CN110113282A - The method that probability class PAPR based on polarization code auxiliary inhibits - Google Patents

Abstract

The present invention provides a kind of method that the probability class PAPR based on polarization code auxiliary inhibits, and can solve the problem of existing ofdm system inevitably and will appear excessively high PAPR to influence system performance.The method that probability class PAPR based on polarization code auxiliary of the invention inhibits, comprising: signal to be encoded is divided into multiple pieces；The block after division is encoded using polarization code, obtains a plurality of transmission signal；The peak-to-average power ratio of transmission signal is calculated, the transmission signal for selecting a peak-to-average power ratio minimum is transmitted.

Description

Probabilistic PAPR Suppression Method Based on Polar Code Assist

technical field

The invention relates to a method for probabilistic PAPR suppression based on polar code assistance.

Background technique

With the popularity of various data services, especially video services, people's demand for access network capacity is increasing, with a transmission rate of 40/100Gbps. Orthogonal frequency division multiplexing (OFDM)-PON system has the characteristics of strong anti-dispersion ability, high spectral efficiency, flexible bandwidth allocation and strong scalability, so it has become one of the research hotspots of next-generation PON. Compared with coherent light detection, the direct detection OFDM-PON system has the advantage of low cost.

However, since OFDM is a multi-carrier technology, the superimposed transmission of each carrier will lead to a higher peak-to-average power ratio (PAPR), which will require network devices to have a wider linear range, thereby increasing the cost of the system.

Therefore, in order to enable OFDM-PON to operate stably and safely, it is imperative to suppress PAPR. Many scholars put into research to inhibit PAPR. For example, clipping method, coding method, and probability method.

Limiting technology is the simplest method to implement, but this method is a nonlinear processing of the signal, so it will cause in-band interference and out-of-band radiation, which will deteriorate the bit error rate performance of the system.

The basic idea of the coding method is to calculate the PAPR of all code groups generated by different coding methods, compare their sizes respectively, remove the code group with larger PAPR, and select the code group with lower PAPR as the allowable code group for signal transmission, thereby avoiding large signal amplitudes. The processing process of this method is a linear process, which will not cause signal distortion, but the problem of bandwidth utilization will be caused by the decrease of coding rate. In addition, to find the optimal codeword, it is necessary to calculate the PAPR of all code groups, and a large number of look-up tables for encoding and decoding are required, which has high computational complexity, and the process of encoding and decoding is complicated, and the rate of information transmission is limited by Therefore, this type of method is only suitable for use in signals with a small number of subcarriers, and is not suitable for use in signals with a large number of subcarriers.

The probabilistic technology does not suppress the amplitude peak of the signal, but improves the PAPR performance of the system by reducing the probability of the peak. Selective mapping technology (SLM) is divided into several groups of information multiplied by the phase factor, and then through IFFT, the selected peak signal is relatively low; Partial transmission sequence technology (PTS) groups the information of IFFT, and then reduces the PAPR by adjusting the phase factor. The information is multiplied by the phase factor respectively. Compared with the former two methods, this kind of method has higher complexity and requires higher accuracy of sideband information.

Referring to Figure 1, the structure of a traditional SLM algorithm, the signal is subjected to quadrature amplitude modulation (QAM), multiplied by U different phase sequences, and then the IFFT operation is performed and the PAPR is calculated separately, and a group of signals with the smallest PAPR is selected to transfer. The SLM algorithm has a good suppression effect on PAPR, but it needs to transmit additional phase information, which reduces the transmission efficiency of the system. But the SLM algorithm has very high redundant information, which reduces the universality of the use of the SLM algorithm.

The above coding method can be found that all code groups need to be calculated, so it is only suitable for some signals with a small number of carriers. For a system with a large number of carriers, the complexity will be extremely high. In addition, because the code groups used have been screened, it is Add some redundancy. Although the probabilistic algorithm does not limit its use due to the number of carriers, it will generate additional sideband signals. This additional sideband information is extremely important for recovering the signal, so it will cost a lot of additional information to transmit the signal.

The optimal situation of these methods is to produce no distortion to the signal, and cannot correct the error of the signal.

Contents of the invention

The present invention aims at the above-mentioned technical problem that exists in the prior art, provides a kind of method that can effectively reduce too high PAPR in the system, reduces the bit error rate of the system based on the probability class PAPR suppression method of polar code assistance.

The technical solution adopted to solve the technical problem of the present invention is a method of probabilistic PAPR suppression based on polar code assistance, including:

dividing the signal to be encoded into multiple blocks;

Encoding the divided blocks using polar codes to obtain multiple transmission signals;

Calculate the peak-to-average power ratio of the transmission signal, and select a transmission signal with the lowest peak-to-average power ratio for transmission.

Optionally, said dividing the signal to be encoded into multiple blocks includes:

determining whether the length of the signal to be encoded is an odd number or an even number,

If it is an even number, the signal to be encoded is divided into a first block and a second block, wherein the lengths of the first block and the second block are equal;

If it is an odd number, the signal is equally divided into a third block and a fourth block, wherein the length of the third block is one unit larger than the length of the fourth block.

Optionally, encoding the signal to be encoded by using a polar code includes:

Carry out parallel U-time encoding on the signal to be encoded, perform quadrature amplitude modulation QAM and inverse fast Fourier transform IFFT, and sequence a plurality of transmission signals, wherein U is a positive integer.

Optionally, it also includes the step of decoding the transmission signal, including:

Received signal sequence, the formula is expressed as For each y _i , compute

y _i represents the receiving end information;

This formula is the ratio of the probability that the signal appears 0 and 1;

Calculate the likelihood value of the i-th bit at the sender in order Indicates that before the receiving end encodes the sending end The evaluation sequence for is:

like then, otherwise

The present invention is based on the method for the probability class PAPR suppression of polar code assistance at least has the following advantages:

By dividing the signal into blocks and encoding the signal with a polar code, multiple different signals are generated, and the PAPR of these signals are calculated respectively, and a signal with the lowest PAPR is selected for transmission, and the PAPR of the signal is analyzed by simulation. Inhibition situation. And use a direct detection OFDM system as the experimental system to conduct experiments and analyze the information correction situation, compare the traditional OFDM signals after back-to-back and optical fiber transmission respectively, after the SLM algorithm and FPC (50%, 20%), the bit error rate FPC algorithm is 50% In the case of redundancy, compared with the SLM algorithm, the increase is 4.2dB, and the FPC redundancy is increased by 1 dB by 20%.

Description of drawings

Fig. 1 is the SLM algorithm diagram of Fig. 1;

Figure 2 is a 2-order polar code structure;

Figure 3 Schematic diagram of N-order polar code encoding;

Fig. 4 is a PC suppression algorithm diagram;

Fig. 5 is PC block coding;

6 is a flow chart of a method for probabilistic PAPR suppression based on polar code assistance;

Figure 7 is a verification structure diagram of the polar code-assisted PAPR suppression algorithm; FPC flexible block coding; QAM quadrature amplitude modulation; S/P serial/parallel conversion; IFFT inverse Fourier transform; CP cyclic redundancy; AWG arbitrary Waveform generator; EDFA erbium-doped fiber amplifier; VOA variable optical attenuator; SSMF standard fiber amplifier; PD photodetector; BER bit error rate;

Figure 8 is the cumulative distribution function and signal;

Figure 9a shows the variation of bit error rate curves under different blocking methods when the redundancy of simulation analysis is 20%;

Figure 9b shows the variation of bit error rate curves under different blocking modes when the simulation analysis redundancy is 50%;

Figure 10 is the bit error rate curve after back-to-back and optical fiber transmission of traditional signals, SLM, and FPC.

Detailed ways

In order for those skilled in the art to better understand the technical solutions of the present invention, a display unit, a display substrate, a driving method thereof, and a display device provided by the present invention will be further described in detail below in conjunction with the drawings and specific embodiments.

Example 1:

This embodiment provides a method for probabilistic PAPR suppression based on polar code assistance, including:

dividing the signal to be encoded into multiple blocks;

Encoding the divided blocks using polar codes to obtain multiple transmission signals;

Calculate the peak-to-average power ratio of the transmission signal, and select a transmission signal with the lowest peak-to-average power ratio for transmission.

Referring to Figure 6, the specific process of the above method includes:

Step1 first initializes the information, the length N of PC code encoding, the length K before encoding, the number of parallel encoding times is U, and the position of the starting point is P.

Step2 performs PC encoding for the two separate blocks, and saves the starting point position

Step3 Calculate the PAPR values of OFDM signals at different starting point positions respectively

Step4 select the minimum PAPR, and save the starting point position

Polar codes are currently the only ones that can be theoretically proven to reach the Shannon limit, and have clear and simple encoding and decoding algorithms. The core of polar code construction is through the processing of "channel polarization". At the encoding end, the coding method is used to make each sub-channel show different reliability. When the code length continues to increase, some channels will tend to have a capacity close to 1 perfect channel (no bit error), the other part of the channel tends to be a pure noise channel with a capacity close to 0, choose to directly transmit information on a channel with a capacity close to 1 to approach the channel capacity. At the decoding end, the polarized channel can be decoded by a simple successive interference cancellation (SC) method, which can obtain performance close to that of maximum natural decoding with low implementation complexity.

Optionally, for a polar code with any given code length, due to its regular butterfly basic unit transformation, it is specified that its code length N satisfies a power of 2. When the code length N tends to infinity, the channel is polarized and transformed into equivalent completely ideal noise-free virtual sub-channels and completely noisy virtual sub-channels. In this way, it is ensured that the information bits are transmitted on the channel with good channel quality, and the fixed bits known by the sending and receiving ends are transmitted on the channel with poor channel quality, so as to achieve the Shannon tolerance as much as possible. The generator matrix GN plays a crucial and decisive role in the encoding of polar codes. Referring to Figures 2 and 3, as far as a basic transformation unit is concerned, it includes a two-input modulo-plus operation. This transformation is equivalent to an input matrix undergoing a regular matrix transformation. The equivalent input is:

A transform unit represents the smallest coding unit for polar code encoding. The information block signal is generated after being divided into blocks. The presence of information chunks will precede encoding.

(u ₁ , u ₂ )(x ₁ , x ₂ )=(u ₁ , u ₂ )G ₁

Among them, G2 is a generating matrix:

For a generator matrix of order N:

in For the tensor product, RN is a flip matrix.

Optionally, referring to Fig. 5, for an initial OFDM signal, directly divide the signal into two parts Block1 and Block2. Assuming that the length of the OFDM signal is K, assuming that the signal length is an even number, then the length of Block1 is K/2, and the length of Block2 is K/2, so that the signal is equally divided into two parts. Then the initial signal can be expressed as:

S={B ₁ ,B ₂ }

Where B ₁ ={χ ₁ ,χ ₂ ...χ _K/2 }, B ₂ ={χ _K/2+1 ,χ _K/2+2 ...χ _K }

It is stipulated that the Block1 signal is divided into two parts by Block2. As shown in the figure, you can slide Bolck2 back and forth to get L/2 different signal combinations. but

S'＝{B _p1 ,B ₂ ,B _p2 }

Where B _p1 ={χ ₁ ,χ ₂ ...χ _P }, B _p2 ={χ _P+K/2+1 ,χ _P+K/2+1 ...χ _K }, B ₁ ={χ _P+1 ,χ _P+2 ...χ _P+K/2 }

The position K of the first data of Block2 in each position will be reserved, and the value range of K is 1, 2,..., L/2-1, and then PC code encoding is performed on the signal divided into two blocks , for the encoded data, it is stipulated that the encoded data of Block1 is placed in the front, and the data of Block2 is placed in the back. but

Encode(S)＝{Encode([B _p1 ,B _p2 ]),Encode(B _p2 )}

Where Encode represents polar code encoding.

Table 1 Information length after algorithm processing

algorithm type C_OFDM SLM FPC length (information) K/M K/M+K/(M+E) K/M+(N-K)/M+(log2(k/2)/E

Among them (log2(k/2)/E is much smaller than the length of K/(M+E), so in the case of ignoring this small redundancy, assuming that the two algorithms have the same redundancy, it can be expressed as:

From the above formula, it can be seen that the redundancy of the signal is closely related to the modulation method used in the secure channel transmission signal.

When the safety channel adopts E=1, E/(E+1)=0.5, when the coding redundancy of the PC code is 50%, it has the same redundancy as the traditional SLM algorithm. When E=4, E/(E+1)=0.8, when the coding redundancy of the PC code is 20%, it has the same redundancy as the traditional SLM algorithm.

For the PC code, the signal is changed by encoding, and the signal before and after encoding is regular, so the signal can be recovered by calculation without additional side information. Compared with the traditional SLM algorithm, it does not require a lot of redundant information. Compared with the encoding method, it is not necessary to calculate all the code groups, only need to find the required code groups. See Figure 4 for the basic flow of the FPC algorithm. It can be found that the traditional SLM algorithm first modulates, then multiplies the modulated signal by U different phase sequences, and then calculates the phase sequence with the smallest PAPR. The FPC algorithm first performs a parallel multiple encoding on the original signal, the number of times is also U times, then performs QAM modulation operation and IFFT, and then calculates the smallest PAPR value in these code groups, and sends it as a signal at the transmitting end.

Optionally, the decoding process includes:

suppose Indicates the sequence obtained by the receiving end, for each y _i , can be calculated Calculate the likelihood value of the i-th bit at the sender in order

Indicates that before the receiving end encodes the sending end The evaluation sequence for is:

like then, otherwise

The following analyzes the comparison of the polar code-assisted probabilistic PAPR suppression method compared with related technologies in the above scheme by means of simulation:

See Figure 7, the experimental structure of the polar code-assisted PAPR suppression algorithm. At the transmitting end of OFDM, a PRBS15 signal acts as the source information, as the initial signal, and then the PAPR suppression algorithm-related algorithm will be applied here to suppress the signal. PAPR, the modulation method adopts 16QAM, in which the carrier number of the signal is 512, and the effective data carrier is 128. After the above operations are processed offline, an electrical signal is generated by a 12.5 arbitrary waveform generator (AWG), and passed through Mach The Zender modulator (MZM) modulates to the optical carrier, using an external cavity laser (ECL) with a wavelength of 1552nm. Before the fiber link, an amplifier and a variable optical attenuator are used to control the input fiber power. The fiber input power of the above solution is 6.5dBm, and the fiber link is a standard single-mode fiber with a length of 30km, and the attenuation is 0.2dB per kilometer at 1550nm. An attenuator and an amplifier form a noise control module at the OFDM receiving end, which can adjust the signal-to-noise ratio of the system and facilitate the test of the relationship between the bit error rate and the optical signal-to-noise ratio. The input power to the photodetector is then controlled by a fixed optical attenuator. The final signal is transmitted to a 50G sampling oscilloscope, and signal processing is performed on the obtained signal. And through a series of digital signal processing operations mirrored by the transmitter, for example, Fourier transform, channel estimation, de-FPC operation, modulation, etc., and compare the transmitted signal with the received signal to calculate the bit error rate.

Cumulative Distribution Function (Cumulative Distribution Function), also known as distribution function, is the integral of probability density function, which can fully describe the probability distribution of a real random variable X. In this paper, by calculating the CCDF of different algorithms, it is used to judge the suppression of the PAPR of the OFDM signal by the algorithm. Among them, 10000 OFDM symbols are used to simulate different algorithms. The result is shown in Figure 8. The pink line represents the C-OFDM signal, and when the CCDF is 10^-3, the corresponding PAPR value is 13.5dB. Among them, the red curve, green curve and blue curve represent the curves corresponding to the cumulative distribution function of the FPC and SLM algorithms when the random number is 50, 100, and 150, respectively. It can be seen from the figure that when the number of random times is fixed, the suppression capabilities of the SLM algorithm and the FPC algorithm are close to coincident. In addition, see Figure 8, which shows the signal diagrams of the algorithms under different algorithms. From top to bottom, the third line represents the traditional OFDM signal. It is obvious that the signal has many prominent peaks. The first and second pictures show the OFDM signal suppressed by the SLM algorithm and the FPC algorithm. It can be clearly seen that the signal is more average and the high peak signal is obviously suppressed.

Referring to Figure 9, a signal with an initial length of 410 (20%) and 256 (50%) is used as the initial signal, and the length after PC encoding is 512. In order to test that the FPC block structure proposed in this paper is the optimal structure, this paper is in When the length is 410, it is simulated that Block1 occupies 40%, 45% and 50% respectively, and Block2 occupies the remaining 60%, 55% and 50%. The red line in the figure indicates that Block1 occupies 50%, and the green line indicates that Block1 occupies 45%. The blue line indicates that Block1 occupies 40%. From the simulation results, it can be found that the scheme occupying 50% is the optimal scheme. At the same time, when the initial length is 256, three schemes (30%, 70%), (40%, 60%), (50%, 50%) are also designed, and the average allocation scheme has the optimal solution.

Referring to Figure 10, the comparison shows the comparison results of the two algorithms SLM algorithm and PC algorithm after suppressing the SLM algorithm. The two pink lines represent the BER curve of the traditional OFDM signal, and the red line represents the BER curve of the signal after the SLM algorithm. Blue and green represent the BER curves of the signal when the FPC redundancy is 20% and 50%, respectively. In addition, the circular circle represents the BTB signal, and the square represents the signal after passing through the 80Km optical fiber. It is obvious from the figure that the SLM algorithm only has a gain of 0.5dB under 10^-3, while the FPC algorithm proposed in this paper has a gain of 4dB when the redundancy is 20%, and has a gain of 0.5 when the redundancy is 0.5. 6.5dB of gain.

It can be understood that, the above embodiments are only exemplary embodiments adopted for illustrating the principle of the present invention, but the present invention is not limited thereto. For those skilled in the art, various variations and improvements can be made without departing from the spirit and essence of the present invention, and these variations and improvements are also considered as the protection scope of the present invention.

Claims (4)

1. A method based on polar code-assisted probabilistic PAPR suppression, characterized in that, comprising: dividing the signal to be encoded into multiple blocks; encoding the divided blocks by using polar codes to obtain multiple transmission signals; Calculate the peak-to-average power ratio of the transmission signal, and select a transmission signal with the lowest peak-to-average power ratio for transmission. 2. the method for suppressing the probability class PAPR based on polar code assistance according to claim 1, is characterized in that, described signal to be coded is divided into a plurality of blocks, comprising: determining whether the length of the signal to be encoded is an odd number or an even number, If it is an even number, dividing the signal to be encoded into a first block and a second block, wherein the lengths of the first block and the second block are equal; If it is an odd number, the signal is equally divided into a third block and a fourth block, wherein the length of the third block is one unit larger than the length of the fourth block. 3. The method for suppressing the probabilistic PAPR based on polar code assistance according to claim 2, wherein the encoding of the signal to be encoded using a polar code comprises: Perform U times of parallel encoding on the signal to be encoded, perform quadrature amplitude modulation QAM and inverse fast Fourier transform IFFT, and sequence multiple transmission signals, wherein U is a positive integer. 4. The method for suppressing probabilistic PAPR based on polar code assistance according to claim 1, further comprising the step of decoding the transmission signal, comprising: Received signal sequence, the formula is expressed as For each y _i , compute y _i represents the receiving end information; This formula is the ratio of the probability that the signal appears 0 and 1; Calculate the likelihood value of the i-th bit at the sender in order Indicates that before the receiving end encodes the sending end The evaluation sequence for is: like then, otherwise