right\},$

Comprising the frequency sampling point in four sample variation symbol: f measurement bandwidth that is described broadband wireless communications channel; T is the time sampling point in the Measuring Time;

The vector position of sampled point between the transmitting terminal multi-antenna space during for many antenna measurements;

The vector position of sampled point between the receiving terminal multi-antenna space during for many antenna measurements.Comprise five parameter symbol: τ in addition _lBe the delay parameter of multipath l, α _lBe the complex magnitude parameter of multipath l, v _lDoppler frequency shift parameter for multipath l;

For the velocity of wave of multipath l leaves direction vector, comprise that velocity of wave leaves pitching angle theta ₁Leave azimuth φ with velocity of wave ₁Two parameters;

Velocity of wave arrival direction vector for multipath l comprises that velocity of wave arrives pitching angle theta ₂With velocity of wave arrival bearing angle φ ₂Two parameters; Amount to seven channel parameters.

Suppose the always total L bar of the multipath that receives, then the frequency domain parameter model of radio communication channel is:

$H(f,t,{\stackrel{\→}{r}}_1,{\stackrel{\→}{r}}_2|\mathrm{\τ},\mathrm{\α},v,{\stackrel{\→}{\mathrm{\Ω}}}_1,{\stackrel{\→}{\mathrm{\Ω}}}_2)=\underset{l=1}{\overset{L}{\mathrm{\Σ}}}{\mathrm{\α}}_l\exp \{-j2\mathrm{\πf}{\mathrm{\τ}}_l\}\exp \{-{j2\mathrm{\πv}}_{l}t\}$

$\exp \{-j\frac{2\mathrm{\π}}{\mathrm{\λ}}({\stackrel{\→}{\mathrm{\Ω}}}_{1,l}\·{\stackrel{\→}{r}}_1)\}\exp \{-j\frac{2\mathrm{\π}}{\mathrm{\λ}}({\stackrel{\→}{\mathrm{\Ω}}}_{2,l}\·{\stackrel{\→}{r}}_2)\}.$

Step 2 ': the received signal frequency domain image data Y (f) of transmit frequency domain data X (f) and radio communication channel measurement that the input radio communication channel is measured, by matrix division arithmetic H (f)=Y (f)/X (f), obtain the real-time frequency domain impulse response H (f) of measured channel, wherein f is for measuring the frequency sampling point in the bandwidth.By the time-frequently-empty multidimensional measure, can obtain channel multidimensional frequency domain sample signal data By computing, can obtain corresponding multidimensional measure channel frequency domain impulse response data again More than each variable implication be: f is for measuring the frequency sampling point in the bandwidth; T is the time sampling point in the Measuring Time;

During for many antenna measurements, the vector position of sampled point between the transmitting terminal multi-antenna space;

During for many antenna measurements, the vector position of sampled point between the receiving terminal multi-antenna space.For example: measure the frequency spectrum data X (f) that transmits and be the matrix of M * 1, sampled signal data Be the four-matrix of M * K * P * Q, H (f)=Y (f)/X (f) is calculated as: will

Four-matrix takes out the one dimension matrix Y (f) relevant with frequency domain bandwidth successively, and its size is M * 1, is divided by with X (f) corresponding element, and obtaining H (f) is the matrix of M * 1; Cycle calculations is finished

All dimensions, acquisition

Four-matrix for M * K * P * Q.

The frequency domain parameter model of the radio communication channel that step 3 ': according to step 1 ' is set up utilizes step 2 ' the real-time frequency domain impulse response of the measured channel that obtains, each parameter value of initialization frequency-domain model.The flow process of seven channel parameter initial methods as shown in Figure 4 in the present embodiment.

Step 301 ', according to step 1 ' the frequency domain parameter model of the radio communication channel set up, make complex magnitude parameter normalizing, order other parameters except that delay parameter are 0, with delay parameter for doing the maximal possibility estimation variable, utilize the real-time frequency domain impulse response of measured channel to make maximal possibility estimation, obtain the delay parameter initialization value of every multipath.Wherein the maximal possibility estimation function is that the real-time frequency domain impulse of described channel rings and the corresponding corresponding frequency point data of the channel estimating adaptation function delivery square of afterwards suing for peace that multiplies each other.

Step 302 ', according to step 1 ' the frequency domain parameter model of the radio communication channel set up, make that time delay is a step 301 ' the delay parameter value estimated, make complex magnitude parameter normalizing, order other parameters except that the Doppler frequency shift parameter are 0, for doing the maximal possibility estimation variable, utilize the real-time frequency domain impulse response of measured channel to make maximal possibility estimation with the Doppler frequency shift parameter, obtain the Doppler frequency shift parameter initialization value of every multipath.Wherein the maximal possibility estimation function is that the real-time frequency domain impulse of described channel rings and the corresponding corresponding frequency point data of the channel estimating adaptation function delivery square of afterwards suing for peace that multiplies each other.

Step 303 ', according to step 1 ' the frequency domain parameter model of the radio communication channel set up, make that time delay is a step 301 ' the delay parameter value estimated, make that Doppler frequency shift is a step 302 ' the Doppler frequency shift parameter value estimated, make complex magnitude parameter normalizing, order removes wave beam arrival direction vector, and (that is: velocity of wave arrives the angle of pitch, wave beam arrival bearing angle) other parameters beyond the parameter are 0, with wave beam arrival direction vector parameters for doing the maximal possibility estimation variable, utilize the real-time frequency domain impulse response of measured channel to make maximal possibility estimation, obtain the wave beam arrival direction vector parameters initialization value of every multipath; Wherein the maximal possibility estimation function is that the real-time frequency domain impulse of described channel rings and the corresponding corresponding frequency point data of the channel estimating adaptation function delivery square of afterwards suing for peace that multiplies each other.

Step 304 ', according to step 1 ' the frequency domain parameter model of the radio communication channel set up, make that time delay is a step 301 ' the delay parameter value estimated, make that Doppler frequency shift is a step 302 ' the Doppler frequency shift parameter value estimated, make that wave beam arrival direction vector is a step 303 ' the wave beam arrival direction vector parameters value estimated, make complex magnitude parameter normalizing, with wave beam departure direction vector parameters for doing the maximal possibility estimation variable, utilize the real-time frequency domain impulse response of measured channel to make maximal possibility estimation, obtain the wave beam departure direction vector parameters initialization value of every multipath.Wherein the maximal possibility estimation function is that the real-time frequency domain impulse of described channel rings and the corresponding corresponding frequency point data of the channel estimating adaptation function delivery square of afterwards suing for peace that multiplies each other.

Step 305 ', according to step 1 ' the frequency domain parameter model of the radio communication channel set up, the delay parameter value, Doppler frequency shift parameter value, wave beam arrival direction vector parameters value, the wave beam departure direction vector parameters value that adopt above 301 ' to 304 ' step to estimate, make complex magnitude parameter normalizing, reconstruct channel frequency domain impulse response, and with the real-time frequency domain impulse response of the channel frequency domain impulse response after the reconstruct, obtain the complex magnitude value of every multipath divided by measured channel.

Step 306 ', judge whether each parameter value satisfies the loop ends condition, if satisfy condition then export the initiation parameter result, if do not satisfy condition then proceed step 307 '.The loop ends condition is: whether the number of times that judging circulates carries out need equals the multipath number of estimation;

Step 307 ', according to step 1 ' the frequency domain parameter model of the radio communication channel set up, the delay parameter value, Doppler frequency shift parameter value, wave beam arrival direction vector parameters value, the wave beam departure direction vector parameters value that adopt above 301 ' to 305 ' step to estimate, the complex magnitude parameter value, reconstruct channel frequency domain impulse response, and with the real-time frequency domain impulse response of the channel frequency domain impulse response negative feedback after the reconstruct to measured channel, obtain the real-time frequency domain impulse response of the measured channel of renewal, estimate in order to the next round initialization.

Above step is carried out in circulation, until step 306 ' judge the end initialization, wherein loop ends condition can be: whether the number of times that the judgement circulation is carried out need equals the multipath number of estimation.

The frequency domain parameter model of the radio communication channel that step 4 ': according to step 1 ' is set up, the real-time frequency domain impulse response and the step 3 of measured channel ' obtain ' each the initiation parameter value that obtains of utilizing step 2, iteration is estimated each parameter value of frequency-domain model, until satisfying the condition of convergence.The flow process of seven channel parameter iteration methods of estimation as shown in Figure 5 in the present embodiment.

Step 401 ', the frequency domain parameter model of the radio communication channel of setting up according to step 1, order treats that the multidimensional parameter value of all multipaths that iteration is estimated is last iteration results estimated value, if be iteration first, make then that described each estimated parameter value is the corresponding initialization value of step 3 output, reconstruct channel frequency domain impulse response, and will give the real-time frequency domain impulse response of measured channel except that the reconstruct channel frequency domain impulse response negative feedback of other multipaths the multipath that this iteration is estimated, obtain the real-time frequency domain impulse response H of the measured channel of renewal _Update(f), estimate in order to iteration.

Step 402 ', according to step 1 ' the frequency domain parameter model of the radio communication channel set up, order other each estimated parameter values except that delay parameter are that last iteration results estimated value is (if be iteration first, then making other each estimated parameter values is step 3 ' output corresponding initialization value), complex magnitude parameter normalizing, with the delay parameter is the maximal possibility estimation variable, utilize the real-time frequency domain impulse response of the measured channel of upgrading to make maximal possibility estimation, obtain the delay parameter value of the multipath of this iteration estimation.Wherein the maximal possibility estimation function is that the real-time frequency domain impulse of described channel rings and the corresponding corresponding frequency point data of the channel estimating adaptation function delivery square of afterwards suing for peace that multiplies each other.

Step 403 ', according to step 1 ' the frequency domain parameter model of the radio communication channel set up, order other each estimated parameter values except that the Doppler frequency shift parameter are that last iteration results estimated value is (if be iteration first, then making other each estimated parameter values is step 3 ' output corresponding initialization value), complex magnitude parameter normalizing, with the Doppler frequency shift parameter is the maximal possibility estimation variable, utilize the real-time frequency domain impulse response of the measured channel of upgrading to make maximal possibility estimation, obtain the Doppler frequency shift parameter value of the multipath of this iteration estimation.Wherein the maximal possibility estimation function is that the real-time frequency domain impulse of described channel rings and the corresponding corresponding frequency point data of the channel estimating adaptation function delivery square of afterwards suing for peace that multiplies each other.

Step 404 ', according to step 1 ' the frequency domain parameter model of the radio communication channel set up, order other each estimated parameter values except that wave beam arrival direction vector parameters are that last iteration results estimated value is (if be iteration first, then making other each estimated parameter values is step 3 ' output corresponding initialization value), complex magnitude parameter normalizing, with wave beam arrival direction vector parameters is the maximal possibility estimation variable, utilize the real-time frequency domain impulse response of the measured channel of upgrading to make maximal possibility estimation, obtain the wave beam arrival direction vector parameters value of the multipath of this iteration estimation.Wherein the maximal possibility estimation function is that the real-time frequency domain impulse of described channel rings and the corresponding corresponding frequency point data of the channel estimating adaptation function delivery square of afterwards suing for peace that multiplies each other.

Step 405 ', according to step 1 ' the frequency domain parameter model of the radio communication channel set up, order other each estimated parameter values except that wave beam departure direction vector parameters are that last iteration results estimated value is (if be iteration first, then making other each estimated parameter values is step 3 ' output corresponding initialization value), complex magnitude parameter normalizing, with wave beam departure direction vector parameters is the maximal possibility estimation variable, utilize the real-time frequency domain impulse response of the measured channel of upgrading to make maximal possibility estimation, obtain the wave beam departure direction vector parameters value of the multipath of this iteration estimation.Wherein the maximal possibility estimation function is that the real-time frequency domain impulse of described channel rings and the corresponding corresponding frequency point data of the channel estimating adaptation function delivery square of afterwards suing for peace that multiplies each other.

Step 406 ', according to step 1 ' the frequency domain parameter model of the radio communication channel set up, adopt step 402 ' to 405 ' delay parameter value, Doppler frequency shift parameter value, wave beam arrival direction vector parameters value, the wave beam departure direction vector parameters value of estimating, make complex magnitude parameter normalizing, reconstruct channel frequency domain impulse response, and with the real-time frequency domain impulse response H of the measured channel of channel frequency domain impulse response after the reconstruct and renewal _Update(step 401 ' the output result) be divided by, obtain the complex magnitude value of the multipath of this iteration estimation.

Step 407 ', judge whether each parameter value of the multipath that this iteration is estimated satisfies the condition of convergence, if satisfy condition then export the parameter Estimation result of this iteration; If do not satisfy condition then return step 402 ', wherein, the condition of convergence can be: whether the iterations of judging each parameter need equals the multipath number estimated.

Step 5 ': whether the parameter Estimation result of determining step 4 ' output satisfies stable iterated conditional, if satisfy stable iterated conditional, then exports the estimated channel model parameter value; If do not satisfy stable iterated conditional, execution in step 4 ' then circulates.

The present invention can choose the part in following seven parameters according to the actual requirements, and these seven parameters are: time delay, complex magnitude, Doppler frequency shift, wave beam leave the angle of pitch, wave beam leaves azimuth, the velocity of wave arrival angle of pitch, wave beam arrival bearing angle.

Here description of the invention and application is illustrative, is not to want with scope restriction of the present invention in the above-described embodiments.Here the distortion of disclosed embodiment and change are possible, and the various parts of the replacement of embodiment and equivalence are known for those those of ordinary skill in the art.Those skilled in the art are noted that under the situation that does not break away from spirit of the present invention or substantive characteristics, and the present invention can be with other forms, structure, layout, ratio, and realize with other elements, material and parts.

Claims (10)

1. broadband wireless communications channel frequency domain multidimensional parametric modeling method is characterized in that described modeling method comprises:

Step 1 utilizes the received signal frequency domain image data of the transmit frequency domain data and the radio communication channel measurement of radio communication channel measurement to calculate the real-time frequency domain impulse response that obtains channel;

Step 2, as required the time-frequently-the multidimensional parameter in spatial domain makes up channel estimating adaptation function and frequency domain parameter model;

Step 3 is utilized described channel estimating adaptation function and frequency domain parameter model, searches out multipath from the real-time frequency domain impulse response of described channel, and the multidimensional parameter of every multipath of initialization;

Step 4 for every multipath, is utilized the real-time frequency domain impulse response of described channel and channel estimating adaptation function to calculate maximal possibility estimation, and is estimated to obtain the multidimensional parameter value of every multipath by iteration;

Step 5 if the multidimensional parameter value of every multipath satisfies the condition of convergence, then stops iteration and estimates, exports the final multidimensional parameter value of estimating of every multipath; Otherwise return step 4.

2. broadband wireless communications channel frequency domain multidimensional parametric modeling method according to claim 1 is characterized in that: the multidimensional parameter of described every multipath is that delay parameter, Doppler frequency shift parameter, wave beam arrive the angle of pitch, wave beam arrival bearing angle, wave beam and leave the angle of pitch, wave beam and leave partial parameters or whole parameters in azimuth, the complex magnitude parameter.

3. broadband wireless communications channel frequency domain multidimensional parametric modeling method according to claim 2, it is characterized in that, the implementation method of described step 1 is: the received signal frequency domain image data Y (f) of transmit frequency domain data X (f) and radio communication channel measurement that radio communication channel is measured is by the real-time frequency domain impulse response H (f) of matrix division arithmetic H (f)=Y (f)/X (f) acquisition channel, and wherein f is the interior frequency sampling point of measurement bandwidth of described broadband wireless communications channel.

4. broadband wireless communications channel frequency domain multidimensional parametric modeling method according to claim 2 is characterized in that the implementation method of described step 2 is: choose the channel estimating adaptation function that delay parameter τ and complex magnitude parameter alpha make up every multipath l Wherein f is the interior frequency sampling point of measurement bandwidth of described broadband wireless communications channel, τ _lBe the delay parameter of multipath l, α _lComplex magnitude parameter for multipath l; If the always total L bar of multipath, then the frequency domain parameter model of radio communication channel is

5. broadband wireless communications channel frequency domain multidimensional parametric modeling method according to claim 4 is characterized in that described step 3 comprises following detailed step:

Steps A 1 according to the frequency domain parameter model of described radio communication channel, makes the complex magnitude parameter alpha of multipath l _lNormalizing, delay parameter τ _lBe the maximal possibility estimation variable, utilize the real-time frequency domain impulse response of described channel to make maximal possibility estimation, obtain the delay parameter initialization value of multipath l; Wherein the maximal possibility estimation function is F wherein ₁..., f _MBe M sample point at frequency domain;

Steps A 2 according to the frequency domain parameter model of described radio communication channel, makes the complex magnitude parameter alpha of multipath l _lNormalizing adopts the delay parameter initialization value reconstruct channel frequency domain impulse response of the multipath l that steps A 1 estimates, and with the channel frequency domain impulse response of the channel frequency domain impulse response after the reconstruct before divided by reconstruct, obtains the complex magnitude parameter initialization value of multipath l;

Steps A 3 judges whether many delay parameter initialization value and complex magnitude parameter initialization values through l satisfy the loop ends condition, if satisfy condition then export the initiation parameter result, if do not satisfy condition then proceed steps A 4; The loop ends condition is: whether the number of times that judging circulates carries out need equals the multipath number of estimation;

Steps A 4, frequency domain parameter model according to described radio communication channel, the delay parameter initialization value and the complex magnitude parameter initialization value reconstruct channel frequency domain impulse response that adopt steps A 1, steps A 2 to estimate, and with the real-time frequency domain impulse response of the channel frequency domain impulse response negative feedback after the reconstruct to channel, obtain the real-time frequency domain impulse response H (f) of the measured channel of renewal, return steps A 1.

6. broadband wireless communications channel frequency domain multidimensional parametric modeling method according to claim 5 is characterized in that described step 4 comprises following detailed step:

Step B1, frequency domain parameter model according to described radio communication channel, order treats that the multidimensional parameter value of all multipaths that iteration is estimated is last iteration results estimated value, if be iteration first, make then that described each estimated parameter value is the corresponding initialization value of step 3 output, reconstruct channel frequency domain impulse response, and will give the real-time frequency domain impulse response of channel except that the reconstruct channel frequency domain impulse response negative feedback of other multipaths of this multipath that needs iteration to estimate, obtain the real-time frequency domain impulse response H of the channel of renewal _Update(f);

Step B2, according to the frequency domain parameter model of described radio communication channel, the complex magnitude parameter alpha of the multipath l that iteration is estimated is treated in order _lNormalizing, the delay parameter τ of multipath l _lAs the variable of maximal possibility estimation, utilize the real-time frequency domain impulse response H of the measured channel of the described renewal of step B1 _Update(f) make maximal possibility estimation, obtain the delay parameter value of the multipath l of this iteration estimation; Wherein the maximal possibility estimation function is

f ₁..., f _MBe M sample point at frequency domain;

Step B3, according to the frequency domain parameter model of described radio communication channel, the complex magnitude parameter alpha of the multipath l that iteration is estimated is treated in order _lNormalizing adopts the delay parameter value reconstruct channel frequency domain impulse response of this iteration of the multipath l that step B2 obtains, and with the real-time frequency domain impulse response H of the channel frequency domain impulse response after the reconstruct divided by the measured channel of steps A 4 described renewals _Update(f), obtain the complex magnitude parameter value of the multipath l of this iteration estimation;

Whether the delay parameter value of the multipath l that step B4, determining step B2 obtain satisfies the condition of convergence, if satisfy the condition of convergence then export delay parameter and complex magnitude parameter Estimation result; If do not satisfy the condition of convergence then return execution in step B1; The condition of convergence is: whether the iterations of judging delay parameter equals the multipath number that needs are estimated.

7. broadband wireless communications channel frequency domain multidimensional parametric modeling method according to claim 2 is characterized in that the implementation method of described step 2 is: choose delay parameter τ, complex magnitude parameter alpha, Doppler frequency shift parameter v, wave beam arrival pitching angle theta ₂, wave beam arrival bearing angle φ ₂, wave beam leaves pitching angle theta ₁, wave beam leaves azimuth φ ₁, make up the channel estimating adaptation function of every multipath l

Wherein f is the interior frequency sampling point of measurement bandwidth of described broadband wireless communications channel, and t is the time sampling point in the Measuring Time,

The vector position of sampled point between the transmitting terminal multi-antenna space during for many antenna measurements, The vector position of sampled point between the receiving terminal multi-antenna space during for many antenna measurements, symbol tau _lBe the delay parameter of multipath l, α _lBe the complex magnitude parameter of multipath l, v _lDoppler frequency shift parameter for multipath l;

For the velocity of wave of multipath l leaves direction vector, comprise that velocity of wave leaves pitching angle theta ₁Leave azimuth φ with velocity of wave ₁Two parameters;

Velocity of wave arrival direction vector for multipath l comprises that velocity of wave arrives pitching angle theta ₂With velocity of wave arrival bearing angle φ ₂Two parameters; If the always total L bar of the multipath that receives, then the frequency domain parameter model of radio communication channel is

。

8. broadband wireless communications channel frequency domain multidimensional parametric modeling method according to claim 7 is characterized in that described step 3 comprises following detailed step:

Steps A 1 ' according to the frequency domain parameter model of described radio communication channel, makes the complex magnitude parameter alpha of multipath l _lNormalizing makes multipath l remove delay parameter τ _lWith the complex magnitude parameter alpha _lOutside other parameters be 0, with delay parameter τ _lBe the maximal possibility estimation variable, utilize the real-time frequency domain impulse sound of described channel to make maximal possibility estimation, obtain the delay parameter initialization value of multipath l; Wherein the maximal possibility estimation function is that the real-time frequency domain impulse of described channel rings and the corresponding corresponding frequency point data of the channel estimating adaptation function delivery square of afterwards suing for peace that multiplies each other;

Steps A 2 ' according to the frequency domain parameter model of described radio communication channel, makes the complex magnitude parameter alpha of multipath l _lNormalizing makes delay parameter τ _lDelay parameter initialization value for steps A 1 ' estimation, make multipath l that other parameters except that complex magnitude parameter, delay parameter and Doppler frequency shift parameter are 0, Doppler frequency shift parameter with multipath l is the maximal possibility estimation variable, utilize the real-time frequency domain impulse response of described channel to make maximal possibility estimation, obtain the Doppler frequency shift parameter initialization value of multipath; Wherein the maximal possibility estimation function is that the real-time frequency domain impulse of described channel rings and the corresponding corresponding frequency point data of the channel estimating adaptation function delivery square of afterwards suing for peace that multiplies each other;

Steps A 3 ', frequency domain parameter model according to described radio communication channel, the delay parameter that makes multipath l is the delay parameter value that steps A 1 ' is estimated, the Doppler frequency shift parameter is the Doppler frequency shift parameter value that steps A 2 ' is estimated, complex magnitude parameter normalizing, make multipath l remove delay parameter, the Doppler frequency shift parameter, other parameters outside complex magnitude parameter and the wave beam arrival direction vector parameters are 0, wave beam arrival direction vector parameters with multipath l is the maximal possibility estimation variable, utilize the real-time frequency domain impulse response of described channel to make maximal possibility estimation, obtain the wave beam arrival direction vector parameters initialization value of multipath l; Wherein the maximal possibility estimation function is that the real-time frequency domain impulse of described channel rings and the corresponding corresponding frequency point data of the channel estimating adaptation function delivery square of afterwards suing for peace that multiplies each other;

Steps A 4 ', frequency domain parameter model according to described radio communication channel, the delay parameter that makes multipath l is the delay parameter value that steps A 1 ' is estimated, the Doppler frequency shift parameter that makes multipath l is the Doppler frequency shift parameter value that steps A 2 ' is estimated, the wave beam arrival direction vector that makes multipath l is the wave beam arrival direction vector parameters value that steps A 3 ' is estimated, make the complex magnitude parameter normalizing of multipath l, with wave beam departure direction vector parameters is the maximal possibility estimation variable, utilize the real-time frequency domain impulse response of described channel to make maximal possibility estimation, obtain the wave beam departure direction vector parameters initialization value of multipath l; Wherein the maximal possibility estimation function is that the real-time frequency domain impulse of described channel rings and the corresponding corresponding frequency point data of the channel estimating adaptation function delivery square of afterwards suing for peace that multiplies each other;

Steps A 5 ', frequency domain parameter model according to described radio communication channel, make the complex magnitude parameter normalizing of multipath l, the delay parameter value, Doppler frequency shift parameter value, wave beam arrival direction vector parameters value, the wave beam departure direction vector parameters value that adopt steps A 1 ' to estimate to A4 ', reconstruct channel frequency domain impulse response, and with the channel frequency domain impulse response of the channel frequency domain impulse response after the reconstruct before, obtain the complex magnitude parameter initialization value of multipath l divided by reconstruct;

Steps A 6 ' judges whether each parameter value satisfies the loop ends condition, if satisfy condition then export the initiation parameter result, if do not satisfy condition then proceed steps A 7 '; The loop ends condition is: whether the number of times that judging circulates carries out need equals the multipath number of estimation;

Steps A 7 ', frequency domain parameter model according to described radio communication channel, the delay parameter value, Doppler frequency shift parameter value, wave beam arrival direction vector parameters value, the wave beam departure direction vector parameters value that adopt steps A 1 ' to estimate to A5 ', complex magnitude parameter value reconstruct channel frequency domain impulse response, and with the real-time frequency domain impulse response of the channel frequency domain impulse response negative feedback after the reconstruct to channel, obtain the real-time frequency domain impulse response H (f) of the channel of renewal, return steps A 1 '.

9. broadband wireless communications channel frequency domain multidimensional parametric modeling method according to claim 7 is characterized in that described step 4 comprises following detailed step:

Step B1 ', frequency domain parameter model according to described radio communication channel, order treats that the multidimensional parameter value of all multipaths that iteration is estimated is last iteration results estimated value, if be iteration first, make then that described each estimated parameter value is the corresponding initialization value of step 3 output, reconstruct channel frequency domain impulse response, and will give the real-time frequency domain impulse response of channel except that the reconstruct channel frequency domain impulse response negative feedback of other multipaths of this multipath that needs iteration to estimate, obtain the real-time frequency domain impulse response H of the channel of renewal _Update(f);

Step B2 ', frequency domain parameter model according to described radio communication channel, order treats that multipath l each estimated parameter value except that delay parameter that iteration is estimated is last iteration results estimated value, if be iteration first, make then that described each estimated parameter value is the corresponding initialization value of step 3 output, complex magnitude parameter normalizing is the maximal possibility estimation variable with the delay parameter, utilizes the real-time frequency domain impulse response H of the channel of described renewal _Update(f) make maximal possibility estimation, obtain the delay parameter value of the multipath l of this iteration estimation; Wherein the maximal possibility estimation function is that the real-time frequency domain impulse of described channel rings and the corresponding corresponding frequency point data of the channel estimating adaptation function delivery square of afterwards suing for peace that multiplies each other;

Step B3 ', frequency domain parameter model according to described radio communication channel, order treats that multipath l each estimated parameter value except that the Doppler frequency shift parameter that iteration is estimated is last iteration results estimated value, if be iteration first, make then that described each estimated parameter value is the corresponding initialization value of step 3 output, complex magnitude parameter normalizing is the maximal possibility estimation variable with the Doppler frequency shift parameter, utilizes the real-time frequency domain impulse response H of the channel of described renewal _Update(f) make maximal possibility estimation, obtain the Doppler frequency shift parameter value of the multipath l of this iteration estimation; Wherein the maximal possibility estimation function is that the real-time frequency domain impulse of described channel rings and the corresponding corresponding frequency point data of the channel estimating adaptation function delivery square of afterwards suing for peace that multiplies each other;

Step B4 ', frequency domain parameter model according to described radio communication channel, order treats that multipath l each estimated parameter value except that wave beam arrival direction vector parameters that iteration is estimated is last iteration results estimated value, if be iteration first, make then that described each estimated parameter value is the corresponding initialization value of step 3 output, complex magnitude parameter normalizing is the maximal possibility estimation variable with wave beam arrival direction vector parameters, utilizes the real-time frequency domain impulse response H of the channel of described renewal _Update(f) make maximal possibility estimation, obtain the wave beam arrival direction vector parameters value of the multipath l of this iteration estimation; Wherein the maximal possibility estimation function is that the real-time frequency domain impulse of described channel rings and the corresponding corresponding frequency point data of the channel estimating adaptation function delivery square of afterwards suing for peace that multiplies each other;

Step B5 ', frequency domain parameter model according to described radio communication channel, order treats that multipath l each estimated parameter value except that wave beam departure direction vector parameters that iteration is estimated is last iteration results estimated value, if be iteration first, make then that described each estimated parameter value is the corresponding initialization value of step 3 output, complex magnitude parameter normalizing is the maximal possibility estimation variable with wave beam departure direction vector parameters, utilizes the real-time frequency domain impulse response H of the channel of described renewal _Update(f) make maximal possibility estimation, obtain the wave beam departure direction vector parameters value of the multipath l of this iteration estimation; Wherein the maximal possibility estimation function is that the real-time frequency domain impulse of described channel rings and the corresponding corresponding frequency point data of the channel estimating adaptation function delivery square of afterwards suing for peace that multiplies each other;

Step B6 ', frequency domain parameter model according to described radio communication channel, the delay parameter value, Doppler frequency shift parameter value, wave beam arrival direction vector parameters value, the wave beam departure direction vector parameters value that adopt step B2 ' to estimate to B5 ', make complex magnitude parameter normalizing, reconstruct channel frequency domain impulse response, and with the real-time frequency domain impulse response H of the channel of channel frequency domain impulse response after the reconstruct and described renewal _Update(f) be divided by, obtain the complex magnitude parameter value of the multipath l of this iteration estimation;

Step B7 ' judges whether to satisfy the condition of convergence, if satisfy the condition of convergence then export the parameter Estimation result of this iteration of multipath l; If do not satisfy condition then return step B2 '; The described condition of convergence is whether the iterations of judging each parameter equals the multipath number that needs are estimated.

10. broadband wireless communications channel frequency domain multidimensional parameterized model is characterized in that described model comprises:

The received signal frequency domain image data that transmit frequency domain data and the radio communication channel that utilizes that radio communication channel measures measured is calculated the actual measurement channel impulse response generation module of the real-time frequency domain impulse response that obtains channel;

Set up the frequency domain parameter model of radio communication channel and the frequency domain coupling MBM of channel estimating adaptation function according to the model parameter demand;

Utilize described frequency domain parameter model and channel estimating adaptation function from the real-time frequency domain impulse response of described channel, to search out multipath, and the frequency domain Multipath searching module of the parameter of every multipath of initialization;

For every multipath, utilize described actual measurement channel frequency domain impulse response and channel estimating adaptation function to calculate maximal possibility estimation, and obtain the frequency-domain model parameter iteration estimation module of the multidimensional parameter value of every multipath by the iteration estimation;

Judge whether to circulate and carry out the estimated parameter determination of stability module that iteration is estimated.

Images