CN103532900A - Dynamic compensation method for reducing LED (Light Emitting Diode) nonlinear distortion of visible light Flip-OFDM (Orthogonal Frequency Division Multiplexing) communication system - Google Patents

Abstract

The invention relates to a dynamic compensation method for reducing LED (Light Emitting Diode) nonlinear distortion of a visible light Flip-OFDM (Orthogonal Frequency Division Multiplexing) communication system. The method comprises the following steps: segmenting a modulated subcarrier data vector into a plurality of non-overlapped subsequences; multiplying by a corresponding weighting coefficient, extending to be a hermitian symmetric sequence and then performing IFFT (Inverse Fast Fourier Transform); selecting a weighting coefficient for keeping the negative signal amplitude of a combined signal to be minimum; decomposing the polarity of the combined signal into a Flip-OFDM positive signal and a flip negative signal and limiting the amplitude; carrying an end cut signal of the positive signal at a zero signal position corresponding to the flip negative signal as a compensation signal; restoring the positive signal by a receiving end through the compensation signal. By using the method, PAPR (Peak to Average Power Ratio) in the Flip-OFDM system can be effectively suppressed, and the LED nonlinear distortion is reduced, so that the usability and the reliability of the system are greatly improved.

Description

Reduce the dynamic compensation method of visible ray Flip-OFDM communication system LED nonlinear distortion

Technical field

The present invention relates to a kind of dynamic compensation method that reduces visible ray Flip-OFDM communication system LED nonlinear distortion.

Background technology

Visible light communication (VLC) have transmitting power high, without electromagnetic interference, without advantages such as frequency spectrum authentication, energy savings, can effectively alleviate Radio Spectrum Resource and be tending towards exhausted problem, be more and more subject to people's attention in recent years.In view of OFDM (OFDM) technology, there is the advantages such as spectrum efficiency is high, bandwidth expansion strong, anti-multipath fading, frequency spectrum resource flexible allocation, therefore OFDM technology is applied to VLC, effectively the transmission rate of elevator system.In VLC, the optical signal modulation mode of intensity modulated/direct-detection requires OFDM (OFDM) signal to be necessary for real signal, and meets unipolarity.Upset light OFDM(Flip-OFDM) by the positive signal in ofdm signal is sent respectively and is realized unipolarity with upset negative signal, it has, and power utilization is high, frequency spectrum resource distributes the advantages such as flexible, is the unipolarity OFDM scheme in VLC with broad prospect of application.

Yet, in VLC system, the signal transfer characteristic of LED presents non-linear, the peak-to-average force ratio that OFDM is higher (PAPR) has very high requirement to the almost linear dynamic range of LED, and when LED almost linear dynamic range can not meet requiring of ofdm signal, signal will distort, the frequency spectrum of superposed signal is changed, thereby cause the orthogonality between each sub-channels to be destroyed, produce phase mutual interference, reduce systematic function.At present, the method that reduces PAPR mainly contains two classes: 1) signal pre-distortion method, and as margining amplitude technique, compression expansion method, but its simple and practical distortion that can bring signal causes distortion; 2) non-signal pre-distortion method, as compiling method, selectivity mapping (SLM) method, partial transmission sequence (PTS) method etc., compiling method suppresses the effective of PAPR, but operational coding pattern negligible amounts, SLM and PTS lay stress on the probability that reduces peak value appearance, be not conceived to reduce the maximum of signal amplitude, can not avoid the appearance of high peak valve signal in system completely.Said method is all for the PAPR Design of Problems of general ofdm communication system, while being applied in visible ray Flip-OFDM communication system, do not consider the characteristic of its single monopole, the in the situation that in system subcarrier order of modulation, higher or sub-carrier number being larger, PAPR rejection is limited, cause LED nonlinear distortion serious, thereby seriously restricted the performance of communication system.

Summary of the invention

For above the deficiencies in the prior art, the technical problem to be solved in the present invention is to provide a kind of dynamic compensation method that reduces visible ray Flip-OFDM communication system LED nonlinear distortion, it can effectively avoid the appearance of high peak valve signal in system, the PAPR of reduction system, reduce the distortion of the non-linear introducing of LED, the bit error rate of improvement system (BER) performance, and then availability and the reliability of assurance communication system.

Technical scheme of the present invention is:

A dynamic compensation method that reduces visible ray Flip-OFDM communication system LED nonlinear distortion, comprises the following steps:

(1) at the transmitting terminal of Flip-OFDM system, the bit stream of input is gone here and there and changed and modulate, obtain data vector X={X (0), X (1) ..., X (N/2-1) }, wherein N is the sub-carrier number of Flip-OFDM system;

(2) data vector X is divided into the subsequence X of a plurality of non-overlapping copies _v, each subsequence is expanded, make it meet close conjugate symmetry in distress, i.e. X _v(0)=X _v(N/2)=0, X _v(N-k)=X _v(k) ^*, k=0 wherein, 1,2 ... N/2-1, X _v(k) ^*represent subsequence X _vthe conjugation of k item; And introduce weight coefficient b _v, wherein

above-mentioned v=1,2 ..., V, the subsequence number that wherein V is partial transmission sequence; Use weight coefficient b _vremove weighting expansion each subsequence X afterwards _v, then each subsequence obtaining is carried out respectively suing for peace after fast Fourier transform, according to the linear behavio(u)r of fast Fourier transform, can obtain composite signal x and be:

$x=\underset{v=1}{\overset{V}{\mathrm{\Σ}}}\mathrm{IFFT}\left(b_v{X}_v\right)=\underset{v=1}{\overset{V}{\mathrm{\Σ}}}{b}_v\·\mathrm{IFFT}\left(X_v\right)=\underset{v=1}{\overset{V}{\mathrm{\Σ}}}{b}_v{x}_v$

(3) select weight coefficient, make the negative signal amplitude of composite signal minimum, the selected scope of weight coefficient set be ± 1, ± i}, initialization b _v=1, v=1 wherein, 2 ..., V, through searching loop method, finds out the b of the negative signal amplitude minimum that makes composite signal _vvalue:

$\{b_1,b_2,...,b_V\}=\underset{\{b_1,{\mathrm{<figure-callout\; id="2"\; label="b"\; filenames="HDA0000388542190000022.png"\; state="\{\{state\}\}">b</figure-callout>}}_2,...,b_V\}}{\arg \min }\left(\underset{\underset{v=1}{\overset{V}{\mathrm{\&Sigma;}}}{b}_v{x}_v<\mathrm{0,1}\&le;k\&le;N}{\max }{\left|\underset{v=1}{\overset{V}{\mathrm{\&Sigma;}}}{b}_v{x}_v\right|}^2\right)$

Argmin(wherein) judgment condition that representative function is used while obtaining minimum value;

(4) composite signal x corresponding to above-mentioned weight coefficient carried out to polarity decomposition, obtain the positive signal of Flip-OFDM

with upset negative signal

$x_k^{+}=\left\{\begin{array}{cc}{x}_k& x_k\&GreaterEqual;0\\ 0& x_k<0\end{array}\right.k=\mathrm{0,1},...,N-1$

$x_k^{-}=\left\{\begin{array}{cc}{-x}_k& x_k<0\\ 0& x_k\&GreaterEqual;0\end{array}\right.k=\mathrm{0,1},...,N-1$

X wherein _kthe k item that represents signal x;

On Flip-OFDM signal, add corresponding direct current biasing signal, according to the almost linear region scope of LED, Flip-OFDM signal is carried out to amplitude limit, and by positive signal cut top signal by way of compensation signaling bearer upset zero-signal position corresponding to negative signal, upset negative signal is carried out to delay process, then the positive signal after processing and upset negative signal are luminous as former and later two ofdm signal driving LED, and signal is sent to light wireless channel in the mode of luminous intensity;

(5) at the receiving terminal of Flip-OFDM system, utilize photoelectricity testing part that light intensity signal is converted to the signal of telecommunication, remove direct current biasing signal and carry out analog-to-digital conversion and obtain positive signal and reception signal corresponding to upset negative signal $y^{+}=\{y_k^{+},k=\mathrm{0,1},...,N-1\},$ $y^{-}=\{y_k^{-},k=\mathrm{0,1},...,N-1\},$ To y ⁺and y ^-carry out negative signal wave absorption, part reduces noise effect, and the operating process of negative signal wave absorption is as follows:

${\tilde{y}}_k^{+}=\left\{\begin{array}{c}{y}_k^{+},y_k^{+}\&GreaterEqual;0\\ 0,y_k^{+}<0\end{array}\right.k=\mathrm{0,1},...,N-1$

${\tilde{y}}_k^{-}=\left\{\begin{array}{c}{y}_k^{-},y_k^{-}\&GreaterEqual;0\\ 0,y_k^{-}<0\end{array}\right.k=\mathrm{0,1},...,N-1$

(6) will adjudicate with dynamic compensation and obtain receiving signal y={y through the signal of negative signal wave absorption _k, k=0,1 ..., N-1}, i.e. contrast with

getting wherein larger is signal, and less is noise or compensating signal, if

?

for signal,

for noise,

if

set noise effect thresholding γ, when

time be considered as noise, when

time be considered as compensating signal, above-mentioned judgement compensation method is specially:

(7) to signal y={y _k, k=0,1 ... N-1} carries out fast Fourier transform, obtains getting its front N/2 frequency-region signal value as receiving signal Y={Y (0), Y (1) after frequency-region signal,, Y (N/2-1) }, Y is divided into the subsequence Y of a plurality of non-overlapping copies according to the partitioning scheme of step (2) _v, v=1 wherein, 2 ..., V, and respectively divided by the weight coefficient b of the negative signal amplitude minimum that makes composite signal of choosing in step (3) _v, then carry out sequence merging, this collating sequence is the frequency domain data signal that receiving terminal receives; Again this frequency domain data signal is separated to mediation parallel-serial conversion, finally obtained the bit stream of Flip-OFDM system output.

Above-mentioned dynamic compensation method, data vector X is divided in the method for a plurality of subsequences in step (2) wherein, partitioning scheme adopts adjacent cutting apart, and wherein the value of V is determined by the required reducing amount reaching of peak-to-average force ratio of Flip-OFDM system signal.

Above-mentioned dynamic compensation method, wherein the value of V is 4 or 8.

Above-mentioned dynamic compensation method, while wherein Flip-OFDM signal being carried out to amplitude limit in step (4), is limited in its envelope peak within the scope of almost linear region, and wherein the dynamic range in almost linear region is { x|V _l.min≤ x≤V _l.max, V _l.maxthe be as the criterion upper bound of the range of linearity, V _l.minthe lower bound of the range of linearity is as the criterion.

Above-mentioned dynamic compensation method wherein, in step (4), adds direct current biasing signal on Flip-OFDM signal, carries out amplitude limit, and the concrete steps that compensating signal is carried on upset negative signal are: the direct current biasing signal V that chooses interpolation<sub TranNum="144">bias</sub>meet V<sub TranNum="145">bias</sub>=V<sub TranNum="146">l.min</sub>thereby the signal adding after direct current biasing meets { x|x>=V<sub TranNum="147">l.min</sub>, now the condition of system works in almost linear region is that Flip-OFDM signal meets { x|0<x≤V<sub TranNum="148">max</sub>, V wherein<sub TranNum="149">max</sub>meet V<sub TranNum="150">max</sub>=V<sub TranNum="151">l.max</sub>-V<sub TranNum="152">bais</sub>, the Flip-OFDM positive signal after amplitude limit<img TranNum="153" file="BDA0000388542180000041.GIF" he="68" img-content="drawing" img-format="GIF" inline="yes" orientation="portrait" wi="50"/>be expressed as:

$s_k^{+}=\left\{\begin{array}{cc}{V}_{\max }& x_k^{+}>V_{\max }\\ x_k^{+}& 0\&le;x_k^{+}\&le;V_{\max }\end{array}\right.k=\mathrm{0,1},...,N-1$

Compensating signal χ _kbe expressed as:

${\mathrm{\&chi;}}_k=\left\{\begin{array}{cc}(x_k^{+}-V_{\max })& x_k^{+}>{\mathrm{<figure-callout\; id="0"\; label="V\; max"\; filenames="HDA0000388542190000031.png"\; state="\{\{state\}\}">V</figure-callout>}}_{\max }\\ 0& 0\&le;x_k^{+}\&le;V_{\max }\end{array}\right.k=\mathrm{0,1},...,N-1$

Upset negative signal

be expressed as:

$s_k^{-}=x_k^{-}+{\mathrm{\&chi;}}_k,k=\mathrm{0,1},...,N-1$

Above-mentioned dynamic compensation method, for the thresholding γ described in step (6), when SNR<4.5dB, γ=∞, when SNR>=4.5dB, γ is expressed as:

$\frac{\mathrm{\&gamma;}}{{\mathrm{\&sigma;}}_z}=0.75\frac{{(\mathrm{SNR}-4.5)}^n+a}{{(\mathrm{SNR}-4.5)}^m+b}$

Wherein (a, b, m, n)=(0.9336,0.03341,0.4875,0.3982), σ _zfor noise criteria poor.

The invention has the beneficial effects as follows:

1, the present invention is by adopting above-mentioned dynamic compensation method, choose the weight coefficient of the negative signal amplitude minimum that makes composite signal, composite signal is decomposed to the Flip-OFDM signal obtaining through polarity and carry out amplitude limiting processing, thereby effectively avoided the appearance of OFDM high peak valve signal, reduced system PAPR.Simultaneously, because upset negative signal amplitude is less, its corresponding clipped noise is also little, and the cutting top signal and can carry out dynamic compensation by compensating signal at receiving terminal of positive signal, thereby greatly reduces the distortion of the non-linear introducing of LED, has improved the BER performance of system;

2, the present invention passes through Flip-OFDM signal limitations within the scope of the almost linear region of LED, thereby can be further level and smooth Flip-OFDM time-domain signal, reduces the distortion that the variation due to LED operating current, frequency is introduced output harmonic wave and intermodulation component dynamic change;

3, the present invention is in visible ray Flip-OFDM model of communication system, the reduction LED nonlinear distortion of take is starting point, utilize the unipolarity characteristic of Flip-OFDM system signal, in conjunction with PTS and Clipping method, propose a kind of dynamic compensation method of the LED of reduction nonlinear distortion, effectively reduced peak-to-average force ratio and the error rate of system.

Accompanying drawing explanation

Below in conjunction with accompanying drawing, the specific embodiment of the present invention is described in further detail.

Fig. 1 is visible ray Flip-OFDM communication system physical layer transceiver block diagram;

Fig. 2 is the LED Voltage-current conversion characteristic curve of visible light communication system;

Fig. 3 is that the present invention is in conjunction with the dynamic compensation method schematic diagram of PTS and Clipping;

Fig. 4 is the CCDF simulation curve of the invention process dynamic compensation method;

Fig. 5 is the BER simulation curve of the invention process dynamic compensation method.

Embodiment

In specific implementation process, the physical layer fundamental block diagram of model Flip-OFDM system.As shown in Figure 1, provided the transmitting-receiving block diagram of Flip-OFDM system, the message processing flow of Flip-OFDM system physical layer is: at transmitting terminal, the bit stream of input is carried out obtaining frequency domain data stream after parallel-serial conversion, modulation, after frequency domain data is extended further to close conjugate symmetric sequence in distress, carry out IFFT conversion, then the OFDM real signal obtaining is decomposed into positive signal and sends with upset negative signal; At receiving terminal, receive positive signal and upset negative signal, the two converts, goes close conjugate symmetry in distress, demodulation, parallel-serial conversion through FFT after subtracting each other, and obtains output bit flow.

As shown in Figure 2, wherein there is one section of almost linear region in visible light communication system LED Voltage-current conversion characteristic curve, the dynamic range in almost linear region can be expressed as { x|V in the middle of characteristic curve _l.min≤ x≤V _l.max, V wherein _l.maxthe be as the criterion upper bound of the range of linearity, V _l.minthe lower bound of the range of linearity is as the criterion.Should make ofdm signal be operated in as far as possible almost linear region, the present invention solves the signal causing because of the high PAPR of Flip-OFDM system to exceed almost linear region scope, causes the problem of LED nonlinear distortion.

The dynamic compensation method schematic diagram of bound fraction transfer sequence (PTS) method of the present invention and amplitude limit (Clipping) method as shown in Figure 3, the method is the new method proposing on the basis of conventional P TS and Clipping, it makes composite signal have the weight coefficient of minimal negative signal amplitude by selection, then composite signal is carried out to amplitude limit, by positive signal cut top signal by way of compensation signaling bearer in zero-signal position corresponding to negative signal, at receiving terminal, by compensating signal, recover positive signal.

The dynamic compensation method that reduces visible ray Flip-OFDM communication system LED nonlinear distortion for this, it specifically comprises the following steps:

(1) at the transmitting terminal of Flip-OFDM system, the bit stream of input is gone here and there and changed and modulate, obtain data vector X={X (0), X (1) ..., X (N/2-1) }, wherein N is the sub-carrier number of Flip-OFDM system;

(2) data vector X is divided into the subsequence X of a plurality of non-overlapping copies _v, each subsequence is expanded, make it meet close conjugate symmetry in distress, i.e. X _v(0)=X _v(N/2)=0, X _v(N-k)=X _v(k) ^*, k=0 wherein, 1,2 ... N/2-1, X _v(k) ^*represent subsequence X _vthe conjugation of k item; And introduce weight coefficient b _v, wherein

above-mentioned v=1,2 ..., V, the subsequence number that wherein V is partial transmission sequence; Use weight coefficient b _vremove weighting expansion each subsequence X afterwards _v, then each subsequence obtaining is carried out respectively suing for peace after fast Fourier transform, according to the linear behavio(u)r of fast Fourier transform, can obtain composite signal x, that is:

$x=\underset{v=1}{\overset{V}{\mathrm{\&Sigma;}}}\mathrm{IFFT}\left(b_v{X}_v\right)=\underset{v=1}{\overset{V}{\mathrm{\&Sigma;}}}{b}_v\&CenterDot;\mathrm{IFFT}\left(X_v\right)=\underset{v=1}{\overset{V}{\mathrm{\&Sigma;}}}{b}_v{x}_v$

Partitioning scheme in this step can adopt adjacent cutting apart, and wherein the value of V is that the reducing amount being needed by the peak-to-average force ratio of Flip-OFDM system signal is determined, generally gets 4 or 8.

(3) select weight coefficient, make the negative signal amplitude of composite signal minimum, the selected scope of weight coefficient set be ± 1, ± i}, initialization b _v=1, v=1 wherein, 2 ..., V, through searching loop method, finds out the b of the negative signal amplitude minimum that makes composite signal _vvalue:

$\{b_1,{\mathrm{<figure-callout\; id="2"\; label="b"\; filenames="HDA0000388542190000022.png"\; state="\{\{state\}\}">b</figure-callout>}}_2,...,b_V\}=\underset{\{b_1,{\mathrm{<figure-callout\; id="2"\; label="b"\; filenames="HDA0000388542190000022.png"\; state="\{\{state\}\}">b</figure-callout>}}_2,...,b_V\}}{\arg \min }\left(\underset{\underset{v=1}{\overset{V}{\mathrm{\&Sigma;}}}{b}_v{x}_v<\mathrm{0,1}\&le;k\&le;N}{\max }{\left|\underset{v=1}{\overset{V}{\mathrm{\&Sigma;}}}{b}_v{x}_v\right|}^2\right)$

Argmin(wherein) judgment condition that representative function is used while obtaining minimum value;

(4) composite signal corresponding to above-mentioned weight coefficient carried out to polarity decomposition, obtain the positive signal of Flip-OFDM

with upset negative signal

$x_k^{+}=\left\{\begin{array}{cc}{x}_k& x_k\&GreaterEqual;0\\ 0& x_k<0\end{array}\right.k=\mathrm{0,1},...,N-1$

$x_k^{-}=\left\{\begin{array}{cc}{-x}_k& x_k<0\\ 0& x_k\&GreaterEqual;0\end{array}\right.k=\mathrm{0,1},...,N-1$

X wherein _kthe k item that represents signal x.

On Flip-OFDM signal, add corresponding direct current biasing signal, according to the almost linear region scope of LED, Flip-OFDM signal is carried out to amplitude limit, its envelope peak is limited within the scope of almost linear region, and by positive signal cut top signal by way of compensation signaling bearer upset zero-signal position corresponding to negative signal.Choose the direct current biasing signal V of interpolation<sub TranNum="223">bias</sub>meet V<sub TranNum="224">bias</sub>=V<sub TranNum="225">l.min</sub>thereby the signal adding after direct current biasing meets { x|x>=V<sub TranNum="226">l.min</sub>, now the condition of system works in almost linear region is that Flip-OFDM signal meets { x|0<x≤V<sub TranNum="227">max</sub>, V wherein<sub TranNum="228">max</sub>meet V<sub TranNum="229">max</sub>=V<sub TranNum="230">l.max</sub>-V<sub TranNum="231">bais</sub>, the Flip-OFDM positive signal after amplitude limit<img TranNum="232" file="BDA0000388542180000068.GIF" he="74" img-content="drawing" img-format="GIF" inline="yes" orientation="portrait" wi="50"/>be expressed as:

$s_k^{+}=\left\{\begin{array}{cc}{V}_{\max }& x_k^{+}>V_{\max }\\ x_k^{+}& 0\&le;x_k^{+}\&le;V_{\max }\end{array}\right.k=\mathrm{0,1},...,N-1$

Compensating signal χ _kbe expressed as:

${\mathrm{\&chi;}}_k=\left\{\begin{array}{cc}(x_k^{+}-V_{\max })& x_k^{+}>{\mathrm{<figure-callout\; id="0"\; label="V\; max"\; filenames="HDA0000388542190000031.png"\; state="\{\{state\}\}">V</figure-callout>}}_{\max }\\ 0& 0\&le;x_k^{+}\&le;V_{\max }\end{array}\right.k=\mathrm{0,1},...,N-1$

Upset negative signal be expressed as:

$s_k^{-}=x_k^{-}+{\mathrm{\&chi;}}_k,k=\mathrm{0,1},...,N-1$

Upset negative signal is carried out to delay process, and then the positive signal after processing and upset negative signal are luminous as former and later two ofdm signal driving LED, and signal is sent to light wireless channel in the mode of luminous intensity;

(5) at the receiving terminal of Flip-OFDM system, utilize photoelectricity testing part that light intensity signal is converted to the signal of telecommunication, remove direct current biasing signal and carry out analog-to-digital conversion and obtain positive signal and reception signal corresponding to upset negative signal $y^{+}=\{y_k^{+},k=\mathrm{0,1},...,N-1\},$ $y^{-}=\{y_k^{-},k=\mathrm{0,1},...,N-1\},$ To y ⁺and y ^-carry out negative signal wave absorption, part reduces noise effect, and the operating process of negative signal slicing is as follows:

${\tilde{y}}_k^{+}=\left\{\begin{array}{c}{y}_k^{+},y_k^{+}\&GreaterEqual;0\\ 0,y_k^{+}<0\end{array}\right.k=\mathrm{0,1},...,N-1$ ${\tilde{y}}_k^{-}=\left\{\begin{array}{c}{y}_k^{-},y_k^{-}\&GreaterEqual;0\\ 0,y_k^{-}<0\end{array}\right.k=\mathrm{0,1},...,N-1$

(6) will adjudicate with dynamic compensation and obtain receiving signal y={y through the signal of negative signal wave absorption _k, k=0,1 ..., N-1}, i.e. contrast

with

getting wherein larger is signal, and less is noise or compensating signal, if

?

for signal, for noise,

if set noise effect thresholding γ, when

time be considered as noise, when time be considered as compensating signal, above-mentioned judgement compensation method is specially:

The value of above-mentioned thresholding γ is relevant with SNR, by corresponding training sequence, obtains, and when SNR<4.5dB, γ=∞, when SNR>=4.5dB, γ can be expressed as:

$\frac{\mathrm{\&gamma;}}{{\mathrm{\&sigma;}}_z}=0.75\frac{{(\mathrm{SNR}-4.5)}^n+a}{{(\mathrm{SNR}-4.5)}^m+b}$

Wherein (a, b, m, n)=(0.9336,0.03341,0.4875,0.3982), σ _zfor noise criteria poor.

(7) to signal y={y _k, k=0,1 ... N-1} carries out fast Fourier transform, obtains getting its front N/2 frequency-region signal value as receiving signal Y={Y (0), Y (1) after frequency-region signal,, Y (N/2-1) }, Y is divided into the subsequence Y of a plurality of non-overlapping copies according to the partitioning scheme of step (2) _v, v=1 wherein, 2 ..., V, and respectively divided by the weight coefficient b of the negative signal amplitude minimum that makes composite signal of choosing in step (3) _v, then carry out sequence merging, this collating sequence is the frequency domain data signal that receiving terminal receives; Again this frequency domain data signal is separated to mediation parallel-serial conversion, finally obtained the bit stream of Flip-OFDM system output.

For further verifying, for above-mentioned dynamic compensation method, be given in the embodiment under following design parameter scheme.Wherein, design parameter scheme is: Flip-OFDM system subcarrier number is 256, modulation system is QAM, subsequence number is 4, LED is typical OSRAMSFH4230 large power white light LED, and almost linear dynamic range is about 1.4V～2V, and the selected scope of weight coefficient set is { ± 1, ± i}, carries out Monte Carlo simulation 10000 times.

For above-mentioned l-G simulation test, Fig. 4 is conventional P TS-Clipping method and the dynamic compensation method CCDF curve when subcarrier adopts 32QAM modulation, Fig. 5 be conventional P TS-Clipping method and dynamic compensation method at subcarrier the BER curve in different Q AM order of modulation situation.By above-mentioned two figure, can be found out, adopt dynamic compensation method of the present invention effectively to reduce PAPR and the BER of system, thereby guarantee availability and the reliability of system.

By reference to the accompanying drawings the preferred embodiments of the disclosure and embodiment are explained in detail above, but the present invention is not limited to the above-described embodiment and examples, in the ken possessing those skilled in the art, can also make without departing from the inventive concept of the premise various variations.

Claims (6)

1. a dynamic compensation method that reduces visible ray Flip-OFDM communication system LED nonlinear distortion, is characterized in that comprising the following steps:

(1) at the transmitting terminal of Flip-OFDM system, the bit stream of input is gone here and there and changed and modulate, obtain data vector X={X (0), X (1) ..., X (N/2-1) }, wherein N is the sub-carrier number of Flip-OFDM system;

(2) data vector X is divided into the subsequence X of a plurality of non-overlapping copies _v, each subsequence is expanded, make it meet close conjugate symmetry in distress, i.e. X _v(0)=X _v(N/2)=0, X _v(N-k)=X _v(k) ^*, k=0 wherein, 1,2 ..., N/2-1, X _v(k) ^*represent subsequence X _vthe conjugation of k item; And introduce weight coefficient b _v, wherein

above-mentioned v=1,2 ..., V, the subsequence number that wherein V is partial transmission sequence; Use weight coefficient b _vremove weighting expansion each subsequence X afterwards _v, then each subsequence obtaining is carried out respectively suing for peace after fast Fourier transform, according to the linear behavio(u)r of fast Fourier transform, can obtain composite signal x and be:

(3) select weight coefficient, make the negative signal amplitude of composite signal minimum, the selected scope of weight coefficient set be ± 1, ± i}, initialization b _v=1, v=1 wherein, 2 ..., V, through searching loop method, finds out the b of the negative signal amplitude minimum that makes composite signal _vvalue:

Argmin(wherein) judgment condition that representative function is used while obtaining minimum value;

(4) composite signal x corresponding to above-mentioned weight coefficient carried out to polarity decomposition, obtain the positive signal of Flip-OFDM

with upset negative signal

X wherein _kthe k item that represents signal x;

On Flip-OFDM signal, add corresponding direct current biasing signal, according to the almost linear region scope of LED, Flip-OFDM signal is carried out to amplitude limit, and by positive signal cut top signal by way of compensation signaling bearer upset zero-signal position corresponding to negative signal, upset negative signal is carried out to delay process, then the positive signal after processing and upset negative signal are luminous as former and later two ofdm signal driving LED, and signal is sent to light wireless channel in the mode of luminous intensity;

(5) at the receiving terminal of Flip-OFDM system, utilize photoelectricity testing part that light intensity signal is converted to the signal of telecommunication, remove direct current biasing signal and carry out analog-to-digital conversion and obtain positive signal and reception signal corresponding to upset negative signal

to y ⁺and y ^-carry out negative signal wave absorption, part reduces noise effect, and the operating process of negative signal wave absorption is as follows:

(6) will adjudicate with dynamic compensation and obtain receiving signal y={y through the signal of negative signal wave absorption _k, k=0,1 ..., N-1}, i.e. contrast with getting wherein larger is signal, and less is noise or compensating signal, if

?

for signal,

for noise, if

set noise effect thresholding γ, when

time be considered as noise, when

time be considered as compensating signal, above-mentioned judgement compensation method is specially:

(7) to signal y={y _k, k=0,1 ... N-1} carries out fast Fourier transform, obtains getting its front N/2 frequency-region signal value as receiving signal Y={Y (0), Y (1) after frequency-region signal,, Y (N/2-1) }, Y is divided into the subsequence Y of a plurality of non-overlapping copies according to the partitioning scheme of step (2) _v, v=1 wherein, 2 ..., V, and respectively divided by the weight coefficient b of the negative signal amplitude minimum that makes composite signal of choosing in step (3) _v, then carry out sequence merging, this collating sequence is the frequency domain data signal that receiving terminal receives; Again this frequency domain data signal is separated to mediation parallel-serial conversion, finally obtained the bit stream of Flip-OFDM system output.

2. dynamic compensation method as claimed in claim 1, it is characterized in that: data vector X is divided in the method for a plurality of subsequences in described step (2), partitioning scheme adopts adjacent cutting apart, and wherein the value of V is determined by the required reducing amount reaching of peak-to-average force ratio of Flip-OFDM system signal.

3. dynamic compensation method as claimed in claim 2, is characterized in that: the value of described V is 4 or 8.

4. dynamic compensation method as claimed any one in claims 1 to 3, it is characterized in that: described in step (4) Flip-OFDM signal is carried out to amplitude limit time, its envelope peak is limited within the scope of almost linear region, and wherein the dynamic range in almost linear region is { x|V _l.min≤ x≤V _l.max, V _l.maxthe be as the criterion upper bound of the range of linearity, V _l.minthe lower bound of the range of linearity is as the criterion.

5. dynamic compensation method as claimed in claim 4, it is characterized in that: described in step (4), on Flip-OFDM signal, add direct current biasing signal, carry out amplitude limit, and the concrete steps that compensating signal is carried on upset negative signal are: the direct current biasing signal V that chooses interpolation<sub TranNum="351">bias</sub>meet V<sub TranNum="352">bias</sub>=V<sub TranNum="353">l.min</sub>thereby the signal adding after direct current biasing meets { x|x>=V<sub TranNum="354">l.min</sub>, now the condition of system works in almost linear region is that Flip-OFDM signal meets { x|0<x≤V<sub TranNum="355">max</sub>, V wherein<sub TranNum="356">max</sub>meet V<sub TranNum="357">max</sub>=V<sub TranNum="358">l.max</sub>-V<sub TranNum="359">bais</sub>, the Flip-OFDM positive signal after amplitude limit<img TranNum="360" file="FDA0000388542170000031.GIF" he="73" id="ifm0022" img-content="drawing" img-format="GIF" inline="no" orientation="portrait" wi="46"/>be expressed as:

Compensating signal χ _kbe expressed as:

Upset negative signal be expressed as:

。

6. the dynamic compensation method as described in claim 1 or 2 or 5, is characterized in that: for the thresholding γ described in step (6), and when SNR<4.5dB, γ=∞, when SNR>=4.5dB, γ is expressed as:

Wherein (a, b, m, n)=(0.9336,0.03341,0.4875,0.3982), σ _zfor noise criteria poor.

Images