Background technology
In transmission of broadband communications system, especially in wireless broadband communication transmission system, the narrow-band interference signal such as the same frequency of analog or digital or adjacent frequency interference, the performance of transmission can be had a strong impact on.Therefore, filtering just becomes that the one of electronic information process field is most basic and extremely important technology, utilizes filtering technique can extract required signal from complicated signal, suppresses noise or interference simultaneously.Wave filter has actually also just become a kind of frequency-selective system, and some frequency signal is given only small decay by it, and other unwanted signals give very big decay, stops unwanted signal to pass through as far as possible.Being a kind of basic element circuit at electronic system median filter, use a lot, technology is also complex, simultaneously the good and bad performance directly determining product of wave filter, so a lot of country pays much attention to filter theory research and product development.
If interference signal is more weak, the not synchronization of influential system, it is possible in conjunction with data, adopt multiple method, for instance the modes such as traditional feed forward equalization or feedback equalization, soft mapping, the equilibrium of iteration equalizing, even turbo are come anti-interference, can obtain good performance.For CDMA and ofdm system, there is many discussions being related to these anti-disturbance methods in recent some.But when disturbing signal very strong, just influence whether the synchronization of system.
In more than 30 year of past, there has been proposed many circuit or method to strengthen useful signal, or separate/eliminate noise and interference signal.In these methods, different circuit structures suppresses or eliminates the ability difference of interference signal.Although those utilize the method for training sequence or known pilot signal to have certain effect, but blind processing mode seems more attractive.The method of the blind process of one type is exactly utilize wave trap to go to eliminate this interference signal after the mid frequency obtaining interference signal.This estimation utilizes power spectrum method (after this signal is carried out FFT, seek its energy spectrum), even if carrying out FFT estimation continually, the change following the tracks of interference signal is still excessively poor.
Other kinds of blind processing method mainly adopts Adaptive Signal Processing.Sef-adapting filter belongs to Modern Filter category, sef-adapting filter is for fixed filters, and fixed filters belongs to classical filter device, and its frequency filtering is fixing, the frequency of sef-adapting filter filtering is adaptive input signal and changes, so its scope of application is wider.When there is no any priori about signal and noise, sef-adapting filter utilizes previous moment to obtain filter parameter to be automatically adjusted the filter parameter of now, with the statistical property of adaptation signal and noise the unknown or change at random, thus realizing optimal filter.Self adaptive wave filter is exactly substantially that one can regulate himself transmission characteristic and reached optimized Wiener filter.
Generally speaking, sef-adapting filter is made up of two parts, one is filter construction, two is the adaptive algorithm adjusting filter coefficient, the structure of sef-adapting filter adopts FIR or IIR structure, due to iir filter existence and stability problem, therefore it is generally adopted the FIR filter structure as sef-adapting filter.
Fig. 1 is the structural representation of existing conventional sef-adapting filter, wherein x (n) is input signal, after inputting signal x (n) delay, Error Calculation is carried out again with output signal y (n), thus obtain control information e (n) send into afterwards sef-adapting filter be filtered parameter regulate, it may be assumed that
hn(i)=hn-1(i)+μ·conj(x(i))·e(i)
E (i)=x (n-D)-y (n), hnI () is channel response coefficient, μ is constant, and conj seeks conjugation, elongatedness when D is, the purpose of adjustment is so that error signal e (i) is minimum.
As in figure 2 it is shown, it is the structural representation of conventional lateral type wave filter, sef-adapting filter, except including a wave filter according to certain structural design, also has a set of adaptive algorithm, and adaptive algorithm designs according to certain judgement.The algorithm of sef-adapting filter is mainly using various criterions as calculating basis.Generally there are two kinds of criterions: least mean-square error criterion and method of least square criterion.For least mean-square error (LeastMeanSquare, LMS) algorithm, it is that the mean-square value making output error sequence e (i)=x (n-D)-y (n) minimizes, and carrys out power of amendment coefficient according to this criterion.
Additionally, be the american documentation literatures such as 4052559,4238746,005325204,4673982,4524424,4420815,6976044 in the patent No.;IEEE journal the 63rd volume of in December, 1975 publication, 12nd phase, in 1692-1716 page and Widrow et al. the paper about " adaptive noise cancel-ation ", it is directed to the algorithm that the real-time lowest mean square to wiener-Hoff wave filter is approximate, all these documents are all adopt IIR or FIR filter to eliminate interference signal, and update the coefficient of wave filter according to the output of wave filter and the result of the time delay cross-correlation of the signal of input.But the convergence rate of these methods is very slow, and can only resist the interference signal with specific frequency spectrum shape, it means that these methods can not be widely used in scene more usually, if any the fast-changing environment of mid frequency of frequency hopping interference or interference signal.
Therefore, in the urgent need to a kind of more excellent sef-adapting filter.
Detailed description of the invention
Referring to Fig. 3, the blind adaptive filtering device that can eliminate arrowband interference of the present invention mainly comprises the steps that sef-adapting filter, the first delay cell and error calculation unit etc..
Described sef-adapting filter will be for being exported y (n) after input signal x (n) filtering, and it can be FIR filter, it is also possible to be iir filter.The internal structure of FIR filter and iir filter is known to those skilled in the art to be known, therefore is not described in detail in this;And, those skilled in the art can select different sef-adapting filters voluntarily according to different application scenarios.
Described first delay cell is coupled in described sef-adapting filter, for the output signal of sef-adapting filter according to the first time delay length delay preset.Generally, when first, elongatedness can be determined according to actual application scenarios, during as being applied to many subbands situation in communication system and single frequency band scenario, can adopt different time delays.And the output signal postponed can be current output signal y (n), the signal exported before sef-adapting filter also can be comprised.
Described error calculation unit is coupled in described sef-adapting filter and the first delay cell, for extracting the first control information after the output signal of the current output signal of described sef-adapting filter Yu described first delay cell is carried out related operation, so that described first control information is fed back to described sef-adapting filter, adjust its parameter for described sef-adapting filter, and then eliminate the interference in the rear input signal that described sef-adapting filter accesses.When calculating target function, the norm of more than second order or second order can be adopted.It should be noted that the norm exponent number adopted is more high, anti-interference is more good, but implementing circuit can be more complicated, therefore, during practical application, the norm of suitable exponent number can be adopted as required to carry out correlation computations.In the present embodiment, for instance:
Then output y (n) of sef-adapting filter is represented by:
Wherein N is sef-adapting filter length, and T is transposed operator, and * is adjoint operator, for making control information (i.e. object function) minimize, namely needs to make C (n1, n2..., nk)=E{y (n) y (n+n1)…y(n+nk) norm minimum, in other words
F=E{ ‖ C (n1, n2..., nk)‖p}→minf
Wherein, ‖ ‖pBeing p rank norms, especially, ‖ ‖ is second order norm.
Calculate for simplifying, select second-order statistic definition:
F=E{ ‖ y (n) y*(n-S/2) ‖ } → minf,
Therefore:
Fori=0,1 ..., N-1
According to least mean square algorithm LMS, following result can be obtained under FIR framework:
Error vector:
E (i)=y (n-1)*‖y(n-S/2-1)‖x(n-i)+y(n-1-S/2)*‖ y (n-1) ‖ x (n-i-S/2), i=0 ..., N-1, h (i)=h (i)-μ e (i).
If output vector is:In order to weigh treatment effect, then
The correlated results of output vector is just:
In addition, it should be noted that, above-mentioned is adopt least mean square algorithm (i.e. LMS) at time-domain calculation minimum error, may be used without recurrent least square method (i.e. RLS), or the algorithm of other forms, and it is not limited only at time-domain calculation, it is possible to be calculated at Transformation Domain, as in frequency-domain calculations etc., no longer described in detail one by one at this.
Have again, at some application scenarios, also one second delay cell can be set up in a device again, for by input described sef-adapting filter input signal x (n) according to sending into described error calculation unit after the second default time delay length delay to carry out obtaining corresponding second control information after related operation adjusting its parameter with output signal y (n) of described sef-adapting filter for described sef-adapting filter, much more no longer this is similar with prior art, therefore also to state at this.
Also have, if the dependency of sef-adapting filter output signal y (n) is very weak, for example lower than the threshold value of a certain setting, adaptive narrow-band interference cancellation circuit can be bypassed to save power consumption in the case, thus, one judging unit can be set in a device, for judge calculated the first control information of described error calculation unit whether lower than default threshold value to decide whether that input signal is sent into described sef-adapting filter carries out interference filtering, if the first calculated control information is lower than default threshold value, then input signal x (n) is sent directly into the unit such as subsequent decoding and is made directly decoding, filter without through sef-adapting filter.
The above-mentioned blind adaptive filtering device that can eliminate arrowband interference is widely used, for instance, (ACI) filter combination can be disturbed with adjacent frequency, to eliminate arrowband and adjacent frequency interference, it is possible to be applied in multi-sub-band system, for eliminating the arrowband interference in each subband.
As shown in Figure 4, it is a kind of transmission of broadband communications system schematic, in the system, comprise the above-mentioned blind adaptive filtering device that can eliminate arrowband interference, before this blind adaptive filtering device that can eliminate arrowband interference is arranged on adjacent frequency interference filter, input signal x (n) is admitted to demodulator, in order to demodulate the communication information after sequentially passing through arrowband interference and adjacent frequency interference process.
Again as shown in Figure 5, it is another kind of transmission of broadband communications system schematic, different from the transmission of broadband communications system shown in Fig. 4 it is in that, can eliminate arrowband interference blind adaptive filtering device be arranged on adjacent frequency interference filter after, after input signal x (n) first passes through adjacent frequency interference process, filtered arrowband interference by the blind adaptive filtering device that can eliminate arrowband interference again, then demodulated by demodulator.
nullRefer to Fig. 6 again,It is many subbands transmission of broadband communications system schematic,This many subbands transmission of broadband communications system includes: multiple subband access units,I.e. subband access unit 1、Subband access unit 2、Subband access unit 3 ...、Subband access unit N,It is respectively used to access different subband signals,Each subband access unit all can eliminate the blind adaptive filtering device of arrowband interference and couple with one,The blind adaptive filtering device 1 that can eliminate arrowband interference is coupled in subband access unit 1、The blind adaptive filtering device 2 that can eliminate arrowband interference is coupled in subband access unit 2、The blind adaptive filtering device 3 that can eliminate arrowband interference is coupled in subband access unit 3 ... the blind adaptive filtering device N that can eliminate arrowband interference is coupled in subband access unit N,Eliminate arrowband interference each subband signal be admitted to assembled unit combination after,Form full communication information.
It should be noted that, communication system shown in above-mentioned Fig. 4-Fig. 6, the structure of each self-contained blind adaptive filtering device that can eliminate arrowband interference is all as previously mentioned, namely sef-adapting filter, the first delay cell and error calculation unit etc. are all included, certainly, it is possible to set up the second delay cell and judging unit etc. again according to reality.
Referring to Fig. 7 again, it is the workflow schematic diagram of the method that blind adaptive eliminates arrowband interference in transmission of broadband communications system.nullFirst sef-adapting filter (being such as FIR filter or iir filter etc.) exports y (n) after being filtered by current input signal x (n),Then,During with first, elongatedness (it can be determined by actual application scenarios) postpones the signal (including current output signal and the signal exported before) of described sef-adapting filter output,Signal after signal y (n) currently exported by described sef-adapting filter again and delay carries out related operation (such as can in time domain,Also LMS or RLS scheduling algorithm can be adopted to calculate the norm of more than second order or second order at other Transformation Domains) to obtain control information e (n) (LMS or RLS can be adopted to extract),Then,Described control information e (n) is fed back to described sef-adapting filter,Filter parameter is adjusted according to the control information e (n) fed back to by the filtering of described sef-adapting filter,So as the interference eliminated in follow-up input signal x (n+i).
In sum, the blind adaptive filtering device that can eliminate arrowband interference of the present invention carries out related calculation by sef-adapting filter exports the delay signal of signal and output signal, thus extract control information, adjusting the filtering parameter of sef-adapting filter, it is applicable to China Mobile multimedia broadcasting (CMMB) receptor and other DTV receivers;Can also be used for broadband receivers, including DTV and telecommunication receiver etc., for the situation having multiple stronger arrowband to disturb, it is also possible to be suitable for;For the multiple stronger arrowband interference that sweep rate is several ms, it is also possible to opposing.The present invention is compared with existing adaptive approach, and it is less to the channel effect of superposition, and receiver performance damage is less;Secondly, it has response speed faster, has certain tracking ability.And compared with traditional notch filter, it need not predict the spectral shape of interference signal, also can avoid the flase drop problem in detection simultaneously.
Above-described embodiment only listing property illustrates principles of the invention and effect, not for the restriction present invention.Above-described embodiment all can without departing from the spirit and scope of the present invention, be modified by any one skilled in the art.Therefore, the scope of the present invention, should as listed by claims.