Peak-to-average power ratio suppression method applied to high-energy-efficiency visible light FBMC system

Abstract

The invention discloses a peak-to-average power ratio suppression method applied to a high-energy-efficiency visible light FBMC system, and the method comprises the steps: firstly carrying out the grouping and pulse amplitude modulation mapping of user binary data or a pseudo-random binary sequence, then multiplying a real number precoding matrix, and carrying out the phase shift, conjugate symmetry and N-point inverse fast Fourier transform, thereby achieving the suppression of the peak-to-average power ratio. The upper branch IFFT output signal is subjected to asymmetric amplitude limiting processing, then is subjected to multi-phase network processing with the lower branch IFFT output signal, and then is multiplied by respective power factors for power adjustment, and the lower branch signal is subjected to N / 2 data delay and then is added with the upper branch signal subjected to power adjustment to generate a real baseband FBMC signal; and finally, after digital-to-analog conversion and electro-optical conversion, a high-energy-efficiency optical FBMC signal is obtained. The real number precoding technology provided by the invention has lower calculation complexity, and can reduce the peak-to-average power ratio of the signal and the signal-to-noise ratio of the balanced data subcarrier, reduce the implementation complexity of system hardware and improve the system performance.

Description

Technical Field

[0001] The present invention relates to the field of optical communication technology, and in particular to a peak-to-average power ratio suppression method applied to a high-energy-efficiency visible light FBMC system. Background Art

[0002] Visible light communication uses visible light in the wavelength range of 380nm to 780nm as an information carrier. It has the advantages of large bandwidth, strong anti-interference ability, and no electromagnetic interference. It has broad application prospects in aerospace, deep-sea exploration, medical communications and other fields. By adding data transmission auxiliary functions to LED lighting facilities, a visible light wireless communication system can be built to achieve high-speed wireless information transmission. However, the traditional commercial LED modulation bandwidth is narrow, and the transmission rate of the system is therefore limited. In order to make full use of the limited bandwidth and increase the transmission rate of the system, high spectral efficiency multi-carrier modulation formats have become a research hotspot.

[0003] Among various physical layer modulation technologies, orthogonal frequency-division multiplexing (OFDM) and filter bank multi-carrier (FBMC) modulation have attracted widespread attention due to their high spectrum efficiency and strong anti-multipath interference capabilities. Compared with OFDM, FBMC technology uses filters with excellent time-frequency focusing characteristics to not only effectively reduce inter-subcarrier interference and adjacent channel leakage, but also improve spectrum efficiency by saving cyclic prefixes.

[0004] There are mainly two forms of visible light FBMC systems, namely direct current biased optical FBMC (DCO-FBMC) and asymmetrically clipped optical FBMC (ACO-FBMC). The DCO-FBMC system has high spectral efficiency but low energy efficiency; while the ACO-FBMC system has high energy efficiency but low spectral efficiency. In order to balance spectral efficiency and improve the energy efficiency of the system, a high-energy-efficiency visible light FBMC scheme based on asymmetric clipping is proposed. For details, see the literature: S.Niu, P.Wang, S.Chi, Z.Liu, W.Pang and L.Guo, Enhanced Optical OFDM / OQAM for VisibleLight Communication Systems, IEEE Wireless Communications Letters, vol.10, no.3, pp.614-618, 2021. However, the system still suffers from the problem of high peak-to-average power ratio (PAPR) and uneven distribution of signal-to-noise ratio (SNR) on data subcarriers due to the limited LED modulation bandwidth, which affects the transmission performance of the high-efficiency visible light FBMC system.

[0005] Precoding technology can effectively suppress the PAPR of FBMC signals and equalize the SNR on data subcarriers, which can significantly improve the bit error rate performance of the receiver. However, traditional precoding technology usually uses a complex precoding matrix, which has high computational complexity. In addition, due to the inconsistency in the number of digital subcarriers in the upper and lower branches of the high-efficiency visible light FBMC system, traditional complex precoding or low-complexity real precoding technology cannot be directly applied to the system. Therefore, it is of certain application value to explore the low-complexity real precoding technology applied to the high-efficiency visible light FBMC system to reduce the symbol PAPR and equalize the data subcarrier SNR, reduce the complexity of system hardware implementation and improve system performance. Summary of the invention

[0006] In order to solve the above technical problems, the present invention provides a peak-to-average power ratio suppression method for a high energy-efficiency visible light FBMC system with a simple algorithm.

[0007] The technical solution of the present invention to solve the above technical problem is: a peak-to-average power ratio suppression method applied to a high energy efficiency visible light FBMC system, comprising the following steps:

[0008] S1: In the FBMC baseband transmitter, the user binary data or pseudo-random binary sequence is divided into two groups, the upper branch and the lower branch, and mapped into corresponding frequency domain symbols according to the pulse amplitude modulation (PAM) method;

[0009] S2: After further grouping the mapped frequency domain symbols according to the number of valid data subcarriers in the upper and lower branches, they are multiplied by the predefined M-order real precoding matrix corresponding to the upper branch and the 2M-order real precoding matrix corresponding to the lower branch to complete the real precoding operation;

[0010] S3: Phase shift, conjugate symmetry and N-point inverse fast Fourier transform IFFT are performed on the symbols pre-coded by the upper and lower branches respectively; the upper branch IFFT output signal is subjected to asymmetric limiting processing, and then subjected to multi-phase network processing together with the lower branch IFFT output signal;

[0011] S4: The upper and lower branch signals output by the multiphase network processing are multiplied by their respective power factors to achieve power adjustment. The lower branch signal after power adjustment is added to the upper branch signal after power adjustment after N / 2 data delays to generate a real baseband FBMC signal, and then a high-efficiency visible light FBMC signal is obtained after digital-to-analog conversion and electro-optical conversion.

[0012] In the peak-to-average power ratio suppression method for a high-efficiency visible light FBMC system, in step S1, in the FBMC baseband transmitter, the serial user binary data or pseudo-random binary sequence is divided into two groups, an upper branch and a lower branch, and mapped into frequency domain symbols by pulse amplitude modulation. and and are the mth frequency domain symbols in the upper and lower branches respectively.

[0013] The peak-to-average power ratio suppression method applied to the high-efficiency visible light FBMC system, in step S2, the number of valid data subcarriers in the upper branch is M, the number of valid data subcarriers in the lower branch is 2M, and the column vector corresponding to each group of M frequency domain symbols after the upper branch is grouped is X u , after the lower branch is grouped, the column vector corresponding to each group of 2M frequency domain symbols is X l :

[0014]

[0015] Where T is the matrix transpose operation;

[0016] X u The M-order real number precoding matrix P corresponding to the predefined upper branch M×M Multiply to get the upper branch precoded symbol Y u , X lThe 2M-order real precoding matrix P corresponding to the lower branch 2M×2M Multiply to get the symbol Y of the lower branch precoding l , complete the real number precoding operation, Y u and Y l They are:

[0017]

[0018] The peak-to-average power ratio suppression method applied to the high-efficiency visible light FBMC system, in the step S2, the real number precoding technology includes Walsh-Hadamard transform WHT, discrete cosine transform DCT and Hartley transform DHT, wherein the precoding matrix used in WHT precoding is Hadamard matrix, and the element located in the rth row and cth column in the W-order precoding matrix used in DCT and DHT precoding is and Respectively expressed as:

[0019]

[0020] The peak-to-average power ratio suppression method applied to the high-efficiency visible light FBMC system, in step S3, phase shifts are performed on the pre-coded symbols on the upper and lower branches respectively to meet the real domain orthogonality requirements of the FBMC modulation format, and the specific operations are as follows:

[0021]

[0022] In the formula χ and η are the phase shift factor vectors of the upper and lower branches respectively, Z u With Z l are the matrix signals after phase shift of the upper and lower branches, Represents the Hadamard product of two matrices.

[0023] The peak-to-average power ratio suppression method for high-efficiency visible light FBMC system is applied to the step S3. After the pre-coded symbols on the upper and lower branches are phase-shifted, the upper and lower branches respectively perform conjugate symmetry and N-point inverse fast Fourier transform processing on the phase-shifted data. The k-th group of upper branch IFFT output signal matrix The kth group of lower branch IFFT output signal matrix for:

[0024]

[0025] In the formula For the matrix For every n elements in For the matrix The nth element in the array, where n ranges from 0 to N-1.

[0026] The peak-to-average power ratio suppression method for high energy efficiency visible light FBMC system is applied to the step S3, wherein the upper branch IFFT output signal is subjected to asymmetric limiting processing, i.e., AC processing to improve the energy efficiency of the system, and the upper branch k-th group n-th point discrete signal after AC Written as:

[0027]

[0028] The peak-to-average power ratio suppression method for the high-efficiency visible light FBMC system is applied to the step S3. After the AC upper branch signal is processed by the multiphase network, the dth output discrete signal is: After the IFFT output signal of the lower branch is processed by the multiphase network, the dth output discrete signal is and They are:

[0029]

[0030] Where h(n) is the impulse response sequence of the prototype filter, and d%N represents the remainder of d divided by N.

[0031] In the peak-to-average power ratio suppression method for a high-efficiency visible light FBMC system, in step S4, the power factors α and β are used to adjust the power of the upper and lower branch multiphase network output signals respectively to optimize the system performance; the adjusted lower branch signal is added to the upper branch signal after power adjustment after N / 2 data delays to generate a real baseband FBMC signal s d for:

[0032]

[0033] Define the power ratio PR as 20log 10 (α / β), unit: dB. The real baseband FBMC signal is converted by digital-to-analog conversion and electro-optical conversion to obtain a high-efficiency visible light FBMC signal. After being transmitted through the visible light channel, it is converted by photoelectric conversion and analog-to-digital conversion at the receiving end, and finally the receiver digital signal processing algorithm is used to recover the transmitted binary data.

[0034] The beneficial effect of the present invention is that: the present invention first groups user binary data or pseudo-random binary sequence in the FBMC baseband transmitter and performs pulse amplitude modulation mapping, then multiplies it with a real precoding matrix, and then undergoes phase shift, conjugate symmetry and N-point inverse fast Fourier transform (IFFT). The upper branch IFFT output signal is processed by asymmetric limiting (AC) and the lower branch IFFT output signal is processed by a polyphase network (PPN), and then multiplied by their respective power factors for power adjustment. The power-adjusted lower branch signal is delayed by N / 2 data and then added to the power-adjusted upper branch signal to generate a real baseband FBMC signal. Finally, a high-efficiency optical FBMC signal is obtained after digital-to-analog conversion and electro-optical conversion. Compared with the traditional precoding technology using a complex precoding matrix (i.e., the matrix elements are complex numbers), the real precoding technology has lower computational complexity, can reduce symbol PAPR and equalized data subcarrier SNR, reduce system hardware implementation complexity and improve system performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a flowchart of the overall process of the present invention.

[0036] Figure 2 4 is a comparison diagram of PAPR complementary cumulative distribution function curves in an embodiment of the present invention.

[0037] Figure 3 4 is a signal-to-noise ratio comparison diagram in an embodiment of the present invention.

[0038] Figure 4 4 is a comparison diagram of bit error rates in an embodiment of the present invention. DETAILED DESCRIPTION

[0039] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0040] like Figure 1 As shown, a peak-to-average power ratio suppression method applied to a high energy efficiency visible light FBMC system comprises the following steps:

[0041] S1: In the FBMC baseband transmitter, the user binary data or pseudo-random binary sequence is divided into two groups, upper and lower branches, and mapped into corresponding frequency domain symbols according to the pulse amplitude modulation (PAM) method.

[0042] In step S1, in the FBMC baseband transmitter, the serial user binary data or pseudo-random binary sequence is divided into two groups, upper and lower branches, and mapped into frequency domain symbols by pulse amplitude modulation. and and are the mth frequency domain symbols in the upper and lower branches respectively.

[0043] S2: After further grouping the mapped frequency domain symbols according to the number of valid data subcarriers in the upper and lower branches, they are multiplied with the predefined M-order real precoding matrix corresponding to the upper branch and the 2M-order real precoding matrix corresponding to the lower branch to complete the real precoding operation.

[0044] In step S2, the number of valid data subcarriers in the upper branch is M, the number of valid data subcarriers in the lower branch is 2M, and the column vector corresponding to each group of M frequency domain symbols after the upper branch is grouped is X u , after the lower branch is grouped, the column vector corresponding to each group of 2M frequency domain symbols is X l :

[0045]

[0046] Where T is the matrix transpose operation;

[0047] X u The M-order real number precoding matrix P corresponding to the predefined upper branch M×M Multiply to get the upper branch precoded symbol Y u , X l The 2M-order real precoding matrix P corresponding to the lower branch 2M×2M Multiply to get the symbol Y of the lower branch precoding l , complete the real number precoding operation, Y u and Y l They are:

[0048]

[0049] Real number precoding technologies include Walsh-Hadamard transform WHT, discrete cosine transform DCT and Hartley transform DHT. The precoding matrix used by WHT precoding is the Hadamard matrix, while the element located in the rth row and cth column of the W-order precoding matrix used by DCT and DHT precoding is and Respectively expressed as:

[0050]

[0051] S3: Phase shift, conjugate symmetry and N-point inverse fast Fourier transform IFFT are performed on the symbols precoded by the upper and lower branches respectively; after asymmetric limiting processing, the upper branch IFFT output signal and the lower branch IFFT output signal are both subjected to multi-phase network processing.

[0052] In step S3, the precoded symbols on the upper and lower branches are phase shifted to meet the real domain orthogonality requirement of the FBMC modulation format. The specific operation is as follows:

[0053]

[0054] In the formula χ and η are the phase shift factor vectors of the upper and lower branches respectively, Z u With Z l are the matrix signals after phase shift of the upper and lower branches, Represents the Hadamard product of two matrices.

[0055] After the pre-coded symbols on the upper and lower branches are phase-shifted, the upper and lower branches perform conjugate symmetry and N-point inverse fast Fourier transform processing on the phase-shifted data respectively. The k-th group of upper branch IFFT output signal matrix The kth group of lower branch IFFT output signal matrix for:

[0056]

[0057] In the formula For the matrix For every n elements in For the matrix The nth element in the array, where n ranges from 0 to N-1.

[0058] The upper branch IFFT output signal is processed by asymmetric limiting, that is, AC processing to improve the energy efficiency of the system. The upper branch k-th group n-th discrete signal after AC Written as:

[0059]

[0060] After the AC upper branch signal is processed by the multiphase network, the dth output discrete signal is After the IFFT output signal of the lower branch is processed by the multiphase network, the dth output discrete signal is and They are:

[0061]

[0062] Where h(n) is the impulse response sequence of the prototype filter, and d%N represents the remainder of d divided by N.

[0063] S4: The upper and lower branch signals output by the multiphase network processing are multiplied by their respective power factors to achieve power adjustment. The lower branch signal after power adjustment is added to the upper branch signal after power adjustment after N / 2 data delays to generate a real baseband FBMC signal, and then a high-efficiency visible light FBMC signal is obtained after digital-to-analog conversion and electro-optical conversion.

[0064] In step S4, the power factors α and β are used to adjust the power of the upper and lower branch multiphase network output signals respectively to optimize the system performance; the adjusted lower branch signal is added to the upper branch signal after power adjustment after N / 2 data delays to generate a real baseband FBMC signal s d for:

[0065]

[0066] Define the power ratio PR as 20log 10 (α / β), unit: dB. The real baseband FBMC signal is converted by digital-to-analog conversion and electro-optical conversion to obtain a high-efficiency visible light FBMC signal. After being transmitted through the visible light channel, it is converted by photoelectric conversion and analog-to-digital conversion at the receiving end, and finally the receiver digital signal processing algorithm is used to recover the transmitted binary data.

[0067] Compared with the complex precoding matrix of traditional precoding technology (i.e., the matrix elements are complex numbers), real precoding technology has lower computational complexity, and can reduce symbol PAPR and equalized data subcarrier SNR, reduce system hardware implementation complexity and improve system performance.

[0068] To verify the effectiveness of the present invention, the digital signal processing algorithm of the FBMC transmitter end is implemented based on software programming, and numerical simulation analysis is performed, where the FBMC signal parameters are as follows: the number of IFFT points is 256, the number of valid data subcarriers in the upper and lower branches is 48 and 96 respectively, the other subcarriers are set to zero, and the PPN uses the PHYDYAS prototype filter.

[0069] Figure 2 The complementary cumulative distribution function curve of the signal peak-to-average power ratio is obtained by statistically analyzing 100,000 randomly generated FBMC symbols after 4 times oversampling. -4 When using precoding technology, the peak-to-average power ratio of the signal can be reduced by up to 4dB compared to not using precoding technology.

[0070] To further evaluate the subcarrier SNR equalization capability of the present invention, a first-order low-pass filter is used to simulate a bandwidth-limited LED model (-3dB bandwidth is 50MHz), and signal transmission only considers the additive white Gaussian noise channel, the signal bandwidth is 375MHz, and the SNR is set to 25dB. Figure 3 Figure 1 is the SNR distribution diagram of different data subcarriers estimated by the receiving end after a frame of FBMC signal (including 1000 FBMC symbols) is transmitted through the channel when the PR is 4dB. Figure 3It can be seen that when precoding is not used, the two branches have a signal-to-noise ratio fluctuation of about 15dB due to the frequency selective power fading caused by the limited bandwidth of the LED. The precoding technology of the present invention can reduce the signal-to-noise ratio fluctuation to below 3dB, and obtain a better subcarrier SNR equalization effect.

[0071] Figure 4 The figure is a distribution diagram of the bit error rate calculated after a frame of FBMC signal is transmitted through the channel as the PR value of the upper and lower branch signal power ratio changes. It can be seen from this that the real number precoding technology proposed in the present invention can significantly reduce the BER. When the PR is 4dB, compared with the signal without precoding, the BER of the FBMC signal generated by the real number precoding technology is reduced by about one order of magnitude.

[0072] In summary, the present invention can not only effectively reduce the peak-to-average power ratio of the signal in the high-energy-efficiency visible light FBMC system, thereby reducing nonlinear distortion, but also balance the signal-to-noise ratio distribution on the data subcarriers and improve the bit error rate performance of the system.

Claims

1. A peak-to-average power ratio suppression method for a high energy efficiency visible light FBMC system, characterized in that: The following steps are involved: S1: In the FBMC baseband transmitter, the user binary data or pseudo-random binary sequence is divided into two groups, the upper branch and the lower branch, and mapped into corresponding frequency domain symbols according to the pulse amplitude modulation (PAM) method; S2: After further grouping the mapped frequency domain symbols according to the number of valid data subcarriers in the upper and lower branches, they are multiplied by the predefined M-order real precoding matrix corresponding to the upper branch and the 2M-order real precoding matrix corresponding to the lower branch to complete the real precoding operation; S3: Phase shift, conjugate symmetry and N-point inverse fast Fourier transform IFFT are performed on the symbols pre-coded by the upper and lower branches respectively; the upper branch IFFT output signal is subjected to asymmetric limiting processing, and then subjected to multi-phase network processing together with the lower branch IFFT output signal; S4: The upper and lower branch signals output by the multiphase network processing are multiplied by their respective power factors to achieve power adjustment. The lower branch signal after power adjustment is added to the upper branch signal after power adjustment after N / 2 data delays to generate a real baseband FBMC signal, and then a high-efficiency visible light FBMC signal is obtained after digital-to-analog conversion and electro-optical conversion.

2. The peak-to-average power ratio suppression method for a high energy efficiency visible light FBMC system according to claim 1, characterized in that: In step S1, in the FBMC baseband transmitter, the serial user binary data or pseudo-random binary sequence is divided into two groups, upper and lower branches, and mapped into frequency domain symbols by pulse amplitude modulation. and and are the mth frequency domain symbols in the upper and lower branches respectively.

3. The peak-to-average power ratio suppression method for a high energy efficiency visible light FBMC system according to claim 1, characterized in that: In step S2, the number of valid data subcarriers in the upper branch is M, the number of valid data subcarriers in the lower branch is 2M, and the column vector corresponding to each group of M frequency domain symbols after the upper branch is grouped is X u , after the lower branch is grouped, the column vector corresponding to each group of 2M frequency domain symbols is X l : Where T is the matrix transpose operation; X u The M-order real number precoding matrix P corresponding to the predefined upper branch M×M Multiply to get the upper branch precoded symbol Y u , X l The 2M-order real precoding matrix P corresponding to the lower branch 2M×2M Multiply to get the symbol Y of the lower branch precoding l , complete the real number precoding operation, Y u and Y l They are:

4. The peak-to-average power ratio suppression method for a high energy efficiency visible light FBMC system according to claim 3, characterized in that: In step S2, the real number precoding technology includes Walsh-Hadamard transform WHT, discrete cosine transform DCT and Hartley transform DHT, wherein the precoding matrix used in WHT precoding is Hadamard matrix, and the element located in the rth row and cth column in the W-order precoding matrix used in DCT and DHT precoding is and Respectively expressed as:

5. The peak-to-average power ratio suppression method for a high energy efficiency visible light FBMC system according to claim 3, characterized in that: In step S3, the precoded symbols on the upper and lower branches are phase shifted to meet the real domain orthogonality requirement of the FBMC modulation format. The specific operation is as follows: In the formula χ and η are the phase shift factor vectors of the upper and lower branches respectively, Z u With Z l are the matrix signals after phase shift of the upper and lower branches, Represents the Hadamard product of two matrices.

6. The peak-to-average power ratio suppression method for a high energy efficiency visible light FBMC system according to claim 5, characterized in that: In step S3, after the pre-coded symbols on the upper and lower branches are phase-shifted, the upper and lower branches respectively perform conjugate symmetry and N-point inverse fast Fourier transform processing on the phase-shifted data, and the k-th group of upper branch IFFT output signal matrix The kth group of lower branch IFFT output signal matrix for: In the formula For the matrix For every n elements in For the matrix The nth element in the array, where n ranges from 0 to N-1.

7. The peak-to-average power ratio suppression method for a high energy efficiency visible light FBMC system according to claim 6, characterized in that: In step S3, the upper branch IFFT output signal is subjected to asymmetric limiting processing, that is, AC processing to improve the energy efficiency of the system. The upper branch k-th group n-th point discrete signal after AC Written as:

8. The peak-to-average power ratio suppression method for a high energy efficiency visible light FBMC system according to claim 7, characterized in that: In step S3, after the AC upper branch signal is processed by a multiphase network, the dth output discrete signal is After the IFFT output signal of the lower branch is processed by the multiphase network, the dth output discrete signal is and They are: Where h(n) is the impulse response sequence of the prototype filter, and d%N represents the remainder of d divided by N.

9. The peak-to-average power ratio suppression method for a high energy efficiency visible light FBMC system according to claim 8, characterized in that: In step S4, the power factors α and β are used to adjust the power of the upper and lower branch multiphase network output signals respectively to optimize the system performance; the adjusted lower branch signal is added to the upper branch signal after power adjustment after N / 2 data delays to generate a real baseband FBMC signal s d for: Define the power ratio PR as 20log 10 (α / β), unit: dB. The real baseband FBMC signal is converted by digital-to-analog conversion and electro-optical conversion to obtain a high-efficiency visible light FBMC signal. After being transmitted through the visible light channel, it is converted by photoelectric conversion and analog-to-digital conversion at the receiving end, and finally the receiver digital signal processing algorithm is used to recover the transmitted binary data.

Images