A channel estimation method and device for low-orbit satellite OFDM communication system

Through pilot extraction and transform domain processing, the channel estimation method of the OFDM communication system of the low-orbit satellite is optimized, solving the problem of insufficient accuracy of inter-satellite channel estimation, and improving receiver performance and resource utilization efficiency.

CN115883294BActive Publication Date: 2024-11-05THE 54TH RESEARCH INSTITUTE OF CHINA ELECTRONICS TECHNOLOGY GROUP CORPORATION +1
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202211508121.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-11-28
Publication Date
2024-11-05
Estimated Expiration
2042-11-28

AI Technical Summary

Technical Problem

The channel estimation accuracy of low-orbit satellite OFDM communication system is insufficient under the conditions of inter-satellite communication links, which affects receiver performance and resource utilization efficiency.

Method used

A channel estimation method is adopted, including pilot extraction, transform domain feature signal generation, initial denoising, energy empowerment and final channel estimation. The channel estimation results are optimized through conjugate multiplication, peak extraction window and energy mean calculation.

Benefits of technology

Improves the accuracy and robustness of channel estimation, especially under low signal-to-noise ratio conditions, and simplifies processing steps and is suitable for resource-limited hardware environments.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115883294B_ABST
    Figure CN115883294B_ABST
Patent Text Reader

Abstract

The present invention belongs to the field of low-orbit satellite communications, and discloses a channel estimation method and device for a low-orbit satellite OFDM communication system, which relates to the design technology of constellation payload gateways and terminals for low-orbit satellite mobile communication systems. Aiming at the characteristics of long transmission distance, low signal strength, limited signal-to-noise ratio, and the presence of frequency and time deviation in low-orbit satellite communications, the present invention performs an initial separation estimate of the noise by superimposing the energy window of the pilot symbols, and completes secondary iterative denoising through energy weighting, thereby realizing channel estimation of the received signal and completing signal demodulation. The present invention has the characteristics of satellite communication links such as strong demodulation capability, simple structure, and adaptability to large frequency and time deviations and different communication scenarios, and is particularly suitable for demodulation of low-orbit satellite mobile OFDM communication systems.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of low-orbit satellite communications, and relates to a low-orbit satellite OFDM communication system channel estimation method and device, which is particularly suitable for the high-performance estimation and demodulation requirements of the constellation payload of the low-orbit satellite mobile communication system and the ground terminal equipment for the receiver signal under the satellite-to-ground communication link conditions. Background Art

[0002] With the continuous evolution of satellite mobile communication system, low-orbit constellation satellite communication system has become an important development direction with its wide and flexible connectivity and global coverage. However, due to the long transmission distance, low signal strength and limited signal-to-noise ratio of low-orbit satellite communication, reliable transmission depends on the good demodulation performance of the receiver.

[0003] For low-orbit satellite OFDM communication systems, the lower the receiver demodulation threshold, the higher the transmission quality under the same transmission conditions, and the higher the transmission rate can be achieved; on the other hand, the stronger the demodulation capability of the receiver, the less resources and smaller antenna size can be used to achieve the same design performance. Whether it is the onboard payload or the ground terminal equipment, the application of more accurate channel estimation methods is conducive to obtaining better receiver performance. Therefore, a low-orbit satellite OFDM communication system channel estimation method and device that adapts to the conditions of the satellite-to-ground communication link is needed, which is particularly beneficial to the development and equipment deployment of my country's independent low-orbit satellite communication system. Summary of the invention

[0004] The technical problem to be solved by the present invention is to solve the problem in the background technology and achieve better receiver performance and resources. In order to adapt to the conditions of the satellite-to-ground communication link, the present invention proposes a channel estimation method and device for a low-orbit satellite OFDM communication system.

[0005] The technical solution adopted by the present invention is:

[0006] A low-orbit satellite OFDM communication system channel estimation method comprises the following steps:

[0007] A. Convert the received AD sampling signal into a frequency domain signal, extract the subcarrier signal occupied by the pilot, discard the non-pilot signal, and obtain the frequency domain characteristic signal of each pilot OFDM symbol;

[0008] B. Generate a local pilot signal according to preset parameters of the received pilot, calculate the conjugate multiplication result of the received pilot frequency domain characteristic signal and the local pilot frequency domain characteristic signal, and transform the conjugate multiplication result back to the time domain signal to obtain the transform domain response characteristic signal of each pilot OFDM symbol;

[0009] C. For the transform domain response characteristic signal of each pilot OFDM symbol, according to different communication scenarios, find the top N peaks with the largest energy, and extract energy with the superimposed peak extraction window to obtain the initial denoising result of the pilot in the transform domain; where N is the number of peaks in the characteristic energy domain that should be retained in the current communication scenario;

[0010] D. Calculate the difference between the transform domain characteristic signal of each pilot OFDM symbol and the initial denoising result, and calculate the energy mean of the difference result to obtain the initial estimated noise energy value;

[0011] E. For the initial denoising result of the pilot in the transform domain, the energy of the transform domain sample points that are less than the initial estimated noise energy value is set to zero, and the transform domain sample points that are greater than the initial estimated noise energy value are multiplied by the energy weight coefficient; and the secondary denoising result of the pilot in the transform domain is obtained;

[0012] F. The secondary denoising result is transformed into a frequency domain signal, the difference between the frequency domain signal and the conjugate multiplication result in step B is calculated, and the energy mean of the difference result is calculated to obtain the final estimated noise energy value. The final estimated noise energy value and the above frequency domain signal together constitute the channel estimation result of each pilot OFDM symbol.

[0013] Wherein, step C comprises:

[0014] C1. For the transform domain response characteristic signal of each pilot OFDM symbol, the real part and the imaginary part are squared respectively, and then the square results are added to obtain the characteristic energy of each point on the transform domain, and the points are arranged in order according to the energy size to obtain the arranged energy sequence;

[0015] C2. According to different communication scenarios and preset lookup tables, the number of peaks N in the characteristic energy domain that should be retained in the current communication scenario is obtained. For the first N peaks, a filter window shape is assigned with the point where the peak is located as the center, and all filter windows are superimposed in the transform domain to obtain a superimposed peak extraction window;

[0016] C3. In the transform domain, the transform domain response characteristic signal of each pilot OFDM symbol is multiplied by the peak extraction window to obtain the initial denoising result of the pilot in the transform domain.

[0017] Wherein, step E comprises:

[0018] E1. For the initial denoising result of the pilot in the transform domain, compare the energy value of each sample point in the transform domain with the result obtained in step D;

[0019] E2. For transform domain samples whose energy values ​​are less than the noise energy value estimated initially, set the energy to zero; for transform domain samples whose energy values ​​are greater than the noise energy value estimated initially, calculate the energy weight coefficient, multiply the real part and the imaginary part by the energy weight coefficient respectively, and compound them into a complex result;

[0020] E3. According to the output result of step E2, a secondary denoising result of the pilot signal in the transform domain is obtained.

[0021] A low-orbit satellite OFDM communication system channel estimation device, comprising: a pilot extraction module, a pilot transform domain characteristic signal generation module, a pilot primary denoising module, a noise energy primary estimation module, a pilot secondary denoising and energy weighting module and a channel estimation result output module;

[0022] The pilot extraction module is used to transform the received AD sampling signal into a frequency domain signal, extract the subcarrier signal occupied by the pilot, discard the non-pilot signal, obtain the frequency domain characteristic signal of each pilot OFDM symbol, and output it to the noise energy initial estimation module;

[0023] The pilot transform domain characteristic signal generation module is used to generate a local pilot signal according to the preset parameters of the received pilot, calculate the conjugate multiplication result of the received pilot frequency domain characteristic signal and the local pilot frequency domain characteristic signal, and transform the conjugate multiplication result back to the time domain signal to obtain the transform domain response characteristic signal of each pilot OFDM symbol, and output it to the pilot primary denoising module;

[0024] The pilot primary denoising module is used to calculate the top N peaks with the largest energy for the transform domain response characteristic signals of each pilot OFDM symbol according to different communication scenarios, and extract energy with a superimposed peak extraction window to obtain the primary denoising result of the pilot in the transform domain, and output it to the noise energy primary estimation module and the pilot secondary denoising and energy weighting module; where N is the number of peaks in the characteristic energy domain that should be retained in the current communication scenario;

[0025] The noise energy initial estimation module is used to calculate the difference between the transform domain characteristic signal of each pilot OFDM symbol and the initial denoising result, and calculate the energy mean of the difference result to obtain the initial estimated noise energy value, which is output to the pilot secondary denoising and energy weighting module;

[0026] The pilot secondary denoising and energy weighting module is used to set the energy of the transform domain sample points that are less than the initial estimated noise energy value to zero for the initial denoising result of the pilot in the transform domain, and multiply the transform domain sample points that are greater than the initial estimated noise energy value by the energy weight coefficient; obtain the secondary denoising result of the pilot in the transform domain, and output it to the channel estimation result output module;

[0027] The channel estimation result output module is used to transform the secondary denoising result into a frequency domain signal, calculate the difference between the frequency domain signal and the conjugate multiplication result in step B, and calculate the energy mean of the difference result to obtain the final estimated noise energy value. The final estimated noise energy value and the corresponding frequency domain signal together constitute the channel estimation result of each pilot OFDM symbol.

[0028] Compared with the prior art, the present invention has the following advantages:

[0029] The present invention has accurate estimation results, especially under the condition of low signal-to-noise ratio channel between satellite and ground, and optimizes the channel estimation effect through composite window weighting and iterative noise reduction processing of transform domain energy. At the same time, the processing methods of each step are combined in a simple multiplication and accumulation, division, lookup table and other ways, which are suitable for practical applications. The device structure of the present invention is simple and very suitable for application environments with limited hardware resources such as low-orbit satellites and user terminals. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 The basic flow chart of the method of the present invention is shown in FIG.

[0031] Figure 2 Schematic diagram of the specific process of step C of the present invention. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0033] A channel estimation method for a low-orbit satellite OFDM communication system, such as Figure 1 As shown, the following steps are included:

[0034] A. Convert the received AD sampling signal into a frequency domain signal, extract the subcarrier signal occupied by the pilot, discard the non-pilot signal, and obtain the frequency domain characteristic signal of each pilot OFDM symbol;

[0035] B. Generate a local pilot signal according to preset parameters of the received pilot, calculate the conjugate multiplication result of the received pilot frequency domain characteristic signal and the local pilot frequency domain characteristic signal, and transform the conjugate multiplication result back to the time domain signal to obtain the transform domain response characteristic signal of each pilot OFDM symbol;

[0036] C. For the transform domain response characteristic signal of each pilot OFDM symbol, according to different communication scenarios, find the first N peaks with the largest energy, and extract the energy with the superimposed peak extraction window to obtain the initial denoising result of the pilot in the transform domain; where N is the number of peaks in the characteristic energy domain that should be retained in the current communication scenario; Figure 2As shown, the specific steps include:

[0037] C1. For the transform domain response characteristic signal of each pilot OFDM symbol, square its real and imaginary parts respectively, and then add the square results to obtain the characteristic energy value P_rs of each sample point in the transform domain. trans (i)

[0038] P_rs trans (i) = [h_rs real (i)] 2 +[h_rs imag (i)] 2 ,

[0039] Where i = 1, 2, ..., K represents the index of each point in the transform domain, K represents the number of subcarriers occupied by the pilot, that is, the number of sampling points in the transform domain, h_rs real (i) represents the real part of the transform domain response characteristic signal at the i-th point, h_rs imag (i) represents the imaginary part of the transform domain response characteristic signal of the i-th point. Further, they are arranged in order of energy value;

[0040]

[0041] Where sort[] means to arrange the sequence from large to small, and get the arranged energy sequence P_sort trans (i).

[0042] C2. According to different communication scenarios and preset lookup tables, the number of peaks N in the characteristic energy domain that should be retained in the current communication scenario is obtained. For the first N peaks, the points where the peaks are located are taken as the center, and filter windows are respectively assigned. These windows are superimposed in the transform domain to obtain superimposed peak extraction windows.

[0043] For the nth peak, the location of its sample point is recorded as P n , its filter window W k (m) can be expressed as:

[0044]

[0045] Where m = 1, 2, ... K, W shape is the preset window sequence, N s Represents the length of the window sequence. Further, by superimposing these window shapes in the transform domain, the peak extraction window obtained can be expressed as:

[0046]

[0047] C3, multiplying the transform domain response characteristic signal of the pilot by the peak extraction window to obtain the initial denoising result h_dn1(i) of the pilot in the transform domain;

[0048] h_dn1(i)=h_dn 1,real (i)+j×h_dn 1,imag (i),

[0049] in, h_dn 1,real (i) = h_rs real (i)×W all (i) represents the real part of the initial denoising result, h_dn 1,imag (i) = h_rs imag (i)×W all (i) represents the imaginary part of the initial denoising result.

[0050] D. Calculate the difference between the transform domain characteristic signal of each pilot OFDM symbol and the initial denoising result, and calculate the energy mean of the difference result to obtain the initial estimated noise energy value;

[0051] E. For the initial denoising result of the pilot in the transform domain, the energy of the transform domain samples whose noise energy value is less than the initial estimate is set to zero, and the transform domain samples whose noise energy value is greater than the initial estimate is multiplied by the energy weight coefficient; and the secondary denoising result of the pilot in the transform domain is obtained; specifically comprising the following steps:

[0052] E1. For the initial denoising result of the pilot, compare the energy value of each sample point in its transform domain with the result obtained in step D to obtain the difference comparison result P diff_value (i)

[0053]

[0054] in, represents the energy value of each sample point of the initial denoising result of the pilot in the transform domain, P noise,trans (i) represents the initial estimated noise energy value obtained in step D.

[0055] E2, for the transform domain samples whose energy value of the initial denoising result is less than or equal to the noise energy value, set the corresponding transform domain response characteristic signal to zero;

[0056] h_dn2(i)=0,

[0057] Among them, i satisfies P diff_value (i)≤0

[0058] For transform domain samples that are greater than the noise energy value, calculate their energy weights, and the weight coefficients are:

[0059]

[0060] Multiply the real part and imaginary part by the weight coefficient respectively, and compound them into a complex result;

[0061] h_dn2(i)=h_dn 2,real (i)+j×h_dn 2,imag (i),

[0062] Among them, i satisfies P diff_value (i)>0,h_dn 2,real (i) = h_dn 1,real (i)×W p (i) represents the real part of the second denoising result, h_dn 2,imag (i) = h_dn 1,imag (i)×W p (i) represents the imaginary part of the second denoising result.

[0063] E3. According to the output of step E2, obtain the secondary denoising result h_dn2(i) of the pilot in the transform domain;

[0064] In summary, the secondary denoising result of the pilot in the transform domain can be expressed as:

[0065]

[0066] Among them, stP diff_value (i)>0 means that P diff_value (i)>0, the corresponding equation holds. Similarly, stP diff_value (i)≤0 means that P diff_value Under the condition (i)≤0, the corresponding equation holds.

[0067] F. The secondary denoising result is transformed into a frequency domain signal, the difference between the frequency domain signal and the conjugate multiplication result in step B is calculated, and the energy mean of the difference result is calculated to obtain the final estimated noise energy value. The final estimated noise energy value and the corresponding frequency domain signal together constitute the channel estimation result of each pilot OFDM symbol.

[0068] A low-orbit satellite OFDM communication system channel estimation device, comprising: a pilot extraction module, a pilot transform domain characteristic signal generation module, a pilot primary denoising module, a noise energy primary estimation module, a pilot secondary denoising and energy weighting module and a channel estimation result output module;

[0069] The pilot extraction module is used to transform the received AD sampling signal into a frequency domain signal, extract the subcarrier signal occupied by the pilot, discard the non-pilot signal, obtain the frequency domain characteristic signal of each pilot OFDM symbol, and output it to the noise energy initial estimation module;

[0070] The pilot transform domain characteristic signal generation module is used to generate a local pilot signal according to the preset parameters of the received pilot, calculate the conjugate multiplication result of the received pilot frequency domain characteristic signal and the local pilot frequency domain characteristic signal, and transform the conjugate multiplication result back to the time domain signal to obtain the transform domain response characteristic signal of each pilot OFDM symbol, and output it to the pilot primary denoising module;

[0071] The pilot primary denoising module is used to calculate the top N peaks with the largest energy for the transform domain response characteristic signals of each pilot OFDM symbol according to different communication scenarios, and extract energy with a superimposed peak extraction window to obtain the primary denoising result of the pilot in the transform domain, and output it to the noise energy primary estimation module and the pilot secondary denoising and energy weighting module; where N is the number of peaks in the characteristic energy domain that should be retained in the current communication scenario;

[0072] The noise energy initial estimation module is used to calculate the difference between the transform domain characteristic signal of each pilot OFDM symbol and the initial denoising result, and calculate the energy mean of the difference result to obtain the initial estimated noise energy value, which is output to the pilot secondary denoising and energy weighting module;

[0073] The pilot secondary denoising and energy weighting module is used to set the energy of the transform domain sample points that are less than the initial estimated noise energy value to zero for the initial denoising result of the pilot in the transform domain, and multiply the transform domain sample points that are greater than the initial estimated noise energy value by the energy weight coefficient; obtain the secondary denoising result of the pilot in the transform domain, and output it to the channel estimation result output module;

[0074] The channel estimation result output module is used to transform the secondary denoising result into a frequency domain signal, calculate the difference between the frequency domain signal and the conjugate multiplication result in step B, and calculate the energy mean of the difference result to obtain the final estimated noise energy value. The final estimated noise energy value and the corresponding frequency domain signal together constitute the channel estimation result of each pilot OFDM symbol.

[0075] Example:

[0076] In the OFDM system, a subframe consists of 8 time slots, each of which has 12 OFDM symbols, and its index value is defined as t sym =[0,1,2,…,11], where 2 OFDM symbols carry pilot information and their symbol index is t pilot =[2,9], the other 10 OFDM symbols carry data information.

[0077] The present invention is applied to the receiving end of the satellite payload and can also be used for the receiving end of the ground terminal.

[0078] The function of the pilot extraction module is to transform the received AD sampling signal into a frequency domain signal through FFT, and extract the subcarrier signal occupied by the pilot symbol to obtain the frequency domain characteristic signal of each pilot OFDM symbol. Wherein k=1, 2, ...K represents the kth subcarrier, and K is the number of subcarriers occupied by the pilot symbol.

[0079] The pilot transform domain characteristic signal generation module mainly obtains the transform domain response characteristic signal of each pilot symbol. According to the preset parameters of the received pilot, a local pilot signal is generated. Its conjugate signal is expressed as The received pilot frequency domain characteristic signal is multiplied by the local pilot conjugate signal to remove the modulation information, and the result is subjected to IFFT to obtain the transform domain response characteristic signal of each pilot OFDM symbol. Among them, i=1,2,…,K represents the index of each point on the transform domain. At the same time, the operation of the subsequent modules on each pilot OFDM symbol is the same. Therefore, taking the transform domain response characteristic signal h_rs(i) of the first pilot OFDM symbol as an example, the subsequent modules are explained.

[0080] The function of the pilot initial denoising module is to obtain the pilot initial denoising result in the transform domain based on the energy peak and peak extraction window of the pilot transform domain response characteristic signal. The specific steps are as follows:

[0081] (1) First, the real and imaginary parts of the transform domain response characteristic signal of the pilot are squared respectively, and then the square results are added to obtain the characteristic energy value P_rs of each point in the transform domain. trans (i) Further, the energy values ​​are arranged in order from large to small to obtain the arranged energy sequence P_sort trans (i)

[0082] (2) According to different communication scenarios and preset lookup tables, the number of characteristic energy peaks N that should be retained in the current communication scenario is obtained. For the nth peak, the position of the sample point where it is located is recorded as pos(n), and the point where the peak is located is taken as the center, and a filter window shape W is assigned. n (m), namely:

[0083]

[0084] Where m = 1, 2, ... K, W shape is the preset window sequence, N s Indicates the window sequence length. Further, the above filter window shapes are superimposed in the transform domain, and the peak extraction window obtained can be expressed as:

[0085]

[0086] (3) Multiply the transform domain response characteristic signal of the pilot by the peak extraction window to obtain the initial denoising result h_dn1(i) of the pilot in the transform domain;

[0087] h_dn1(i)=h_rs real (i)×W all (i)+j×[h_rs imag (i)×W all (i)].

[0088] The noise energy initial estimation module first compares the transform domain response characteristic signal of the pilot with its initial denoising result. diff (i) = h_rs(i)-h_dn1(i), and then calculate the energy mean based on the difference result to obtain the initial estimated noise energy value,

[0089] The pilot secondary denoising and energy weighting module first sets the energy of the transform domain samples that are less than or equal to the initial estimated noise energy value to zero, and multiplies the transform domain samples that are greater than the initial estimated noise energy value by the energy weight coefficient to obtain the pilot secondary denoising result h_dn2(i) in the transform domain:

[0090]

[0091] Among them, stP diff_value (i)>0 means that P diff_value (i)>0, the corresponding equation holds. Similarly, stP diff_value (i)≤0 means that P diff_value Under the condition (i)≤0, the corresponding equation holds.

[0092] The function of the channel estimation result output module is to transform the secondary denoising result of the pilot in the transform domain to the frequency domain through FFT, calculate the difference between the frequency domain signal and the conjugate multiplication result in step B, and calculate the energy mean of the difference result to obtain the final estimated noise energy value. The final estimated noise energy value and the corresponding frequency domain signal together constitute the channel estimation result of each pilot OFDM symbol.

Claims

1. A channel estimation method for a low-orbit satellite OFDM communication system. It is characterized in that The following steps are involved: A. Convert the received AD sampling signal into a frequency domain signal, extract the subcarrier signal occupied by the pilot, discard the non-pilot signal, and obtain the frequency domain characteristic signal of each pilot OFDM symbol; B. Generate a local pilot signal according to preset parameters of the received pilot, calculate the conjugate multiplication result of the received pilot frequency domain characteristic signal and the frequency domain characteristic signal of the local pilot signal, and transform the conjugate multiplication result back to the time domain signal to obtain the transform domain response characteristic signal of each pilot OFDM symbol; C. For the transform domain response characteristic signal of each pilot OFDM symbol, according to different communication scenarios, find the top N peaks with the largest energy, and extract energy with the superimposed peak extraction window to obtain the initial denoising result of the pilot in the transform domain; where N is the number of peaks in the characteristic energy domain that should be retained in the current communication scenario; D. Calculate the difference between the transform domain response characteristic signal of each pilot OFDM symbol and the initial denoising result, and calculate the energy mean of the difference result to obtain the initial estimated noise energy value; E. For the initial denoising result of the pilot in the transform domain, the energy of the transform domain sample points that are less than the initial estimated noise energy value is set to zero, and the transform domain sample points that are greater than the initial estimated noise energy value are multiplied by the energy weight coefficient; and the secondary denoising result of the pilot in the transform domain is obtained; F. The secondary denoising result is transformed into a frequency domain signal, the difference between the frequency domain signal and the conjugate multiplication result in step B is calculated, and the energy mean of the difference result is calculated to obtain the final estimated noise energy value. The final estimated noise energy value and the above frequency domain signal together constitute the channel estimation result of each pilot OFDM symbol.

2. A channel estimation method for a low-orbit satellite OFDM communication system according to claim 1, It is characterized in that Step C includes: C1. For the transform domain response characteristic signal of each pilot OFDM symbol, the real part and the imaginary part are squared respectively, and then the square results are added to obtain the characteristic energy of each point on the transform domain, and the points are arranged in order according to the energy size to obtain the arranged energy sequence; C2. According to different communication scenarios and preset lookup tables, the number of peaks N in the characteristic energy domain that should be retained in the current communication scenario is obtained. For the first N peaks, a filter window shape is assigned with the point where the peak is located as the center, and all filter windows are superimposed in the transform domain to obtain a superimposed peak extraction window; C3. In the transform domain, the transform domain response characteristic signal of each pilot OFDM symbol is multiplied by the peak extraction window to obtain the initial denoising result of the pilot in the transform domain.

3. The channel estimation method for a low-orbit satellite OFDM communication system according to claim 1, It is characterized in that Step E includes: E1. For the initial denoising result of the pilot in the transform domain, compare the energy value of each sample point in the transform domain with the result obtained in step D; E2. For transform domain samples whose energy values ​​are less than the noise energy value estimated initially, set the energy to zero; for transform domain samples whose energy values ​​are greater than the noise energy value estimated initially, calculate the energy weight coefficient, multiply the real part and the imaginary part by the energy weight coefficient respectively, and compound them into a complex result; E3. According to the output result of step E2, a secondary denoising result of the pilot signal in the transform domain is obtained.

4. A channel estimation device for a low-orbit satellite OFDM communication system, It is characterized in that include: A pilot extraction module, a pilot transform domain characteristic signal generation module, a pilot primary denoising module, a noise energy primary estimation module, a pilot secondary denoising and energy weighting module, and a channel estimation result output module; The pilot extraction module is used to transform the received AD sampling signal into a frequency domain signal, extract the subcarrier signal occupied by the pilot, discard the non-pilot signal, obtain the frequency domain characteristic signal of each pilot OFDM symbol, and output it to the noise energy initial estimation module; The pilot transform domain characteristic signal generation module is used to generate a local pilot signal according to the preset parameters of the received pilot, calculate the conjugate multiplication result of the received pilot frequency domain characteristic signal and the frequency domain characteristic signal of the local pilot signal, and transform the conjugate multiplication result back to the time domain signal to obtain the transform domain response characteristic signal of each pilot OFDM symbol, and output it to the pilot primary denoising module; The pilot primary denoising module is used to calculate the top N peaks with the largest energy for the transform domain response characteristic signals of each pilot OFDM symbol according to different communication scenarios, and extract energy with a superimposed peak extraction window to obtain the primary denoising result of the pilot in the transform domain, and output it to the noise energy primary estimation module and the pilot secondary denoising and energy weighting module; where N is the number of peaks in the characteristic energy domain that should be retained in the current communication scenario; The noise energy initial estimation module is used to calculate the difference between the transform domain response characteristic signal of each pilot OFDM symbol and the initial denoising result, and calculate the energy mean of the difference result to obtain the initial estimated noise energy value, which is output to the pilot secondary denoising and energy weighting module; The pilot secondary denoising and energy weighting module is used to set the energy of the transform domain sample points that are less than the initial estimated noise energy value to zero for the initial denoising result of the pilot in the transform domain, and multiply the transform domain sample points that are greater than the initial estimated noise energy value by the energy weight coefficient; obtain the secondary denoising result of the pilot in the transform domain, and output it to the channel estimation result output module; The channel estimation result output module is used to transform the secondary denoising result into a frequency domain signal, calculate the difference between the frequency domain signal and the conjugate multiplication result in step B, and calculate the energy mean of the difference result to obtain the final estimated noise energy value. The final estimated noise energy value and the corresponding frequency domain signal together constitute the channel estimation result of each pilot OFDM symbol.

Citation Information

Patent Citations

  • Channel estimation method and device in multiple input multiple output OFDM (Orthogonal Frequency Division Multiplexing) system

    CN101848183A

  • Carrier synchronization method of low-earth-orbit satellite broadband OFDM communication system

    CN111147123A

  • Frequency offset estimation method for decimal subcarrier spacing of low-orbit broadband satellite system

    CN113098819A

  • Satellite mobile OFDM communication system channel estimation method and device

    CN113824666A

  • Method and device of channel estimation for OFDM system

    US20110268206A1