CN106533590A - Uplink channel quality measurement method based on receiving end EVM - Google Patents

Abstract

The invention provides an uplink channel quality measurement method based on receiving end EVM, used for solving the technical problem of low system throughput of the existing channel quality measurement method. The method comprises the following steps: constructing an adaptive uplink system model, and establishing a mathematic relation of frequency domain signals between a transmitting end and a receiving end of the model; extracting a pilot signal from the frequency domain signals of the receiving end according to a pilot configuration mode in a 3GPP standard; estimating the system frequency domain channel response; estimating a measurement signal of a channel quality measurement algorithm based on the receiving end EVM; decoding the Turbo code of an estimation value of the frequency domain signals of the transmitting end to obtain decoded bit data, and performing re-encoding and re-modulation on the bit data to obtain an ideal reference signal; calculating the receiving end EVM by using an error vector amplitude calculation formula according to the measurement signal and the ideal reference signal; and selecting a modulation encoding mode MCS. The uplink channel quality measurement method provided by the invention is high in channel quality measurement accuracy, low in computation complexity, high in robustness and is suitable for wireless communication systems.

Description

Uplink channel quality measuring method based on receiving end EVM

Technical Field

The invention belongs to the field of wireless communication, relates to a channel quality measuring method, and particularly relates to a channel quality measuring method of an uplink adaptive system based on receiving end EVM, which is suitable for wireless communication systems such as a ground cellular communication system, a satellite communication system, a Massive-MIMO system, a point-to-point link transmission system and the like.

Background

In a wireless communication system, a channel is a time-varying channel with a multipath effect, and if a transmitting terminal knows prior information of the channel, the transmitting terminal can select a modulation mode and a coding rate more suitable for channel transmission to transmit data, so that the influence of various factors on a received signal in the transmission process is reduced, the transmission condition of the variable channel is better adapted, and the system performance is improved. Therefore, in order to improve the performance of the communication system, the research on link adaptation is necessary. The adaptation of the uplink includes power control techniques and rate control techniques. The power control technology maintains a certain signal-to-noise ratio of a receiving end by dynamically adjusting the transmitting power, thereby ensuring the transmission quality of a link. The rate control technology is a technology mainly adopted by link adaptation, namely a frequently-called Adaptive Modulation and Coding (AMC) technology, and an eNodeB dynamically selects a Modulation and Coding Scheme (MCS) according to information of channel environment change provided by a user terminal, so that the BLER limit of a system is met, the performance of the system is maximally improved, and the transmission quality of a link is ensured. Specifically, users with better channel conditions, which are closer to the base station, are assigned a higher order modulation mode and a higher coding rate; for a user far away from the base station and with poor channel conditions, the receiving end needs more redundant information to ensure correct demodulation, but more redundant information reduces the coding rate, so that a low-order modulation method and a lower coding rate are allocated.

In an uplink adaptive system, a receiving end needs to feed back channel state information to a transmitting end, which needs to perform channel quality measurement, and the existing method generally adopts a signal to interference and noise ratio (SINR) to characterize the transmission quality of a channel, that is, the SINR is used as a measurement standard for channel quality measurement, the receiving end selects an MCS according to the estimated SINR and then feeds back the selected MCS to the transmitting end, and the transmitting end performs the optimal configuration of next transmission parameters according to the fed back channel quality information. The basic idea of the research on this aspect is to estimate the SINR of each subcarrier, then map the SINR values of these subcarriers to an effective SINR that can reflect the average performance of the entire link by a certain mapping method, and select the MCS according to the effective SINR. Commonly used methods are the Exponential Effective SINR Mapping (EESM) algorithm, the Mutual Information Effective SINR Mapping (MIESM) algorithm, the average mutual information per bit (MMIB) algorithm, the Average Effective SINR Mapping (AESM) algorithm, and the harmonic-mean algorithm (Harm-mean). These two algorithms, EESM and MIESM, are similar and are derived from the Effective SINR Mapping (ESM), except that they use different mapping functions, namely an exponential mapping function and a related mutual information mapping function. However, in both algorithms, the mapping function contains a tuning factor, which needs to be obtained by offline simulation, and the parameter value depends on MCS, channel state and antenna configuration, so that it is difficult to achieve the optimum and the robustness is poor, thereby causing the loss of system throughput performance. The MMIB refers to average mutual information between coded bits and log-likelihood ratios (LLRs) thereof, MMIB channel quality measurement refers to equivalence of a wireless channel into a plurality of parallel bit LLR channels, effective SINR and bit-level average mutual information establish a one-to-one correspondence relationship by calculating mutual information between the coded bits and corresponding receiving-end LLR values, but the accuracy of the MMIB algorithm is low because the calculation of the mutual information is very complicated and an accurate mapping formula is difficult to obtain, in addition, the performance of the algorithm depends on modulation orders to a great extent, that is, the algorithm is very sensitive to the change of the modulation orders and is not suitable for the wireless channel with time-varying characteristics. The Harm-mean algorithm is to take the harmonic mean of SINRs of a plurality of subcarriers as the equivalent SINR, while the AESM algorithm considers that in SC-FDMA systems, SINRs of all subcarriers are approximately equal, so averaging can be used to simplify the calculation. However, the two methods have certain limitations and are only suitable for scenes with less serious frequency selection, and although the complexity of the two algorithms is small, the calculation accuracy is correspondingly poor.

From the above analysis, it can be seen that the above methods are equivalent to a many-to-one mapping, so that the effective SINRs of two different sub-channels may be the same, the difference between the different sub-channels is ignored, and the same MCS is selected, and the MCS may not be optimal for the sub-channels, so that the measurement accuracy is low, the robustness is poor, thereby affecting the system performance and causing the reduction of the throughput. Therefore, the SINR is used as a metric of channel quality, and the accuracy of the measurement value is very critical to the improvement of system throughput performance. However, in wireless communication, there are many factors that affect the change of the SINR value, such as fading channel, interference, noise, and the like, so it is difficult to ensure the accuracy of the SINR value, which is challenging.

In the uplink, the signal changes due to noise or interference when the signal passes through the channel. If the channel condition is good, the signal changes less through the channel, otherwise, if the channel condition is poor, the signal changes more through the channel. Therefore, in addition to the traditional SINR-based channel quality measurement, we can also judge the channel quality by measuring the degree of change of the signal passing through the channel, and the Error Vector Magnitude (EVM) is an index for measuring the change, because the influence of interference and noise on the signal can be visually expressed by the deviation of the signal from the standard constellation point on the constellation diagram. The receiving-end EVM is defined as a ratio between a square root value of the average power of the error vector signal at the receiving end and a square root value of the average power of the reference signal at the receiving end. Therefore, the size of the receiving end EVM can reflect the channel quality, when the channel condition is good, the signal is less influenced by the channel, and the receiving end EVM is smaller; when the channel condition is poor, the signal is greatly influenced by the channel, and the receiving end EVM is large. And the receiving end EVM of the system can be obtained more easily than the SINR of each subcarrier is estimated.

Disclosure of Invention

The invention aims to overcome the defects of the prior art and provides an uplink channel quality measuring method based on a receiving end EVM (evolution-modulation), wherein the receiving end obtains a mapping table between the receiving end EVM and the MCS by utilizing off-line simulation according to the instantaneous EVM of the system to select a proper MCS and feeds the mapping table back to a transmitting end for configuring the optimal parameters for next transmission, and the method aims to solve the technical problem of low system throughput caused by low channel quality measuring accuracy and high calculation complexity while meeting the BLER (block error rate) limitation of an uplink adaptive link.

In order to achieve the purpose, the technical scheme adopted by the invention comprises the following steps:

(1) constructing an uplink adaptive system model, and establishing a mathematical relation of frequency domain signals between a transmitting end and a receiving end of the model, wherein the method comprises the following steps:

(1a) adopting a multiple access technology to construct an uplink adaptive system model of a ground cellular communication system, or a satellite communication system, or a Massive-MIMO system or a point-to-point link transmission system: the MIMO channel comprises a receiving end and an MIMO channel, wherein the receiving end comprises a Turbo coding module, a modulation module, a layer mapping module, a DFT conversion module and a symbol generation module, the receiving end comprises an FFT conversion module, a channel estimation module, a frequency domain equalization module, an IDFT conversion module, a demodulation module and a decoding module, and the MIMO channel comprises a channel quality measurement module and an adaptive modulation and coding module based on the receiving end EVM.

(1b) Establishing a mathematical relationship of frequency domain signals between a transmitting end and a receiving end of an uplink adaptive system modelObtaining a receiving end frequency domain signal matrix y_f：

Where Y is the receiving end frequency domain signal, H is the system frequency domain channel response, F_MFor a normalized DFT transform of size M,k,l＝1,2,...,M，N_Tin order to determine the number of the transmitting antennas,is dimension N_TX is the modulation signal of the transmitting end, V is the mean value of 0, and the variance isWhite gaussian noise.

(2) At the receiving end, according to the pilot configuration form in the 3GPP standard, the pilot signal Y is extracted from the frequency domain signal Y at the receiving end of the uplink self-adaptive system model_p：

Y_p＝H_pS_p+V_p

Wherein S is_p、H_pAnd V_pRespectively, a pilot signal at the transmitting end, a pilot channel response, and white gaussian noise.

(3) At a receiving end, estimating the system frequency domain channel response H to obtain a system frequency domain channel response estimated value

(3a) Using a channel estimation algorithm based on the pilot signal Y_pResponse to pilot channel H_pEstimating to obtain pilot channel response estimation value

(3b) Using time domain interpolation algorithm, based on pilot channel responseComputing system frequency domain channel response

(4) Estimating a measurement signal Z based on a receiving end EVM channel quality measurement algorithm:

(4a) adopting a frequency domain equalization algorithm, and obtaining a receiving end frequency domain signal matrix y in the step (1b)_fAnd the system frequency domain channel response obtained in step (3)Calculating an estimated value of a frequency domain signal of a transmitting end

(4b) Estimation value of frequency domain signal of transmitting terminalIFFT conversion is carried out to obtain the time domain signal estimation value of the transmitting terminalI.e. a measurement signal Z based on the receiving-end EVM channel quality measurement algorithm.

(5) At the receiving end, the time domain signal estimation value of the transmitting end is carried outAnd decoding the Turbo code to obtain decoded bit data, and recoding and remodulating the bit data to obtain the ideal reference signal R.

(6) And (5) calculating the receiving end EVM according to the measurement signal Z obtained in the step (4) and the ideal reference signal R obtained in the step (5) by adopting an error vector magnitude calculation formula.

(7) At the receiving end, a modulation coding scheme MCS is selected:

(7a) and performing off-line simulation on the uplink self-adaptive system model to obtain a mapping table between the EVM and the MCS of the receiving end.

(7b) And simulating the uplink self-adaptive system model for multiple times, selecting a corresponding MCS from a mapping table between the receiving end EVM and the MCS according to the instantaneous receiving end EVM value in each simulation cycle, and feeding back the MCS to the transmitting end for configuring the optimal parameters of the next transmission.

Compared with the prior art, the invention has the following advantages:

(1) the invention obtains the mapping table between the receiving end EVM and the MCS through off-line simulation, and selects the corresponding MCS from the mapping table between the receiving end EVM and the MCS according to the system instantaneous receiving end EVM, and the mapping relation is one-to-one, so the channel quality measurement accuracy is higher, and the throughput performance of the system is effectively improved under the condition of meeting the BLER limit of the system;

(2) the invention adopts the channel quality measuring method based on the receiving end EVM when measuring the channel quality of the uplink self-adaptive system, has small calculation complexity and high robustness, can be better suitable for the uplink self-adaptive system, and further improves the throughput performance of the system.

Drawings

FIG. 1 is a block diagram of an implementation flow of the present invention;

FIG. 2 is a schematic diagram of an uplink adaptive system model of the present invention;

FIG. 3 is a schematic diagram of a channel quality measurement indicator EVM employed by the present invention;

FIG. 4 is a flowchart of obtaining a mapping table between EVM and MCS at a receiving end in the present invention;

FIG. 5 is a graph of EVM and BLER simulation at the receiving end of the present invention;

fig. 6 is a diagram of a simulation of system BLER and throughput for the present invention.

Detailed Description

The invention is described in further detail below with reference to the figures and specific examples.

Referring to fig. 1, the implementation steps of the present invention are:

step 1, an uplink adaptive system model of a ground cellular communication system, or a satellite communication system, or a Massive-MIMO system or a point-to-point link transmission system is constructed by using an SC-FDMA multiple access technology, and the LTE system in ground cellular communication is constructed in this embodiment, as shown in fig. 2, and includes a transmitting end including Turbo coding, modulation, layer mapping, DFT transformation and SC-FDMA symbol generation modules, a receiving end including FFT transformation, channel estimation, frequency domain equalization, IDFT transformation, demodulation and decoding, and an MIMO channel including an adaptive feedback module including channel quality measurement based on receiving end EVM and Adaptive Modulation Coding (AMC).

Step 2, establishing the mathematical relation of the frequency domain signals between the transmitting end and the receiving end of the constructed uplink self-adaptive system model

WhereinIs an IDFT transform of size M, H_r,n＝diag{[H₁]_r,n,[H₂]_r,n,...,[H_M]_r,n}，[H_m]_r,nRefers to the frequency domain channel response of the m sub-carrier under the nth transmitting antenna and the r receiving antenna, v_rMean value of 0 and variance of frequency domainOf the noise vector x_n＝[x_n,1,x_n,2,...,x_n,M]^TIs a time domain signal N of a transmitting terminal after M-QAM modulation_TNumber of transmitting antennas, N_RFor the number of receive antennas, h is the system time domain channel response.

Representing the receiving end frequency domain signal as a matrix form y_f：

Wherein Y is the receiving end frequency domain signal, H is the system frequency domain channel response, V is the mean value 0, and the variance isWhite gaussian noise.

Step 3, at the receiving end, according to the pilot frequency configuration form of the 3GPP LTE standard, extracting the pilot frequency signal Y from the frequency domain signal Y of the receiving end of the uplink self-adaptive system model_p：

Within a slot, different SC-FDMA symbols may have different CP (cyclic prefix) lengths, and after passing through the channel, the signal is received and processed at the receiving end. Receiving end is receivingAfter the uplink signal passing through the channel, the time-frequency synchronization is carried out on the received signal, and a pilot signal Y is extracted from the frequency domain signal Y of the receiving end which passes through the time-frequency synchronization_p：

Y_p＝H_pS_p+V_p

Wherein S is_p、H_pAnd V_pRespectively, the pilot signal at the transmitting end, the pilot channel response, and the white gaussian noise. Pilot signal S of transmitting terminal_pAre known.

Step 4, at the receiving end, estimating the system frequency domain channel response H to obtain the system frequency domain channel response estimated value

Step 4a, based on the pilot signal Y extracted in step 3_pResponding H to pilot channel by using LS channel estimation algorithm or MMSE channel estimation algorithm_pEstimating to obtain pilot channel response estimation value

If the LS channel estimation is adopted, the channel response estimation coefficient is as follows:

if MMSE channel estimation is adopted, the channel response estimation coefficient is as follows:

wherein,is pilot channel response and receptionThe covariance matrix of the received signal is determined,is the received signal auto-covariance matrix.

The LS channel estimation algorithm is widely used because it is simple and easy to implement and does not require the statistical properties of the channel, but the estimation result of the LS algorithm is easily affected by noise, and particularly when the signal-to-noise ratio is low, the estimation accuracy is greatly reduced. The MMSE algorithm has good inhibition effect on inter-subcarrier interference and Gaussian white noise. In this embodiment, an MMSE channel estimation algorithm is employed.

Step 4b, using time domain interpolation algorithm, according to pilot channel responseComputing system frequency domain channel responseThe method comprises the following steps:

step 4b1, using FFT/IFFT method to the pilot channel response estimation value obtained in step 4aIFFT conversion is carried out to obtain a time domain pilot channel response estimated value

Wherein,is of size N_pIDFT of, N_pAs to the number of pilot channel responses,n＝0,1,...,N_p-1, T is a matrix transpose.

Step 4b2, estimating the time domain pilot channel responseInterpolation is carried out to obtain the estimated value of the time domain channel response of the system

Where N is the number of system channel responses.

Step 4b3, estimating the system time domain channel responseFFT operation is carried out to obtain the system frequency domain channel response

Wherein F_NFor a DFT transform of size N,k＝0,1,...,N-1。

step 5, estimating a measurement signal Z based on the EVM channel quality measurement algorithm of the receiving end:

step 5a, adopting a ZF frequency domain equalization algorithm or an MMSE frequency domain equalization algorithm, and according to the receiving end frequency domain signal matrix y in the step 2_fAnd the system frequency domain channel response obtained in step 4Calculating an estimated value of a frequency domain signal of a transmitting end

If ZF frequency domain equalization algorithm is adopted, the equalization coefficient isThe estimated value of the frequency domain signal of the transmitting terminal is as follows:

if MMSE equalization algorithm is adopted, the equalization coefficient isThe estimated value of the frequency domain signal of the transmitting terminal is as follows:

in this embodiment, a ZF frequency domain equalization algorithm is employed.

Step 5b, estimating the frequency domain signal of the transmitting terminalIFFT conversion is carried out to obtain the time domain signal estimation value of the transmitting terminalNamely, the measurement signal Z based on the receiving end EVM channel quality measurement algorithm:

step 6, estimating the time domain signal of the transmitting terminalPerforming Turbo code decoding to obtain decoded bit data, and performing re-encoding and re-modulation on the bit data to obtain an ideal reference signal R:

step 6a, firstly, the time domain signal estimation value of the transmitting terminal is carried outTurbo code logarithm MAP decoding is carried out to obtain the time domain signal estimated value of the transmitting terminalSoft information L(s) of bit data of_k)：

Then the soft information is processedHard decision is carried out to obtain a time domain signal estimation value of a transmitting terminalBit data ofThe decision formula is:

step 6b, comparisonSpecial dataAnd performing recoding and remodulation to obtain an ideal reference signal R.

And 7, calculating the EVM of the receiving end according to the measurement signal Z obtained in the step 5 and the ideal reference signal R obtained in the step 6 by adopting an error vector magnitude calculation formula:

the receiving-end EVM is defined as a ratio between a square Root value of the average power of the error vector signal at the receiving end and a square Root value of the average power of the reference signal, that is, a ratio between a Root mean square value (RMS) of the error vector signal and the reference signal, as shown in fig. 3.

Step 7a, adopting an error vector magnitude calculation formula to calculate the EVM of each subframe of the receiving end in the uplink shared data channel_i：

Step 7b, calculating EVM of all sub-frame receiving ends under one frame_iObtaining the receiving end EVM by the average value of the data, wherein the realization formula is as follows:

wherein i is the subframe number, and M is the total number of subframes.

Step 8, selecting a modulation coding scheme MCS:

step 8a, performing offline simulation on the uplink adaptive system model to obtain a mapping table between the receiving end EVM and the MCS, and performing offline simulation on the uplink adaptive system model using the SISO AWGN channel to obtain the mapping table, where a specific implementation flow is shown in fig. 4, where the MCS used in this embodiment is 1-15:

in step 8a1, MCS is set to 1.

And 8a2, performing off-line cyclic simulation on the uplink adaptive system model adopting the SISO AWGN channel, and finding out the value range of SNR which can ensure that BLER is uniformly distributed between 0 and 1 from the simulation result.

Step 8a3, performing off-line loop simulation on the uplink adaptive system model using the SISO AWGN channel at each SNR within the SNR value range to obtain receiving ends EVM and BLER corresponding to each loop, and connecting the receiving ends EVM and BLER to obtain a simulation curve between the receiving ends EVM and BLER, as shown in fig. 5.

Step 8a4, on the simulation curve between receiving end EVM and BLER, finding and storing the receiving end EVM corresponding to BLER equal to 0.1, and storing the current MCS.

Step 8a5, which is to make MCS equal to MCS +1, repeat steps 8a 2-8 a4 until MCS becomes 15, and extract the mapping table between the receiving-end EVM and the corresponding MCS from the stored multiple receiving-end EVMs and MCSs, as shown in table 1:

TABLE 1

And 8b, simulating the uplink self-adaptive system model for multiple times, selecting a corresponding MCS from the table 1 according to the instantaneous receiving end EVM value in each simulation cycle, and feeding back the MCS to the transmitting end for configuring the optimal parameters of the next transmission.

The effects of the present invention can be further illustrated by the following simulations:

1. simulation conditions

Simulation software: adopting Matlab;

simulation scene: the parameter setting in the aspect of transmission is based on the 3GPP LTE standard, the 1.4MHz system bandwidth, the number of transmitting antennas is 1, the number of receiving antennas is 2, VehA channels are assumed to have no problems of time-frequency synchronization, I/Q imbalance and the like. The specific simulation parameters are shown in table 2:

table 2: simulation parameter table

Parameter(s)	Numerical value
		Cell	7 cells
Radius of cell	500m
		Carrier frequency/system bandwidth	2GHz/1.4MHz
Channel model	VehA
		SNR(dB)	5:5:40
Target BLER	Less than or equal to 10 percent
		Antenna arrangement	1 sending and 2 receiving (1 × 2)
BS receiver type	ZF receiver
		Channel estimation	MMSE

2. Simulation content and result analysis

By using the simulation conditions, the method simulates the existing channel quality measurement algorithm, such as the EESM algorithm, the MIESM algorithm, the AESM algorithm and the Harm-mean algorithm, and obtains a BLER and throughput performance comparison graph of the method and the existing channel quality measurement algorithm under different SNR, as shown in figure 6.

The abscissa in fig. 6(a) and 6(b) represents the SNR in the current scene in units of dB. The ordinate of fig. 6(a) represents the system BLER, and the ordinate of fig. 6(b) represents the system throughput in bits/s. In fig. 6, a curve marked by a small vertical bar is a BLER and throughput simulation result curve of the EESM algorithm; the curve marked by a square is used as a BLER and throughput simulation result curve of the MIESM algorithm; the curve marked by the five-pointed star is a BLER and throughput simulation result curve of the AESM algorithm; a curve marked by a diamond is a BLER and throughput simulation result curve of an Hram-mean algorithm; the curve marked by a circle is used as a simulation result curve of BLER and throughput of the system when the method is applied.

As can be seen from fig. 6(a), for all channel quality measurement algorithms, both system BLER and throughput performance improve as the SNR increases. For EESM, MIESM, AESM and Harm-mean algorithms, when SNR exceeds 25dB, the target BLER is reached, and for the algorithm provided by the invention, when SNR exceeds 13dB, the target BLER can be reached, therefore, the invention has better BLER performance and can better meet the limitation of system BLER. As can be seen from fig. 6(b), as the SNR increases, the measurement accuracy is higher and the throughput performance is significantly improved compared to the existing channel quality measurement algorithm. Therefore, compared with the existing channel quality measurement algorithms such as EESM, MIESM, AESM and Harm-mean, the throughput performance of the system can be more effectively improved under the condition of meeting the BLER requirement of the system.

Claims (9)

1. A method for measuring the quality of an uplink channel based on a receiving end EVM comprises the following steps:

(1) constructing an uplink adaptive system model, and establishing a mathematical relation of frequency domain signals between a transmitting end and a receiving end of the model, wherein the method comprises the following steps:

(1a) adopting a multiple access technology to construct an uplink adaptive system model of a ground cellular communication system, or a satellite communication system, or a Massive-MIMO system or a point-to-point link transmission system: the MIMO channel comprises a receiving end and an adaptive feedback module, wherein the receiving end comprises a Turbo coding module, a modulation module, a layer mapping module, a DFT conversion module and a symbol generation module, the receiving end comprises an FFT conversion module, a channel estimation module, a frequency domain equalization module, an IDFT conversion module, a demodulation module and a decoding module, and the adaptive feedback module comprises a channel quality measurement module based on an EVM (error vector magnitude) of the receiving end and an adaptive modulation coding module;

(1b) establishing a mathematical relationship of frequency domain signals between a transmitting end and a receiving end of an uplink adaptive system modelObtaining a receiving end frequency domain signal matrix y_f：

$y^f=Y=H(F_M\⊗I_{N_T})x+V$

Where Y is the receiving end frequency domain signal, H is the system frequency domain channel response, F_MFor a normalized DFT transform of size M,k,l＝1,2,...,M，N_Tin order to determine the number of the transmitting antennas,is dimension N_TX is the modulation signal of the transmitting end, V is the mean value of 0, and the variance isWhite gaussian noise of (1);

(2) at the receiving end, according to the pilot configuration form in the 3GPP standard, the pilot signal Y is extracted from the frequency domain signal Y at the receiving end of the uplink self-adaptive system model_p：

Y_p＝H_pS_p+V_p

Wherein S is_p、H_pAnd V_pRespectively indicating a pilot signal of a transmitting end, pilot channel response and Gaussian white noise;

(3) at a receiving end, estimating the system frequency domain channel response H to obtain a system frequency domain channel response estimated value

(3a) Using a channel estimation algorithm based on the pilot signal Y_pResponse to pilot channel H_pEstimating to obtain pilot channel response estimation value

(3b) Using time domain interpolation algorithm, based on pilot channel responseComputing system frequency domain channel response

(4) Estimating a measurement signal Z based on a receiving end EVM channel quality measurement algorithm:

(4a) adopting a frequency domain equalization algorithm, and obtaining a receiving end frequency domain signal matrix y in the step (1b)_fAnd the system frequency domain channel response obtained in step (3)Calculating an estimated value of a frequency domain signal of a transmitting end

(4b) Estimation value of frequency domain signal of transmitting terminalCarrying out IFFT conversion is carried out to obtain a time domain signal estimated value of a transmitting terminalNamely a measurement signal Z based on the EVM channel quality measurement algorithm of a receiving end;

(5) at the receiving end, the time domain signal estimation value of the transmitting end is carried outDecoding the Turbo code to obtain decoded bit data, and recoding and remodulating the bit data to obtain an ideal reference signal R;

(6) calculating the EVM of the receiving end according to the measurement signal Z obtained in the step (4) and the ideal reference signal R obtained in the step (5) by adopting an error vector magnitude calculation formula;

(7) at the receiving end, a modulation coding scheme MCS is selected:

(7a) performing off-line simulation on the uplink adaptive system model to obtain a mapping table between the EVM and the MCS of the receiving end;

(7b) and simulating the uplink self-adaptive system model for multiple times, selecting a corresponding MCS from a mapping table between the receiving end EVM and the MCS according to the instantaneous receiving end EVM value in each simulation cycle, and feeding back the MCS to the transmitting end for configuring the optimal parameters of the next transmission.

2. The receiving-end EVM-based uplink channel quality measuring method as claimed in claim 1, wherein the mathematical relationship of the frequency domain signals between the transmitting end and the receiving end of the uplink adaptive system model in step (1)The expression is as follows:

$\begin{array}{c}{y}_i^f=\underset{n=1}{\overset{N_T}{\mathrm{\Σ}}}{F}_M(h_{r,n}{x}_n+v_n)\\ =\underset{n=1}{\overset{N_T}{\mathrm{\Σ}}}\left(F_M{h}_{r,n}{F}_M{F}_M^H{x}_n+F_M{v}_n\right)\\ =\underset{n=1}{\overset{N_T}{\mathrm{\Σ}}}{H}_M{F}_M{x}_n+v_r\end{array}$

whereinIs an IDFT transform of size M, H_r,n＝diag{[H₁]_r,n,[H₂]_r,n,...,[H_M]_r,n}，[H_m]_r,nRefers to the frequency domain channel response of the m sub-carrier under the nth transmitting antenna and the r receiving antenna, v_rMean value of 0 and variance of frequency domainOf the noise vector x_n＝[x_n,1,x_n,2,...,x_n,M]^TIs a time domain signal N of a transmitting terminal after M-QAM modulation_TNumber of transmitting antennas, N_RFor the number of receive antennas, h is the system time domain channel response.

3. The method of claim 1, wherein the channel estimation algorithm in step (3a) is an LS channel estimation algorithm or an MMSE channel estimation algorithm: if the LS channel estimation algorithm is adopted, the channel response estimation coefficient isIf MMSE channel estimation algorithm is adopted, the channel response estimation coefficient isWherein,is the covariance matrix of the pilot channel response and the received signal,is the received signal auto-covariance matrix.

4. The receiving-end EVM-based uplink channel quality measuring method as claimed in claim 1, wherein the system frequency domain channel response in step (3b)The acquisition steps are as follows:

(3b1) using FFT/IFFT method to estimate the pilot channel response obtained in step (3a)IFFT conversion is carried out to obtain a time domain pilot channel response estimated value

${\hat{G}}_p=F_{N_p}^H{\hat{H}}_p={\[{\hat{G}}_p\left(0\right),{\hat{G}}_p\left(1\right),...,{\hat{G}}_p\left(n\right)\]}^T$

Wherein,is of size N_pIDFT of, N_pAs to the number of pilot channel responses,n＝0,1,...,N_p-1, T is a matrix transpose;

(3b2) for time domain pilot channel response estimation valueInterpolation is carried out to obtain the estimated value of the time domain channel response of the system

${\hat{G}}_n=\left\{\begin{array}{cc}{\hat{G}}_p\left(n\right)& 0\&le;n\&le;N_p/2\\ 0& N_p/2<n\&le;N-\&lsqb;(N_p/2)+1\&rsqb;\\ {\hat{G}}_p(n-N+N_p)& n-\&lsqb;(N_p/2)+1\&rsqb;<n\&le;N-1\end{array}\right.$

${\hat{G}}_n={\&lsqb;{\hat{G}}_n\left(0\right),{\hat{G}}_n\left(1\right),...,{\hat{G}}_n(N-1)\&rsqb;}^T$

Wherein N is the number of system channel responses;

(3b3) channel response estimation value for system time domainFFT operation is carried out to obtain the system frequency domain channel response

$\hat{H}=F_N{\hat{G}}_n={\&lsqb;\hat{H}\left(0\right),\hat{H}\left(1\right),...,\hat{H}\left(k\right)\&rsqb;}^T$

Wherein F_NOf size NThe DFT of the signal is transformed,k＝0,1,...,N-1。

5. the uplink channel quality measuring method based on receiving-end EVM according to claim 1, wherein the frequency domain equalization algorithm in step (4a) adopts ZF frequency domain equalization algorithm or MMSE frequency domain equalization algorithm: if ZF frequency domain equalization algorithm is adopted, the equalization coefficient isThe estimated value of the frequency domain signal at the transmitting end isIf MMSE equalization algorithm is adopted, the equalization coefficient is The estimated value of the frequency domain signal at the transmitting end is

6. The receiving-end EVM-based uplink channel quality measurement method according to claim 1, wherein the measurement signal Z based on the receiving-end EVM channel quality measurement algorithm in step (4b) is implemented by the following formula:

$Z=\hat{x}={\tilde{F}}^H\hat{X}.$

7. the method as claimed in claim 1, wherein the step of obtaining the ideal reference signal R in step (5) comprises:

(5a) firstly, the time domain signal estimation value of a transmitting terminal is carried outTurbo code logarithm MAP decoding is carried out to obtain the time domain signal estimated value of the transmitting terminalSoft information L(s) of bit data of_k)：

$L\left({\hat{s}}_k\right)=ln\frac{P(s_k=1|{\hat{x}}_k)}{P(s_k=0|{\hat{x}}_k)}$

Then the soft information is processedHard decision is carried out to obtain a time domain signal estimation value of a transmitting terminalBit data ofThe decision formula is:

${\hat{s}}_k=\left\{\begin{array}{c}1,L\left({\hat{s}}_k\right)\&GreaterEqual;0\\ 0,L\left({\hat{s}}_k\right)<0\end{array}\right.$

(5b) for bit dataAnd performing recoding and remodulation to obtain an ideal reference signal R.

8. The method as claimed in claim 1, wherein the step (6) of calculating the receiver-side EVM comprises the steps of:

(6a) calculating EVM of each subframe of a receiving end in an uplink shared data channel by adopting an error vector magnitude calculation formula_i：

${\mathrm{EVM}}_i=\frac{RMS\left(\right|Z-R\left|\right)}{RMS\left(\right|R\left|\right)}=\sqrt{\frac{\frac{1}{N}\underset{n\&Element;N}{\&Sigma;}|R_n-Z_n{|}^2}{\frac{1}{N}\underset{n\&Element;N}{\&Sigma;}|R_n{|}^2}}$

(6b) Calculating EVM of all sub-frame receiving ends under one frame_iObtaining the receiving end EVM by the average value of the data, wherein the realization formula is as follows:

$EVM=\sqrt{\frac{1}{M}\underset{i=1}{\overset{M}{\&Sigma;}}{\left({\mathrm{EVM}}_i\right)}^2}$

wherein i is the subframe number, and M is the total number of subframes.

9. The method of claim 1, wherein the step (7a) of mapping table between EVM and MCS comprises the steps of:

(7a1) initializing, and setting MCS as 1;

(7a2) performing off-line cyclic simulation on an uplink adaptive system model adopting a SISO AWGN channel, and finding out a value range of SNR (signal to noise ratio) which can enable BLER (block error ratio) to be uniformly distributed between 0 and 1 from a simulation result;

(7a3) performing off-line circulation simulation on an uplink adaptive system model adopting a SISO AWGN channel under each SNR in an SNR value range to obtain receiving end EVM and BLER corresponding to each circulation, and connecting the receiving end EVM and BLER to obtain a simulation curve between the receiving end EVM and the BLER;

(7a4) finding out and storing a receiving end EVM corresponding to BLER (0.1) on a simulation curve between the receiving end EVM and BLER, and simultaneously storing the current MCS;

(7a5) the steps (7a2) to (7a4) are repeated until MCS becomes 28, and a mapping table between the receiver EVM and the corresponding MCS is extracted from the stored plurality of receiver EVMs and MCSs.