A method for reducing background noise based on spectrum analyzer

Abstract

The present invention relates to a noise reduction method based on a spectrum analyzer, which belongs to the field of electronic test and measurement, including: performing spectrum analyzer noise reduction modeling to obtain the noise power NP read on the spectrum analyzer interface; calculating the back-end noise power density N2R; setting the center frequency of the spectrum analyzer to the starting value of the current frequency conversion channel, obtaining the parameter compensation value corresponding to the center frequency, reading the noise power NP3, obtaining the front-end noise N1 and the front-end noise density N1R; sequentially stepping from the frequency starting value to the end value of the current frequency conversion channel, and then changing the frequency conversion channel of the spectrum analyzer to obtain the noise parameters of the spectrum analyzer in all cases; obtaining the displayed average noise level by N1R and N2R obtained by the current setting parameters of the spectrum analyzer for the specific frequency conversion channel and the center frequency, and calculating the input signal power after eliminating the noise. The present invention can effectively eliminate the influence of the noise of the spectrum analyzer itself on the measurement and search for small signals.

Description

Technical Field

[0001] The invention relates to the field of electronic testing and measurement, and in particular to a background noise reduction method based on a spectrum analyzer. Background Art

[0002] One of the main functions of a spectrum analyzer is to find and measure small signals, and the main limitation is the noise generated by the spectrum analyzer itself. When the signal power is close to the noise power, the measurement result will be affected by the noise. In order to reduce the noise floor of the spectrum analyzer, the commonly used methods on the instrument are to reduce the RBW (resolution bandwidth), adjust the internal attenuator to 0dB, use adaptive filters to reduce noise, etc. However, reducing the RBW will greatly increase the measurement time and reduce the overall test efficiency; adjusting the internal attenuator to 0dB will make the original matching worse and reduce the accuracy of the measured amplitude value; and the adaptive filter noise reduction method also has the problem of slow convergence time, which greatly increases the measurement time. Summary of the invention

[0003] The purpose of the present invention is to overcome the shortcomings of the prior art, provide a noise reduction method based on a spectrum analyzer, and solve the deficiencies of the prior art.

[0004] The object of the present invention is achieved by the following technical solution: a noise reduction method based on a spectrum analyzer, the noise reduction method comprising:

[0005] Step 1: Calibrate the spectrum analyzer, measure the parameters that affect the background noise, and build a model for the background noise of the spectrum analyzer based on the obtained parameters. Calculate the noise power NP read on the spectrum analyzer interface based on the modeling results.

[0006] Step 2: Change the parameters configured in the hardware of the spectrum analyzer to obtain the back-end noise power density N2R;

[0007] Step 3: Set the center frequency of the spectrum analyzer to the starting value of the current frequency conversion channel, obtain the parameter compensation value of the corresponding center frequency obtained by the previous whole machine calibration, and read the noise power NP3 under the current conditions to obtain the front-end noise N1 and front-end noise density N1R related to the center frequency;

[0008] Step 4: First, step from the frequency starting value to the ending value of the current frequency conversion channel in sequence, then change the frequency conversion channel of the spectrum analyzer to obtain the noise parameters of the spectrum analyzer in all cases;

[0009] Step 5: Determine the current frequency conversion channel and the frequency range that needs to be noise-reduced based on the set spectrum analyzer interface parameters, and use this to find the previously recorded N1R and N2R to obtain the displayed average noise level DANL, and calculate the input signal power after eliminating the background noise of the spectrum analyzer itself.

[0010] The step 4 specifically includes the following contents:

[0011] A1. Change the center frequency of the spectrum analyzer from the starting value to the ending value of the frequency of the current frequency conversion channel according to the set step, repeat the operation of step 3, and record the N1R and N2R corresponding to each center frequency of the current frequency conversion channel;

[0012] A2. Change the frequency conversion channel of the spectrum analyzer, and operate from step 3 to step A1 again. Record the results completely according to each frequency conversion channel, a series of center frequencies corresponding to the frequency conversion channel, and N1R and N2R corresponding to the center frequencies to obtain the noise parameters of the spectrum analyzer in all cases.

[0013] The parameters affecting the background noise include the mechanical attenuator attenuation value MATT, the electronic attenuator attenuation value EATT, the preamplifier amplification value PreAmp, the tuning frequency amplitude correction Flatness, the system gain SystemGain, the intermediate frequency attenuation AifAtt and the resolution bandwidth RBW.

[0014] The noise power NP read on the spectrum analyzer interface calculated according to the modeling situation includes:

[0015] According to the modeling situation, the formula for calculating the noise power NP read on the spectrum analyzer interface is: ;

[0016] Divide both sides of the formula by get , where NPM is the noise power at the mixer, that is, the noise power read from the spectrum analyzer interface divided by The result obtained.

[0017] The step 2 includes the following contents:

[0018] Change the spectrum analyzer configuration to the hardware intermediate frequency attenuation AifAtt, obtain the first intermediate frequency attenuation value AifAtt1 and the second intermediate frequency attenuation value AifAtt2, set , ;

[0019] The noise power read from the spectrum analyzer interface when the intermediate frequency attenuation is the first intermediate frequency attenuation value AifAtt1 is divided by The result of npm1, , and the noise power read from the spectrum analyzer interface when the intermediate frequency attenuation is the second intermediate frequency attenuation value AifAtt2 divided by The result of npm2, ;

[0020] Joint formula and The back-end noise power N2 is obtained as , and normalize it to get the back-end noise power density N2R: , where EnbwRatio is the equivalent noise bandwidth ratio of the resolution bandwidth filter.

[0021] The step three includes the following contents:

[0022] Set the center frequency of the spectrum analyzer to the starting value of the current frequency conversion channel, obtain the Flatness3 and AifAtt3 compensation values ​​corresponding to the starting value frequency obtained from the previous machine calibration, and then read the noise power NP3 under the current conditions from the spectrum analyzer interface and divide it by The result is expressed as NPM3, and set ;

[0023] Then we can get the front-end noise related to the center frequency. and front-end noise density .

[0024] The step five includes the following contents:

[0025] According to the set spectrum analyzer interface parameters, determine the current frequency conversion channel and the frequency range that needs to be noise reduced, and use this to find the N1R and N2R recorded previously, and bring them into the formula The ratio of the noise power at the mixer to the equivalent bandwidth of RBW, NPMR, is calculated as , and then according to the current RBW value, the NPM at this time is ;

[0026] According to the current mechanical attenuator attenuation value MATT, electronic attenuator attenuation value EATT, and preamplifier amplification value PreAmp of the spectrum analyzer, DANL is calculated as follows: ;

[0027] Display signal power according to the original interface and Calculate the input signal power after eliminating the background noise of the spectrum analyzer itself for .

[0028] When the spectrum analyzer parameters change, the process returns to step 5 to recalculate the input signal power under the current parameter conditions.

[0029] The present invention has the following advantages: a noise reduction method based on a spectrum analyzer can effectively eliminate the influence of the spectrum analyzer's own noise on the measurement and search for small signals, and can also effectively improve the signal-to-noise ratio when measuring low-level signals and the overall sensitivity of the spectrum analyzer. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 A schematic diagram of the noise model structure of a spectrum analyzer constructed in the present invention;

[0031] Figure 2 The signal power curve before and after noise reduction. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the detailed description of the embodiments of the present application provided below in conjunction with the drawings is not intended to limit the scope of protection of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application. The present invention is further described below in conjunction with the drawings.

[0033] The present invention specifically relates to a noise reduction method based on a spectrum analyzer, which divides the noise components of the spectrum analyzer into front-end noise (FrontEnd Noise) and back-end noise (BackEnd Noise), wherein the front-end noise is the input noise related to the center frequency of the spectrum analyzer, and the back-end noise is the noise independent of the frequency; specifically, it includes the following contents:

[0034] S1. Calibrate the spectrum analyzer itself and measure the important parameters that will affect the background noise. These parameters include: mechanical attenuator attenuation value MATT, electronic attenuator attenuation value EATT, preamplifier amplification value PreAmp, tuning frequency amplitude correction Flatness, system gain SystemGain, intermediate frequency attenuation AifAtt, and resolution bandwidth RBW. Combined with the parameters measured after the above calibration, the background noise of the spectrum analyzer is modeled, including the model when the configuration and working conditions of the spectrum analyzer change. The specific model is as follows: Figure 1 As shown in the figure, after completing the spectrum analyzer noise floor modeling, we actually calculate the front-end and rear-end noise on the spectrum analyzer. The steps are as follows:

[0035] A. Disconnect the front end of the spectrum analyzer from the signal, select the frequency conversion channel, set the center frequency to the center value of the current frequency conversion channel, set the spectrum analyzer span to zero, the detector to the average detector, the average mode to power average, the display points to 1001 points, and set the values ​​of MATT, EATT, PreAmp and RBW;

[0036] B. Get the corresponding Flatness, SystemGain, and AifATT compensation values ​​in the current spectrum analyzer.

[0037] C. Obtain the background noise power of the spectrum analyzer under the current situation, corresponding to Figure 1 Noise Power in

[0038] D. From the above model, we can get the noise power NP (Noise Power):

[0039] (1),

[0040] Where N1 is the front-end noise power FrontEnd Noise, N2 is the back-end noise power BackEndNoise; the noise power NP is reversely calculated to the mixer Mixer, and both ends are divided by , we get the following formula:

[0041] (2),

[0042] Where NPM is the noise power at the mixer, which is the noise power NP read from the spectrum analyzer interface divided by The result obtained.

[0043] E. Change the spectrum analyzer configuration to the hardware intermediate frequency attenuation AifAtt, obtain AifAtt1 and AifAtt2, and use , NPM1 is the noise power value read from the spectrum analyzer interface when the intermediate frequency attenuation is set to AifAtt1 divided by The result obtained, NPM2 is the noise power value read from the spectrum analyzer interface when the intermediate frequency attenuation is set to AifAtt2 divided by The results obtained can transform formula (2) into the following two formulas:

[0044] (3),

[0045] (4),

[0046] The back-end noise power can be calculated from formulas (3) and (4):

[0047] (5),

[0048] Considering the influence of the resolution bandwidth RBW in the digital intermediate frequency on the measured noise power, it is normalized to obtain the back-end noise power density:

[0049] (6),

[0050] where EnbwRatio is the equivalent noise bandwidth ratio of the resolution bandwidth filter.

[0051] F. Set the center frequency of the spectrum analyzer to the starting value of the current frequency conversion channel, and obtain the frequency corresponding to the starting value obtained from the previous calibration of the whole machine. , Compensation value. Note that the Flatness and AifAtt above are the center values ​​of the corresponding current frequency conversion channel, so they are not the same here. SystemGain is only related to the frequency conversion channel, so it remains unchanged. Then read the noise power value NP3 of the current condition from the spectrum analyzer interface and divide it by Get the result NPM3, and , N2R has been obtained before. Substituting them into formula (2), combined with N2 obtained by formula (6), the front-end noise related to the center frequency can be calculated:

[0052] (7),

[0053] Normalize it to get the front-end noise power density corresponding to the center frequency:

[0054] (8),

[0055] G. Change the center frequency of the spectrum analyzer in a certain step from the starting value to the ending value of the frequency of the current frequency conversion channel, repeat the operation in step F, and record the N1R and N2R corresponding to each center frequency value of the current frequency conversion channel.

[0056] H. Change the frequency conversion channel of the spectrum analyzer, and perform the operation from step A to step G again. The results are completely recorded according to each frequency conversion channel, a series of center frequency values ​​corresponding to the frequency conversion channel, and the front-end and rear-end noise power densities N1R and N2R corresponding to the center frequency values. In this way, the noise parameters of this spectrum analyzer in all cases are obtained.

[0057] S2. In actual use, the current frequency conversion channel and the frequency range that needs to be noise-reduced are determined according to the spectrum analyzer interface parameters set by the user, and the N1R and N2R recorded previously are found and substituted into formula (2). The ratio NPMR of the noise power at the mixer to the equivalent bandwidth of the RBW can be calculated, and then NPM can be obtained according to the current RBW value, where x is the frequency range within which noise reduction is required, and the formula is as follows:

[0058] (9),

[0059] (10),

[0060] According to the current parameters of the spectrum analyzer: mechanical attenuator attenuation value MATT, electronic attenuator attenuation value EATT, preamplifier amplification value PreAmp, DANL (displayed average noise level) can be calculated:

[0061] (11),

[0062] S3. Display signal power according to the original interface DANL(x) is used to obtain the input signal power after eliminating the background noise of the spectrum analyzer itself. :

[0063] (12),

[0064] S4. If the parameters of the spectrum analyzer change, it should return to S2 to recalculate the noise floor power under the current parameter conditions. The original signal power measured by the present invention is superimposed with the noise floor power of the spectrum analyzer itself. Therefore, the signal power value of each display point after eliminating the influence of the noise floor of the spectrum analyzer itself can be obtained by subtracting the calculated noise floor power of the spectrum analyzer itself from the original signal power. In this way, the noise floor level displayed on the interface will be reduced, and the real signal power can be obtained. The effect is as follows: Figure 2 As shown in the figure, the input center frequency is 1G, the amplitude is -107dbm. Before noise reduction, the signal power is measured to be -104dbm because the noise is superimposed on the signal. After noise reduction, the correct signal power is measured to be -107dbm, and its average noise floor level is reduced from -107dbm to -117dbm. It can be seen that this method has a good effect in reducing the noise floor of the spectrum analyzer.

[0065] Furthermore, to achieve the best accuracy, some kind of averaging is required, such as setting a VBW (video filter bandwidth) narrower than the selected RBW, using average detection over long sweeps, trace averaging, etc. The more averaging operations are performed, the more accurate the result will be.

[0066] The above is only a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used for various other combinations, modifications and improvements, and can be modified within the scope of the concept described herein through the above teachings or the technology or knowledge of the relevant field. The changes and modifications made by those skilled in the art do not deviate from the spirit and scope of the present invention, and should be within the scope of protection of the claims attached to the present invention.

Claims

1. A noise reduction method based on a spectrum analyzer, characterized in that: The noise reduction method comprises: Step 1: Calibrate the spectrum analyzer, measure the parameters that affect the background noise, and build a model for the background noise of the spectrum analyzer based on the obtained parameters. Calculate the noise power NP read on the spectrum analyzer interface based on the modeling results. Step 2: Change the parameters configured in the hardware of the spectrum analyzer to obtain the back-end noise power density N2R; Step 3: Set the center frequency of the spectrum analyzer to the starting value of the current frequency conversion channel, obtain the parameter compensation value of the corresponding center frequency obtained by the previous whole machine calibration, and read the noise power NP3 under the current conditions to obtain the front-end noise N1 and front-end noise density N1R related to the center frequency; Step 4: First, step from the frequency starting value to the ending value of the current frequency conversion channel in sequence, then change the frequency conversion channel of the spectrum analyzer to obtain the noise parameters of the spectrum analyzer in all cases; Step 5: Determine the current frequency conversion channel and the frequency range that needs to be noise-reduced according to the set spectrum analyzer interface parameters, and use this to find the recorded N1R and N2R, obtain the displayed average noise level DANL, and calculate the input signal power after eliminating the background noise of the spectrum analyzer itself; The parameters affecting the background noise include mechanical attenuator attenuation value MATT, electronic attenuator attenuation value EATT, preamplifier amplification value PreAmp, tuning frequency amplitude correction Flatness, system gain SystemGain, intermediate frequency attenuation AifAtt and resolution bandwidth RBW; The noise power NP read on the spectrum analyzer interface calculated according to the modeling situation includes: According to the modeling situation, the formula for calculating the noise power NP read on the spectrum analyzer interface is: , N1 is the front-end noise, N2 is the back-end noise; Divide both sides of the formula by get , where NPM is the noise power at the mixer, that is, the noise power read from the spectrum analyzer interface divided by The results obtained; The step 2 includes the following contents: Change the spectrum analyzer configuration to the hardware intermediate frequency attenuation AifAtt, obtain the first intermediate frequency attenuation value AifAtt1 and the second intermediate frequency attenuation value AifAtt2, set , ; The noise power read from the spectrum analyzer interface when the intermediate frequency attenuation is the first intermediate frequency attenuation value AifAtt1 is divided by The result of NPM1, where , and the noise power read from the spectrum analyzer interface when the intermediate frequency attenuation is the second intermediate frequency attenuation value AifAtt2 divided by The result of npm2, where ; Joint formula and The back-end noise power N2 is obtained as , and normalize it to get the back-end noise power density N2R: , where EnbwRatio is the equivalent noise bandwidth ratio of the resolution bandwidth filter; The step three includes the following contents: Set the center frequency of the spectrum analyzer to the starting value of the current frequency conversion channel, obtain the Flatness3 and AifAtt3 compensation values ​​corresponding to the starting value frequency obtained from the previous machine calibration, and then read the noise power NP3 under the current conditions from the spectrum analyzer interface and divide it by The result is expressed as NPM3, and set , Flatness3 is the tuning frequency amplitude correction corresponding to the starting value frequency, and AifAtt3 is the intermediate frequency attenuation corresponding to the starting value frequency; Then we can get the front-end noise related to the center frequency. and front-end noise density ; The step five includes the following contents: According to the set spectrum analyzer interface parameters, the current frequency conversion channel and the frequency range that needs to be noise reduced are judged, and the recorded N1R and N2R are found and put into the formula The ratio of the noise power at the mixer to the equivalent bandwidth of RBW, NPMR, is calculated as , and then according to the current RBW value, the NPM at this time is ,in x is the frequency range within which noise reduction is required; According to the current mechanical attenuator attenuation value MATT, electronic attenuator attenuation value EATT, and preamplifier amplification value PreAmp of the spectrum analyzer, DANL is calculated as follows: ; Display signal power according to the original interface and Calculate the input signal power after eliminating the background noise of the spectrum analyzer itself for .

2. The method for reducing background noise based on a spectrum analyzer according to claim 1, characterized in that: The step 4 specifically includes the following contents: A1. Change the center frequency of the spectrum analyzer from the starting value to the ending value of the frequency of the current frequency conversion channel according to the set step, repeat the operation of step 3, and record the N1R and N2R corresponding to each center frequency of the current frequency conversion channel; A2. Change the frequency conversion channel of the spectrum analyzer, and operate from step 3 to step A1 again. Record the results completely according to each frequency conversion channel, a series of center frequencies corresponding to the frequency conversion channel, and N1R and N2R corresponding to the center frequencies to obtain the noise parameters of the spectrum analyzer in all cases.

3. A method for reducing background noise based on a spectrum analyzer according to claim 1 or 2, characterized in that: When the spectrum analyzer parameters change, the process returns to step 5 to recalculate the input signal power under the current parameter conditions.

Images