CN1988524A - Method for reducing transmission signal peak average ratio of multiple carrier TD-SCDMA down link - Google Patents

Abstract

The invention discloses a method to lower PAPR of multi-carrier TD-SCDMA downlink transmission signals, including RRC filtering and QPSK modulation on each carrier separately, merging carriers of different routes, shearing the carrier signals, which is processed after the mergence combining the hard and soft shears. The invention is a shear method that can lower PAPR effectively and at the same time meet 3GPP launch indicators.

Description

A kind of method that reduces multi-carrier TD-SCDMA downlink transmission signal peak-to-average ratio

Technical field

The present invention relates to 3G (Third Generation) Moblie field multi-carrier TD-SCDMA base station transmit signals and handle, relate to a kind of method of improving the downlink transmission signal peak-to-average ratio specifically.

Background technology

For satisfying the growing market demand of mobile communication, people have proposed the 3G (Third Generation) Moblie standard, and wherein WCDMA and CDMA2000 have obtained commercialization.TD-SCDMA also is one of 3G (Third Generation) Moblie international standard, will enter pre-commercial stage very soon.After in a single day the TD-SCDMA standard obtains commercialization, will face the problem that how to continue to increase power system capacity, carrying out the multicarrier covering in same cell/section will be the important means that addresses this problem.Although formally do not define the multicarrier emission as yet in present TD-SCDMA standard, existing how tame manufacturer has carried out the suggestion of this respect, and multicarrier covering scheme is expected to include in very soon in the TD-SCDMA standard.

For the big capacity base station transceiver (BTS) that adopts multi-carrier scheme, for supporting more wireless carrier and more baseband signal processing unit, power consumption must increase, and power amplifier has occupied most of power consumption, and the power consumption that therefore reduces power amplifier is to realize the key of the big capacity of multicarrier base station.

At first the multi-carrier TD-SCDMA down link signal is described.

Single carrier TD-SCDMA down link signal spread spectrum model as shown in Figure 1.

The spread spectrum operation is divided into spread spectrum and two steps of scrambling, and wherein spread spectrum coefficient is between 1～16.Spreading code is a kind of ovsf code, and to guarantee the spreading code quadrature of different spreading factors on the unified time slot, this operation can be regarded as and is used for distinguishing different user in the same time slot.For reducing the PAR peak to average ratio in many yards whens transmission, for each channel code d ₁ ^{(k, i)}, d ₂ ^{(k, i)},

A relevant phase coefficient w is all arranged _Q ^kW is taken advantage of in the weighting of channel code data _Q ^kAfterwards, be Q with the data after the weighting through length _kChannel code C ^(k)Spread spectrum obtains data w _Q ^(k)d ₁ ^{(k, i)}(c ₁ ^{(k, i)}, c ₂ ^{(k, i)}... c _Qk ^{(k, i)}) w _Q ^(k)d ₂ ^{(k, i)}(c ₁ ^{(k, i)}, c ₂ ^{(k, i)}... c _Qk ^{(k, i)}), $\·\·\·w_Q^{\left(k\right)}{d}_{\frac{Q_{\mathrm{MAX}}}{Q_k}}^{(k,i)}(c_1^{(k,i)},c_2^{(k,i)},...c_{Q_k}^{(k,i)}),$ Then by the specific scrambler sequence in sub-district V=( V ₁, V ₂..., V ₁₆) carry out the position and take advantage of scrambling, obtain data V ₁, V ₂,, V _Qk, V _Qk+1, V _2Qk, V _QMAX-Qk+1, V _QMAX, i.e. data after spread spectrum and the scrambling.This operation can be regarded as and is used for distinguishing different districts.

Single carrier TD-SCDMA down link signal pooled model as shown in Figure 2.

Behind the spread spectrum, each complex value channel (user 1,2 ... K) multiply by weighted factor G respectively _j(j=1,2, K), if comprise synchronous channel SCH in the time slot, its weighted factor then is G _SCHLike this, we have obtained base band TD-SCDMA signal.Pulse shaping through common can obtain being with notch shape.

If realize the multicarrier emission on a BTS, each carrier wave needs independently to modulate, and then the radio frequency merging, makes an uproar but the power merging can increase the end, and the radio frequency multi-carrier signal peak equal also will further worsen than (PAR), thereby reduces the efficient of power amplifier.Multi-carrier signal being carried out base-band digital handle, is to reduce the effective means of making an uproar and controlling multi-carrier signal PAR in the end.The range of linearity of the signal pair amplifier that PAR is high requires higher, otherwise signal will suffer hard shearing, thereby causes intermodulation distortion, reduces effect.

Secondly, the test to the base station transmitter regulation in the 3GPP standard is described.

(Error Vector Magnitude EVM) is difference between desirable transmitted waveform and the actual transmission waveform to the error vector value, and therefore in fact this value is equivalent to average error rate (BER), is less than 12.5% in accordance with regulations; Peak Code Domain Error (Peak Code Domain Error PCDE) is the tolerance of crosstalking between the TD-SCDMA different channels, in other words, tolerance be orthogonality between different code channels, spreading factor is to be less than-28dB in 16 o'clock in accordance with regulations; (Adjacent Channel Leakage Ratio is transmitting power and its ratio at the measured value of neighboring trace ACLR) to ACLR, is better than 40dB in accordance with regulations.

Stipulated test (calculating) flow process at these indexs in the 3GPP TDD uniformity test standard,, earlier output signal and reference signal have been made explanations for describing these flow processs better.

Tx output signal to be measured: in the Tx output signal input measurement instrument to be measured,, store to treat further processing through RRC filtering.Physical signalling availability vector Z between whole measurement zone represents that this vector comprises N=ns * sf+ma duplicate sample originally, and wherein, ns is a symbolic number interior between measurement zone, and sf is the number of chips in each symbol, and ma is the midamble number of chips.

Reference signal: can after same matched filter filtering, when having intersymbol interference, not store by measuring instrument construction reference signal according to relevant Tx standard.Reference signal availability vector R between whole measurement zone represents that this vector comprises N=ns * sf+ma duplicate sample originally.

The Tx measurement result can be divided into two classes in the channel: deviation class and residual error class.

Deviation class: do not having the value non-zero of parameter under the situation of error.These parameter representation signals, such as the RF frequency, power (under the solid size situation), code domain power (under the many yards situations), sequential (only to UE) etc.

The residual error class: in the value that does not have parameter under the situation of error is zero.The difference of these parameter representation signals is such as error vector value (EVM) and Peak Code Domain Error (pCDE).

Change reference signal and measured signal in the residual error class parameter, measured signal after changing and the root mean square between reference signal (RMS) difference hour just reach best fit approximation.Measured signal after the change is with Z ' expression, and the reference signal after the change is with R ' expression, and then error vector can be expressed as E=Z '-R '.

At parameter to be assessed, E can take following two kinds multi-form:

EVM form (the physical error signal in representing between whole measurement zone)

A vectorial E comprises N=ns * sf+ma duplicate sample originally;

Wherein, ns, sf, the definition of ma is the same.

PCDE form (between the sign field, taking a sample) by the derivation of EVM form

Ns seasonal effect in time series vector e originally is made of sf duplicate sample between each sign field.

The calculation procedure of EVM is as follows:

1) the error vector E by above-mentioned definition calculates its root-mean-square value RMS (E).

2) the reference signal R by above-mentioned definition calculates its root-mean-square value RMS (R).

3) be calculated as follows EVM

$\mathrm{EVM}=\frac{\mathrm{RMS}\left(E\right)}{\mathrm{RMS}\left(R\right)}\×100\%$

The calculation procedure of PCDE is as follows:

1) gets the error vector e of above-mentioned definition.

2) get orthogonal vectors among the spreading code set C, do normalized, obtain Cnorm=C/sqrt (sf) by vector norm.

3) to the e descrambling, obtain e '.

4) inner product of calculating e ' and Cnorm to all interior between measurement zone symbols and multiple this process of all code weights in the code space, obtains the array of a k * ns thus, and each element is represented the error vector of a special symbol and special code in the array.Wherein K is the sum of sign indicating number in the code space, and ns is the symbolic number between measurement zone.

5) a calculating K RMS value, each RMS value are (these values can be described as " absolute code EVM " (unit: volt)) that is obtained by ns symbol in the sign indicating number.

6) from K " absolute code EVM ", find out peak value (this value can be described as " absolute peak sign indicating number EVM ").

7) be calculated as follows PCDE:

Once more, the existing method that suppresses signal peak is described.

Be to suppress signal peak, the normal method that adopts of people be hard shearing (Hard clipping, HC).Three carrier signals are sheared typical scenario as shown in Figure 3 firmly, and its step is mainly:

1), to each carrier wave, statistics (or estimation) peak-peak is sheared by certain percentage.

2), each carrier wave after the shearing is respectively through RRC (Root Raised Cosine, root raised cosine) filtering and QPSK (Quadrature Phase Shift Keying, Quadrature Phase Shift Keying) modulation.

3), each road carrier wave merges the back up-conversion, launches through power amplifier.

The hard advantage of shearing is to realize that circuit is simple, and it is little to delay time, and can effectively control and be with inside and outside leakage.Shortcoming is to have destroyed frequency spectrum, if carry out filtering again after shearing, then can make peak value regeneration again, can not effectively reduce multi-carrier signal PAR.

Summary of the invention

Technical problem to be solved by this invention is, on the basis of the hard cutting method of multi-carrier signal, satisfy soft or hard shearing method (the Soft-Hard Clipping of 3GPP emission index when having proposed a kind of effective reduction multi-carrier signal PAR, SHC), this method is soft shearing and the hard method that combines of shearing.

In order to address the above problem, the method of reduction TD-SCDMA multi-carrier downlink link signal peak-to-average force ratio of the present invention, be that a kind of soft or hard is sheared the method for (SHC), comprise: each carrier wave is carried out RRC filtering and QPSK modulation respectively, shear with each road carrier wave merging with to carrier signal, it is characterized in that, describedly carrier signal is carried out cutting steps carrying out after each road carrier wave merges, is the soft shearing of each the road carrier signal after being combined and the shearing that hard shearing combines.

Above-mentionedly carrier signal carried out cutting steps comprise:

Step a) is judged signal peak and position thereof;

Step b) is set and is sheared threshold value, the number of pulses in the control Preset Time window;

Step c) after the pulse train of the signal phase information that is maintained, is carried out low pass FIR filtering to it;

Step d) deducts filtered sequence from primary signal, to suppress selected peak value;

Said method is characterized in that, described step a) adopts adjacent signals peakedness ratio method.

Said method, its characteristics are that the initial value of threshold value is set at 80% of signal peak-peak in the described step b), dynamically adjust by observing shear effect.

Said method, its characteristics are that the number of pulses in the described step b) in the Preset Time window is controlled to be 3.

Said method, its characteristics are, also further comprise step e), with same threshold value in the step b), burst are sheared once more firmly, guarantee to reduce peak-to-average force ratio.

Method of the present invention can reduce the peak-to-average force ratio PAR of multi-carrier signal significantly when guaranteeing to satisfy ACLR, PCDE and EVM.

Description of drawings

Fig. 1 is a single carrier TD-SCDMA down link signal spread spectrum model framework chart;

Fig. 2 is a single carrier TD-SCDMA down link signal pooled model block diagram;

Fig. 3 is that three carrier signals are sheared the typical scenario block diagram firmly;

Fig. 4 is that three carrier signal soft or hards are sheared the typical scenario block diagram;

Fig. 5 is PAR-EVM simulation result figure under hard shearing and the soft or hard shearing scheme of the present invention;

Fig. 6 is ACLR-EVM simulation result figure under hard shearing and the soft or hard shearing scheme of the present invention;

Fig. 7 is PCDE-EVM simulation result figure under hard shearing and the soft or hard shearing scheme of the present invention.

Embodiment

Describe embodiments of the invention with reference to the accompanying drawings in detail.

As shown in Figure 4, be that three carrier signal soft or hards are sheared the typical scenario block diagram.Compare with Fig. 3, the shearing scheme of using soft or hard shearing method of the present invention comprises step (is example with three carrier waves):

Step 101, to each carrier wave through RRC filtering;

Step 102 is modulated through QPSK each carrier wave;

Step 103 merges each road carrier wave;

Step 104 is carried out soft or hard with the carrier signal after merging and is sheared.

Launch through power amplifier then.

Wherein, soft or hard shearing method of the present invention specifically comprises:

Step a) is judged signal peak and position thereof;

Step b) is set and is sheared threshold value, the number of pulses in the control Preset Time window;

Step c) after the pulse train of the signal phase information that is maintained, is carried out low pass FIR filtering to it;

Step d) deducts filtered sequence from primary signal, to suppress selected peak value;

Step e) is used same threshold value, and burst is sheared once more firmly, guarantees to reduce PAR.

Judge that the signal peak ratio is easier to, and can adopt adjacent signals peakedness ratio method.Key is to control treats the shear signal number of pulses in the Preset Time window (filter length), if quantity is too much, can cause excessive shear, destroys frequency spectrum; If quantity is very few, do not reach the effect that suppresses peak value again.For being easy to realize, the shear pulse quantity for the treatment of in the suggestion time window is 3.

The setting of threshold value can realize by calibration, can be made as 80% of peak-peak such as the initial threshold value, observes shear effect by measuring instrument, and the situation of showing is dynamically adjusted.

Application example below by simulation result shows soft or hard shearing method of the present invention.Investigate the situation of TD-SCDMA down link three carrier signals.The bandwidth of each carrier wave is 1.6MHz, and spreading factor is 16, and over-sampling rate is 8,48 rank RRC filters, and emulation length is 20ms.Set different threshold values and carry out emulation, getting and making the threshold value of PAR minimum when can satisfy the EVM requirement is final threshold value, and under this threshold value, calculates signal ACLR and PCDE value after shearing.Adopt hard shearing method and soft or hard shearing method to carry out signal suppressing respectively, simulation result is seen Fig. 5～7.When getting EVM=12.5% (3GPP requirement), each parameter value and 3GPP setting relatively see the following form under two kinds of shearing schemes.

The effect comparison of two kinds of method for peak suppression of table 1

	PAR(dB)	ACLR(dB)	PCDE(dB)
				HC	8.8	117.4	-38.0
SHC	4.8	59.5	-37.5
				The 3GPP requirement	/	＞40	＜-28

Easily know by result in the table, the soft or hard shearing method satisfy 3GPP about ACLR, PCDE and EVM numerical value regulation in, compare the peak-to-average force ratio PAR that hard shearing method can reduce multi-carrier signal significantly.

Certainly; the present invention also can have other various embodiments; under the situation that does not deviate from spirit of the present invention and essence thereof; those of ordinary skill in the art work as can make various corresponding changes and modification according to the present invention, but these corresponding changes and modification all should belong to the protection range of the appended claim of the present invention.

Claims (6)

1, a kind of method that reduces multi-carrier TD-SCDMA downlink transmission signal peak-to-average ratio, comprise each carrier wave is carried out RRC filtering and QPSK modulation respectively, each road carrier wave is merged, with carrier signal is sheared, it is characterized in that, describedly carrier signal is carried out cutting steps carrying out after each road carrier wave merges, is the shearing of each the road carrier signal after being combined.

2, method according to claim 1 is characterized in that, describedly carrier signal is carried out cutting steps comprises:

Step a) is judged signal peak and position thereof;

Step b) is set and is sheared threshold value, the number of pulses in the control Preset Time window;

Step c) after the pulse train of the signal phase information that is maintained, is carried out low pass FIR filtering to it;

Step d) deducts filtered sequence from primary signal, to suppress selected peak value.

3, method according to claim 2 is characterized in that, described step a) adopts adjacent signals peakedness ratio method.

4, method according to claim 2 is characterized in that, the initial value of threshold value is set at 80% of signal peak-peak in the described step b), dynamically adjusts by observing shear effect.

5, method according to claim 2 is characterized in that, the number of pulses in the described step b) in the Preset Time window is controlled to be 3.

6, method according to claim 2 is characterized in that, also further comprises step e), with same threshold value in the step b), burst is sheared once more firmly, guarantees to reduce peak-to-average force ratio.

Images