A wireless communication frequency offset compensation method in a super-high speed state

Abstract

This invention discloses a method for frequency offset compensation in wireless communication under hypersonic conditions, belonging to the field of frequency offset estimation technology. By simultaneously performing frequency offset estimation at different ranges, multiple frequency offset estimation results are obtained, and the final frequency offset estimation result is obtained by combining joint decisions to complete frequency offset compensation. When the method of this invention is implemented on an FPGA, its parallel processing structure can be used to significantly shorten the frequency offset compensation processing time and significantly improve the accuracy of frequency offset compensation.

Description

Technical Field

[0001] This invention belongs to the field of frequency offset estimation technology, and specifically relates to a method for frequency offset compensation in wireless communication under hypersonic conditions. Background Technology

[0002] In high-speed communication scenarios, the relative motion between the transmitting and receiving ends generates significant Doppler frequency offsets, preventing the receiver from correctly demodulating the signal. Traditional frequency offset estimation algorithms either have limited estimation ranges, failing to cover the large frequency offset range in high-speed scenarios, or employ long sequence correlations to ensure estimation accuracy under large frequency offsets, resulting in a multi-fold increase in hardware processing latency, failing to meet the low latency requirements of high-speed scenarios, and causing subsequent channel estimation and decoding modules to malfunction. Ultimately, this leads to a sharp drop in communication link throughput, a surge in bit error rate, and even link interruption. Moreover, the frequency offset change rate is also accelerated in high-speed scenarios, further compressing the time window for frequency offset estimation and tracking, placing higher demands on the algorithm's real-time performance and dynamic tracking capabilities.

[0003] Taking communication systems based on Orthogonal Frequency Division Multiplexing (OFDM) technology as an example, OFDM has become one of the core technologies for broadband wireless communication due to its high spectral efficiency and strong resistance to multipath fading. However, OFDM systems are extremely sensitive to synchronization errors, especially carrier frequency offset. Frequency offset can destroy the orthogonality between subcarriers, introduce severe inter-subcarrier interference, and cause a sharp deterioration in the system demodulation performance (van de Beek JJ, Sandell M, Börjesson P O. ML estimation of time and frequency offset in OFDM systems[J]. IEEE Transactions on Signal Processing, 1997). In the receiving link, accurate and fast frequency offset estimation and compensation are prerequisites for ensuring correct signal demodulation. As a key part of the synchronization link of OFDM receiver, the accuracy and robustness of the frequency offset estimation algorithm directly determine the bit error rate performance of the received signal. Once the frequency offset estimation fails and compensation errors occur, it will cause continuous interference, making the entire OFDM receiving system unable to work properly.

[0004] Field-Programmable Gate Arrays (FPGAs) are currently the mainstream programmable devices, offering advantages such as ease of use, high compatibility, and parallel processing. They overcome the drawbacks of traditional application-specific integrated circuits (ASICs) with their long development cycles and high costs, and can serve as a platform for implementing and verifying synchronization algorithms (H. Li, Z. Wang and H. Wang. Research and implementation of high speed parallel carrier synchronization algorithm [C]. International Conference on Communication Software and Networks (ICCSN), 2017: 28-32). The scenario involves wireless communication using electromagnetic waves as the transmission medium while the terminal is in hypersonic motion. In this case, the electromagnetic wave frequency between the transmitting and receiving terminals will exhibit a significant frequency offset due to the Doppler effect. Therefore, frequency offset estimation requires multiple sequences of varying lengths. Traditional frequency offset estimation requires multiple serial repetitions after initial estimation, which introduces significant processing delays when implemented using FPGAs, resulting in failure to meet system requirements.

[0005] Therefore, how to consider the parallel computing concept implemented in FPGA and shorten the processing latency of multiple frequency offset estimations are technical challenges that urgently need to be overcome in this field. Summary of the Invention

[0006] To address the technical problems existing in the prior art, the present invention provides a method for frequency offset compensation in wireless communication under hypersonic conditions. By utilizing the parallel processing structure of FPGA, the frequency offset compensation processing time is significantly shortened, and by using joint decision-making, the accuracy of frequency offset compensation is significantly improved.

[0007] The technical solution adopted in this invention is as follows:

[0008] A method for frequency offset compensation in wireless communication under hypersonic conditions includes the following steps:

[0009] Step 1: Assuming the maximum normalized frequency offset of the hypersonic scenario is M, generate a signal waveform for frequency offset estimation at the transmitting end. Specifically, select... ZC (Zadoff Chu) sequences with the same segment length and root exponent are used as frequency offset estimation sequences; among them, ;

[0010] Step 2: At the receiving end, the received signal is synchronized at a specific time. Then, the synchronization header is estimated in parallel n times for different frequency offset ranges. The estimation range increases sequentially with the number of frequency offset estimations. Let the nth estimation range be denoted as... The estimation range of the second frequency offset is Its corresponding normalized frequency offset estimate is This yields n normalized frequency offset estimates; among which, , Indicates the first The upper limit of the normalized frequency offset estimate for the second frequency offset estimate; ;

[0011] Step 3: Make a joint decision based on n normalized frequency offset estimates. The final frequency offset estimation result is obtained within the range, and the specific process is as follows:

[0012] Step 3.1, Order the round of judgment For the first round of judgments, the first normalized frequency offset estimate is used. ,exist Obtain the set of all possible first-order normalized frequency offset estimates within the range. ;in, It is an integer and satisfies ;

[0013] Step 3.2, let ;judge Is it greater than n? If yes, go to step 3.4; otherwise, go to step 3.3.

[0014] Step 3.3, for the first Round of judgment, take the first Subnormalized frequency offset estimate ,exist Obtain all possible first numbers within the range Subnormalized frequency offset estimation set ;in, ; It is an integer and satisfies ; Let represent the independent variable that minimizes the objective function. That is, to make Take the minimum value ;

[0015] Return to step 3.2;

[0016] Step 3.4, the set of normalized frequency offset estimates in the nth iteration. Take in The normalized frequency offset estimate within the range, i.e., the final frequency offset estimate result. ,in, ;

[0017] Step 4: Based on the final frequency offset estimation results Frequency offset compensation is performed.

[0018] Furthermore, the normalized frequency offset estimate in step 2 The specific calculation process is as follows:

[0019] Step 2.1, calculate the first... Phase difference of secondary frequency offset estimation :

[0020]

[0021] In the formula, For the first Index for second frequency offset estimation; For the first The sequence bias time estimated by the second frequency offset; Indicates the received signal; For the first The sequence length estimated by the second frequency offset; Indicates taking the conjugate;

[0022] Step 2.2, calculate the first... Frequency offset of the second frequency offset estimation :

[0023]

[0024] In the formula, Indicates taking the complex argument;

[0025] Step 2.3: Calculate the normalized frequency offset estimate. :

[0026]

[0027] In the formula, It represents the reciprocal of the carrier frequency.

[0028] Furthermore, according to In the calculation formula The properties of the function, to obtain ;in, .

[0029] Furthermore, when At that time, due to the final frequency offset estimation result obtained in step 3 exist Within this scope, to simplify the FPGA hardware implementation process, step 3 is simplified to the following process:

[0030] Step A: Determine the round of judgment For the first round of judgments, the first normalized frequency offset estimate is used. ;

[0031] Step B, let ;judge Is it greater than n? If yes, go to step E; otherwise, go to step C.

[0032] Step C, for the first Round of judgment, take the first Subnormalized frequency offset estimate ;like Then let the intermediate variable ;like Then let the intermediate variable ;

[0033] Step D: Set the judgment conditions: , ,as well as If any of the judgment conditions are met, then let the selection flag be set. Otherwise, let the selection flag be set. ;

[0034] Return to step B;

[0035] Step E: Based on the selection flags obtained from all decision rounds, calculate the normalized frequency offset estimate to be corrected. ;like The final frequency offset estimation result ;like The final frequency offset estimation result Otherwise, the final frequency offset estimation result .

[0036] Furthermore, the wireless communication frequency offset compensation method under hypersonic conditions is applicable to OFDM wireless communication systems.

[0037] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0038] This invention proposes a frequency offset compensation method for wireless communication under hypersonic conditions. By simultaneously performing frequency offset estimation at different ranges, multiple frequency offset estimation results are obtained. The final frequency offset estimation result is obtained by combining the results with a joint decision, thereby completing the frequency offset compensation. When the method of this invention is implemented on an FPGA, its parallel processing structure can be used to significantly shorten the frequency offset compensation processing time and significantly improve the accuracy of frequency offset compensation. Attached Figure Description

[0039] Figure 1 This is a flowchart of the wireless communication frequency offset compensation method proposed in Example 2 under hypersonic conditions;

[0040] Figure 2 This is a flowchart of the joint decision algorithm in Example 2;

[0041] Figure 3This is a flowchart of the frequency offset estimation calculation module of the FPGA in Example 2;

[0042] Figure 4 This is a flowchart of the frequency offset compensation module implementation of the FPGA in Example 2;

[0043] Figure 5 The data provided are simulation results of the frequency offset estimation algorithm in Example 2 under different signal-to-noise ratios. Detailed Implementation

[0044] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0045] Example 1

[0046] For OFDM wireless communication systems, this embodiment proposes a frequency offset compensation method for wireless communication under hypersonic conditions. Taking a scenario where the normalized frequency offset is less than 4 as an example, five frequency offset estimations are performed, where the time difference between the first and last sequence is... The second sequence deviation time is The time deviation between the third and subsequent sequences is... The time deviation between the fourth and subsequent sequences is... The time deviation of the sequence before and after the fifth iteration is... That is, respectively in , , , and Frequency offset estimation is performed within five ranges.

[0047] The method specifically includes the following steps:

[0048] Step 1: The maximum normalized frequency offset in this scenario is 4. At the transmitting end, a signal waveform for frequency offset estimation is generated. Specifically, 24 ZC sequences with the same length and the same root exponent are selected as the frequency offset estimation sequence.

[0049] Step 2: At the receiving end, the received signal is synchronized at a specific time. Then, the synchronization header is estimated in parallel five times within different ranges. The estimation ranges are as follows: , , , and Its corresponding normalized frequency offset estimate is Thus, five normalized frequency offset estimates were obtained;

[0050] The normalized frequency offset estimate The specific calculation process is as follows:

[0051] Step 2.1, calculate the first... Phase difference of secondary frequency offset estimation :

[0052] Equation (1)

[0053] In the formula, For the first Index for second frequency offset estimation; For the first The sequence bias time estimated by the second frequency offset; Indicates the received signal; For the first The sequence length estimated by the second frequency offset; Indicates taking the conjugate;

[0054] Step 2.2, calculate the first... Frequency offset of the second frequency offset estimation :

[0055] Equation (2)

[0056] In the formula, Indicates taking the complex argument;

[0057] Step 2.3: Calculate the normalized frequency offset estimate. :

[0058] Equation (3)

[0059] In the formula, Represents the reciprocal of the carrier frequency;

[0060] Step 3: Make a joint decision based on the five normalized frequency offset estimates. The final frequency offset estimation result is obtained within the range, and the specific process is as follows:

[0061] Step 3.1, Order the round of judgment For the first round of judgments, the first normalized frequency offset estimate is used. ,exist Obtain the set of all possible first-order normalized frequency offset estimates within the range. ;in, It is an integer and satisfies ;

[0062] Step 3.2, let ;judge Is it greater than n? If yes, go to step 3.4; otherwise, go to step 3.3.

[0063] Step 3.3, for the first Round of judgment, take the first Subnormalized frequency offset estimate ,exist Obtain all possible first numbers within the range Subnormalized frequency offset estimation set ;in, ; It is an integer and satisfies ; Let represent the independent variable that minimizes the objective function. That is, to make Take the minimum value ;

[0064] Return to step 3.2;

[0065] Step 3.4, in the 5th normalized frequency offset estimation set In China Normalized frequency offset estimate within the range (taking the value closest to 0) ,in, ;

[0066] Step 4: Based on the normalized frequency offset estimate Frequency offset compensation is performed.

[0067] Example 2

[0068] For OFDM wireless communication systems, this embodiment proposes a method for frequency offset compensation in hypersonic conditions, which can... That is, when the time difference between the two estimated sequences is twice, the implementation difficulty of FPGA is greatly simplified.

[0069] Taking a scenario where the normalized frequency offset is less than 4 as an example, five frequency offset estimations are performed, where the time of the first sequence bias is... The second sequence deviation time is The time deviation between the third and subsequent sequences is... The time deviation between the fourth and subsequent sequences is... The time deviation of the sequence before and after the fifth iteration is... That is, respectively in , , , and Frequency offset estimation is performed within five ranges.

[0070] The method specifically includes the following steps:

[0071] Step 1: The maximum normalized frequency offset in this scenario is 4. At the transmitting end, a signal waveform for frequency offset estimation is generated. Specifically, 24 ZC sequences with the same length and the same root exponent are selected as the frequency offset estimation sequence.

[0072] Step 2: At the receiving end, the received signal is synchronized at a specific time. Then, the synchronization header is estimated in parallel five times for frequency offset within different ranges. The estimation range of the second frequency offset is In this embodiment, the ranges of the five estimations are as follows: , , , and Its corresponding normalized frequency offset estimate is Thus, five normalized frequency offset estimates were obtained;

[0073] The normalized frequency offset estimate The specific calculation process is as follows:

[0074] Step 2.1, calculate the first... Phase difference of secondary frequency offset estimation :

[0075] Equation (1)

[0076] In the formula, For the first Index for second frequency offset estimation; For the first The sequence bias time estimated by the second frequency offset; Indicates the received signal; For the first The sequence length estimated by the second frequency offset; Indicates taking the conjugate;

[0077] Step 2.2, calculate the first... Frequency offset of the second frequency offset estimation :

[0078] Equation (2)

[0079] In the formula, Indicates taking the complex argument;

[0080] Step 2.3: Calculate the normalized frequency offset estimate. :

[0081] Equation (3)

[0082] In the formula, Represents the reciprocal of the carrier frequency;

[0083] Step 3: Make a joint decision based on the five normalized frequency offset estimates. The final frequency offset estimation result is obtained within the range, and the specific process is as follows:

[0084] Step 3.1, Order the round of judgment For the first round of judgments, the first normalized frequency offset estimate is used. ;

[0085] Step 3.2, let ;judge Is it greater than n? If yes, go to step 3.5; otherwise, go to step 3.3.

[0086] Step 3.3, for the first Round of judgment, take the first Subnormalized frequency offset estimate ;like Then let the intermediate variable ;like Then let the intermediate variable ;

[0087] Step 3.4: Set the judgment conditions: , ,as well as If any of the judgment conditions are met, then let the selection flag be set. Otherwise, let the selection flag be set. ;

[0088] Return to step 3.2;

[0089] Step 3.5: Based on the selection flags obtained from all decision rounds, calculate the normalized frequency offset estimate to be corrected. ;like The final frequency offset estimation result ;like The final frequency offset estimation result Otherwise, the final frequency offset estimation result .

[0090] Regarding the wireless communication frequency offset compensation method under the above-mentioned hypersonic state, this embodiment adopts FPGA design and implementation, specifically using Verilog HDL as the development language, Vivado 2020.2 as the development environment, and the FPGA model is xcvu13p-fhgb2104-2-i.

[0091] At the transmitting end, MATLAB is used to generate a coe file from the baseband OFDM modulation waveform, which is then loaded into the FPGA using a ROM IP core and transmitted via a digital-to-analog converter (DAC) module.

[0092] The general processing flow of the receiving module is as follows: Figure 2 As shown, the frequency offset compensation section mainly comprises two modules: a frequency offset estimation calculation module and a frequency offset compensation module. The frequency offset estimation calculation module mainly consists of two parts: parallel phase difference calculation and joint decision-making of frequency offset estimates. These parts are used to obtain frequency offset estimation results within different frequency offset estimation ranges and to jointly decide the final frequency offset estimation result. The frequency offset compensation module mainly consists of two parts: compensation signal generation and compensation of the original signal. After the corresponding compensation signal is generated using the DDS IP core, the original signal can be compensated.

[0093] The implementation steps for these two modules are explained below.

[0094] Step A, the flowchart of the frequency offset estimation calculation module is as follows: Figure 3 As shown, this step first requires five frequency offset estimation calculations with different estimation ranges, but the operation logic of each calculation is the same. Therefore, only one channel of code needs to be written. The other channels only need to modify the data delay Ti and the unit conversion of the CORDIC core output phase. Then, the results of multiple frequency offset estimations are sent to the joint decision module to obtain the final frequency offset estimate.

[0095] (1) Implementation of the multiplier: First, two data streams need to be read from the RAM IP core. These two data streams have a certain time deviation in practice (the specific time deviation depends on the frequency offset estimation range). Then, one of the data streams is conjugate and then fed into the Xilinx official Complex Multiplier IP core to complete the complex multiplication.

[0096] (2) Implementation of the accumulator: The real and imaginary parts of the output of the multiplier are fed into two Xilinx official Accumulator IP cores respectively to complete the accumulation operation of the real and imaginary parts respectively;

[0097] (3) Implementation of phase extraction: The phase value of the accumulated result can be obtained by feeding the accumulated result of the real part and the imaginary part of the accumulator into the official CORDICIP core of Xilinx;

[0098] (4) Implementation of unit conversion: After processing according to formulas (2) and (3), the result values ​​of each frequency offset estimation can be obtained;

[0099] (5) Implementation of joint decision: The five frequency offset estimates after unit conversion are sent to the joint decision module. The decision criteria of the joint decision module are as follows: Figure 1 As shown.

[0100] Step B, the flowchart of the frequency offset compensation module is as follows: Figure 4As shown, this step requires inputting the frequency offset estimate obtained by the frequency offset estimation algorithm module into the Xilinx official DDS IP core to generate the compensation signal. Then, the compensation signal and the received signal are input together into the Xilinx official Complex Multiple IP core to obtain the final frequency offset compensated signal.

[0101] The advantage of this embodiment lies in its implementation of multi-path parallel frequency offset estimation, ultimately obtaining a single frequency offset estimation result through joint decision-making. This significantly reduces the processing latency caused by the traditional multiple-stage frequency offset estimation method, which requires frequency offset compensation after each estimation before proceeding to the next. Furthermore, compared to traditional frequency offset estimation algorithms, this method also improves frequency offset estimation accuracy to a certain extent. The performance curves of MALTAB under different signal-to-noise ratios are shown below. Figure 5 As shown.

[0102] It should be noted that this is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions, and improvements made by those skilled in the art within the scope of the technology disclosed in the present invention, and within the spirit and principles of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for frequency offset compensation in wireless communication in a hypersonic regime, the method comprising: receiving a signal from a satellite; determining a Doppler shift of the signal; and compensating for the Doppler shift by adjusting a frequency of a receiver. Includes the following steps: Step 1: Assuming the maximum normalized frequency offset of the hypersonic scenario is M, generate a signal waveform for frequency offset estimation at the transmitting end. Specifically, select multiple ZC sequences of the same length and root exponent as the frequency offset estimation sequences. Step 2: At the receiving end, the received signal is synchronized at a specific time. Then, the synchronization header is estimated in parallel n times for different frequency offset ranges. The estimation range increases sequentially with the number of frequency offset estimations. Let the nth estimation range be denoted as... The estimation range of the second frequency offset is Its corresponding normalized frequency offset estimate is This yields n normalized frequency offset estimates; among which, , Indicates the first The upper limit of the normalized frequency offset estimate for the second frequency offset estimate; ; Step 3: Make a joint decision based on n normalized frequency offset estimates. The final frequency offset estimation result is obtained within the specified range. The specific process is as follows: Step 3.1, Order the round of judgment For the first round of judgments, the first normalized frequency offset estimate is used. ,exist Obtain the set of all possible first-order normalized frequency offset estimates within the range. ;in, It is an integer and satisfies ; Step 3.2, let ;judge Is it greater than n? If yes, go to step 3.4; otherwise, go to step 3.

3. Step 3.3, for the first Round of judgment, take the first Subnormalized frequency offset estimate ,exist Obtain all possible first numbers within the range Subnormalized frequency offset estimation set ;in, ; It is an integer and satisfies ; Let represent the independent variable that minimizes the objective function. That is, to make Take the minimum value ; Return to step 3.2; Step 3.4, the set of normalized frequency offset estimates in the nth iteration. Take in The normalized frequency offset estimate within the range, i.e., the final frequency offset estimate result. ,in, ; Step 4: Based on the final frequency offset estimation results Frequency offset compensation is performed.

2. The wireless communication frequency offset compensation method under hypersonic conditions according to claim 1, characterized in that, Normalized frequency offset estimate in step 2 The specific calculation process is as follows: Step 2.1, calculate the first... Phase difference of secondary frequency offset estimation : In the formula, For the first Index for second frequency offset estimation; For the first The sequence bias time estimated by the second frequency offset; Indicates the received signal; For the first The sequence length estimated by the second frequency offset; Indicates taking the conjugate; Step 2.2, calculate the first... Frequency offset of the second frequency offset estimation : In the formula, Indicates taking the complex argument; Step 2.3: Calculate the normalized frequency offset estimate. : In the formula, It represents the reciprocal of the carrier frequency.

3. The wireless communication frequency offset compensation method under hypersonic conditions according to claim 2, characterized in that, according to In the calculation formula The properties of the function, to obtain ;in, .

4. The wireless communication frequency offset compensation method under hypersonic conditions according to claim 2, characterized in that, when At that time, due to the final frequency offset estimation result obtained in step 3 exist Within this scope, to simplify the FPGA hardware implementation process, step 3 is simplified to the following process: Step A: Determine the round of judgment ; For the first round of judgments, the first normalized frequency offset estimate is used. ; Step B, let ;judge Is it greater than n? If yes, go to step E; otherwise, go to step C. Step C, for the first Round of judgment, take the first Subnormalized frequency offset estimate ;like Then let the intermediate variable ;like Then let the intermediate variable ; Step D: Set the judgment conditions: , ,as well as If any of the judgment conditions are met, then let the selection flag be set. ; Otherwise, let the selection flag be set. ; Return to step B; Step E: Based on the selection flags obtained from all decision rounds, calculate the normalized frequency offset estimate to be corrected. ;like The final frequency offset estimation result ;like The final frequency offset estimation result ; Otherwise, the final frequency offset estimation result .

5. The wireless communication frequency offset compensation method under hypersonic conditions according to any one of claims 1 to 4, characterized in that, The wireless communication frequency offset compensation method under hypersonic conditions is applicable to OFDM wireless communication systems.

Images