CN102546036A - Compensation method of error vector magnitude (EVM) - Google Patents

Abstract

The invention provides a compensation method of error vector magnitude (EVM). The method comprises the following steps: according to a mathematical expression of a digital vector, analyzing technical parameters which are related to the EVM of vector demodulation, determining parameters which are generated after vector demodulation and influence the EVM, respectively calculating all the technical parameters influencing the EVM so as to get technical components influencing the EVM value, and calculating a theoretical estimated value of the EVM by square-sum-root calculation. According to the compensation method disclosed by the invention, the scientific and reasonable compensation of the EVM can be realized and a great effect of evaluating the measurement results of measuring the EVM by various vector signal analyzers can be realized. In the technical field of metering, the compensation method has greater practical value, plays a role in guiding technical development and production and further has the advantage of extensive application prospect.

Description

A kind of compensation method of Error Vector Magnitude

Technical field

When the present invention relates to the digital vector signal analysis and measuring to the theoretical calculation of Error Vector Magnitude.Be specifically related to a kind of theoretical calculation method of Error Vector Magnitude.

Background technology

In the continuous developing progress of digital communication technology, more and more deep for the analysis of digital vector signal measurement, wherein, be related to the reasonability of assessment of the measurement result for the measurement of Error Vector Magnitude (EVM) and simulation theory estimation.Because the vector analysis instrument that each apparatus manufacturer of the whole world is produced all is a comprehensive high-performance analytical equipment; Has multiple power of test; Also more complicated is formed in its inside; Therefore, also there is certain otherness in the measurement result for this index of Error Vector Magnitude (EVM) of measuring the digital vector signal.For credibility and confidential interval that can rational evaluation Error Vector Magnitude (EVM) result, especially crucial for the theoretical calculation of Error Vector Magnitude.

Summary of the invention

Problem to be solved by this invention is through VSA measure error amplitude of the vector EVM, and actual measured results and fiducial value are analyzed, and obtains the measurement result of Error Vector Magnitude.The present invention proposes a kind of compensation method of Error Vector Magnitude; This method adopts and sets out based on the I/Q baseband performance of vector analysis, through analyzing the technical performance of each digital vector demodulation parameter, accurately analyzes the amount of influence for the measure error amplitude of the vector; Analyze, decompose each demodulation link performance parameters of VSA step by step; Final realize clear, reasonably influence contributor component source, thereby reach the theoretical calculation of Error Vector Magnitude (EVM), implementation step is following:

Step 1, according to the numeral expression formula of digital vector, analyze the related technical parameter of vector demodulation Error Vector Magnitude (EVM).

Step 2, confirm the parameter that influences Error Vector Magnitude that produced after the vector demodulation.

Step 3, each technical parameter that will influence Error Vector Magnitude calculate respectively, obtain the technological component of image error amplitude of the vector value.

Step 4, the theoretical calculation value through side and root error of calculation amplitude of the vector.

The invention solves the theoretical calculation method problem of Error Vector Magnitude in the vector analysis, realized the evaluation of scientific and reasonable Error Vector Magnitude, good effect has been played in the evaluation of the measurement result of all kinds of VSA measure error amplitude of the vectors.In field of measuring techniques, have bigger practical value, and to technology development with produce the effect of the directiveness that plays, be with a wide range of applications.

Description of drawings

The connection block diagram of Fig. 1 vector analysis appearance measure error amplitude of the vector

Embodiment

In practical application; VSA carries out digital sample after being mixed to intermediate frequency to the RF digital vector signal, resolves into the I road and the Q road signal of two-way quadrature through quadrature demodulator; Again through means such as amplitude and phase compensations; Through a series of data processing and calculating, I road and Q way word base-band information are shown respectively on time domain and frequency domain respectively, also can see its radio spectrum and demodulation parameter and value.Measure the digital vector signal through VSA, confirm each parameter value of demodulation, the technical performance of analyzing, decomposing each demodulation parameter of VSA step by step, the theoretical valuation of error of calculation amplitude of the vector, concrete steps are described below.As shown in Figure 1, be the connection block diagram of vector analysis appearance measure error amplitude of the vector.

Step 1, according to the numeral expression formula of digital vector, analyze the related technical parameter of vector demodulation Error Vector Magnitude (EVM);

Step 2, confirm the parameter that influences Error Vector Magnitude that produced after the vector demodulation;

Confirm the parameter that influences the Error Vector Magnitude measured value that produced after the vector demodulation.Spuious and parameters compositions such as nonlinear distortion, harmonic distortion in main I/Q imbalance, parasitic amplitude modulation, phase noise, signal to noise ratio snr, the band;

Step 3, each technical parameter that will influence Error Vector Magnitude calculate respectively, obtain the technological component of image error amplitude of the vector value;

(1) I/Q is uneven

The disequilibrium of modulator and demodulator has comprised unbalance in phase and amplitude imbalance.Owing in the I/Q modulator analytic function of VSA, comprised content such as signal filtering, compensation in demodulating process, make the signal optimization after the I/Q demodulation, its disequilibrium is minimum.

Citing document is analyzed modulator and demodulator for the influence of amplitude imbalance α and unbalance in phase φ, amplitude imbalance and unbalance in phase is made a distinction, and distinguished the unbalance in phase amount on I road and Q road.Calculate through series of theories, the EVM component of introducing can resolve into amplitude and two components of phase place calculate.Be formulated as:

Wherein, α is the amplitude imbalance component,

φ is the unbalance in phase component.

For this imbalance problem, can these imbalances not minimized with matching through indemnifying measure.At present, the product of comparative maturity, the disequilibrium of modulator can be accomplished very little, reaches the EVM value less than 0.05% degree.

(2) parasitic amplitude modulation

We know that in the noise for signal source, the main noise of existence is AM noise and phase noise, below analyze earlier the influence of AM noise to EVM.

Parasitic amplitude modulation is to have occurred the amplitude modulation(PAM) that should not occur on the carrier signal, and the carrier signal of the original permanent width of cloth has been become the carrier signal that has changes in amplitude, and therefore, the amplitude of modulation signal is referred to as " parasitic amplitude modulation " with the ratio of carrier amplitude.

In the noise for signal source, can know that with reference to domestic and international technical data AM noise is smaller, occupy the main status of signal source noise with the phase modulation noise, so when considering the EVM source, the component that parasitic amplitude modulation causes can be ignored.

From the method for VSA demodulation digital vector signal, owing to taked means such as amplitude compensation, in the signal source noise, the influence of AM noise extremely is little, can ignore.As shown in Figure 5, when we were the NADC standardized digital signal of 870MHz at the measurement carrier frequency, can measure parasitic amplitude modulation was 0.00dB, EVM ₂=0.

(3) signal to noise ratio and phase noise

Under the negligible situation of the parasitic amplitude modulation of in preamble, mentioning, can think does not have amplitude modulation(PAM), i.e. constant envelope signal, and therefore, we mainly consider the influence of phase noise for EVM.

In the system that transmitter and receiver are formed, Error Vector Magnitude EVM and phase place, signal to noise ratio are directly related, drill the apply-official formula in the citing document 1:

${\mathrm{EVM}}_3^2=\frac{1}{\mathrm{SNR}}+2[1-\cos \left({\mathrm{\Φ}}_c\right)]---\left(2\right)$

Can know the relation of signal to noise ratio, phase modulation phase place and Error Vector Magnitude EVM from formula (2).

(a) signal to noise ratio

Signal to noise ratio is our desired signal and the ratio of background noise.Suppose not have phase modulation phase place Ф _cThe time, signal to noise ratio can be expressed as with mathematic(al) representation the influence of EVM:

${\mathrm{EVM}}_{31}=\sqrt{\frac{1}{\mathrm{SNR}}}\×100\%---\left(3\right)$

Wherein, SNR is a signal to noise ratio.

Usually in the actual signal test, each interchannel always has certain bandwidth, and signal is big more with the ratio of noise in bandwidth, and signal to noise ratio is good more so, and the ability to accept of receiver is strong more; Otherwise signal reception is poor more.In the practical application, can reduce the noise level amplitude, signal receiving performance power is improved through means such as filtering.When we utilized VSA to measure the digital vector signal, also energy measurement obtained the such test value of SNR.

(b) phase noise

Suppose that under the good situation of signal to noise ratio snr, the phase modulation phase place can be expressed as with mathematic(al) representation the influence of EVM so:

${\mathrm{EVM}}_{32}=\sqrt{2-2\cos \left[\mathrm{\θ}\left(t\right)\right]}---\left(4\right)$

Wherein, θ (t) is the phase modulation phase place.

Because Measurement Phase is very difficult thing under the situation of digital modulation signals; Can know that according to document 1 relation between phase jitter mean-square value

and the monolateral relatively band power spectral density L (f) is:

${\left[\mathrm{\θ}\left(t\right)\right]}^2={\mathrm{\σ}}_{\mathrm{\Δ\Φ}}^2=2{\∫}_{f_2}^{f_1}L\left(f\right)\mathrm{df}---\left(5\right)$

Can convert monolateral relatively band power spectral density to the phase jitter mean-square value through formula (5); Can easily phase noise be changed into phase jitter; Be phase problem, thus can using formula (4), just can calculate the influence amount of phase modulation phase place for EVM.

In the formula (5), L (f) is monolateral band phase noise spectral density, can test the parameter that obtains easily through our existing equipment.In our actual measurement, can measure the phase noise of the CF signal of digital vector signal source through testing equipments such as phase noise measurement system such as 3048A, FSUP, PN9000.

So ${\mathrm{EVM}}_3=\sqrt{{\mathrm{EVM}}_{31}^2+{\mathrm{EVM}}_{32}^2}---\left(6\right)$

The EVM component that signal to noise ratio and phase noise are introduced can calculate through formula (6).

(4) carrier wave leakage

Carrier wave leakage is an average power signal and the ratio of the carrier power of leakage; With symbol

expression, unit is dB.List of references 2, carrier wave leakage can be expressed as with mathematic(al) representation the influence of EVM:

${\mathrm{EVM}}_4=\frac{|C+S-S|}{S}={10}^{-\mathrm{\ξ}/20}\×100\%---\left(7\right)$

Wherein, S is the mean square power value of signal,

C is a carrier wave leakage power RMS effective value.

Usually in the actual signal test, can take various technological means to make carrier wave leakage as far as possible little, still, always can have this part energy.The EVM component of introducing can calculate through formula (7).

The EVM component that carrier wave leakage is introduced can calculate through formula (7).

(5) nonlinear distortion

In actual measurement, all be to use high performance VSA to carry out the measurement of digital vector signal, nonlinear distortions such as the distortion of its third order intermodulation, harmonic distortion are all smaller.At the power 20dB that the corresponding input power of three rank cut-off points of active device is imported greater than us, non-linear influence to EVM is ignored, i.e. EVM ₅≈ 0%.

Step 4, the theoretical calculation value through side and root error of calculation amplitude of the vector.

Total Error Vector Magnitude EVM can be approximately equal to the root of the quadratic sum of each EVM amount of influence, that is:

Claims (1)

1. the compensation method of an Error Vector Magnitude is characterized in that:

Step 1, according to the numeral expression formula of digital vector, analyze the related technical parameter of vector demodulation Error Vector Magnitude (EVM);

Step 2, confirm the parameter that influences the Error Vector Magnitude measured value that produced after the vector demodulation, comprise spuious and parameters such as nonlinear distortion, harmonic distortion in main I/Q imbalance, parasitic amplitude modulation, phase noise, signal to noise ratio snr, the band;

Step 3, each technical parameter that will influence Error Vector Magnitude calculate respectively, obtain the technological component of image error amplitude of the vector value, and its process is following:

(1) I/Q is uneven

In demodulating process, give signal filtering, compensation, make the signal optimization after the I/Q demodulation, its disequilibrium is minimum.

Distinguish the unbalance in phase amount on I road and Q road, calculate through series of theories, the EVM component of introducing can resolve into amplitude and two components of phase place calculate, and is formulated as:

Wherein, α is the amplitude imbalance component,

φ is the unbalance in phase component.

(2) parasitic amplitude modulation

The carrier signal of the original permanent width of cloth is become the carrier signal that has changes in amplitude, and amplitude modulation is 0.00dB, EVM ₃=0;

(3) signal to noise ratio and phase noise are quoted the apply-official formula of drilling:

${\mathrm{EVM}}_3^2=\frac{1}{\mathrm{SNR}}+2[1-\cos \left({\mathrm{\Φ}}_c\right)]---\left(2\right)$

Can know the relation of signal to noise ratio, phase modulation phase place and Error Vector Magnitude EVM from formula (2).

(a) signal to noise ratio

Signal to noise ratio is expressed as with mathematic(al) representation the influence of EVM:

${\mathrm{EVM}}_{31}=\sqrt{\frac{1}{\mathrm{SNR}}}\×100\%---\left(3\right)$

Wherein, SNR is a signal to noise ratio.

(b) phase noise is expressed as with mathematic(al) representation:

${\mathrm{EVM}}_{32}=\sqrt{2-2\cos \left[\mathrm{\θ}\left(t\right)\right]}---\left(4\right)$

Wherein, θ (t) is the phase modulation phase place.

Relation between phase jitter mean-square value

and the monolateral relatively band power spectral density L (f) is:

${\left[\mathrm{\θ}\left(t\right)\right]}^2={\mathrm{\σ}}_{\mathrm{\Δ\Φ}}^2=2{\∫}_{f_2}^{f_1}L\left(f\right)\mathrm{df}---\left(5\right)$

Can convert monolateral relatively band power spectral density to the phase jitter mean-square value through formula (5); In the formula (5); L (f) is monolateral band phase noise spectral density, can test the parameter that obtains easily through our existing equipment, and the EVM component that signal to noise ratio and phase noise are introduced can pass through formula

${\mathrm{EVM}}_3=\sqrt{{\mathrm{EVM}}_{31}^2+{\mathrm{EVM}}_{32}^2}---\left(6\right)$

(4) carrier wave leakage

Carrier wave leakage is an average power signal and the ratio of the carrier power of leakage; Represent with symbol ; Unit is dB, and carrier wave leakage adopts mathematic(al) representation to be expressed as to the influence of EVM:

${\mathrm{EVM}}_4=\frac{|C+S-S|}{S}={10}^{-\mathrm{\ξ}/20}\×100\%---\left(7\right)$

Wherein, S is the mean square power value of signal,

C is a carrier wave leakage power RMS effective value.

(5) nonlinear distortion

The input power of three rank cut-off points correspondences of active device is greater than the power 20dB of input, and non-linear influence to EVM is ignored, i.e. EVM ₅≈ 0%.

Step 4, the theoretical calculation value through side and root error of calculation amplitude of the vector.

Total Error Vector Magnitude EVM can be approximately equal to the root of the quadratic sum of each EVM amount of influence, that is:

Images