CN110166394B - SC-FDMA signal PAPR reduction method - Google Patents

Abstract

The invention discloses a method for reducing PAPR of SC-FDMA signal, which comprises the steps of introducing additive predistortion vector to carry out iterative processing on original modulation constellation point symbols causing time domain SC-FDMA signal peak values, minimizing in-band distortion of original modulation symbols at a sending end under the limiting condition of ensuring the PAPR threshold of a system during each iteration to avoid the performance loss of bit error rate at a receiving end of the system as much as possible, and then solving corresponding optimization problem to obtain the optimal additive predistortion vector.

Description

SC-FDMA signal PAPR reduction method

Technical Field

The invention relates to a medical electronic technology, in particular to an SC-FDMA signal PAPR reduction method.

Background

The PAPR of the input signal has an important influence on the power amplifier at the transmitting end, which results in a large dynamic range of the signal when the PAPR of the input signal is large, and the power amplifier at the transmitting end works in a saturation region, thereby introducing nonlinear distortion to reduce the transmission performance of the system. In-band distortion is introduced to the current user, while out-of-band radiation also introduces interference to neighboring users. Therefore, when the transmitted signal has a large PAPR, a certain power back-off is introduced to reduce the non-linear distortion so as to ensure that the power amplifier operates in the linear amplification region as much as possible, but this will greatly reduce the power efficiency of the power amplifier and thus the system efficiency. Besides reducing the energy efficiency of the system, when the transmitting signal with a larger PAPR works in the front-end power amplifier, the output signal power is also reduced, thereby limiting the coverage area of the wireless mobile communication system. Bringing difficulties to the application and implementation of the corresponding physical layer technology in the uplink. Uplink users are powered by batteries with limited energy, and it is desirable to use as little energy as possible to obtain the desired user experience, with more stringent requirements in terms of power consumption and energy efficiency. Therefore, for the uplink, it is more important to realize the multi-user low PAPR information transmission, reduce the PAPR of the SC-FDMA signal, and avoid the system energy efficiency and spectral efficiency loss as much as possible, which has important meaning and application prospect for the uplink multi-user low PAPR information transmission.

Disclosure of Invention

The present invention is directed to a PAPR reduction method for SC-FDMA signals to solve the above-mentioned problems in the background art.

In order to achieve the purpose, the invention provides the following technical scheme:

an SC-FDMA signal PAPR reduction method comprises the following steps:

A. designing an optimal additive predistortion vector based on all influence factors on the generation of the maximum peak value;

B. and (3) performing iterative processing on the signals with the peaks: the method comprises the following steps: in the ith iteration, firstly aiming at the end of the last iteration, the obtained time domain SC-FDMA symbol vector x^(i-1)∈C^NCarrying out PAPR calculation; and comparing the calculated PAPR with a PAPR threshold required by the system, stopping processing if the current PAPR is smaller than a set threshold value, and taking the SC-FDMA signal at the moment as a final sending signal. If the current PAPR value is larger than the threshold value, the following processing is continuously carried out on the signal: firstly, the peak search is carried out on the SC-FDMA signal to be processed currently to determine the position where the maximum peak occurs, which is marked as n_maxFound by the following formula;

wherein d is^(i-1)∈C^MThe symbol vector of input data of a user in an M-point DFT module after the last iteration is finished, IDFT_N(.) and DFT_M(.) are normalized N-point IDFT and M-point DFT, respectively; for peak suppression of time domain signals, in a vector of data symbols d^(i-1)Introducing an additive predistortion vector C⁽ⁱ⁾∈C^MWherein the vector of predistortion C⁽ⁱ⁾For an M-dimensional vector containing M elements, the data symbols that introduce additive predistortion at this time may be updated as: d⁽ⁱ⁾＝d^(i-1)+C⁽ⁱ⁾(ii) a Optimizing additive predistortion vectors, jointly designing PAPR and BER, minimizing in-band distortion by EVM, and enabling SC-FDMA symbols to meet PAPR limit, wherein the process of optimizing the predistortion vectors can be modeled as follows:

s.t.c⁽ⁱ⁾＝IDFT_N(DFT_M(C⁽ⁱ⁾))

gamma is a threshold value.

As a further technical scheme of the invention: in the process of optimizing predistortion vectors

Instead of the former

Order to

Simplified to the following form:

s.t.c^(m)＝IDFT_N(DFT_M(C^(m)))

wherein lambda is a Lagrange factor, and the convex optimization problem in the formula (9) is solved by a KKT condition; to obtain

Solving the above KKT condition yields:

k＝0,1,...,M-1

wherein the content of the first and second substances,

is that

So that the signal is output

Can be updated as:

compared with the prior art, the invention has the beneficial effects that: the invention carries out iterative processing on the original modulation constellation point symbol causing the time domain SC-FDMA signal peak value by introducing the additive predistortion vector, avoids the bit error rate performance loss of the system receiving end as much as possible by minimizing the in-band distortion of the original modulation symbol of the sending end under the limiting condition of ensuring the PAPR threshold of the system during each iteration, and then solves the corresponding optimization problem to obtain the optimal additive predistortion vector.

Drawings

FIG. 1 is a first simulation of the present invention.

FIG. 2 is a second simulation of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Referring to fig. 1-2, an SC-FDMA signal PAPR reduction method,

firstly, introducing an SC-FDMA signal model;

let one SC-FDMA data symbol d ═ d₀,d₁,...,d_M-1]^TAfter the normalized DFT of M points, the frequency domain data is changed into X ═ X₀,X₁,...,X_M-1]^T。

To X_kAfter the IDFT of N points, the formula is a time domain signal x (N)

Combining the formulas (1) and (2) to obtain the following formula:

wherein g (n) is represented by

The PAPR of the time-domain signal x ═ x (0), x (1),.., x (N-1) ] can be defined as:

SC-FDMA signal PAPR reduction technology provided by the invention

We have previously analyzed that the peak-producing factor of SC-FDMA signals is due to the coherent superposition of the main and side lobes of the individual pulses to which the original modulation symbols correspond. An optimal additive predistortion vector can be designed if all the factors influencing the generation of the maximum peak are taken into account. The peak appearing signal is then iteratively processed. In particular, in the ith iteration, the time domain SC-FDMA symbol vector x obtained for the end of the last iteration is first obtained^(i-1)∈C^NPAPR calculation is performed. Comparing the calculated PAPR with the PAPR threshold required by the system, if the current PAPR is smaller than the set threshold, stopping the processing, and comparing the PAPR with the set thresholdAs a final transmission signal. If the current PAPR value is larger than the threshold value, the following processing is continuously carried out on the signal. First, a peak search is performed on the SC-FDMA signal currently to be processed to determine the position where the maximum peak occurs, denoted as n_maxThe formula is found out by the following formula,

wherein d is^(i-1)∈C^MThe symbol vector of input data of a user in an M-point DFT module after the last iteration is finished, IDFT_N(.) and DFT_M(.) are normalized N-point IDFT and M-point DFT, respectively.

For peak suppression of time domain signals, in a vector of data symbols d^(i-1)Introducing an additive predistortion vector C⁽ⁱ⁾∈C^MWherein the vector of predistortion C⁽ⁱ⁾For an M-dimensional vector containing M elements, the data symbols that introduce additive predistortion at this time may be updated as:

d⁽ⁱ⁾＝d^(i-1)+C⁽ⁱ⁾

and optimizing the additive predistortion vector, jointly designing the PAPR and the BER, minimizing the in-band distortion by using the EVM, and enabling the SC-FDMA symbol to meet the PAPR limit. The process of optimizing the predistortion vector can be modeled as:

s.t.c⁽ⁱ⁾＝IDFT_N(DFT_M(C⁽ⁱ⁾))

the above problem is a non-convex optimization problem because

Non-convex, the following alternatives can be made in order to solve the problem with convex optimization knowledge. The energy of the introduced additive predistortion vector is weak and can be approximately used

Instead of the former

At the same time, in order to simplify the calculation

The original optimization problem is changed into a convex optimization problem, and can be simplified into the following form:

s.t.c^(m)＝IDFT_N(DFT_M(C^(m)))

the lagrange function is constructed from the above problem as:

wherein λ is lagrangian factor, the convex optimization problem in equation (9) is solved by KKT condition, and can be obtained:

solving the above KKT condition yields:

k＝0,1,...,M-1

wherein the content of the first and second substances,

is that

So that the signal is output

Can be updated as:

compared with the conventional SC-FDMA system, the extra computational complexity required by PAPR reduction of the proposed algorithm is mainly focused on time domain signal PAPR calculation and solution of the optimization problem.

The working principle of the invention is as follows: the performance verification is carried out on the proposed method for reducing the PAPR of the SC-FDMA signal through computer simulation. The method adopts 16QAM modulation, the number of subcarriers of each SC-FDMA symbol is 1024, the number of subcarriers of each user is 72, the PAPR threshold set by the system is 6.5dB, and the maximum iteration number is 3. Simulation results as in fig. 1-2 can be obtained.

From simulation results, compared with the conventional SC-FDMA system, the method has the advantage that when the threshold is set to be 6.5dB, the value of a Complementary Cumulative Distribution Function (CCDF) curve is 10^-4The corresponding PAPR reduction gain is about 2dB with little loss in BER performance.

It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.

Furthermore, it should be understood that although the present description refers to embodiments, not every embodiment may contain only a single embodiment, and such description is for clarity only, and those skilled in the art should integrate the description, and the embodiments may be combined as appropriate to form other embodiments understood by those skilled in the art.

Claims (2)

1. A method for PAPR reduction of an SC-FDMA signal, comprising the steps of:

A. designing an optimal additive predistortion vector based on all influence factors on the generation of the maximum peak value;

B. and (3) performing iterative processing on the signals with the peaks: the method comprises the following steps: in the ith iteration, firstly aiming at the end of the last iteration, the obtained time domain SC-FDMA symbol vector x^(i-1)∈C^NCarrying out PAPR calculation; comparing the calculated PAPR with a PAPR threshold required by the system, if the current PAPR is smaller than a set threshold, stopping processing, taking the SC-FDMA signal at the moment as a final sending signal, and if the current PAPR is larger than the threshold, continuing to perform the following processing on the signal: firstly, the peak search is carried out on the SC-FDMA signal to be processed currently to determine the position where the maximum peak occurs, which is marked as n_maxFound by the following formula;

wherein d is^(i-1)∈C^MThe symbol vector of input data of a user in an M-point DFT module after the last iteration is finished, IDFT_N(.) and DFT_M(.) are normalized N-point IDFT and M-point DFT, respectively; for peak suppression of time domain signals, in a vector of data symbols d^(i-1)Introducing an additive predistortion vector C⁽ⁱ⁾∈C^MWherein the vector of predistortion C⁽ⁱ⁾For vectors of M dimensions containing M elements, in which case addition is introducedThe data symbols of the sexual predistortion can be updated as: d⁽ⁱ⁾＝d^(i-1)+C⁽ⁱ⁾(ii) a Optimizing additive predistortion vectors, jointly designing PAPR and BER, minimizing in-band distortion by EVM, and enabling SC-FDMA symbols to meet PAPR limit, wherein the process of optimizing the predistortion vectors can be modeled as follows:

s.t.c⁽ⁱ⁾＝IDFT_N(DFT_M(C⁽ⁱ⁾))

gamma is a threshold value.

2. An SC-FDMA signal PAPR reduction method according to claim 1 wherein the pre-distortion vector is optimized using

Instead of the former

Order to

Simplified to the following form:

s.t.c^(m)＝IDFT_N(DFT_M(C^(m)))

wherein lambda is a Lagrange factor, and the convex optimization problem is solved by a KKT condition; to obtain

Solving the above KKT condition yields:

wherein the content of the first and second substances,

is that

So that the signal is output

Can be updated as:

Images