CN102546510A - Method for decreasing peak-to-average power ratio of orthogonal frequency division multiplexing (OFDM) signal - Google Patents

Abstract

The invention discloses a method for decreasing a peak-to-average power ratio (PAPR) of an orthogonal frequency division multiplexing (OFDM) signal. The method comprises the following steps of: clipping a time-domain sequence x, projecting a clipped signal to a linear space consisting of all time-domain sub-sequences, calculating a phase rotation factor al<(v)> of each time-domain sub-sequence x<(v)>, multiplying each time-domain sub-sequence x<(v)> by using a corresponding phase rotation factor al<(v)>, superposing results, calculating a new linear combination signal xSLC, judging whether PAPRSLC is less than or equal to PAPR0 or not according to the PAPRSLC of the xSLC, finishing iteration if PAPRSLC is less than or equal to PAPR0, performing repeated iteration if PAPRSLC is not less than or equal to PAPR0, performing parallel/serial conversion on the time-domain sequence of which the PAPR is decreased, and transmitting the time-domain sequence to a receiver. The phase rotation factors are generated by iteration and projection, the PAPR can be decreased to a certain extent by the iteration of each time, and computational complexity is linearly increased along with the increasing of the number V of the phase rotation factors; and compared with the conventional method, the invention can greatly reduce required computational complexity when the PAPR is required to be decreased to a great extent.

Description

A kind of method that reduces the ofdm signal peak-to-average power ratio

Technical field

The invention belongs to the wireless communication technology field that adopts orthogonal frequency-division multiplex singal (OFDM), be specifically related to a kind of method of the reduction ofdm signal peak-to-average power ratio based on grouping linear combination.

Background technology

In mobile radio channel, signal arrives reception antenna from transmitting antenna through a time-variant multipath channel, can decline of generation time selectivity and frequency selective fading.The time-varying characteristics of channel cause the broadening of signal spectrum, cause Doppler (Doppler) effect, cause signal to be the selectivity decline in time.According to the frequency selective fading characteristic that multipath channel shows in frequency domain, people have proposed OFDM (OFDM) technology.The OFDM technology is meant frequency domain is divided into a plurality of subchannels, and is mutually orthogonal overlapping between each adjacent sub-channel, thereby improves the utilization ratio of frequency spectrum.It is thus clear that OFDM can overcome frequency selective fading on the one hand; On the other hand, get less than a period of time of coherence time at interval as duration of an OFDM symbol, thereby can reduce of the influence of the time selective fading of channel greatly to transmission system.

Yet the OFDM technology still exists some major issues not solved well.One of them difficult point and key technology are exactly how to the control of high peak-to-average power ratio (ratio of peak power and average power).Because ofdm signal is to be formed by stacking a plurality of sub-carrier signals, compares with constant envelope signal such as single-carrier signal, the peak-to-average power ratio of ofdm signal will be a height very.If the peak-to-average power ratio of signal is very high, the linear dynamic range of transmitter power amplifier will require very wide.The power amplifier that linear dynamic range is wide is with making the cost of transmitter increase greatly.Though the power amplifier that linear dynamic range is low can reduce cost, send seriously distortion of signal, thereby cause the serious decline of systematic function.Therefore, in order to reduce the requirement of ofdm signal, need to reduce the peak-to-average power ratio of ofdm signal to transmit power amplifier.

At present, people have proposed the peak-to-average power ratio that a lot of methods reduce ofdm signal.Wherein, partial sequence transmission (Partial Transmit Sequences) technology is a kind of effective method.The sub-carrier number of supposing ofdm system is N, and then an OFDM frequency domain signal X comprises N modulation symbol, X=[X (0), X (1) ..., X (N-1)].The main thought of partial transmission sequence can be summarized as: modulated frequency domain signal X is divided into V mutually disjoint frequency domain subsequence, is { X ^(v), v=0,1 ..., V-1}, the length of each subsequence still is N, promptly each subsequence comprises

Individual non-vanishing modulation symbol, other all use zero to fill, promptly

$X=\underset{v=0}{\overset{V-1}{\mathrm{\&Sigma;}}}{X}^{\left(v\right)}$

Here, subsequence is counted V >=2, and the value of V is to confirm that by the reduction amount of the peak-to-average power ratio needs of ofdm system signal generally speaking, V gets 4,8 or 16.Then, each frequency domain subsequence is carried out inverse fast Fourier transform (IFFT) computing that N is ordered, i.e. x ^(v)=IFFT (X ^(v)).To each x ^(v)Multiply by phase rotation coefficient b ^(v)And summation, then

$x=\underset{v=0}{\overset{V-1}{\mathrm{\&Sigma;}}}{b}^{\left(v\right)}{x}^{\left(v\right)}$

For the ease of phase rotation coefficient being sent to receiving terminal, b ^(v)Generally in that { 1 ,-1, j chooses in four centrifugal pumps of-j}, wherein

Travel through all possible 4 ^VIndividual phase rotation coefficient b ^(v)Value, thereby obtain different time-domain signal x, in these time-domain signals, select the minimum signal of peak-to-average power ratio to send.

It is thus clear that; Though partial transmission sequence method can reduce peak-to-average power ratio effectively,, when needs obtain bigger peak-to-average power ratio reduction amount; Just need to increase packet count V; Thereby produce a lot of candidate signals, and calculate the peak-to-average power ratio of each candidate signal, select minimum conduct to send.Because computation complexity is exponential increase with packet count V, can make computation complexity very high like this.

Summary of the invention

To the high shortcoming of method complexity that reduces peak-to-average power ratio in the prior art, the present invention proposes a kind of method of new reduction ofdm system signal peak-to-average power power ratio, this method greatly reduces computation complexity when reducing peak-to-average power ratio.

The invention provides a kind of method that reduces ofdm system signal peak-to-average power power ratio, this method may further comprise the steps:

(1) bit stream of input is encoded, interweaved and modulates, obtain the frequency domain data signal;

(2) in the frequency domain data signal, insert frequency pilot sign, obtain frequency domain sequence X={X (k) according to the pectination pattern, k=0,1 ..., N-1}, wherein, N is total sub-carrier number of ofdm system, k representes the subcarrier sequence number; And frequency domain sequence X carried out the inverse transformation of N point quick Fourier, obtain time domain sequences x=[x (0), x (1) ..., x (N-1)];

(3) with frequency domain sequence X after string and conversion, be divided into V frequency domain subsequence { X of equal in length according to the mode of adjoining ^(v), v=0,1 ..., V-1}, wherein, v representes the sequence number of frequency domain subsequence, V representes the number of frequency domain subsequence, 2≤V＜N;

X ^(v)＝[X ^(v)(0)，X ^(v)(1)，...，X ^(v)(N-1)]，

(4) to each frequency domain subsequence X ^(v)Carry out the inverse transformation of N point quick Fourier, obtain corresponding time domain subsequence x ^(v)=[x ^(v)(0), x ^(v)(1) ... x ^(v)(N-1)];

(5) time domain sequences x is reduced peak-to-average power ratio and handles, specifically comprise following substep:

(5.1) make x _l=x, iterations l=0;

(5.2) to x _lShear, obtain shear signal x _{C, l}=[x _{C, l}(0), x _{C, l}(1) ..., x _{C, l}(N-1)];

(5.3) with shear signal x _{C, l}Project to by all time domain subsequence { x ^(v), v=0,1 ..., the linear space L (x that V-1} forms ⁽⁰⁾, x ⁽¹⁾...., x ^(V-1)) on, calculate by following formula afterwards, obtain each time domain subsequence x ^(v)Phase rotation coefficient a _l ^(v),

${a_l}^{\left(v\right)}=\frac{\left(\underset{n=0}{\overset{N-1}{\mathrm{\&Sigma;}}}{x}_{c,l}\left(n\right){\left(x^{\left(v\right)}\left(n\right)\right)}^{*}\right)/\left(\underset{n=0}{\overset{N-1}{\mathrm{\&Sigma;}}}{\left[\right|x^{\left(v\right)}\left(n\right)\left|\right]}^2\right)}{|\left(\underset{n=0}{\overset{N-1}{\mathrm{\&Sigma;}}}{x}_{c,l}\left(n\right){\left(x^{\left(v\right)}\left(n\right)\right)}^{*}\right)/\left(\underset{n=0}{\overset{N-1}{\mathrm{\&Sigma;}}}{\left[\right|x^{\left(v\right)}\left(n\right)\left|\right]}^2\right)|}$

Wherein, n=0,1 ..., N-1, * represents conjugation;

(5.4) with each time domain subsequence x ^(v)Multiply by corresponding phase rotation coefficient a respectively _l ^(v), and result stack obtained the linear combination signal

(5.5) calculate new linear composite signal $x_{\mathrm{SLC}}=x_l+{\mathrm{\&lambda;}}_l({\hat{x}}_{\mathrm{SLC}}-x_l);$ Relaxation factor λ _lCalculate by following formula:

Wherein, f (n)=x _{C, 1}(n)-and x (n), n=0,1 ..., N-1, the iteration maximum times of I for setting, λ ^cBe a constant, λ ^c∈ [1,2];

(5.6) calculate new linear composite signal x _SLCPeak-to-average power ratio PAPR _SLC, judge whether PAPR _SLC≤PAPR ₀, PAPR wherein ₀Be the target peak-to-average power ratio of setting,, get into step (6) if then iteration stops; If not, then make l=l+1, x _l=x _SLC, change step (5.2) over to;

The time domain sequences that (6) will reduce peak-to-average power ratio sends to receiving terminal after also string is changed.

The present invention proposes a kind of method that reduces the signal peak-to-average power power ratio of ofdm system; This method produces the phase rotation coefficient sequence through iterative projection; Each iteration can have certain peak-to-average power ratio reduction amount, and the several times iteration obtains bigger peak-to-average power ratio reduction amount and traditional partial sequence transmission method ratio later on; When the big peak-to-average power ratio reduction amount of needs, required computation complexity has very big reduction.At receiving terminal, utilize the method for existing segmentation channel estimating to recover the phase rotation coefficient sequence that transmitting terminal uses, guaranteed that the error rate is unaffected.Through emulation experiment checking, when the big peak-to-average power ratio reduction amount of needs, the computation complexity of the method for the invention can access bigger reduction.

Description of drawings

Fig. 1 is the system block diagram of ofdm system transmitter end of the present invention;

Fig. 2 is the block diagram that the iteration described in the present invention reduces ofdm signal peak-to-average power ratio method;

Fig. 3 is the system block diagram of ofdm system receiver end of the present invention.

Embodiment

Below in conjunction with accompanying drawing and instance the present invention is further specified.

As shown in Figure 1, total sub-carrier number of establishing this ofdm system is N, and k representes the subcarrier sequence number; The define system frequency-region signal is X={X (k), k=0,1; ..., N-1}, this signal are divided into V sub-block (2≤V＜N); The value of V is to confirm that by the reduction amount of the peak-to-average power ratio needs of ofdm system signal general V gets 4,8 or 16.

The method of the equal peak of reduction ofdm system of the present invention power ratio comprises the steps:

(1) bit stream of input is encoded, interweaved and modulates, obtain the frequency domain data signal;

(2) in the frequency domain data signal, insert frequency pilot sign, promptly when

(i=1,2 according to pectination pattern (comb type); ..., in the time of 2V), insert frequency pilot sign; Obtain frequency domain sequence X; X is carried out the inverse transformation of N point quick Fourier, obtain time domain sequences x=[x (0), x (1); ..., x (N-1)];

(3) with X after string and conversion, be divided into V frequency domain subsequence X of equal in length according to the mode of adjoining ^(v), X wherein ^(v)=[X ^(v)(0), X ^(v)(1) ..., X ^(v)(N-1)],

Therefore have

Frequency domain subsequence X ^(v)In comprise data-signal, frequency pilot sign and 0.

(4) each frequency domain subsequence is carried out the inverse fast Fourier transform that N is ordered, obtain corresponding time domain subsequence x ^(v), wherein

x ^(v)＝[x ^(v)(0)，x ^(v)(1)，...x ^(v)(N-1)]；

(5) as shown in Figure 2, according to step (5.1) to (5.6) time domain sequences x is reduced peak-to-average power ratio and handle.If PAPR ₀Be the target peak-to-average power ratio, setting range is 5～9dB usually; A ₀Be predefined shearing thresholding, setting range is 5～15dB usually, specifically confirms that according to selected modulation system I is corresponding iteration maximum times, generally confirms λ through emulation experiment ^cBe a constant, satisfy λ ^c∈ [1,2];

(5.1) make x _l=x, iterations l=0;

(5.2) to time series x _lCarry out shear treatment and obtain shear signal x _{C, l}, x _{C, l}=[x _{C, l}(0), x _{C, l}(1) ..., x _{C, l}(N-1)], described shear treatment is meant:

$x_{c,l}\left(n\right)=\left\{\begin{array}{cc}{x}_l\left(n\right),& \left|x_l\left(n\right)\right|\&le;A_0\\ A_0{e}^{\mathrm{j\&theta;}\left(n\right)},& \left|x_l\left(n\right)\right|>A_0\end{array}\right.$

Wherein, x _l(n) be x _lIn n element, n=0,1 ..., N-1, N represent total sub-carrier number of ofdm system, θ (n) is x _l(n) phase place, A ₀For shearing thresholding;

x _{C, l}(n) for shearing later time domain sequences.

(5.3) with shear signal x _{C, 1}Project to by all time domain subsequence { x ^(v), v=0,1 ..., the linear space L (x that V-1} forms ⁽⁰⁾, x ⁽¹⁾...., x ^(V-1)) on, press the phase rotation coefficient that following formula calculates each time domain subsequence afterwards:

${a_l}^{\left(v\right)}=\frac{\left(\underset{n=0}{\overset{N-1}{\mathrm{\&Sigma;}}}{x}_{c,l}\left(n\right){\left(x^{\left(v\right)}\left(n\right)\right)}^{*}\right)/\left(\underset{n=0}{\overset{N-1}{\mathrm{\&Sigma;}}}{\left[\right|x^{\left(v\right)}\left(n\right)\left|\right]}^2\right)}{|\left(\underset{n=0}{\overset{N-1}{\mathrm{\&Sigma;}}}{x}_{c,l}\left(n\right){\left(x^{\left(v\right)}\left(n\right)\right)}^{*}\right)/\left(\underset{n=0}{\overset{N-1}{\mathrm{\&Sigma;}}}{\left[\right|x^{\left(v\right)}\left(n\right)\left|\right]}^2\right)|}$

Wherein, n=0,1 ..., N-1, * represents conjugation; || represent absolute value;

(5.4) with each time domain subsequence x ^(v)Multiply by corresponding phase rotation coefficient a respectively _l ^(v), and stack obtains the linear combination signal Promptly ${\hat{x}}_{\mathrm{SLC}}=\underset{v=0}{\overset{V-1}{\mathrm{\&Sigma;}}}{a}_l^{\left(v\right)}{x}^{\left(v\right)};$

(5.5) calculate new linear composite signal $x_{\mathrm{SLC}}=x_l+{\mathrm{\&lambda;}}_l({\hat{x}}_{\mathrm{SLC}}-x_l);$ Relaxation factor λ _lCalculate by following formula,

Wherein, f (n)=x _{C, 1}(n)-and x (n), n=0,1 ..., N-1, λ ^cBe a constant, λ ^c∈ [1,2];

(5.6) calculate new linear composite signal x _SLCPeak-to-average power ratio PAPR _SLC, judge whether PAPR _SLC≤PAPR ₀If,, then iteration stops, and gets into step (6); If not, then make l=l+1, x _l=x _SLC, change step (5.2) over to.

The time domain sequences that (6) will reduce peak-to-average power ratio sends to receiving terminal after also string is changed.

Information processing method to receiving terminal describes below, and is as shown in Figure 3.

(a) to the received signal through string and conversion, fast Fourier transform (FFT), and string conversion after, obtain receiving terminal frequency domain sequence Y,

Y＝{Y(k)，k＝0，1，...，N-1}；

(b) by identical mode in the above-mentioned steps (3), Y is divided into V receiving terminal frequency domain subsequence { Y ^(v), v=0,1 ..., V-1},

Wherein, H (k) is illustrated in k number of sub-carrier upper signal channel frequency domain response value, and W (k) is the frequency domain value of white Gaussian noise on the k number of sub-carrier;

(c) the v value is calculated all interior data symbols of V receiving terminal frequency domain subsequence according to step (c1) to (c2) from 0 to V-1:

(c1) defines the virtual channel frequency response

is

using the formula (c1) is calculated:

$H^{\&prime;}\left(k_i^{\left(v\right)}\right)=Y^v\left(k_i^{\left(v\right)}\right)/P\left(k_i^{\left(v\right)}\right)---\left(c1\right)$

Wherein,

is illustrated in the frequency-region signal value on

number of sub-carrier; I frequency pilot sign (i=1 in v receiving terminal frequency domain subsequence of

expression; 2; ...; 2V), the position of all frequency pilot signs is known with value at receiving terminal;

Through type (c1); Pseudo channel frequency response on all pilot frequency symbol positions in v the receiving terminal frequency domain subsequence is all calculated; Utilize the pseudo channel frequency response values on the pilot frequency symbol position, obtain pseudo channel frequency response through interpolation, if the employing linear interpolation at the subcarrier place at the data symbol place of v receiving terminal frequency domain subsequence; Then the pseudo channel frequency response is obtained by following formula on the data symbol subcarrier of v receiving terminal frequency domain subsequence, promptly

$H^{\&prime;}\left(k^{\left(v\right)}\right)=H^{\&prime;}\left(k_i^{\left(v\right)}\right)+\left(\frac{H^{\&prime;}\left(k_{i+1}^{\left(v\right)}\right)-H^{\&prime;}\left(k_i^{\left(v\right)}\right)}{L}\right)(k^{\left(v\right)}-k_i^{\left(v\right)}),$

$=a_I^{\left(v\right)}[H\left(k_i^{\left(v\right)}\right)+\left(\frac{H\left(k_{i+1}^{\left(v\right)}\right)-H\left(k_i^{\left(v\right)}\right)}{L}\right)(k^{\left(v\right)}-k_i^{\left(v\right)})],(k_i^{\left(v\right)}<k^{\left(v\right)}<k_{i+1}^{\left(v\right)})$

Wherein, k ^vBe the position of data symbol in v the receiving terminal frequency domain subsequence,

If adopt other interpolation methods, computing formula difference to some extent then, the present invention does not do requirement to interpolation method;

(c2) receiving terminal utilizes the pseudo channel frequency response

Calculate the data symbol D ' (k of v receiving terminal frequency domain subsequence ^(v)):

D′(k ^(v))＝Y ^v(k ^(v))/H′(k ^(v))；

(d) data symbol in V the receiving terminal frequency domain subsequence that step (c) is obtained is merged into a sequence according to the size of subcarrier sequence number, again with this sequence through demodulation, the bit stream that deinterleaves and decode and obtain exporting.

Process according to producing phase rotation coefficient in the method for the invention can be found out; The method that the present invention proposes is owing to be the value of directly calculating V phase rotation coefficient; Its complexity is linear growth with the V growth, and through analyzing, each required complexity of iteration is not very high yet; Iterations can be very not big yet, total computation complexity like this can be more much lower with the complexity that the V growth is exponential increase than traditional partial sequence transmission method.

Embodiments of the invention adopt following concrete parameter scheme: the sub-carrier number of ofdm system is 64, and packet count is 8, and the pilot tone number is 16; Input signal is QPSK (Quadrature Phase Shift Keying; QPSK) signal of modulation is sheared thresholding and is made as 2.0, λ ^c=1.95, the target peak-to-average power ratio is PAPR ₀=6.0dB, corresponding maximum iteration time is I=20.

Show through simulation result, at complementary cumulative distribution function C CDF=10 ^-3The time, when the peak-to-average power ratio of reduction ofdm signal was 4.3dB, method of the present invention required real number addition number of times and partial sequence transmission method were than reducing by 99.68%, and required real multiplications number of times and partial sequence transmission method are than reducing by 97.66%.

More than be an instance of the present invention, but the present invention should not be confined to the disclosed content of this instance and accompanying drawing.So everyly do not break away from the equivalence of accomplishing under the disclosed spirit of the present invention or revise, all fall into the scope of the present invention's protection.

Claims (2)

1. a method that reduces ofdm system signal peak-to-average power power ratio is characterized in that, this method may further comprise the steps:

(1) bit stream of input is encoded, interweaved and modulates, obtain the frequency domain data signal;

(2) in the frequency domain data signal, insert frequency pilot sign, obtain frequency domain sequence X={X (k) according to the pectination pattern, k=0,1 ..., N-1, wherein, N is total sub-carrier number of ofdm system, k representes the subcarrier sequence number; And frequency domain sequence X carried out the inverse transformation of N point quick Fourier, obtain time domain sequences x=[x (0), x (1) ..., x (N-1)];

(3) with frequency domain sequence X after string and conversion, be divided into V frequency domain subsequence { X of equal in length according to the mode of adjoining ^(v), v=0,1 ..., V-1}, wherein, 2≤V＜N,

X ^(v)＝[X ^(v)(0)，X ^(v)(1)，...，X ^(v)(N-1)]，

(4) to each frequency domain subsequence X ^(v)Carry out the inverse transformation of N point quick Fourier, obtain corresponding time domain subsequence x ^(v)=[x ^(v)(0), x ^(v)(1) ... x ^(v)(N-1)];

(5) time domain sequences x is reduced peak-to-average power ratio and handles, comprise following substep:

(5.1) make x _l=x, iterations l=0;

(5.2) to x _lShear, obtain shear signal x _{C, l}=[x _{C, l}(0), x _{C, l}(1) ..., x _{C, l}(N-1)];

(5.3) with shear signal x _{C, l}Project to by all time domain subsequence { x ^(v), v=0,1 ..., the linear space L (x that V-1} forms ⁽⁰⁾, x ⁽¹⁾...., x ^(V-1)) on, calculate by following formula afterwards, obtain each time domain subsequence x ^(v)Phase rotation coefficient a _l ^(v),

${a_l}^{\left(v\right)}=\frac{\left(\underset{n=0}{\overset{N-1}{\mathrm{\&Sigma;}}}{x}_{c,l}\left(n\right){\left(x^{\left(v\right)}\left(n\right)\right)}^{*}\right)/\left(\underset{n=0}{\overset{N-1}{\mathrm{\&Sigma;}}}{\left[\right|x^{\left(v\right)}\left(n\right)\left|\right]}^2\right)}{|\left(\underset{n=0}{\overset{N-1}{\mathrm{\&Sigma;}}}{x}_{c,l}\left(n\right){\left(x^{\left(v\right)}\left(n\right)\right)}^{*}\right)/\left(\underset{n=0}{\overset{N-1}{\mathrm{\&Sigma;}}}{\left[\right|x^{\left(v\right)}\left(n\right)\left|\right]}^2\right)|}$

Wherein, n=0,1 ..., N-1, * represents conjugation;

(5.4) with each time domain subsequence x ^(v)Multiply by corresponding phase rotation coefficient a respectively _l ^(v), and result stack obtained linear combination signal x _SLC

(5.5) calculate new linear composite signal x _SLC=x _l+ λ _l(x _SLC-x _l); Relaxation factor λ _lCalculate by following formula,

Wherein, f (n)=x _{C, 1}(n)-and x (n), n=0,1 ..., N-1, the iteration maximum times of I for setting, λ ^cBe a constant, λ ^c∈ [1,2];

(5.6) calculate new linear composite signal x _SLCPeak-to-average power ratio PAPR _SLC, judge whether PAPR _SLC≤PAPR ₀, PAPR wherein ₀Be the target peak-to-average power ratio of setting,, get into step (6) if then iteration stops; If not, then make l=l+1, x _l=x _SLC, change step (5.2) over to;

The time domain sequences that (6) will reduce peak-to-average power ratio sends to receiving terminal after also string is changed.

2. the method for reduction ofdm system signal peak-to-average power power ratio according to claim 1 is characterized in that, according to following formula to time domain sequences x _l(n) shear,

$x_{c,l}\left(n\right)=\left\{\begin{array}{cc}{x}_l\left(n\right),& \left|x_l\left(n\right)\right|\&le;A_0\\ A_0{e}^{\mathrm{j\&theta;}\left(n\right)},& \left|x_l\left(n\right)\right|>A_0\end{array}\right.$

Wherein, n=0,1 ..., N-1; N representes total sub-carrier number of ofdm system, θ _nBe x _l(n) phase place, A ₀For shearing thresholding;

x _{C, l}(n) for shearing later time domain sequences.

Images