阅读说明：本技术 一种用于无线光通信系统的幅相ofdm调制方法 (Amplitude-phase OFDM modulation method for wireless optical communication system ) 是由 廉杰 王彤 廉殿斌 高妍 于 2021-08-14 设计创作，主要内容包括：本发明公开了一种用于无线光通信系统的幅相OFDM调制方法,称为幅度相位光学OFDM(MPO-OFDM)。在这种方法中,不需要OFDM调制器的输入端的Hermitian对称数据来生成实信号。取而代之的是,常规复数值OFDM信号的幅度和相位分量来连续发送。MPO-OFDM对所发送的相位信息进行量化,并使用M阶脉冲幅度调制(M-PAM)进行编码发送。使用M-PAM,可以有效减小加性噪声对相位分量的影响。本发明在此基础上优化了控制发射信号幅度的调制系数,从而使得幅度分量的信噪比(SNR)最大化。与DCO,ACO,U-OFDM和基于极性的OFDM相比,本发明所提出的MPO-OFDM可以提供更好的误码率(BER)性能。(The invention discloses an amplitude-phase OFDM modulation method for a wireless optical communication system, which is called amplitude-phase optical OFDM (MPO-OFDM). In this approach, Hermitian symmetric data at the input of the OFDM modulator is not required to generate a real signal. Instead, the amplitude and phase components of a conventional complex-valued OFDM signal are transmitted continuously. MPO-OFDM quantizes the transmitted phase information and encodes the transmission using M-order pulse amplitude modulation (M-PAM). By using M-PAM, the influence of additive noise on the phase component can be effectively reduced. The invention optimizes the modulation coefficient for controlling the amplitude of the transmitting signal on the basis of the amplitude component, thereby maximizing the signal-to-noise ratio (SNR) of the amplitude component. Compared with DCO, ACO, U-OFDM and polarity-based OFDM, MPO-OFDM provided by the invention can provide better Bit Error Rate (BER) performance.)

1. An amplitude-phase OFDM modulation method for a wireless optical communication system, comprising the steps of:

step 1: designing a signal transmitter;

step 1-1: suppose X_iIs M-QAM data modulated on the ith subcarrier; without using Hermitian symmetry, X is_iThe k-th component after the inverse fast fourier transform IFFT is represented as:

wherein N represents the number of carriers;

step 1-2: suppose X_i，Are independently and identically distributed and satisfy | X_iA random variable with | less than or equal to 1; x [ k ] is]After the parallel-to-serial converter, the mth time-domain sample of the OFDM symbol is represented as:

when N is present>At 64 times, x_s[m]The real part and the imaginary part of the equation are modeled as independent Gaussian distribution variables, the mean value is zero, and the variance is

Step 1-3: then x is put_s[m]After Cartesian to polar converter, x_s[m]Is represented as | x_s[m]I and < x_s[m]；x_s[m]The amplitude and phase information amplitude and phase components need to be transmitted separately, so they are real and non-negative;

step 2: receiving a signal and receiver design;

step 2-1: at the receiver, if channel loss is ignored, the SNR of the received M-PAM phase information is approximated toWherein N is₀And R_sRespectively representing the noise spectral density and the transmitted OFDM symbol rate; p_maxRepresents the maximum emission power of the employed light source; therefore, the bit error rate BER of the binary number obtained by demodulating the M-PAM phase information is approximately calculated as:

where erfc (·) is a complementary error function defined asM_pThe modulation order when M-PAM is used for modulating phase information is shown;

after reconstructing the phase information, equivalent additive noiseHas a variance of approximately

Wherein, L represents the binary digit number used when quantizing the phase information, Δ p represents the quantization precision when quantizing the phase information, and j and L represent the indication algebra in the calculation process respectively;

step 2-2: after sampling the input signal and using a polar to cartesian converter, the nth sample in one OFDM symbol is represented as:

whereinAndrespectively representing equivalent additive noise and quantization noise loaded on the phase information;andhas a variance of σ² _pAnd σ² _qu＝(△p)²12; n represents the nth sample, β represents the modulation index of the signal, and this parameter is used to adjust the amplitude of the signal; w is a_T[n]The nth sample, representing the noise added to the amplitude component, contains two parts, denoted w_T[n]＝w_m[n]+w_c[n]Wherein w is_m[n]And w_c[n]Additive noise and truncation noise due to peak power limitation representing the amplitude part, respectively; w is a_m[n]And w_c[n]Can be modeled as a Gaussian random variable with a mean value of 0, w_m[n]Has a variance of δ² _m＝N₀R_s；

Based on the Bussgang theorem, the power loss rate due to peak power clipping is expressed as α, which is estimated by equation (5):

wherein the function Var (x) represents the variance of x;to model the nonlinear response of the transmitter, it is expressed as:

the probability density function of β | x [ k ] | is derived as:

wherein sigma² _xRepresents x [ k ]]，The variance of (a);

therefore, the variance of the signal truncation noise due to the finite emission power of the light source is calculated as:

step 2-3: after the r [ n ] signal is subjected to FFT, the M-QAM data of the q sub-carrier is expressed as:

whereinG_qConsidered as a random variable, its variance is calculated as:

therefore, the SNR of the reconstructed M-QAM signal on a single subcarrier at the receiving end is represented as:

wherein E {. denotes Ideal, E { | λ (l) & gt²The calculation formula is:

wherein

Equation (12) is substituted for equation (11), and the BER of the data is approximated as:

wherein M is_dRepresenting the modulation order of the QAM data.

Technical Field

The invention belongs to the technical field of optical communication, and particularly relates to an amplitude-phase OFDM modulation method.

Background

In recent years, intensity modulation and direct detection (IM/DD) communication systems such as free space optical communication, visible light communication, and infrared communication have attracted much attention. IM/DD optical systems have many advantages over rate-frequency (RF) communications, including high security, high data rates, low power consumption, and no spectral modulation, using Light Emitting Diodes (LEDs) or Laser Diodes (LDs) as transmitters. However, since signals transmitted and received in the IM/DD procedure must be non-negative and real, the modulation scheme needs to be designed when signal modulation is performed in such a communication system.

Since a non-negative real signal needs to be transmitted when the IM/DD system is used, the conventional OFDM cannot be directly applied to the IM/DD system. Prior work has proposed some OFDM improvements for wireless optical communication systems that use modulation signals with Hermitian symmetry as a basis to generate real signals directly. Meanwhile, in order to convert a bipolar real OFDM signal into a unipolar one, direct current biased optical OFDM (DCO-OFDM) is often employed. In DCO-OFDM, the added DC offset is typically half the peak transmit power. However, light sources with peak power constraints introduce more signal truncation distortion due to the DC offset required to add additional transmit power and due to the offset of the DC signal. Further, some researchers have proposed asymmetrically amplitude limited optical OFDM (ACO-OFDM), which modulates only subcarriers of odd frequencies to generate a unipolar OFDM signal, however, this algorithm has a low bandwidth utilization efficiency. In conjunction with ACO-OFDM and pulse position modulation, some researchers have proposed a technique called fractional reverse polarity optical OFDM that allows for dimming control in visible light communication applications. Unipolar OFDM (U-OFDM), also known as Flip-OFDM, is a new OFDM modulation algorithm recently proposed that avoids increasing dc bias by generating a unipolar signal by continuously transmitting the positive and negative portions of a bipolar original signal in two time slots. However, due to the use of additional signal slots, U-OFDM requires twice the bandwidth as DCO-OFDM when transmitting the same OFDM symbol rate.

The existing algorithm for generating the unipolar real OFDM signal often needs to modulate a signal with Hermitian symmetry, so that the spectrum utilization rate is reduced. And on the basis of constructing a modulation signal with Hermitian symmetry by adopting DCO-, ACO-and U-OFDM algorithms, direct current bias is respectively increased, fewer subcarriers are adopted, transmission time slots are increased, and the transmission of bipolar OFDM signals is avoided. Resulting in further reduction of power usage and spectrum utilization of the system.

Disclosure of Invention

To overcome the deficiencies of the prior art, the present invention provides an amplitude-phase OFDM modulation method, referred to as amplitude-phase optical OFDM (MPO-OFDM), for wireless optical communication systems. In this approach, Hermitian symmetric data at the input of the OFDM modulator is not required to generate a real signal. Instead, the amplitude and phase components of a conventional complex-valued OFDM signal are transmitted continuously. MPO-OFDM quantizes the transmitted phase information and encodes the transmission using M-order pulse amplitude modulation (M-PAM). By using M-PAM, the influence of additive noise on the phase component can be effectively reduced. The invention optimizes the modulation coefficient for controlling the amplitude of the transmitting signal on the basis of the amplitude component, thereby maximizing the signal-to-noise ratio (SNR) of the amplitude component. Compared with DCO, ACO, U-OFDM and polarity-based OFDM, MPO-OFDM provided by the invention can provide better Bit Error Rate (BER) performance.

The technical scheme adopted by the invention for solving the technical problem comprises the following steps:

step 1: designing a signal transmitter;

step 1-1: suppose X_iIs M-QAM data modulated on the ith subcarrier; without using Hermitian symmetry, X is_iThe k-th component after the inverse fast fourier transform IFFT is represented as:

wherein N represents the number of carriers;

step 1-2: suppose thatAre independently and identically distributed and satisfy | X_iA random variable with | less than or equal to 1; x [ k ] is]After passing through the parallel-to-serial converter,the mth time-domain sample of the OFDM symbol is represented as:

when N is present>At 64 times, x_s[m]The real part and the imaginary part of the equation are modeled as independent Gaussian distribution variables, the mean value is zero, and the variance is

Step 1-3: then x is put_s[m]After Cartesian to polar converter, x_s[m]Is represented as | x_s[m]I and < x_s[m]；x_s[m]The amplitude and phase information amplitude and phase components need to be transmitted separately, so they are real and non-negative;

step 2: receiving a signal and receiver design;

step 2-1: at the receiver, if channel loss is ignored, the SNR of the received M-PAM phase information is approximated toWherein N is₀And R_sRespectively representing the noise spectral density and the transmitted OFDM symbol rate; p_maxRepresents the maximum emission power of the employed light source; therefore, the bit error rate BER of the binary number obtained by demodulating the M-PAM phase information is approximately calculated as:

where erfc (·) is a complementary error function defined asM_pThe modulation order when M-PAM is used for modulating phase information is shown;

after reconstructing the phase information, equivalent additive noiseHas a variance of approximately

Wherein, L represents the binary digit number used when quantizing the phase information, Δ p represents the quantization precision when quantizing the phase information, and j and L represent the indication algebra in the calculation process respectively;

step 2-2: after sampling the input signal and using a polar to cartesian converter, the nth sample in one OFDM symbol is represented as:

whereinAndrespectively representing equivalent additive noise and quantization noise loaded on the phase information;andhas a variance of σ² _pAnd σ² _qu＝(△p)²12; n represents the nth sample, β represents the modulation index of the signal, and this parameter is used to adjust the amplitude of the signal; w is a_T[n]The nth sample, representing the noise added to the amplitude component, contains two parts, denoted w_T[n]＝w_m[n]+w_c[n]Wherein w is_m[n]And w_c[n]Additive noise and truncation noise due to peak power limitation representing the amplitude part, respectively; w is a_m[n]And w_c[n]Can be modeled as a Gaussian with mean 0Random variable, w_m[n]Has a variance of δ² _m＝N₀R_s；

Based on the Bussgang theorem, the power loss rate due to peak power clipping is expressed as α, which is estimated by equation (5):

wherein the function Var (x) represents the variance of x;to model the nonlinear response of the transmitter, it is expressed as:

the probability density function of β | x [ k ] | is derived as:

wherein sigma² _xRepresentsThe variance of (a);

therefore, the variance of the signal truncation noise due to the finite emission power of the light source is calculated as:

step 2-3: after the r [ n ] signal is subjected to FFT, the M-QAM data of the q sub-carrier is expressed as:

whereinG_qConsidered as a random variable, its variance is calculated as:

therefore, the SNR of the reconstructed M-QAM signal on a single subcarrier at the receiving end is represented as:

wherein E {. denotes Ideal, E { | λ (l) & gt²The calculation formula is:

wherein

Equation (12) is substituted for equation (11), and the BER of the data is approximated as:

wherein M is_dRepresenting the modulation order of the QAM data.

The invention has the following beneficial effects:

the MPO-OFDM system for visible light communication provided by the invention adopts a mode of respectively sending the phase and amplitude information of a complex signal, thereby avoiding the problem that the prior wireless optical OFDM system has low spectrum utilization rate because a modulation signal with Hermitian symmetry is required to be used.

Drawings

FIG. 1 is a schematic block diagram of a transmitter in MPO-OFDM system according to the present invention.

FIG. 2 is a diagram of an MPO-OFDM signal according to an embodiment of the present invention; wherein (a) is M-PAM transmission phase for M-Q, (b) is M-PAM transmission phase for M-Q/2, and (c) is 2-PAM for transmission phase.

FIG. 3 is a schematic block diagram of a receiver in the MPO-OFDM system of the present invention.

Fig. 4 is a diagram of phase quantization using M-PAM encoding according to an embodiment of the present invention.

FIG. 5 is a comparison of BER performance of DCO-, ACO-, U-, MPO-based and monopole-based OFDM using different received peak powers according to an embodiment of the present invention.

Detailed Description

The invention is further illustrated with reference to the following figures and examples.

An amplitude-phase OFDM modulation method for a wireless optical communication system, comprising the steps of:

step 1: signal transmitter design, as shown in fig. 1;

step 1-1: suppose X_iIs M-QAM data modulated on the ith subcarrier; without using Hermitian symmetry, X is_iThe k-th component after the inverse fast fourier transform IFFT is represented as:

wherein N represents the number of carriers;

step 1-2: suppose thatAre independently and identically distributed and satisfy | X_iA random variable with | less than or equal to 1; x [ k ] is]After the parallel-to-serial converter, the mth time-domain sample of the OFDM symbol is represented as:

here a complex value, when N>At 64 times, x_s[m]Real and imaginary part ofThe part is modeled as an independent Gaussian distribution variable, the mean value is zero, and the variance is

Step 1-3: then x is put_s[m]After Cartesian to polar converter, x_s[m]Is represented as | x_s[m]I and < x_s[m]；x_s[m]The amplitude and phase information amplitude and phase components need to be transmitted separately, so they are real and non-negative;

fig. 2 shows an example of representing the phase using different constellation sizes M-PAM. In the figure, the first frame transmits an amplitude component | x_s[m]I, m-0, 1, …, N-1. When its value amplitude exceeds the peak power Pmax, it is truncated, and the other signal frames are used to transmit phase information. Since the phase information can be modeled as a random number between 0 and 2 pi, the phase information is not subject to truncation distortion caused by the light source emission power limitation. The modulation order of M-PAM for the transmission phase can be optimally chosen considering the bandwidth requirement and the influence of noise on the phase.

Step 2: receiving a signal and receiver design;

step 2-1: at the receiver, if channel loss is ignored, the SNR of the received M-PAM phase information is approximated toWherein N is₀And R_sRespectively representing the noise spectral density and the transmitted OFDM symbol rate; p_maxRepresents the maximum emission power of the employed light source; therefore, the bit error rate BER of the binary number obtained by demodulating the M-PAM phase information is approximately calculated as:

where erfc (·) is a complementary error function defined asM_pThe modulation order when M-PAM is used for modulating phase information is shown;

after reconstructing the phase information, equivalent additive noiseHas a variance of approximately

Wherein, L represents the binary digit number used when quantizing the phase information, Δ p represents the quantization precision when quantizing the phase information, and j and L represent the indication algebra in the calculation process respectively;

step 2-2: after sampling the input signal and using a polar to cartesian converter, the nth sample in one OFDM symbol is represented as:

whereinAndrespectively representing equivalent additive noise and quantization noise loaded on the phase information;andhas a variance of σ² _pAnd σ² _qu＝(△p)²12; n represents the nth sample, β represents the modulation index of the signal, and this parameter is used to adjust the amplitude of the signal; w is a_T[n]The nth sample, representing the noise added to the amplitude component, contains two parts, denoted w_T[n]＝w_m[n]+w_c[n]Wherein w is_m[n]And w_c[n]Additive noise and truncation noise due to peak power limitation representing the amplitude part, respectively; w is a_m[n]And w_c[n]Can be modeled as a Gaussian random variable with a mean value of 0, w_m[n]Has a variance of δ² _m＝N₀R_s；

Based on the Bussgang theorem, the power loss rate due to peak power clipping is expressed as α, which is estimated by equation (5):

wherein the function Var (x) represents the variance of x;to model the nonlinear response of the transmitter, it is expressed as:

the probability density function of β | x [ k ] | is derived as:

wherein sigma² _xRepresentsThe variance of (a);

therefore, the variance of the signal truncation noise clipping noise due to the limited transmission power of the light source is calculated as:

step 2-3: after the r [ n ] signal is subjected to FFT, the M-QAM data of the q sub-carrier is expressed as:

whereinG_qConsidered as a random variable, its variance is calculated as:

therefore, the SNR of the reconstructed M-QAM signal on a single subcarrier at the receiving end is represented as:

wherein E {. denotes Ideal, E { | λ (l) & gt²The calculation formula is:

wherein

Equation (12) is substituted for equation (11), and the BER of the data is approximated as:

wherein M is_dRepresenting the modulation order of the QAM data.

The specific embodiment is as follows:

FIG. 3 is a block diagram of an MPO-OFDM transmitter according to the present invention, in which a modulation factor beta is used to control | x_s[m]The magnitude of | is given. Due to exceeding of power limitThe signal is truncated to introduce truncation errors and too small a signal amplitude does not provide sufficient signal-to-noise ratio, so that the beta can be controlled to balance clipping distortion and signal power. In the present invention, β will be controlled to optimize SNR. By using M-PAM, phase information x is subjected to_s[m]Quantization, coding and transmission are performed. The quantization level and modulation order of M-PAM are two design parameters to be optimized in MPO-OFDM, respectively.

Fig. 4 shows the algorithm flow for quantizing and encoding the phase information of MPO-OFDM. In the figure, the phase information is quantized using a quantization level Q and encoded as an L-bit binary number with a correspondence of Q2^L. Thus, the quantization resolution isAnd the M-PAM is used for transmitting quantized L-phase information after quantization.

Fig. 5 compares BER performance of the prior OFDM modulation algorithm with the proposed MPO-OFDM at different peak optical powers. In this result, the modulation factor β is optimized. In general, the MPO-OFDM provided by the invention can provide better BER performance than other testing technologies. MPO-OFDM with 8-PAM encoded phase information has a power advantage over other OFDM modulation techniques of over 2dB for the same bit rate. When the BER performance is similar, MPO-OFDM can achieve a higher transmission bit rate.