CN101150357B - Method for eliminating peak power - Google Patents

Abstract

This invention relates to a method for eliminating power of peak values including: 1, carrying out interpolation filter to each channle signals after modulation and spread to change them to signals of different frequency points, 2, making linear overlapping all the frequency point signals on the time domain, 3, carrying out cancel of peaks to signals needing clipping according to the clipping sequence regulated according to the change of amplitude, which can adaptively regulate peak-clipping sequencies of 1 path and Q to adaptively eliminating power of peak values and the loss of phase information of original signals is the least.

Description

The method of cutting peak power

Technical field

The present invention relates to the communications field, relate in particular to a kind of TDD(time division duplex) in the multifrequency point communication system cutting peak power method and realize the communication ends of cutting peak power.

Background technology

In the residential quarter of TDD communication system (such as TD-SCDMA time division synchronous code division multiple access communication system and OFDM TDD OFDM tdd communication systems), owing to need the multifrequency point signal of emission will merge at digital intermediate frequency, so that sharing a cover transmitter in follow-up radio-frequency channel launches, therefore will occur because the peak-to-peak superposition produces stronger signal peak-to-average ratio PAPR (peak to average power raio, the ratio of peak power and average power) at the antenna transmission end.Peak power is too large, is easy to make radio-frequency transmissions to the inelastic region, thereby produces stronger ACLR(Adjacent Channel Leakage power Ratio, ACLR, the side channel leakage power ratio of transmitter), and then reduce systematic function.If do not want to cause nonlinear distortion, the power that transmits must be less than the 1dB compression point, this falls too low with regard to the average power that requires signal, but in this case, the efficient of power amplifier will reduce, the signal power of simultaneously radio frequency transmission can not reach the dB number that physical layer requires, thereby so that the coverage of base station is dwindled, user's signal power is impaired.And high peak-to-average force ratio can cause the dynamic range of D/A converter to diminish, if use the D/A converter of high exponent number, cost will improve greatly, if use the low D/A converter of exponent number, quantizing noise can strengthen.

Three kinds of methods that reduce peak-to-average force ratio are proposed: slicing method, sequence selection method and phase amplitude converter technique in the prior art.In the TDD multiple frequency point system, the sequence selection method implements complexity, usually will not adopt.The slicing method is that the too high peak value of power is eliminated, and produces easily nonlinear distortion.A kind of time domain peak clipping frequency domain filtering method is wherein arranged, and this kind method is at first used and is firmly declared method, will firmly cut above the data of threshold power, then uses filter to eliminate and firmly cuts the nonlinear distortion that causes.But this method nonlinear distortion can not eliminate fully, still can cause very large spectrum leakage.

Although the phase amplitude converter technique does not produce nonlinear distortion, but if so that the rear peak-to-average force ratio of each time rotation is optimum, must carry out a large amount of calculating and need to transmit side information, time-delay is large and processing is complicated, transmits side information and also produces easily the error code diffusion.

For this reason, in conjunction with two kinds of methods that reduce peak-to-average force ratio, the applicant is the method that discloses a kind of peak power in 200610007411.9 on February 8th, 2006 at application number.In the method, the applicant processes by the peak clipping sequence of each N/2 the data symbol in peak power point front and back and equal length is carried out peak clipping.Because the peak clipping operation only is simple plus and minus calculation, and does not use multiplying, thereby the processing speed of peak clipping is fast and easily realization.Process by peak clipping of the present invention, reduced peak-to-average force ratio.

The TDD system is owing to adopted smart antenna, base station transmission DL(down link) data need to send many sets of data each time, every suit data use an antenna to send, every suit data have the different beam weighting factors to be used for wave beam forming, and the beam weighting factor is to the UL(up link) data carry out channel estimating and obtain.It is that independent each circuit-switched data is carried out peak clipping that peak clipping is processed, and, what prior art generally adopted is the digital interpolation method, the peak-data that namely detects in time domain is added and subtracted mutually with default peak clipping impulse ratio, the digital signal that cans be compared to input is inserted a peak clipping sequence, because existing peak clipping sequence is fixed, and can not according to the adaptive adjusting peak clipping sequence of the variation of amplitude, cause easily the distortion of data phase.If the phase place of having destroyed data is processed in peak clipping, the wave beam forming effect of smart antenna will wreck so.In addition, even if use a single aerial system (for example little base station of TD-SCDMA), if the phase relation of signal destroyed, also can be so that the receiving terminal demodulation causes larger error.

In addition; although peak-to-average force ratio is larger constantly for some; but the average power of signal is less; the peak power of this moment still is lower than the 1dB compression point of power amplifier; and existing scheme also can be carried out the peak clipping processing usually; this peak clipping processing mode has increased processing time and the intractability of whole signal transmission processing process, the most important thing is that carrier data is carried out peak clipping to be processed, and exists the problem of distorted signals.That is to say that prior art only considers that the peak-to-average force ratio of carrier data decides peak clipping, the difficulty of not only increase processing but also have the problem of information distortion.

Also have, prior art is normally adjusted to peak power predefined threshold power when peak clipping is processed.When setting threshold power, do not consider the probability that peak value occurs, so that peak power point arranges is excessive or too small, and then cause the distortion that data are larger.Such as, for power amplifier, if the probability that a certain peak power occurs is very little, for example surpass signal peak-to-average ratio PAR (probability of peak to average power raio:PAR=15dB only has ten thousand/, can not consider peak clipping.Reach one of percentage if surpass the probability of PAR=15dB, need to consider to eliminate.

Summary of the invention

The communication ends that a purpose of the present invention is to provide the method for cutting peak power in a kind of TDD multifrequency point communication system and realizes cutting peak power is fixed to solve the prior art peak clipping sequence, damages thus the problem of data phase.

In order to reach the present invention, the invention provides a kind of method of cutting peak power, comprising: (1) each road signal behind the modulation spread spectrum carries out interpolation filtering, and frequency conversion becomes the different frequent points signal; (2) all frequency signals are in the enterprising line linearity stack of time domain; The peak clipping sequence of (3) regulating according to changes in amplitude needs the signal of peak clipping to carry out cutting after superposeing, the peak-to-average force ratio dB number of the described signal that needs peak clipping and average power dB count sum greater than or more than or equal to default detection power dB number, described detection power is less than or equal to 1dB compression point input power value.

Regulating peak clipping sequence according to changes in amplitude in the step (3) comprises by following steps and calculates this I road peak clipping sequence h _i : a1: utilize formula

Calculate the greatest measure h of I road amplitude adjusted _Imax, wherein, pow_max ₀Be target power, I _{Max_m}Pow_max ₀Corresponding I way

Be predefined I road peak clipping sequence.

Regulating peak clipping sequence according to changes in amplitude in the step (3) comprises by following steps and calculates this Q road peak clipping sequence: b1: utilize formula

Calculate the greatest measure h of Q road amplitude adjusted _Qmax, wherein, pow_max ₀Be target power, Q _{Max_m}Pow_max ₀Corresponding Q way value, I _{Max_m}Be pow_max ₀Corresponding I way value; B2: utilize

Calculate the Q road amplitude adjusted factor scaling_q of this peak clipping, wherein hBe predefined Q road peak clipping sequence; B3: utilize h _q =scaling_q hCalculate this Q road peak clipping sequence, wherein hBe predefined Q road peak clipping sequence.

Step is treated to cutting in (3): if I circuit-switched data corresponding to peak power greater than zero, then the I circuit-switched data of this frequency point data of pre-save symbol and this I road peak clipping sequence h _i Subtract each other corresponding position, otherwise the I circuit-switched data of this frequency point data of pre-save symbol and this I road peak clipping sequence h _i Corresponding position addition; If the Q circuit-switched data that peak power corresponds to is greater than zero, then the Q circuit-switched data of this frequency point data of pre-save symbol and this Q road peak clipping sequence h _q Subtract each other corresponding position, otherwise the Q circuit-switched data of this frequency point data of pre-save symbol and this Q road peak clipping sequence h _q Corresponding position addition.

This frequency point data symbol of pre-save is the data symbol that power surpasses threshold power.

Need the signal of peak clipping to be in the step (3): the peak-to-average force ratio dB number of described signal and average power dB count sum greater than or more than or equal to default detection power dB number, described detection power is less than or equal to 1dB compression point input power value.

Described detection power P _{In, max}For: P _{In, max}=P _{In, 1dB}-X 0≤X≤3dB, wherein, P _{In, 1dB}Be 1dB compression point input power value, X is 1dB rollback value.X is 1dB rollback value, and X is that the side channel leakage power according to system requirements arranges than ACLR, and the ACLR of system requirements is higher, and X arranges larger.

Step (3) also comprises before: according to the error rate being required to arrange this professional corresponding contrary accumulated probability distribution function CCDF value, obtain parameter peak-to-average force ratio corresponding to CCDF value; The described signal of peak clipping that needs in the step (3) also is included as peak-to-average force ratio more than or equal to the signal of described parameter peak-to-average force ratio.

A kind of communication ends that realizes cutting peak power comprises: some interpolation filters: be used for the circuit-switched data behind the ovennodulation spread spectrum is carried out L times of interpolation filtering; Some frequency converters: be used for the data after the interpolation are converted to a corresponding frequency point data; One superimposer: connect all frequency converters, be used for all frequency point data after the frequency conversion are superposeed at the enterprising line linearity of time domain, the signal after the linear superposition divides I circuit-switched data and Q circuit-switched data to carry out respectively buffer memory; The peak clipping judging unit: be used for peak-to-average force ratio and average power sum greater than or just carry out the peak clipping processing more than or equal to the signal of default detection power, described detection power is less than or equal to 1dB compression point input power value; The interpolation despiker comprises: amplitude adjusted maximum value calculation subelement: be used for calculating the greatest measure of I road amplitude adjusted and the greatest measure of Q road amplitude adjusted; Amplitude adjusted factor computation subunit: be used for calculating this I road amplitude adjusted factor and the Q road amplitude adjusted factor; This peak clipping sequence computation subunit: be used for calculating this I road peak clipping sequence and Q road peak clipping sequence; Adder: be used for this frequency point data symbol and this peak clipping sequence of pre-save are carried out plus and minus calculation.

Compared with prior art, the present invention can self adaptation regulate peak clipping sequence at every turn in the peak clipping process, reduce phase distortion and amplitude distortion.Quantize peak clipping EVM is mainly seen in the loss that amplitude and phase place cause.The EVM(vector amplitude error, Error Vector Magnitude) in TS 25.101 V3.2.2, be defined as: the difference between measured waveform and the theoretical value of modulation waveform.EVM is square the taking root in of ratio of the energy of average error vector and average reference signal energy, and provides with the form of a%.EVM is important measurement modulation quality index in the digital communication system.3-G (Generation Three mobile communication system) (WCDMA, CDMA2000, TD-SCDMA) agreement specified modulation precision adopts EVM to weigh.It is the pressure index of equipment network access testing.The present invention regulates peak clipping sequence by self adaptation, can so that the deterioration of EVM improve.

The present invention only to peak-to-average force ratio dB number and average power dB count sum greater than or more than or equal to the signal of default detection power dB number, just carry out peak clipping and process, to the signal that does not satisfy condition, do not carry out the peak clipping processing.By the present invention, reduce the treating capacity of signal transmitting terminal, reduce intractability, reduced simultaneously the distortion of information.And, according to system the requirement of ACPR is arranged threshold power, arrange and more can manage.

The present invention is directed to different business the requirement of the error rate is arranged corresponding CCDF, obtain corresponding parameter peak-to-average force ratio according to CCDF value, when the peak-to-average force ratio of signal during more than or equal to described parameter peak-to-average force ratio, just it is carried out the interpolation method peak clipping, the minimizing distorted signals.

Description of drawings

Fig. 1 is the flow chart of the method for cutting peak power of the present invention;

Fig. 2 is the situation schematic diagram that average power is low and the peak-to-average force ratio transformation range is large;

Fig. 3 is the little situation schematic diagram of average power height and peak-to-average force ratio transformation range;

Fig. 4 be under the high-performance ACPR detection power less than 1dB compression point schematic diagram;

Fig. 5 is the CCDF distribution map of the data-signal of TDD;

Fig. 6 is that the peak clipping sequence of utilizing above-mentioned self adaptation to regulate is carried out the power contrast figure that front and back are processed in peak clipping;

Fig. 7 is the comparison schematic diagram before and after the peak clipping of I circuit-switched data amplitude;

Fig. 8 is the comparison schematic diagram before and after the peak clipping of Q circuit-switched data amplitude;

Fig. 9 is an instance graph of cutting peak power of the present invention;

Figure 10 is the schematic diagram of test adaptive peak clipping sequence method;

Figure 11 is that different peak clipping target value are on the schematic diagram that affects of EVM;

Figure 12 is that fixedly gross power is tested the schematic diagram of the improvement situation of ACPR;

Figure 13 is the schematic diagram that the power of output fixing signal is checked the improvement situation of total consumed power.

Embodiment

Below in conjunction with accompanying drawing, specify the present invention.

See also Fig. 1, it is the flow chart of the method for cutting peak power of the present invention.It comprises:

S110: each the road signal behind the modulation spread spectrum carries out interpolation filtering, and frequency conversion becomes the different frequent points signal.

Before signal carries out interpolation filtering, can plug training sequence code, the phase place deflection that the training sequence code of different frequent points is correlated with at each time slot of each circuit-switched data behind the modulation spread spectrum.

According to existing TDD multifrequency point scheme, a plurality of frequencies of supporting for same residential quarter have following agreement: dominant frequency point uses identical scrambler and basic midamble(training sequence with auxiliary frequency) code, therefore the basic midamble code position of each frequency is complete matching, and the basic midamble code of each frequency is identical binary sequence vector simultaneously.Will occur owing to the peak-to-peak superposition produces larger peak power in the basic midamble code part of transmitting terminal, be far longer than the data power in the time slot.

If the phase rotating that the Midamble code of different frequent points is correlated with in the middle of the power interface of Midamble code be bordering on the power of data in the time slot, the peak-to-average force ratio of a time slot when subsequently peak clipping will be even lower in this time, and the Midamble code is impaired few.

For this reason, only need to carry out different phase place deflection at the Midamble of different frequent points code: the Midamble code data is multiplied by exp (j* (n-1) * π), and j is imaginary symbols, and n is carrier index, n:1～N _Always, N _AlwaysCarrier wave sum for each residential quarter in the TDD multi-carrier communications systems.Accordingly, only needing that at each frequency of receiving terminal the Midamble code data is multiplied by exp (j* (n-1) * π) gets final product.

Such as, support the TDD communication system of three frequencies, can carry out at the Midamble of second frequency code the deflection of 180 degree, and the Midamble code of first frequency and the 3rd frequency carries out 0 degree and 360 degree rotations, as there not being deflection, so this kind phase rotating method is the most simple, and successful.

For another example, supporting the TDD communication system of six frequencies, is that the Midamble code of second frequency, the 4th frequency, the 6th frequency carries out the deflection of 180 degree.

By that analogy, for the TDD communication system of multifrequency point, the Midamble code data that is numbered the even number frequency carries out the deflection of 180 degree, and the Midamble code data that is numbered the odd number frequency is not rotated.Certainly also to carry out corresponding processing at receiving terminal.

The method that the Midamble code of different frequent points carries out different phase place deflection also has a lot, just no longer give an example at this, use the method for phase place deflection so that the peak clipping processing of back can be more for business datum power partly owing to inherited physical layer, and so that rising and falling, the peak value of Midamble code approaches, can be more accurate when carrying out channel estimating, and the Midamble code is impaired little, so that channel estimating is more accurate.

S120: all frequency signals are in the enterprising line linearity stack of time domain.

The physical layer data of TDD communication system is first through ovennodulation, and spectrum-spreading and scrambling forms frame data, and transmission rate is 1.28Mps.Carry out L times of interpolation (L interpolation multiple) at digital intermediate frequency, make it become high-speed data.Be 60 such as L, the speed after the interpolation reaches 76.8Mps.Simultaneously multichannel physical layer data after the interpolation is carried out frequency at frequency domain and move, carry out linear superposition in time domain.So the mathematic(al) representation of multifrequency point communication system signal before entering peak clipping is:

$\underset{\‾}{x}\left(t\right)=\underset{m=1}{\overset{M}{\mathrm{\Σ}}}\underset{k=1}{\overset{K}{\mathrm{\Σ}}}{\mathrm{\λ}}^{(m,k)}\·{\underset{\‾}{d}}_{n,}^{(m,k)}{s}_{(n-1)Q+q}^{(m,k)}\·{\mathrm{Cr}}_0\left(t\right)\·\exp ({\mathrm{j\ω}}_{0}t+{\mathrm{m\ω}}_1).---(4-1)$

n=1…N,q=1…Q

In the formula

The distinctive spreading code of frequency m: spreading code c ^(k)Can be regarded as the distinctive spreading code of a user and residential quarter with the combination of cell scrambling v, different frequencies uses different scramblers:

$s_p^{(m,k)}=c_{1+\left[(p-1)\mathrm{mod}Q\right]}^{\left(k\right)}\·v_{1+\left[(p-1)\mathrm{mod}Q\right]}^m,k=1...K,p=1...\mathrm{NQ}---(4-2)$

Q is the later spreading code length of normalization, and length is that 16, N is a later symbol weight that code channel takies of chnnel coding modulation for TD-SCDMA, and the length after the normalization is 22, λ ^kBeing the amplitude adjusted factor, is to obtain according to power conversion.Cr ₀(t) be at t moment cosine filter coefficient.ω ₀The center frequency point of intermediate frequency, ω ₁Be the shared band bandwidth of each frequency.

Defined peak-to-average force ratio this moment and be the ratio of peak power and its average power within a period of time, according to definition:

$\mathrm{PAPR}=\frac{\underset{0\≤m\≤\mathrm{NQ}}{\max }\left\{{\left|x\right[m\left]\right|}^2\right\}}{\mathrm{mean}\left\{{\left|x\right[m\left]\right|}^2\right\}},0\≤m\≤(2\mathrm{NQ}+L_m)\·N_{\mathrm{TS}}\·N_{\mathrm{interpolate}}---(4-3)$

L _mThe length that is the midamble data is 144chip, N _TSTo participate in PAPR to calculate a timeslot number, N _InterpolateBe the factor of carrying out interpolation at intermediate frequency, the TDD communication system is carried out L (L=60) times interpolation at intermediate frequency in the present embodiment, so N _Interpolate=L.

From formula 4-1 as can be known, the number of the peak power of data symbol and modulation system, logical subscriber number, frequency has relation, and along with being on the increase of frequency number and number of users, peak-to-average force ratio will be larger.

S130: the peak clipping sequence of regulating according to changes in amplitude needs the signal of peak clipping to carry out cutting after superposeing.

Need the signal of peak clipping can refer to peak-to-average force ratio dB number and average power dB count sum greater than or more than or equal to the signal of default detection power dB number, described detection power is less than or equal to 1dB compression point input power value.

Although sometimes peak-to-average force ratio is larger, the average power of signal is less, if the peak power of this moment still is lower than default detection power, such as, peak power is lower than the 1dB compression point of power amplifier, and the peak power of signal had better not eliminate.

See also Fig. 2, it is the situation schematic diagram that average power is low and the peak-to-average force ratio transformation range is large.See also Fig. 3, it is the little situation schematic diagram of average power height and peak-to-average force ratio transformation range.At a time, for power amplifier, this maximum power point (average+peak-to-average force ratio=peak value) constantly importantly namely is not that average power neither peak-to-average force ratio, as long as peak power is not more than default detection power, distorted signals is just very little.If the power of signal input signal has exceeded the 1dB compression point of power amplifier, and the power of input signal continues to increase, the ACPR of signal can sharply worsen so, for example input signal surpasses the 1dB compression point, if signal continues to increase, every increase 1dB, the index of the ACPR in the system may worsen 3dB so.Because signal had just had the distortion of 1dB when input signal arrived the 1dB compression point.This time, ACPR began to have worsened, so the ACPR index of system requirements is very high, the signal of input can only be less than the 1dB compression point so, and only in this way the ACPR of system just can reach good effect.

For this reason, the present invention propose the peak-to-average force ratio PARP (dB) of signal and average power sum greater than or more than or equal to default detection power P _{In, max}, just carry out peak clipping and process.That is,

PARP(dB)+10log ₁₀ pow_mean>P _in，max (4-4)

Or

PARP(dB)+10log ₁₀ pow_mean≥P _in，max （4-5）

Detection power P _{In, max}For

P _in，max=P _in，1dB-X 0≤X≤3dB （4-6）

Wherein, P _{In, 1dB}Be 1dB compression point input power value, X is 1dB rollback value.1dB rollback value X is that the side channel leakage power according to system requirements arranges than ACLR, and the ACLR of system requirements is higher, and X arranges larger.

Because when input signal enters the 1dB compression point, it not linear gain, so for the demanding system of ACPR, be operated in the linear zone of power amplifier, to can or not produce very little distortion, at this moment the performance of ACPR will be got well, so detection power should be less than or equal to the input power of 1dB compression point.See also Fig. 4, its be under the high-performance ACPR detection power less than 1dB compression point schematic diagram.Detection power should be less than P _{In, max}, that is:

PARP(dB)+10log ₁₀ pow_mean>P _in，max (4-7)

Pow_mean=mean{|x[m wherein] | ², 0≤m≤(2NQ+L _m) N _TSN _Interpolate(4-8)

The signal maximum power that the input power amplifier is set this moment is pow_max ₀, 4-7 obtains according to formula:

pow_max ₀(dB)=P _in，max=P _in，1dB-X (4-9)

If X=0dB, maximal input just equals the input power of 1dB compression point so, and X is larger just shows that the ACPR of system requirements is higher.

The signal maximum power of input power amplifier is pow_max ₀, the signal maximum power of usually inputting power amplifier also is referred to as target power.In the peak clipping processing procedure, need to be with the power peak clipping of signal below target power.The present invention provides a threshold power detect_threshold again, be input in the peak clipping processor all be one the series connection data symbol, data of a timing cycle input, if the sampling clock cycle is 76.8MSPS, the cycle of then inputting data symbol is 1/76.8 microsecond, for this reason as long as long as the signal power of judging input greater than threshold power detect_threshold, just need to eliminate pow_max with the power of this signal ₀Below, in many cases detect_threshold and pow_max ₀Numerical value approaches or equates that target setting power of the present invention equates with threshold power.

The present invention can be only to peak-to-average force ratio dB number and average power dB count sum greater than or more than or equal to the signal of default detection power dB number, just carry out peak clipping and process, to the signal that does not satisfy condition, do not carry out the peak clipping processing.By the present invention, reduce the treating capacity of signal transmitting terminal, reduce intractability, reduced simultaneously the distortion of information.And, according to system the requirement of ACPR is arranged threshold power, it is more reasonable to arrange.

Because the length of observation time is different, resulting peak value may different these time, thereby simple consideration PAR has not had too large meaning, what more be concerned about is the statistical information of peak-to-average force ratio, namely need to describe PAPR from the angle of probability, usually use CCDF (Complementary cumulative distribution function) curve can express intuitively the distribution of the PAPR of TDD system.If peak value occurs among the present invention less than certain probability (for example: ten thousand/), then no longer consider this peak point.

CCDF is a kind of probability distribution, if the probability that a certain peak power occurs for power amplifier is very little, the probability that for example surpasses PAR=40dB only has ten thousand/, the distortion that the peak power point of small probability causes for system, worsen very little, the peak power point of this moment just can be ignored or do not considered, all need to consider so this patent arranges peak value under certain probability, and can ignore less than the peak value under certain probability, like this with regard to so that the setting of threshold power more reasonable.

Thresholding peak-to-average force ratio PAPR is set ₀For

${\mathrm{<figure-callout\; id="0"\; label="PAPR"\; filenames="H061B3250120061013E000032.png,H061B3250120061013E000051.png"\; state="\{\{state\}\}">PAPR</figure-callout>}}_0=10\&CenterDot;\log \left(\frac{{\mathrm{<figure-callout\; id="0"\; label="pow"\; filenames="H061B3250120061013E000032.png,H061B3250120061013E000051.png"\; state="\{\{state\}\}">pow</figure-callout>}}_0}{\mathrm{pow}\_\mathrm{mean}}\right)---(4-10)$

Any one is that PAPR constantly is:

$\mathrm{PAPR}=10\&CenterDot;\log \left(\frac{{\mathrm{pow}}_m}{\mathrm{pow}\_\mathrm{mean}}\right)---(4-11)$

So contrary accumulated probability distribution function CCDF:

CCDF=Prob[PAPR>PAPR ₀] (4-12)

According to the CCDF distribution function, the present invention's actual needs of giving chapter and verse is selected a suitable PARP ₀Excursion as transmitting power replaces theoretic maximum PAPR, such as requiring lower voice signal for the error rate, can choose CCDF=10 ^-3That corresponding PAPR ₀Value as to the requirement of radio-frequency (RF) power amplification, and for the demanding transfer of data of the error rate, can select PAPR corresponding to higher CCDF ₀The linearity of amplifier scope that is used as expecting.See also Fig. 5, it is the CCDF distribution map of the data-signal of TDD.Maximum peak-to-average force ratio value PAR does not reach 8.59dB before the peak clipping, arrive the not enough one thousandth of probability of the signal of this peak-to-average force ratio, can ignore its impact when power amplifier.At CCDF=10 ^-3PAR (the PAR=PAPR that this point is corresponding ₀) be 7.98dB, the peak-to-average force ratio of this moment just must be paid attention to, so the purpose of peak clipping is very clear and definite, is exactly will be input signal CCDF=10 ^-3Corresponding peak power eliminates.

That is, also comprise before the peak clipping: according to the error rate being required to arrange this professional corresponding contrary accumulated probability distribution function CCDF value, obtain parameter peak-to-average force ratio corresponding to CCDF value;

Comprise when peak clipping is processed: find peak-to-average force ratio more than or equal to the signal of described parameter peak-to-average force ratio, if described peak power greater than default threshold power, interpolation method eliminates peak value.

Still see also Fig. 5, CCDF=10 before the peak clipping ^-3Corresponding PAR=7.98, CCDF=10 after the peak clipping ^-3Corresponding PAR=5.8 illustrates that then at the one thousandth Probability Point, peak-to-average force ratio has improved about general 7.98-5.8=2.18dB.

Minimum for phase loss before and after the peak clipping, in peak clipping sequence is set, need most and consider that phase place is corresponding, the present invention accomplishes to adjust adaptively peak clipping sequence according to following steps.

In the input signal in a certain period the power of a maximum power point greater than threshold power (peak detection circuit detects peak power point):

pow_max _m>detect_threshold（4-13）

Pow_max wherein _m: pow_max _m=max (pow _m)=I _Max ²+ Q _Max ²(4-14)

Then use the digital interpolation method to carry out digital despicking, it is that I circuit-switched data and Q circuit-switched data deduct respectively interpolation filter coefficient that peak clipping is processed, simultaneously can be according to the Q road of input signal and the I rood phase place to peak power point, the signal that peak clipping sequence is inputted in attention this moment all is plural number, real part is the I road, imaginary part is the Q road, so complex symbol of every input just can obtain the phase place of this complex symbol:

$\tan \mathrm{\&theta;}=\frac{Q_{\max \_m}}{I_{\max \_m}}---(4-15)$

The phase place of input signal peak point also is the phase place of I road and Q road peak clipping sequence, and the phase place of peak clipping sequence maximum power point equates with the signal of peak power point.Each circuit-switched data and peak clipping sequence separately must equal the target power point to its peak power point after cutting simultaneously, so can derive

$\left\{\begin{array}{c}{(I_{\max \_m}-h_{i\max })}^2+{(Q_{\max \_m}-h_{q\max })}^2=\mathrm{pow}\_{\mathrm{<figure-callout\; id="0"\; label="max"\; filenames="H061B3250120061013E000032.png,H061B3250120061013E000051.png"\; state="\{\{state\}\}">max</figure-callout>}}_0\\ \tan \mathrm{\&theta;}=\frac{h_{q\max }}{h_{i\max }}\end{array}\right.---(4-16)$

Can obtain the greatest measure of each road (I road, Q road) amplitude adjusted

$h_{i\max }=I_{\max \_m}-\mathrm{abs}\left(\cos \mathrm{\&theta;}\right)\&CenterDot;\sqrt{{\mathrm{pow}\_\max }_0}---(4-17)$

$h_{q\max }=Q_{\max \_m}-\mathrm{abs}\left(\sin \mathrm{\&theta;}\right)\&CenterDot;\sqrt{{\mathrm{pow}\_\max }_0}---(4-18)$

Use the present invention so that phase loss is minimum, can be proven from following simulation result.Because peak clipping sequence essence is the FIR filter coefficient of a linear phase, numerical value is fixed.But the peak power point of input signal each time all may be different, the performance number that therefore will eliminate each time is different, namely the amplitude that eliminates of each road is different each time, try to achieve the amplitude that each road need to eliminate after, just can be in the hope of the amplitude adjusted factor each time:

$\mathrm{scaling}\_i=\frac{h_{i\max }}{\max \left(\underset{\&OverBar;}{h}\right)}---(4-19)$

$\mathrm{scaling}\_q=\frac{h_{q\max }}{\max \left(\underset{\&OverBar;}{h}\right)}---(4-20)$

According to the amplitude adjusted factor that obtains each time just can adaptive adjusting the amplitude of peak clipping sequence each time, wherein comprising the corresponding relation of phase place at this.

h _i ＝scaling_i· h （4-21）

h _q ＝scaling_q· h （4-22）

That is,

(1) regulating peak clipping sequence according to changes in amplitude comprises by following steps and calculates this I road peak clipping sequence h _i :

A1: utilize formula Calculate the greatest measure h of I road amplitude adjusted _Imax, wherein, pow_max ₀Be target power, I _{Max_m}Pow_max ₀Corresponding I way value,

Q _{Max_m}Be pow_max ₀Corresponding Q way value;

A2: utilize

Calculate the I road amplitude adjusted factor scaling_i of this peak clipping;

A3: utilize h _i =scaling_i hCalculate this I road peak clipping sequence h _i , wherein hBe predefined I road peak clipping sequence.

(2) regulating peak clipping sequence according to changes in amplitude comprises by following steps and calculates this Q road peak clipping sequence:

B1: utilize formula

Calculate the greatest measure h of Q road amplitude adjusted _Qmax, wherein, pow_max ₀Be target power, Q _{Max_m}Pow_max ₀Corresponding Q way value,

I _{Max_m}Be pow_max ₀Corresponding Q way value;

B2: utilize Calculate the Q road amplitude adjusted factor scaling_q of this peak clipping;

B3: utilize h _q =scaling_q hCalculate this Q road peak clipping sequence, wherein hBe predefined Q road peak clipping sequence.

(3) be treated to cutting among the step S130:

If I circuit-switched data corresponding to peak power be greater than zero, then the I circuit-switched data of this frequency point data of pre-save symbol and this I road peak clipping sequence h _i Subtract each other corresponding position, otherwise the I circuit-switched data of this frequency point data of pre-save symbol and this I road peak clipping sequence h _i Corresponding position addition; If the Q circuit-switched data that peak power corresponds to is greater than zero, then the Q circuit-switched data of this frequency point data of pre-save symbol and this Q road peak clipping sequence h _q Subtract each other corresponding position, otherwise the Q circuit-switched data of this frequency point data of pre-save symbol and this Q road peak clipping sequence h _q Corresponding position addition.

This frequency point data symbol of pre-save is the data symbol that power surpasses threshold power.

Realized the function that the amplitude self adaptation of peak clipping sequence (interpolation filter coefficient) is regulated according to top a series of processing.The below is the emulation to said method, when detecting a peak power point, just its power is eliminated target power, and the trend of waviness remains unchanged on every side simultaneously.

See also Fig. 6, it is that the peak clipping sequence of utilizing above-mentioned self adaptation to regulate is carried out the power contrast figure that front and back are processed in peak clipping.As can be known from Figure, process front and back power from peak clipping and contrast the consistency that has satisfied the peak clipping front and back.See also Fig. 7, it is the comparison schematic diagram before and after the peak clipping of I circuit-switched data amplitude.Please refer to Fig. 8, it is the comparison schematic diagram before and after the peak clipping of Q circuit-switched data amplitude.From Fig. 7 and Fig. 8 as can be known, each amplitude also can be carried out adaptive adjustment according to ratio separately.

Self application is regulated the self adaptation adjustment that despiker not only embodies amplitude, destroys minimum to the phase place of input signal simultaneously.Suppose that the amplitude of the Amplitude Ratio Q circuit-switched data of I circuit-switched data is much larger at this moment, if the I circuit-switched data is eliminated identical amplitude with the Q circuit-switched data, even if can just in time eliminate power threshold to peak power point, the phase place of input signal will suffer very large destruction so, if but adopt above-mentioned self adaptation to regulate the method for peak clipping sequence, the proportionate relationship of the amplitude of I road and Q circuit-switched data before according to peak clipping eliminates different amplitudes, and phase place is just destroyed relatively much less after the peak clipping, further improves thus peak-to-average force ratio.

By improving peak-to-average force ratio, can reach following technique effect:

1) improves the transmitting power of multi-carrier signal, in the substantially constant situation of total consumed power, also just improved the efficient of power amplifier; 2) improved the efficient of power amplifier, when needs carry out high-power output, just can greatly reduce the supply power of power supply, saved so on the one hand hardware cost, alleviate to a certain extent simultaneously the heat dissipation problem of base station, also saved the cost of operation maintenance simultaneously.

See also Fig. 9, it implements schematic diagram for the present invention eliminates one of peak method.Signal after the linear superposition divides I circuit-switched data and Q circuit-switched data to carry out respectively buffer memory, obtain peak power (S21), judge that whether described peak power is more than or equal to detection threshold (S22), if, then self adaptation is adjusted peak clipping sequence, and signal is carried out peak clipping process I circuit-switched data and the Q circuit-switched data obtain after the peak clipping, otherwise, directly export I circuit-switched data and Q circuit-switched data.The present invention can arrange threshold power by the requirement of ACPR.The present invention can also described signal peak-to-average force ratio dB number and average power dB count sum greater than or judge whether to carry out peak clipping more than or equal to default detection power dB number and process, perhaps the present invention can also further consider that whether the peak-to-average force ratio of signal is more than or equal to the parameter peak-to-average force ratio that arranges according to the CCDF value, if just carry out peak clipping and process.

Peak clipping is processed and comprised two processes: the first process is to obtain this I road peak clipping sequence and this Q road peak clipping sequence, and the second process is to utilize this I road peak clipping sequence and this Q road peak clipping sequence that this data symbol of preserving is carried out cutting processing.

The first process comprises: obtain first peak power point I road amplitude and Q road amplitude, namely obtain peak power pow_max ₀Corresponding I way value I _{Max_m}With acquisition peak power pow_max ₀Corresponding Q way value Q _{Max_m},, rear utilization

Calculate its phase place; (S23);

Then, generate respectively the I path filter coefficient adjustment factor (formula 4-19) and the Q path filter coefficient adjustment factor (formula 4-20) (S24);

Subsequently, calculate respectively this I road peak clipping sequence (formula 4-21) and this Q road peak clipping sequence (formula 4-22) (S25);

The second process comprises: utilize obtain this I road peak clipping sequence and this Q road peak clipping sequence finish peak clipping sequence to peak power signal to cutting processing (S26):

In order to verify the validity of this self adaptation adjusting peak clipping sequence method, the applicant verifies the method.Please refer to Figure 10, it is the schematic diagram of test adaptive peak clipping sequence method.It, is tested according to following several situations after the enterprising line linearity stack of time domain all frequency signals.The first is that signal is delivered to D/A converter without peak clipping and carried out D/A conversion (being exactly the 9dB that identifies among the figure), the second is that the target power value of peak clipping is 6dB, the third is that the target power value of peak clipping is 7dB, and the 4th kind is that the target power value of peak clipping is 8dB.

For the validity of contrast test index, the contrast test indices does not have the data of peak clipping and the peak clipping data of different target numerical value to compare test.See also Figure 11, it is that different peak clipping target value are on the impact of EVM.As can be known from Figure, EVM worsens to some extent after the peak clipping, but still satisfies 3GPP to the requirement of EVM, less than 12%.

The target of introducing peak clipping according to preamble is to improve total transmitting power, improve the operating efficiency of power amplifier and improve ACPR, also test for these several: fixedly gross power is tested the improvement situation (seeing also Figure 12) of ACPR, and the power of output fixing signal is checked the improvement situation (seeing also Figure 13) of total consumed power for this reason.

According to user's request can self application the adjustment interpolation filter coefficient, accurate cutting peak power point eliminates after the peak point simultaneously each time, and the damage of the phase place of signal is minimum, reduced phase width of cloth damage after the demodulation, also just so that the EVM index worsens improves.The peak clipping sequence automatic regulating equipment is simple, is easy to realize.Simultaneously test specification before and after the peak clipping to the improvement situations of power amplifier and radio-frequency devices.

A kind of communication ends that realizes cutting peak power comprises: some interpolation filters: be used for the circuit-switched data behind the ovennodulation spread spectrum is carried out L times of interpolation filtering; Some frequency converters: be used for the data after the interpolation are converted to a corresponding frequency point data;

One superimposer: connect all converter units, be used for all frequency point data after the frequency conversion in the enterprising line linearity stack of time domain;

The interpolation despiker comprises:

Amplitude adjusted maximum value calculation subelement: be used for calculating the greatest measure of I road amplitude adjusted and the greatest measure of Q road amplitude adjusted;

Amplitude adjusted factor computation subunit: be used for calculating this I road amplitude adjusted factor and the Q road amplitude adjusted factor;

This peak clipping sequence computation subunit: be used for calculating this I road peak clipping sequence and Q road peak clipping sequence;

Adder: be used for this frequency point data symbol and this peak clipping sequence of pre-save are carried out plus and minus calculation.

The present invention also comprises the peak clipping judging unit: be used for peak-to-average force ratio and average power sum greater than or just preserve to carry out the peak clipping processing more than or equal to the signal of default detection power, described detection power is less than or equal to 1dB compression point input power value.

Above-mentioned described interpolation despiker can be realized by programmable logic device FGPA.

More than disclosed only be several specific embodiment of the present invention, but the present invention is not limited thereto, the changes that any person skilled in the art can think of all should drop in protection scope of the present invention.

Claims (10)

1. the method for a cutting peak power is characterized in that, comprising:

(1) each the road signal behind the modulation spread spectrum carries out interpolation filtering, and frequency conversion becomes the different frequent points signal;

(2) all frequency signals are in the enterprising line linearity stack of time domain;

The peak clipping sequence of (3) regulating according to changes in amplitude needs the signal of peak clipping to carry out cutting after superposeing, the peak-to-average force ratio dB number of the described signal that needs peak clipping and average power dB count sum greater than or more than or equal to default detection power dB number, described detection power is less than or equal to 1dB compression point input power value.

2. the method for claim 1 is characterized in that, regulates peak clipping sequence according to changes in amplitude in the step (3) and comprises by following steps and calculate this I road peak clipping sequence h _i :

A1: utilize formula

Calculate the greatest measure h of I road amplitude adjusted _Imax, wherein, pow_max ₀Be target power, I _{Max_m}Pow_max ₀Corresponding I way value,

Q _{Max_m}Be pow_max ₀Corresponding Q way value;

A2: utilize Calculate the I road amplitude adjusted factor scaling_i of this peak clipping, wherein hBe predefined I road peak clipping sequence;

A3: utilize h _i =scaling_i hCalculate this I road peak clipping sequence h _i , wherein hBe predefined I road peak clipping sequence.

3. method as claimed in claim 1 or 2 is characterized in that, regulates peak clipping sequence according to changes in amplitude in the step (3) and comprises by following steps and calculate this Q road peak clipping sequence:

B1: utilize formula

Calculate the greatest measure h of Q road amplitude adjusted _Qmax, wherein, pow_max ₀Be target power, Q _{Max_m}Pow_max ₀Corresponding Q way value,

I _{Max_m}Be pow_max ₀Corresponding I way value;

B2: utilize

Calculate the Q road amplitude adjusted factor scaling_q of this peak clipping, wherein hBe predefined Q road peak clipping sequence;

B3: utilize h _q =scaling_q hCalculate this Q road peak clipping sequence, wherein hBe predefined Q road peak clipping sequence.

4. method as claimed in claim 3 is characterized in that, step is treated to cutting in (3):

If I circuit-switched data corresponding to peak power be greater than zero, then the I circuit-switched data of this frequency point data of pre-save symbol and this I road peak clipping sequence h _i Subtract each other corresponding position, otherwise the I circuit-switched data of this frequency point data of pre-save symbol and this I road peak clipping sequence h _i Corresponding position addition;

If the Q circuit-switched data that peak power corresponds to is greater than zero, then the Q circuit-switched data of this frequency point data of pre-save symbol and this Q road peak clipping sequence h _q Subtract each other corresponding position, otherwise the Q circuit-switched data of this frequency point data of pre-save symbol and this Q road peak clipping sequence h _q Corresponding position addition.

5. method as claimed in claim 4 is characterized in that, this frequency point data symbol of pre-save is the data symbol that power surpasses threshold power.

6. such as claim 1 or 4 described methods, it is characterized in that described detection power P _{In, max}For: P _{In, max}=P _{In, 1dB}-X 0≤X≤3dB, wherein, P _{In, 1dB}Be 1dB compression point input power value, X is 1dB rollback value.

7. such as claim 1 or 4 described methods, it is characterized in that X is 1dB rollback value, is that the side channel leakage power according to system requirements arranges than (ACLR), and the ACLR of system requirements is higher, and X arranges larger.

8. the method for claim 1 is characterized in that,

Step (3) also comprises before: according to the error rate being required to arrange this professional corresponding contrary accumulated probability distribution function (CCDF) value, obtain parameter peak-to-average force ratio corresponding to CCDF value;

The described signal of peak clipping that needs in the step (3) also is included as peak-to-average force ratio more than or equal to the signal of described parameter peak-to-average force ratio.

9. a communication ends that realizes cutting peak power is characterized in that, comprising: some interpolation filters: be used for the circuit-switched data behind the ovennodulation spread spectrum is carried out L times of interpolation filtering;

Some frequency converters: be used for the data after the interpolation are converted to a corresponding frequency point data;

One superimposer: connect all frequency converters, be used for all frequency point data after the frequency conversion are superposeed at the enterprising line linearity of time domain, the signal after the linear superposition divides I circuit-switched data and Q circuit-switched data to carry out respectively buffer memory;

The peak clipping judging unit: be used for peak-to-average force ratio and average power sum greater than or just preserve to carry out the peak clipping processing more than or equal to the signal of default detection power, described detection power is less than or equal to 1dB compression point input power value;

The interpolation despiker comprises:

Amplitude adjusted maximum value calculation subelement: be used for calculating the greatest measure of I road amplitude adjusted and the greatest measure of Q road amplitude adjusted;

Amplitude adjusted factor computation subunit: be used for calculating this I road amplitude adjusted factor and the Q road amplitude adjusted factor;

This peak clipping sequence computation subunit: be used for calculating this I road peak clipping sequence and Q road peak clipping sequence;

Adder: be used for this frequency point data symbol and this peak clipping sequence of pre-save are carried out plus and minus calculation.

10. communication ends as claimed in claim 9 is characterized in that, the interpolation despiker is programmable logic device (FGPA).

Images