CN1863180A - Method for estimating channel shock response effective orders in wireless communication system - Google Patents

Abstract

The invention relates to a method for estimating effective order or channel impulse response in wireless communication system, adopting least square method to obtain the estimated channel impulse response sequence and finding mod of each element in the sequence and obtaining energy sequence of the channel impulse response, selecting a reference element from the energy sequence, summing the reference element and several elements before it, then summing all elements behind it, and defining effective order determining threshold by the ratio of the two sums: if the determining threshold is lower than threshold value,setting adjacent element behind the reference element as a new reference element, continuing summing and determining a new effective order determining threshold; and if higher than threshold value, the sequence number corresponding to the reference element is the effective order of the channel impulse response. And the invention can improve estimating performance of effective order of wireless channel impulse response.

Description

The method of estimation of channel impulse response effective order in the wireless communication system

Technical field

The present invention relates to the method for estimation of channel impulse response effective order in a kind of wireless communication system, in wireless communication system based on OFDM (OFDM) technology, formation characteristics at LS (least square) channel estimation method time domain estimated result, the combining wireless channel impulse response is the characteristics of finite impulse response in time domain simultaneously, result in time domain to the LS channel estimation method is carried out energy measuring, and carries out reverse search to obtain channel effective order estimated value.The invention belongs to wireless communication technology field.

Background technology

Ofdm system utilizes FFT (fast fourier transform) to finish modulation at frequency domain, separates mediation channel equalization.System uses a tap equalizers to realize channel equalization at frequency domain; In order to obtain high bit rate and channel utilization, system modulation is used high-order modulating, and correlation demodulation mode is efficiently used in demodulation; Ofdm system will be finished correlation demodulation and channel equalization, need obtain channel information by channel estimating.Channel estimation method is very many in the ofdm system, and classical algorithm comprises: based on the LS channel estimation method of Pilot (frequency pilot sign), based on MMSE (least mean-square error) channel estimation method of Pilot; Blind channel estimation method; MIMO-OFDM (multi-I/O OFDM) system channel algorithm for estimating etc.In these algorithms, the MMSE channel estimation method is the result based on the LS channel estimation method, and a part of MIMO-OFDM system also uses the LS channel estimation method or carries out channel estimating with it as the basis.In existing OFDM mobile communication system, the DVB-T in Europe and DAB adopt the LS channel estimation method based on Pilot simultaneously.With the closely-related algorithm of LS channel estimation method in, many effective orders that all need to use channel impulse response, know that the channel impulse response effective order can improve the precision of ofdm system estimated frequency skew, can simplify the complexity of MMSE channel estimation method; System can utilize the estimated value of channel impulse response effective order to improve the LS channel estimation method to disturbing the robustness of (ICI) between noise and subchannel.

In the middle of the method for the channel effective order of studying at present, following two kinds of typical methods are arranged:

(1) estimates based on the channel impulse response effective order of channel impulse response monotonic decay

Songping Wu; Bar-Ness, Y. wait the people at " OFDM channel estimation in the presenceof frequency offset and phase noise " (Communications, 2003.ICC ' 03.IEEEInternational Conference on, Volume:5,2003 Page (s): 3366-3370) proposed a kind of channel impulse response effective order estimation approach in the literary composition, will obtain the better estimation of channel impulse response effective order according to the judgement standard in the document, the power of each rank coefficient must have dull decline form in the channel impulse response; Algorithm performance is subjected to The noise easily simultaneously; When the power of each rank coefficient of channel has Rayleigh fading negative exponent power-delay spectrum, the decline of each rank coefficient rises and falls will influence the judgement of this algorithm, and algorithm can not estimate the effective order of channel impulse response well.

Y.Li, L.J.Cimini Jr, and N.R.Sollenberger, Deng the people at " Robust channelestimation for OFDM systems with rapid dispersive fading channels " (IEEE Trans.onCom., vol.46, pp.902-915, Jul.1998.) decision criteria of the middle channel impulse response effective order that proposes can detect the effective order of various channels effectively, and the precondition of algorithm is to obtain the higher autocorrelation matrix of precision.In mobile communication system, because the time variation of channel, the higher autocorrelation matrix of precision is difficult to obtain.

Summary of the invention

The objective of the invention is at the deficiencies in the prior art, the method of estimation of channel impulse response effective order in a kind of wireless communication system is provided, this method do not rely on channel impulse response monotonicity, and be difficult for affected by noisely, can improve the precision that the channel effective order is estimated.

To achieve these goals, the present invention at first adopts least square method to obtain estimated channel impulse response sequence in time domain, and each element in the channel impulse response sequence is asked mould, obtains the energy sequence of channel impulse response.In energy sequence, select a benchmark element then, with benchmark element and several elements summations before thereof, again with whole elements summations of benchmark element back, utilize two and ratio define the effective order decision threshold.If the effective order decision threshold less than threshold value, then is decided to be new benchmark element to the element of original hase eka-element rear adjacent, proceed to sue for peace and definite new effective order decision threshold; If the effective order decision threshold is greater than threshold value, then original selected pairing sequence number of benchmark element is the effective order of channel impulse response.

Method of the present invention comprises following concrete steps:

(1), adopt least square method to obtain estimated channel impulse response sequence according to system's transmitting terminal pilot frequency sequence and receiving terminal and pilot frequency sequence corresponding response sequence.

(2) each element in the channel impulse response sequence of estimating to obtain is done asked modular arithmetic, obtain the energy sequence of channel impulse response.

(3) in the energy sequence of channel impulse response, the element that selection is positioned in the middle of the sequence is the benchmark element, n element before benchmark element and the benchmark element carried out summation operation, with whole element summations of benchmark element back, former and later two are obtained a ratio with comparing again; The sequence number that the channel impulse response sequence length is deducted the benchmark element obtains a difference, and front gained ratio and gained difference are multiplied each other obtains a product, and 1/20th of this product is defined as the effective order decision threshold.Wherein, the n value is 5～10.

(4) determine a threshold value according to the estimated probability of the channel impulse response effective order of expecting,, continue the processing of step 3 if the effective order decision threshold less than threshold value, then is decided to be new benchmark element to the element of original benchmark element rear adjacent of selecting; If the effective order decision threshold is greater than threshold value, then original selected pairing sequence number of benchmark element is the effective order of channel impulse response.

Among the present invention, the selected estimated probability according to the channel impulse response effective order of expecting of threshold value is determined, can select in 2.5～2.6.Threshold value is 2.55 when the estimated probability of channel impulse response effective order of expectation reaches 90%, and threshold value is high more, and false alarm probability is low more, but the estimated probability with equal-wattage also will reduce; Otherwise threshold value reduces, and the estimated probability of channel impulse response coefficient of efficiency with equal-wattage is high more, and false alarm probability also will improve simultaneously.

In the channel impulse response effective order method of estimation of the present invention, the energy that adopts channel impulse response is as estimating object, and the energy of channel impulse response is carried out the part add up, the decline that has reduced each rank coefficient of channel rises and falls to algorithmic decision result's influence, reduced the sensitiveness of algorithm, so the estimated performance of channel effective order is improved to noise.The inventive method can estimate to have the effective order of the mobile telecommunication channel of Rayleigh fading negative exponent power-delay spectrum better, can reduce the influence of noise match exponents estimated result.

Description of drawings

Fig. 1 is the curve chart that the channel impulse response effective order of the embodiment of the invention is estimated object.

Among Fig. 1, each rank coefficient amplitude c=0.08 of channel impulse response.

Fig. 2 compares channel impulse response effective order results estimated channel impulse response effective order results estimated and existing document for the present invention.

Among Fig. 2, abscissa represents that the effective order estimated, ordinate represent the probability of estimating.

Embodiment

Below in conjunction with drawings and Examples technical scheme of the present invention is further described.

Fig. 1 provides the estimation object among the embodiment, each rank coefficient amplitude c=0.08 of channel impulse response.

The method of estimation of the channel impulse response effective order that proposes according to the present invention, concrete implementation step is as follows:

(1) according to the pilot frequency sequence X of system's transmitting terminal _pAnd receiving terminal and pilot frequency sequence corresponding response sequence Y _p, adopt the LS least square method to obtain estimated channel impulse response sequence

According to the LS method, according to Min{ (Y _p-X _pF _pH) H (Y _p-X _pF _pH) } estimate to obtain the channel impulse response sequence ${\hat{h}}_{\mathrm{ls}}=F_P^H{X}_P^{-1}{Y}_P.$ Wherein, X _pBe system's transmitting terminal pilot transmitted sequence; Y _pBe system receiving terminal and the corresponding response sequence of transmitting terminal pilot frequency sequence; F _pBe the fast fourier transform matrix.

(2) the channel impulse response sequence to estimating to obtain In each element do and ask modular arithmetic ${\hat{H}}_{\mathrm{ls}}^s={\left|{\stackrel{)}{h}}_{\mathrm{ls}}\left(s\right)\right|}^2,$ Obtain the energy sequence of channel impulse response ${\stackrel{)}{H}}_{\mathrm{ls}}=\left[\begin{array}{cccc}{\stackrel{)}{H}}_{\mathrm{ls}}^1& {\stackrel{)}{H}}_{\mathrm{ls}}^2& L& {\stackrel{)}{H}}_{\mathrm{ls}}^s\end{array}\right].$

(3) in the energy sequence of channel impulse response, the element that selection is positioned in the middle of the sequence is the benchmark element, n element before benchmark element and the benchmark element carried out summation operation, again with the whole elements summation of benchmark element back, with former and later two with compare.In the present embodiment, the energy sequence art tail of selected distance channel impulse response is that 15 element is the benchmark element, n=9.Totally 10 elements of 9 elements that are about to benchmark element and benchmark element front carry out summation operation, again with whole elements summations of benchmark element back, with former and later two with compare and obtain a ratio; The sequence number that the channel impulse response sequence length is deducted the benchmark element obtains a difference, and front gained ratio and gained difference are multiplied each other obtains a product, and 1/20th of this product is defined as effective order decision threshold k _s, promptly $\frac{\left(\underset{s-9}{\overset{s}{\mathrm{\Σ}}}{\hat{H}}_{\mathrm{LS}}^s\right)/2*10}{\left(\underset{k=s+1}{\overset{N_p}{\mathrm{\Σ}}}{\hat{H}}_{\mathrm{LS}}^k\right)/(2*(N_p-s))}$ The ratio result is designated as k _s

Wherein, N _pBe estimated channel impulse response sequence length, s is the sequence number at benchmark element place in the channel impulse response sequence.

(4) determine a threshold value according to the estimated probability of the channel impulse response effective order of expecting, get 2.55 among the embodiment.If effective order decision threshold k _sLess than threshold value, then the element of original benchmark element rear adjacent of selecting is decided to be new benchmark element, continue the processing of step 3; If effective order decision threshold k _sGreater than threshold value, then original selected pairing sequence number s of benchmark element is the effective order of channel impulse response.

In the present embodiment, obtain as effective order decision threshold k _sThe value of middle s is from N _p-15 beginnings are decremented to 1; The decrement step size of s is taken as 1.Satisfy for the first time k _sS value during 〉=2.55 conditions is judged to the effective order of channel impulse response.

Among the present invention, the selected estimated probability according to the channel impulse response effective order of expecting of threshold value is determined, can select in 2.5～2.6.Threshold value is 2.55 when the estimated probability of channel impulse response effective order of expectation reaches 90%, and threshold value is high more, and false alarm probability is low more, but the estimated probability with equal-wattage also will reduce; Otherwise threshold value reduces, and the estimated probability of channel impulse response coefficient of efficiency with equal-wattage is high more, and false alarm probability also will improve simultaneously.

Fig. 2 provides be the present invention to channel impulse response effective order results estimated, provided the estimated result that adopts existing document that background technology partly mentions to obtain simultaneously based on the channel impulse response effective order method of estimation of channel impulse response monotonic decay.Abscissa represents that the effective order estimated, ordinate represent the probability of estimating among Fig. 2.The result that two kinds of methods are obtained compares as can be seen, and the present invention can effectively estimate that performance is better than the method that existing document is provided to the channel effective order.

In sum, the channel impulse response effective order method of estimation that the present invention proposes can improve the precision of radio channel estimation, strengthens the performance of wireless communication system.

Claims (2)

1, the method for estimation of channel impulse response effective order in a kind of wireless communication system is characterized in that comprising the steps:

1), adopt least square method to obtain estimated channel impulse response sequence according to system's transmitting terminal pilot frequency sequence and receiving terminal and pilot frequency sequence corresponding response sequence;

2) each element in the channel impulse response sequence of estimating to obtain is done asked modular arithmetic, obtain the energy sequence of channel impulse response;

3) in the energy sequence of channel impulse response, the element that selection is positioned in the middle of the sequence is the benchmark element, n element before benchmark element and the benchmark element carried out summation operation, with whole element summations of benchmark element back, former and later two are obtained a ratio with comparing again; The sequence number that the channel impulse response sequence length is deducted the benchmark element obtains a difference, and front gained ratio and gained difference are multiplied each other obtains a product, and 1/20th of this product is defined as the effective order decision threshold; Wherein, the n value is 5～10;

4) determine a threshold value according to the estimated probability of the channel impulse response effective order of expecting,, continue the processing of step 3 if the effective order decision threshold less than threshold value, then is decided to be new benchmark element to the element of original benchmark element rear adjacent of selecting; If the effective order decision threshold is greater than threshold value, then original selected pairing sequence number of benchmark element is the effective order of channel impulse response.

2,, it is characterized in that described threshold value selection 2.55 according to the method for estimation of channel impulse response effective order in the wireless communication system of claim 1.

Images