Pilot design and frequency offset estimation method in OFDM system

Abstract

The application discloses a pilot design and frequency offset estimation method in an OFDM system, which comprises the following steps: high layer determines system parameters; determines pilot sequence length L; calculates upper guard band subcarrier number N ceil and lower guard band subcarrier number N floor ; generates m sequence S(k); generates constant modulus pilot sequence D(k); calculates weighted pilot sequence D ′ '(k); generates cyclic weighted pilot sequence D s '(k') using D'(k); counts total power P all of the weighted pilot sequence; generates time domain pilot signal through IFFT; transmits and receives the signal; carries out power normalization operation on the frequency domain received signal to obtain normalized frequency domain received signal Z(k); extracts L signals from index M ZP +N floor in the normalized frequency domain received signal, and obtains frequency domain received extracted signal Z S (k); calculates correlation coefficient sequence ρ(n); and calculates frequency offset Fo est according to the correlation coefficient sequence. Compared with the prior art, the application significantly saves pilot resources and remarkably improves the frequency offset estimation range, and is not affected by the time delay and phase of the received signal.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, and in particular to a pilot design and frequency offset estimation method in an OFDM system. Background Technology

[0002] In current mobile communication technologies such as LTE (Long Term Evolution, i.e., fourth-generation mobile communication technology) and NR (New Radio, i.e., fifth-generation mobile communication technology), OFDM (Orthogonal Frequency Division Multiplexing) is used as a key technology for air interface transmission. In scenarios where there is frequency offset in air interface transmission, the performance of OFDM system will be severely affected. Therefore, frequency offset estimation and correction technology is the key technology to overcome the impact of frequency offset in OFDM system.

[0003] In scenarios with large system frequency offsets (e.g., in low-Earth orbit satellite communication systems, the system frequency offset can exceed 1MHz, while the typical subcarrier spacing of OFDM systems is 120kHz), the system needs to perform frequency offset estimation over a large range. Existing OFDM system frequency offset estimation schemes for such scenarios typically consist of two parts: fractional frequency offset estimation and integer frequency offset estimation. Fractional frequency offset represents the portion of the system frequency offset that is a fraction of the system subcarrier spacing; integer frequency offset represents the portion of the system frequency offset that is an integer multiple of the system subcarrier spacing. Fractional frequency offset estimation requires configuring multiple pilot symbols in the time domain and estimating the fractional frequency offset by calculating the phase difference between the channel estimation results of different symbols. In scenarios with large frequency offsets, the channel estimation results are poor, resulting in poor accuracy for fractional frequency offset estimation using this method. Existing integer frequency offset estimation methods often require correcting the fractional frequency offset before implementation, so their performance is also unsatisfactory in scenarios with large frequency offsets. In summary, existing frequency offset estimation schemes require configuring multiple pilot symbols and have poor performance in scenarios with large frequency offsets. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a pilot design and frequency offset estimation method for OFDM systems, the specific technical solution of which is as follows:

[0005] A method for pilot design and frequency offset estimation in an OFDM system includes the following steps:

[0006] S1. The higher layer determines the system parameters, including the number of resource blocks used for transmitting pilots in the current OFDM system being M, the system subcarrier spacing being Δf, and the maximum possible frequency offset of the system being fo. max The number of subcarriers in each RB is

[0007] S2. Determine the pilot sequence length L, calculated using the formula:

[0008]

[0009] in Indicates rounding up. Indicates rounding down;

[0010] S3. Calculate the number N of upper guard sideband subcarriers. ceil With the number of lower guard sideband subcarriers N floor ;

[0011] S4. Generate an m-sequence S(k) of length L, where S(k) = 0, 1; k = 0, ..., L-1;

[0012] S5. Generate a constant-mode pilot sequence D(k) of length L, k = 0, ..., L-1;

[0013] S6. Calculate the weighted pilot sequence D′(k), the calculation formula is:

[0014] D′(k)=D(k)S(k),k=0,…,L-1;

[0015] S7. Use D′(k) to generate the cyclically weighted pilot sequence D. s (k′), the calculation formula is:

[0016]

[0017] S8, Statistically weighted pilot sequence total power P all The calculation formula is:

[0018]

[0019] S9. Generate time-domain pilot signals using IFFT;

[0020] S10. Transmit and receive signals;

[0021] S11. Perform power normalization on the frequency domain received signal to obtain the normalized frequency domain received signal Z(k);

[0022] S12, in the normalized frequency domain received signal, from index M ZP +N floor The system first extracts a total of L signals to obtain the frequency domain received decimation signal Z. S (k);

[0023] S13. Calculate the correlation coefficient sequence ρ(n);

[0024] S14. Calculate the frequency offset based on the correlation coefficient sequence.

[0025] Specifically, the number N of the guard sideband subcarriers mentioned in step S3 ceil With the number of lower guard sideband subcarriers N floorThe formula for calculation is:

[0026]

[0027]

[0028] Specifically, the constant-mode pilot sequence D(k) mentioned in step S5 must satisfy the constant-mode property |D(k)| 2 =1,k=0,…,L-1.

[0029] Specifically, step S9 further includes the following sub-steps:

[0030] S91, D s (k ′ By padding with zeros at the beginning and end, a zero-padded cyclically weighted pilot sequence D is obtained. s,zp (k ′ The calculation formula is:

[0031]

[0032] k″=0,…,N FFT -1,

[0033] Where M ZP To satisfy 0≤M ZP ≤N FFT -N floor -N ceil -L is any integer;

[0034] S92. Obtain the time-domain pilot signal x through IFFT operation. t The calculation formula is:

[0035] x t =IFFT(D s,zp ).

[0036] Specifically, step S10 further includes the following sub-steps:

[0037] S101. Add a cyclic prefix to the time-domain pilot signal to obtain a time-domain transmit signal x with CP. t,CP (n), the calculation formula is:

[0038] x t,CP (n)=x t (mod(nN CP N FFT ), n=0,…,N FFT +N CP -1

[0039] S102. Set the received time domain signal to y. t,CP (n), n=0,…,N FFT+N CP -1, the receiver removes the cyclic prefix to obtain the de-cyclic prefixed time-domain received signal y. t (n)=y t,CP (n+N CP ), n=0,…,N FFT -1

[0040] S103. Obtain the frequency domain received signal Y(k), k=0,…,N through FFT operation. FFT -1, Y = FFT(y t ).

[0041] Specifically, the formula for calculating the normalized frequency domain received signal Z(k) by performing power normalization on the frequency domain received signal in step S11 is as follows:

[0042]

[0043] Specifically, step S12 involves obtaining the frequency domain received decimation signal Z. S The formula for calculating (k) is:

[0044] Z S (k)=Z(k+M ZP +N floor ), k=0,…,L-1.

[0045] Specifically, the formula for calculating the correlation coefficient sequence ρ(n) in step S13 is as follows:

[0046]

[0047] Specifically, step S14 further includes the following sub-steps:

[0048] S141. Search for the maximum correlation coefficient index and calculate integer multiples of frequency offset Fo. int The calculation formula is:

[0049]

[0050] S142. Obtain the absolute value of the fractional octave frequency offset Fo by looking up ρ(Idx0) in a table. frac,abs The calculation formula is:

[0051]

[0052] Where T(m) represents the m-th correlation coefficient in the pre-configured table, and F(m) represents the fractional frequency offset corresponding to the m-th correlation coefficient in the table. The table has a total of M rows.

[0053] S143, Calculate the frequency offset estimation result Fo est The calculation formula is:

[0054] .

[0055] The beneficial effects of this invention are:

[0056] The present invention requires only one pilot symbol to estimate both integer and fractional frequency offsets at once, significantly saving pilot resources compared to existing solutions. Furthermore, the system parameters can be configured according to the system's maximum possible frequency offset range, significantly improving the frequency offset estimation range. The present invention only requires the received signal power of different subcarriers to complete the frequency offset estimation, and is not affected by received signal delay and phase, exhibiting higher robustness compared to existing solutions. Detailed Implementation

[0057] To provide a clearer understanding of the technical features, objectives, and effects of the present invention, specific embodiments of the present invention are now described.

[0058] Example 1

[0059] In this embodiment, it is assumed that the number of OFDM points in the system is N. FFT =2048, subcarrier spacing Δf = 120kHz, the maximum possible frequency offset of the system is fo max =2000kHz, then this embodiment mainly includes:

[0060] (1) Higher-level system parameter determination: The current OFDM system uses M = 10 RBs (resource blocks) for transmitting pilots, the system subcarrier spacing is Δf = 120kHz, and the maximum possible frequency offset of the system is fo. max =2000kHz, the number of subcarriers in each RB is

[0061] (2) Determine the pilot sequence length L:

[0062]

[0063] (3) Calculate the number N of upper guard sideband subcarriers. ceil With the number of lower guard sideband subcarriers N floor :

[0064]

[0065]

[0066] (4) Generate an m-sequence S(k) of length L = 63, with primitive polynomial x. 6 +x 3 +1,

[0067]

[0068] (5) Generate a constant-mode pilot sequence D(k), k = 0, ..., 62, of length L = 63. Here, we use...

[0069] (6) Calculate the weighted pilot sequence D′(k), k=0,…,62

[0070] D′(k)=D(k)S(k),k=0,…,62

[0071] (7) Use D′(k) to generate the cyclically weighted pilot sequence D s (k′)

[0072]

[0073] (8) Statistically weighted pilot sequence total power P all

[0074]

[0075] (9) Generate time-domain pilot signals using IFFT. First, convert D... s (k′) is padded with zeros at the beginning and end to obtain the zero-padded cyclic weighted pilot sequence D. s,zp (k′)

[0076]

[0077] Then, the time-domain pilot signal is obtained through IFFT operation.

[0078] x t =IFFT(D s,zp )

[0079] (10) Signal transmission and reception. First, a cyclic prefix is ​​added to the time-domain pilot signal to obtain the time-domain transmit signal x with CP. t,CP (n)

[0080] x t,CP (n)=x t (mod(n-144,2048)),n=0,…,2191

[0081] Assume the receiver receives a time-domain signal y. t,CP (n), n=0,…,2191, the receiver removes the cyclic prefix to obtain the time-domain received signal y. t (n)=y t,CP (n+144), n=0,…,2047, and then the frequency domain received signal Y(k),k=0,…,2047 is obtained through FFT operation.

[0082] Y = FFT(y t )

[0083] (11) Perform a power normalization operation on the frequency domain received signal to obtain the normalized frequency domain received signal Z(k).

[0084]

[0085] (12) Extract a total of 63 signals from the normalized frequency domain received signal, starting from index 28, to obtain the frequency domain received decimation signal Z. S (k)

[0086] Z S (k)=Z(k+28),k=0,…,62-1

[0087] (13) Calculate the correlation coefficient sequence ρ(n)

[0088]

[0089] (14) Calculate the frequency offset Fo based on the correlation coefficient sequence est First, search for the index of the maximum correlation coefficient and calculate the integer multiples of the frequency offset Fo. int

[0090]

[0091] Then, the absolute value of the fractional octave frequency offset Fo is obtained by looking up ρ(Idx0) in Table 1. frac,abs

[0092]

[0093] Then the frequency offset estimation result is calculated as follows:

[0094]

[0095] Table 1 shows the mapping data between the maximum correlation coefficient and the absolute value of the fractional octave frequency offset.

[0096]

[0097]

[0098] Table 1. Mapping table of maximum correlation coefficient and absolute value of fractional octave frequency offset.

[0099] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention.

Claims

1. A method for pilot design and frequency offset estimation in an OFDM system, characterized in that, Includes the following steps: S1. The higher layer determines the system parameters, including the number of resource blocks currently used for transmitting pilots in the OFDM system. The system subcarrier spacing is The system's maximum possible frequency offset is The number of subcarriers in each RB is ; S2. Determine the pilot sequence length. The calculation formula is in Indicates rounding up. Indicates rounding down; S3. Calculate the number of upper guard sideband subcarriers. Number of lower guard sideband subcarriers ; S4. Generate a string of length... m-sequence ; S5, Generation length is constant mode pilot sequence The constant mode pilot sequence described in step S5 Must meet constant model properties ; S6. Calculate the weighted pilot sequence The calculation formula is: ； S7, Use Generate cyclically weighted pilot sequences The calculation formula is: ； S8, Statistically weighted pilot sequence total power The calculation formula is: ； S9. Generate time-domain pilot signals using IFFT; S10. Transmit and receive signals; S11. Perform a power normalization operation on the frequency domain received signal to obtain a normalized frequency domain received signal. ; S12, from the index in the normalized frequency domain received signal. The first extraction was completed. One signal, to obtain the frequency domain received decimated signal. ;in To meet any integer; S13. Calculate the correlation coefficient sequence ; S14. Calculate the frequency offset based on the correlation coefficient sequence. ; Step S14 further includes the following sub-steps: S141. Calculate integer multiples of frequency offset by searching for the maximum correlation coefficient index. The calculation formula is: S142, By The absolute value of the decimal frequency offset is obtained by looking up the table. The calculation formula is: in This indicates the first item in the pre-configured table. One correlation coefficient, Indicates the first in the table The table contains the fractional octave deviations corresponding to each correlation coefficient. OK; S143, Calculate the frequency offset estimation results The calculation formula is: 。 2. The pilot design and frequency offset estimation method in an OFDM system according to claim 1, characterized in that, The number of guard sideband subcarriers mentioned in step S3 Number of lower guard sideband subcarriers The formula for calculation is: 。 3. The pilot design and frequency offset estimation method in an OFDM system according to claim 2, characterized in that, Step S9 further includes the following sub-steps: S91, will By padding with zeros at the beginning and end, a zero-padded cyclically weighted pilot sequence is obtained. The calculation formula is ； in Indicates the number of OFDM points in the system; S92. Obtain the time-domain pilot signal through IFFT operation. The calculation formula is: 。 4. The pilot design and frequency offset estimation method in an OFDM system according to claim 3, characterized in that, Step S10 further includes the following sub-steps: S101. Add a cyclic prefix to the time-domain pilot signal to obtain a time-domain transmit signal with CP. The calculation formula is: S102, Set the receiver to receive the time domain signal as... The receiver removes the cyclic prefix to obtain the decyclic-prefixed time-domain received signal. S103. Obtain the frequency domain received signal through FFT operation. .

5. The pilot design and frequency offset estimation method in an OFDM system according to claim 1, characterized in that, Step S11 describes performing a power normalization operation on the frequency domain received signal to obtain a normalized frequency domain received signal. The formula for calculation is: 。 6. The pilot design and frequency offset estimation method in an OFDM system according to claim 1, characterized in that, Step S12 describes obtaining the frequency domain received decimation signal. The formula for calculation is: 。 7. The pilot design and frequency offset estimation method in an OFDM system according to claim 1, characterized in that, Step S13 describes the calculation of the correlation coefficient sequence. The formula for calculation is: .

Images