CN113381959A - Pseudo-random phase sequence for time-frequency estimation and time-frequency estimation method - Google Patents
Pseudo-random phase sequence for time-frequency estimation and time-frequency estimation method Download PDFInfo
- Publication number
- CN113381959A CN113381959A CN202110648177.2A CN202110648177A CN113381959A CN 113381959 A CN113381959 A CN 113381959A CN 202110648177 A CN202110648177 A CN 202110648177A CN 113381959 A CN113381959 A CN 113381959A
- Authority
- CN
- China
- Prior art keywords
- value
- frequency
- time
- xchirp
- domain
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L27/00—Modulated-carrier systems
- H04L27/26—Systems using multi-frequency codes
- H04L27/2601—Multicarrier modulation systems
- H04L27/2602—Signal structure
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L27/00—Modulated-carrier systems
- H04L27/26—Systems using multi-frequency codes
- H04L27/2601—Multicarrier modulation systems
- H04L27/2647—Arrangements specific to the receiver only
- H04L27/2655—Synchronisation arrangements
- H04L27/2668—Details of algorithms
- H04L27/2673—Details of algorithms characterised by synchronisation parameters
- H04L27/2675—Pilot or known symbols
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Synchronisation In Digital Transmission Systems (AREA)
Abstract
The invention provides a pseudo-random phase sequence for time-frequency estimation, which is characterized by comprising the pseudo-random phase sequence which has linear time-frequency characteristics and good self-correlation and cross-correlation characteristics, so that a group of pilot symbols which are mutually uncorrelated in pairs can be constructed by the pseudo-random phase sequence and used as pilot frequency of a multi-user communication system. Furthermore, the invention provides a method for constructing pilot symbols by utilizing the pseudo-random phase sequence group pair, which comprises parallel pilot symbols and serial pilot symbols. Furthermore, the invention also provides a method for performing time-frequency estimation by adopting frequency domain correlation and time domain circumference correlation based on the pilot frequency symbol, and solves the key technology of the pilot frequency of a multi-user communication system.
Description
Technical Field
The invention relates to a communication technology, in particular to a pseudo-random phase sequence and a time-frequency estimation method for time-frequency estimation.
Background
Because of the characteristics of constant envelope, ideal circular autocorrelation, good cross-correlation and the like, the constant envelope zero autocorrelation sequence (CAZAC) is widely applied to the field of communication, and at present, the constant envelope zero autocorrelation sequence (CAZAC) commonly comprises a Zadoff-chu sequence, a Frank sequence, a Golomb polyphase sequence, a Chirp sequence and the like, and can be regarded as a pseudo-random phase sequence in practice.
The common communication systems use the above characteristics for pilot signals to achieve synchronization estimation.
Patent EP0952713a2 proposes a method for performing time-frequency estimation using Chirp signals, which can estimate frequency offset and time offset at the same time, but since there are only two Chirp signals (i.e. upper and lower Chirp) with specific lengths, it is not suitable for pilot of a multi-user communication system.
The pseudo-random phase sequence for time frequency estimation provided by the invention not only has linear time frequency characteristics like a Chirp signal and can be used for time frequency estimation, but also has a group of pseudo-random phase sequences which are pairwise irrelevant under the same length, so that the pseudo-random phase sequence is suitable for pilot frequency of a multi-user communication system.
Furthermore, the invention designs a series pilot symbol and a parallel pilot symbol for a multi-user communication system by utilizing the pseudo-random phase sequence group pair.
Furthermore, the invention provides a frequency domain correlation method and a time domain circumference correlation method based on the pilot frequency symbol of the multi-user communication system for carrying out time frequency estimation.
Disclosure of Invention
Firstly, a large number of formulas are stated to appear in the invention, the formulas are not methods and rules of intelligence activity, but are used for strictly defining a pseudo-random phase sequence for time-frequency estimation, and after all, the non-strict pseudo-random phase sequence does not have the linear time-frequency characteristic and the autocorrelation and cross-correlation characteristics of the invention.
Firstly, the object of the present invention is to design a pseudo-random phase sequence for multi-user time-frequency estimation, therefore, the pseudo-random phase sequence should have the following properties:
1: the pseudo-random phase sequence has linear time-frequency characteristics, and the linear time-frequency characteristics refer to the linear change of frequency along with time, so that the time-frequency estimation is a linear equation and has low complexity;
2: the pseudo-random phase sequence has good self-correlation and cross-correlation characteristics, and the good cross-correlation characteristics mean that a pairwise non-correlated pseudo-random phase sequence group exists under a specific length, so that the pseudo-random phase sequence is suitable for a multi-user communication system.
In accordance with the above requirements, the present invention provides a pseudo-random phase sequence for time-frequency estimation, which comprises,
according to property 1, the phase of the pseudo-random phase sequence is characterized by a quadratic function, i.e. the product of a first order (linear) function of time and a first order (linear) function of frequency, such pseudo-random phase sequence having a linear time-frequency relationship;
according to property 2, the length of the pseudo-random phase sequence is a prime number P, and since a composite number can be factorized, a cross-correlation peak value with a factor as a period exists during pairwise cross-correlation, which is not an ideal set of pseudo-random phase sequences which are not correlated with each other.
The invention provides a pseudo-random phase sequence satisfying the above conditions, which is expressed by pseudo-code without loss of generality as follows:
xChirp(R)＝exp(-i*pi*R*(m+a).*(m+a+2*b)/P)； (1)
wherein exp is an exponential function; i is an imaginary unit; pi is the circumference ratio; p is the length of the pseudorandom phase sequence and is a prime number; r is the slope of the frequency changing linearly along with the time, called root value, and is a positive integer smaller than P, and the pseudo-random phase sequences with different root values R are not related to each other; m is a time-frequency point, the value range of the m is { -P/2:1: P/2-1}, namely m is an integral multiple of 1/2 and the interval is 1; a. b is an offset, which is an arbitrary integer, a is equivalent to a sequence cyclic shift, b is equivalent to a sequence frequency shift, preferably, a, b are zero; is dot product.
With the pseudo-random phase sequence group with linear time-frequency relationship and no correlation between two pseudo-random phase sequence groups, the pilot frequency symbol for the multi-user communication system can be further constructed.
Because the time delay and the frequency offset are two independent variables, at least two pseudo-random phase sequences with different root values are required to be paired to estimate the time delay and the frequency offset.
The present invention proposes two methods for constructing pilot symbols using the above pseudo-random phase sequence pair, including,
the first type is a parallel pilot symbol mpilot (Rxy), and one pilot symbol is a pair of power normalized superposition of pseudo-random phase sequences xChirp (Rx) and xChirp (Ry) with different root values;
thus, the length of one parallel pilot symbol is P;
the second type is serial pilot symbols sPilot (Rxy), one pilot symbol is a pair of pseudo-random phase sequences xChrip (Rx) and xChrip (Ry) which have different root values and are spliced in sequence;
thus, the length of one serial pilot symbol is 2P;
when the pilot symbols are used for pilot, N consecutive pilot symbols are needed to be used together, where N > is 2, so that the receiving side can always capture a cyclic shifted sample of the pilot symbols.
A parallel pilot symbol is represented by the pseudo code:
mPilot(Rxy)＝(xChirp(Rx)+xChirp(Ry))/sqrt(2) (2)
wherein sqrt is a square root function; xchirp (Rx), xchirp (Ry) are pseudo-random phase sequence pairs with root values of Rx and Ry, respectively, and it is preferable to select Ry ═ P-Rx, that is, xchirp (Ry) is the conjugate of xchirp (Rx), also called Rx and Ry are conjugate root value pairs.
A serial pilot symbol is represented by the pseudo code:
sPilot(Rxy)＝[xChirp(Rx),xChirp(Ry)] (3)
wherein, [, ] is a splicing operation; the other parameters are as defined above.
The parallel pilot symbols are characterized in that the pilot symbols are short (the length is P) but do not have constant envelope characteristics, so that the power amplifier efficiency is reduced, and the parallel pilot symbols are suitable for communication systems which have high speed requirements but are insensitive to power consumption.
The serial pilot symbol has the characteristics that the length of the pilot symbol is doubled (the length is 2P), and the constant envelope characteristic is still kept, so that the power amplifier is fully utilized, and the serial pilot symbol is suitable for a communication system which has low requirement on speed but needs low power consumption.
According to the pilot frequency symbol of the special design, the invention further provides a time-frequency estimation method based on frequency domain correlation and a time-frequency estimation method based on time domain circumference correlation, which comprises the following steps:
a time-frequency estimation method based on frequency domain correlation is characterized by comprising the following steps,
respectively carrying out two kinds of frequency domain correlation on a received baseband sampling signal with the pilot frequency symbol length, and obtaining the position index value of the maximum absolute value of two kinds of correlation results;
and specifically calculating the index values of the two maximum absolute values to respectively obtain a time delay estimation value and a frequency deviation estimation value.
The above-mentioned two kinds of frequency domain correlations are respectively carried out to the received baseband sampling signal with a pilot frequency symbol length, and the position index value of the maximum absolute value of the two kinds of correlation results is obtained, including,
local time domain reference signals xchirp (rm) and xchirp (rn) for frequency domain correlation;
receiving a baseband sampling signal rx with a pilot symbol length;
performing point multiplication on the received signal rx and a local time domain reference signal xChirp (Rm) to obtain a time domain correlation value rx (Rm);
performing point multiplication on a received signal rx and a local time domain reference signal xChirp (Rn) to obtain a time domain correlation value rx (Rn);
performing P-point Fourier transform on the time domain correlation value rx (Rm) to obtain a frequency domain correlation value F (Rm);
performing P-point Fourier transform on the time domain correlation value rx (Rn) to obtain a frequency domain correlation value F (Rn);
searching the maximum absolute value of the frequency domain correlation value F (Rm) and obtaining an index value Im thereof;
searching the maximum absolute value of the frequency domain correlation value F (Rn) and obtaining an index value In of the maximum absolute value;
the frequency domain correlation method is expressed by pseudo codes as follows:
rx(Rm)＝rx.*xChirp(Rm)；
rx(Rn)＝rx.*xChirp(Rn)；
F(Rm)＝fft(rx(Rm),P)；
F(Rn)＝fft(rx(Rn),P)；
where fft is a Fourier transform function.
The maximum absolute value of the frequency domain correlation value is searched, and the index value is obtained and expressed as follows by using a pseudo code:
[～,Im]＝max(abs(F(Rm)))；
[～,In]＝max(abs(F(Rm)))；
wherein max is a maximum function, the first output is the maximum, discard, the second output is the maximum index value, abs is an absolute value function.
The local time domain reference signals xchirp (rm) and xchirp (rn) are the conjugate of xchirp (rx) and the conjugate of xchirp (ry), respectively, when the transmitting end uses the parallel pilot symbols.
The local time domain reference signals xchirp (rm) and xchirp (rn) are the conjugate of xchirp (rx) and xchirp (ry) of two symbols, respectively, when serial pilot symbols are used at the transmitting end. To prevent ambiguity, the pseudo code is expressed as:
xChirp(Rm)＝conj([xChirp(Rx),xChirp(Rx)])；
xChirp(Rn)＝conj([xChirp(Ry),xChirp(Ry)])；
where, conj is a conjugate function.
Therefore, when the sending end uses the serial pilot symbols, the length of the time domain correlation value is 2P, and when the P-point fourier transform is performed, the front and rear P-point time domain correlation values can be respectively subjected to fourier transform, and the transform results are added.
The specific calculation of the index values of the two maximum absolute values is carried out to respectively obtain a time delay estimation value and a frequency deviation estimation value, including,
the time delay estimated value eT is the difference between the maximum absolute value index value Im and In multiplied by iRxy and the P is subjected to modulus extraction, wherein iRxy is the inverse element of the difference modulus P of the root values Rx and Ry;
frequency offset estimation value eFm is obtained by subtracting the product of root Rx and delay estimation value eT from the maximum absolute value index Im and taking the modulus of P, and frequency offset estimation value eFn is obtained by subtracting the product of root Ry and delay estimation value eT from the maximum absolute value index In and taking the modulus of P;
if eFm is equal to eFn, then the estimated frequency offset eF is equal to eFm or eFn, and the current time delay estimate eT and the estimated frequency offset eF are valid, otherwise, time-frequency estimation continues.
The time delay estimation and the frequency offset estimation are expressed by pseudo codes as follows:
where mod is a modulo function.
When a pseudo-random phase sequence pair is preferred (i.e. conjugate root value pair is used), it can be obtained according to the formulas (5), (5'):
eF＝(Im+In)/2；
the above equation (4) can be simplified to:
eT＝mod((Im-In)/2*iRx,P)；
where iRx is the inverse of the root Rx modulo P.
The time-frequency estimation method based on frequency domain correlation only needs one Fourier transform, is simple to implement and is an optimal method, and actually, time-frequency estimation can also be implemented by adopting a time-domain circular correlation method.
A time-frequency estimation method based on time-domain circular correlation is characterized by comprising the following steps,
respectively carrying out two time domain circular correlations on a received baseband sampling signal with a pilot frequency symbol length, and obtaining a position index value of a maximum absolute value of two correlation results;
and specifically calculating the index values of the two maximum absolute values to respectively obtain a time delay estimation value and a frequency deviation estimation value.
The above-mentioned two kinds of time domain circular correlations are respectively carried out to the received baseband sampling signal with a pilot frequency symbol length, and the position index value of the maximum absolute value of the two kinds of correlation results is obtained, including,
local frequency domain reference signals fcirp (rm) and fcirp (rn) for time domain circular correlation;
a received frequency domain baseband sampling signal Frx of a pilot symbol length and transformed to the frequency domain;
performing point multiplication on the frequency domain receiving signal Frx and a local frequency domain reference signal fChirp (Rm) to obtain a frequency domain correlation value F (Rm);
performing point multiplication on the frequency domain receiving signal Frx and a local frequency domain reference signal fChirp (Rn) to obtain a frequency domain correlation value F (Rn);
performing P-point inverse Fourier transform on the frequency domain correlation value F (Rm) to obtain a time domain correlation value T (Rm);
carrying out P-point Fourier inverse transformation on the frequency domain correlation value F (Rn) to obtain a time domain correlation value T (Rn);
searching the maximum absolute value of the time domain correlation value T (Rm) and obtaining an index value Im thereof;
searching the maximum absolute value of the time domain correlation value T (Rn) and obtaining an index value In thereof;
the above time domain circular correlation method is expressed by pseudo code as:
F(Rm)＝Frx.*fChirp(Rm)；
F(Rn)＝Frx.*fChirp(Rn)；
T(Rm)＝ifft(F(Rm),P)；
T(Rn)＝ifft(F(Rn),P)；
where ifft is the inverse fourier transform function.
The above process of searching the maximum absolute value of the time domain correlation value and obtaining the index value is the same as above and is not expressed by a pseudo code.
When the transmitting end uses the parallel pilot symbols, the local frequency domain reference signals fchirp (rm) and fchirp (rn) are respectively xchirp (rx) transformed to the frequency domain and taken as conjugate, and xchirp (ry) transformed to the frequency domain and taken as conjugate.
The local frequency domain reference signals fchirp (rm) and fchirp (rn) are xchirp (rx) of two symbols respectively transformed to the frequency domain and taken as conjugate when the transmitting end uses serial pilot symbols and xchirp (ry) of two symbols transformed to the frequency domain and taken as conjugate. To prevent ambiguity, the pseudo code is expressed as:
fChirp(Rm)＝conj([fft(xChirp(Rx)),fft(xChirp(Rx))])；
fChirp(Rn)＝conj([fft(xChirp(Ry)),fft(xChirp(Ry))])；
the received frequency-domain baseband sampling signal Frx with a pilot symbol length and transformed to the frequency domain is used at the transmitting end to transform the received baseband sampling signal rx with a pilot symbol length to the frequency domain when the parallel pilot symbols are used at the transmitting end.
When the serial pilot symbols are used at the transmitting end, the received frequency-domain baseband sampling signal Frx with a pilot symbol length and converted to the frequency domain is spliced by converting the first P samples of the received baseband sampling signal rx with a pilot symbol length (length of 2P) to the frequency domain and converting the first P samples of the received baseband sampling signal rx with the last P samples of the baseband sampling signal rx to the frequency domain. To prevent ambiguity, the pseudo code is expressed as:
Frx＝[fft(rx(0:P-1)),fft(rx(P:2P-1))]；
similarly, when the sending end uses the serial pilot symbols, the length of the frequency domain correlation value is 2P, and when performing P-point inverse fourier transform, the front and rear P-point time domain correlation values may be subjected to inverse fourier transform respectively, and the transform results may be added.
The specific calculation of the index values of the two maximum absolute values is carried out to respectively obtain a time delay estimation value and a frequency deviation estimation value, including,
the time delay estimated value eT is the difference between the maximum absolute value index value Im multiplied by Rx and the maximum absolute value index value In multiplied by Ry multiplied by iRxy and the P is subjected to modulus taking, wherein iRxy is the inverse element of the difference modulus P of the root values Rx and Ry;
frequency offset estimation value eFm is obtained by subtracting time delay estimation value eT from maximum absolute value index value Im and multiplying by root value Rx, and performing modulo P, and frequency offset estimation value eFn is obtained by subtracting time delay estimation value eT from maximum absolute value index value In and multiplying by root value Ry, and performing modulo P;
if eFm is equal to eFn, then the estimated frequency offset eF is equal to eFm or eFn, and the current time delay estimate eT and the estimated frequency offset eF are valid, otherwise, time-frequency estimation continues.
The time delay estimation and the frequency offset estimation are expressed by pseudo codes as follows:
when a pseudo-random phase sequence pair is preferred (i.e. conjugate root value pair is used), it can be obtained according to the following equations (7), (7'):
eF＝mod((Im-In)/2*Rx,P)；
the above expression (6) can be simplified to:
eT＝(Im+In)/2；
the ranges of the time delay estimation value and the frequency offset estimation value obtained by the frequency domain correlation method and the time domain circumference correlation method are both 0: P-1, and actually, a negative value exists in the frequency offset, so that the frequency offset estimation value needs to be corrected, that is, when the frequency offset estimation value eF is greater than (P-1)/2, the value is eF minus P, that is, the estimation range of the frequency offset is- (P-1)/2 to (P-1)/2. The physical carrier frequency offset cFo is eF times the baseband bandwidth and divided by P, expressed in pseudo-code as:
cFo＝BW*eF/P；
where BW is the baseband bandwidth.
The pseudo-random phase sequence provided by the invention can be used for multi-user pilot frequency to carry out time-frequency estimation, and can also be used in the fields of radar ranging, speed measurement, satellite positioning system distance measurement and the like by actually utilizing the characteristic of time-frequency estimation.
Drawings
FIG. 1 is a time-frequency estimation flow chart of the frequency domain correlation method according to the present invention;
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Fig. 1 shows a time-frequency estimation flow chart of the frequency-domain correlation method of the present invention, a received baseband sampling signal Rx with a pilot symbol length is point-multiplied with local reference signals xchirp (rm) and xchirp (rn) to obtain time-domain correlation results Rx (rm) and Rx (rn), then a P-point fourier transform is performed to obtain frequency-domain correlation results f (rm) and f (rn), maximum absolute values of f (rm) and f (rn) are searched, and index values Im and In are obtained respectively, so that a delay estimation value eT is mod ((Im-In) × rxy, P), an estimation value eF is mod (Im-eT Rx, P), or mod (In-eT Ry, P), and the frequency offset needs to be further corrected and calculated to obtain a real physical frequency offset. Wherein, Rx and Ry are the root values of pseudo-random phase sequence forming pilot frequency, and iRxy is the inverse element of (Rx-Ry) mode P.
Fig. 1 is only an example of a frequency domain correlation method, and in fact, when Ry is preferably equal to P-Rx, a time-frequency estimation algorithm is simplified, and a slightly different flowchart may be obtained.
In summary, the present invention provides a pseudo-random phase sequence for time-frequency estimation, which is characterized in that the pseudo-random phase sequence has a linear time-frequency characteristic and good auto-correlation and cross-correlation characteristics, so that a group of pilot symbols which are uncorrelated with each other can be constructed by using the pseudo-random phase sequence and used as a pilot of a multi-user communication system. Furthermore, the invention provides a method for constructing pilot symbols by using the pseudo-random phase sequence pair, which comprises parallel pilot symbols and serial pilot symbols. Furthermore, the invention also provides a method for performing time-frequency estimation by adopting frequency domain correlation and time domain circumference correlation based on the pilot frequency symbol, and solves the key technology of the pilot frequency of a multi-user communication system.
The foregoing are merely exemplary embodiments of the present invention, which enable those skilled in the art to understand or practice the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A pseudo-random phase sequence for time-frequency estimation, comprising,
the phase characteristics of the pseudo-random phase sequence are quadratic functions;
the length of the pseudo-random phase sequence is a prime number P;
without loss of generality, the pseudo random phase bit sequence is expressed as:
xChirp(R)＝exp(-i*pi*R*(m+a).*(m+a+2*b)/P)； (1)
wherein exp is an exponential function; i is an imaginary unit; pi is the circumference ratio; p is the length of the pseudorandom phase sequence and is a prime number; r is the slope of the frequency changing linearly along with the time, called root value, and is a positive integer smaller than P, and the pseudo-random phase sequences with different root values R are not related to each other; m is a time-frequency point, the value range of the m is { -P/2:1: P/2-1}, namely m is an integral multiple of 1/2 and the interval is 1; a. b is an offset, which is an arbitrary integer, a is equivalent to a cyclic shift of the sequence, and b is equivalent to a frequency shift of the sequence.
2. The pseudo-random phase sequence of claim 1, a parallel pilot symbol for a multi-user communication system, comprising,
a parallel pilot symbol mpilot (rxy) is a power normalized superposition of a pair of the pseudo-random phase sequences xchirp (rx) and xchirp (ry) with different root values;
one parallel pilot symbol is of length P.
3. The pseudo-random phase sequence of claim 1, a serial pilot symbol for a multi-user communication system, comprising,
a serial pilot symbol, splalot (rxy), is a sequential concatenation of a pair of the pseudo-random phase sequences xchirp (rx) and xchirp (ry) with different root values;
one serial pilot symbol is 2P in length.
4. The time-frequency estimation method based on frequency domain correlation according to claim 1, 2 or 3, comprising,
respectively carrying out two kinds of frequency domain correlation on a received baseband sampling signal with the pilot frequency symbol length, and obtaining the position index value of the maximum absolute value of two kinds of correlation results;
and specifically calculating the position index values of the two maximum absolute values to respectively obtain a time delay estimation value and a frequency deviation estimation value.
5. The frequency-domain time-frequency estimation method according to claim 4, wherein the two kinds of frequency-domain correlations are performed on the received baseband sampling signal with a pilot symbol length, and a position index value of a maximum absolute value of the two kinds of correlation results is obtained, including,
local time domain reference signals xchirp (rm) and xchirp (rn) for frequency domain correlation;
receiving a baseband sampling signal rx with a pilot symbol length;
performing point multiplication on the received signal rx and a local time domain reference signal xChirp (Rm) to obtain a time domain correlation value rx (Rm);
performing point multiplication on a received signal rx and a local time domain reference signal xChirp (Rn) to obtain a time domain correlation value rx (Rn);
performing P-point Fourier transform on the time domain correlation value rx (Rm) to obtain a frequency domain correlation value F (Rm);
performing P-point Fourier transform on the time domain correlation value rx (Rn) to obtain a frequency domain correlation value F (Rn);
searching the maximum absolute value of the frequency domain correlation value F (Rm) and obtaining an index value Im thereof;
searching the maximum absolute value of the frequency domain correlation value F (Rn) and obtaining an index value In of the maximum absolute value;
wherein, xChirp (Rm) and xChirp (Rn) are the conjugates of xChirp (Rx) and xChirp (Ry) that form the pilot symbols, or the conjugates of xChirp (Rx) and xChirp (Ry) that form the pilot symbols.
6. The frequency-domain time-frequency estimation method according to claim 4, wherein the specific calculation is performed on the index values of the two maximum absolute values to obtain the time delay estimation value and the frequency offset estimation value, respectively, including,
the time delay estimated value eT is the difference between the index value Im and In of the maximum absolute value and is multiplied by iRxy, and the modulus of P is obtained, wherein iRxy is the inverse element of the difference modulus P of the root values Rx and Ry;
frequency offset estimation value eFm is obtained by subtracting the product of root Rx and delay estimation value eT from index value Im of the maximum absolute value and taking the modulus of P, and frequency offset estimation value eFn is obtained by subtracting the product of root Ry and delay estimation value eT from index value In of the maximum absolute value and taking the modulus of P;
if eFm is equal to eFn, then the estimated frequency offset eF is equal to eFm or eFn, and the current time delay estimate eT and the estimated frequency offset eF are valid, otherwise, time-frequency estimation continues.
7. The time-frequency estimation method based on time-domain circular correlation according to claim 1, 2 or 3, comprising,
respectively carrying out two time domain circular correlations on a received baseband sampling signal with a pilot frequency symbol length, and obtaining a position index value of a maximum absolute value of two correlation results;
and specifically calculating the index values of the two maximum absolute values to respectively obtain a time delay estimation value and a frequency deviation estimation value.
8. The time-domain time-frequency estimation method according to claim 7, wherein said performing two kinds of time-domain circular correlations on the received baseband sampling signal with a pilot symbol length respectively and obtaining the position index value of the maximum absolute value of the two kinds of correlation results comprises,
local frequency domain reference signals fcirp (rm) and fcirp (rn) for time domain circular correlation;
a received frequency domain baseband sampling signal Frx of a pilot symbol length and transformed to the frequency domain;
performing point multiplication on the frequency domain receiving signal Frx and a local frequency domain reference signal fChirp (Rm) to obtain a frequency domain correlation value F (Rm);
performing point multiplication on the frequency domain receiving signal Frx and a local frequency domain reference signal fChirp (Rn) to obtain a frequency domain correlation value F (Rn);
performing P-point inverse Fourier transform on the frequency domain correlation value F (Rm) to obtain a time domain correlation value T (Rm);
carrying out P-point Fourier inverse transformation on the frequency domain correlation value F (Rn) to obtain a time domain correlation value T (Rn);
searching the maximum absolute value of the time domain correlation value T (Rm) and obtaining an index value Im thereof;
searching the maximum absolute value of the time domain correlation value T (Rn) and obtaining an index value In thereof;
wherein, fchirp (rm) and fchirp (rn) are conjugates of xchirp (rx) and xchirp (ry) which form pilot symbols after being transformed into frequency domain, or conjugates of xchirp (rx) and xchirp (ry) which form two symbols after being transformed into frequency domain; the frequency domain received signal Frx is a baseband received signal of length P transformed to the frequency domain, or two consecutive baseband received signals of length P transformed to the frequency domain.
9. The time-domain time-frequency estimation method according to claim 7, wherein the specific calculation is performed on the index values of the two maximum absolute values to obtain the time delay estimation value and the frequency offset estimation value, respectively, including,
the time delay estimated value eT is the difference between the index value Im of the maximum absolute value multiplied by Rx and the index value In of the maximum absolute value multiplied by Ry multiplied by iRxy and the modulus of P is taken, wherein iRxy is the inverse element of the modulus P of the difference between the root value Rx and Ry;
frequency offset estimation value eFm is obtained by subtracting time delay estimation value eT from index value Im of the maximum absolute value and multiplying by root value Rx, and frequency offset estimation value eFn is obtained by subtracting time delay estimation value eT from index value In of the maximum absolute value and multiplying by root value Ry and multiplying by P;
if eFm is equal to eFn, then the estimated frequency offset eF is equal to eFm or eFn, and the current time delay estimate eT and the estimated frequency offset eF are valid, otherwise, time-frequency estimation continues.
10. According to claim 6 or 9, correcting the frequency offset estimation value obtained by the frequency domain correlation method and the time domain circular correlation method and obtaining the physical frequency offset, including,
when the estimated value of frequency offset eF is greater than (P-1)/2, eF minus P, i.e., the estimated range of frequency offset is- (P-1)/2 to (P-1)/2;
physical carrier frequency offset cFo is the corrected frequency offset estimate eF multiplied by the baseband bandwidth BW and divided by P.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110648177.2A CN113381959A (en) | 2021-06-10 | 2021-06-10 | Pseudo-random phase sequence for time-frequency estimation and time-frequency estimation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110648177.2A CN113381959A (en) | 2021-06-10 | 2021-06-10 | Pseudo-random phase sequence for time-frequency estimation and time-frequency estimation method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113381959A true CN113381959A (en) | 2021-09-10 |
Family
ID=77573521
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110648177.2A Pending CN113381959A (en) | 2021-06-10 | 2021-06-10 | Pseudo-random phase sequence for time-frequency estimation and time-frequency estimation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113381959A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114337981A (en) * | 2022-03-11 | 2022-04-12 | 深圳市华普微电子有限公司 | Method for generating pilot signal of ZC-like sequence and time-frequency estimation |
| CN114337982A (en) * | 2022-03-16 | 2022-04-12 | 深圳市华普微电子有限公司 | Chirp pilot signal generation and time-frequency estimation method |
| CN114338296A (en) * | 2021-11-24 | 2022-04-12 | 北京智芯微电子科技有限公司 | Channel estimation and demodulation method, apparatus and medium, receiving device, transmitting device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0952713A2 (en) * | 1998-04-24 | 1999-10-27 | Nec Corporation | A frequency and timing synchronization circuit making use of a chirp signal |
| US20170187557A1 (en) * | 2014-04-16 | 2017-06-29 | Shanghai National Engineering Research Center Of Digital Television Co., Ltd. | Preamble symbol generation and receiving method, and frequency-domain symbol generation method and device |
| CN112564843A (en) * | 2020-11-19 | 2021-03-26 | 无锡泽太微电子有限公司 | Micro frequency hopping method and micro frequency hopping multiple access communication system |
-
2021
- 2021-06-10 CN CN202110648177.2A patent/CN113381959A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0952713A2 (en) * | 1998-04-24 | 1999-10-27 | Nec Corporation | A frequency and timing synchronization circuit making use of a chirp signal |
| US20170187557A1 (en) * | 2014-04-16 | 2017-06-29 | Shanghai National Engineering Research Center Of Digital Television Co., Ltd. | Preamble symbol generation and receiving method, and frequency-domain symbol generation method and device |
| CN112564843A (en) * | 2020-11-19 | 2021-03-26 | 无锡泽太微电子有限公司 | Micro frequency hopping method and micro frequency hopping multiple access communication system |
Non-Patent Citations (3)
| Title |
|---|
| 于凤芹, 曹家麟: "基于分数阶傅里叶变换的多分量Chirp信号的检测与参数估计", 电声技术, no. 01 * |
| 王东辉;庞伟正;刘成德;: "基于连续小波互谱的宽带Chirp信号DOA估计", 应用科技, no. 04 * |
| 赵晶晶;霍凯;刘永祥;杨小琪;: "基于循环前缀的相位编码OFDM雷达多普勒频移估计和补偿", 电子与信息学报, no. 04 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114338296A (en) * | 2021-11-24 | 2022-04-12 | 北京智芯微电子科技有限公司 | Channel estimation and demodulation method, apparatus and medium, receiving device, transmitting device |
| CN114338296B (en) * | 2021-11-24 | 2023-08-18 | 北京智芯微电子科技有限公司 | Channel estimation and demodulation method, device and medium, receiving equipment, and transmitting equipment |
| CN114337981A (en) * | 2022-03-11 | 2022-04-12 | 深圳市华普微电子有限公司 | Method for generating pilot signal of ZC-like sequence and time-frequency estimation |
| CN114337982A (en) * | 2022-03-16 | 2022-04-12 | 深圳市华普微电子有限公司 | Chirp pilot signal generation and time-frequency estimation method |
| CN114337982B (en) * | 2022-03-16 | 2022-07-15 | 深圳市华普微电子有限公司 | Chirp pilot signal generation and time-frequency estimation method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113381959A (en) | Pseudo-random phase sequence for time-frequency estimation and time-frequency estimation method | |
| CN101437005B (en) | Method for estimating integer multiple frequency deviation with timing error during communication synchronization process | |
| EP1980030B1 (en) | Method for improved synchronization and information transmission in a communication system | |
| CN101599938B (en) | A Joint Time-Domain Synchronization Method for Receivers of Orthogonal Frequency Division Multiplexing UWB System | |
| EP1843501B1 (en) | Method for synchronising a signal frame in a telecommunication system | |
| CN102271108B (en) | Rapid calculation method and device for discrete Fourier transform of Zadoff-Chu sequence | |
| CN102223226B (en) | SC-FDM system time-frequency synchronization device and synchronization method | |
| JP2002539667A5 (en) | ||
| CN104125190A (en) | OFDM (orthogonal frequency division multiplexing) system symbol timing synchronization realizing method suitable for low-signal-to-noise-ratio channel environments | |
| CN103580718A (en) | Quick time-frequency synchronizing method under low signal-to-noise ratio | |
| US20150029997A1 (en) | Method and apparatus for generating code sequence in a communication system | |
| CN114337981A (en) | Method for generating pilot signal of ZC-like sequence and time-frequency estimation | |
| EP1441463A1 (en) | A method and device for carrier frequency synchronization | |
| CN102255722B (en) | Improve the method for synchronizing in communication system and information transmission | |
| CN104125188A (en) | OFDM (Orthogonal Frequency Division Multiplexing) frequency synchronizing method based on Zadoff-Chu sequence | |
| CN106453192B (en) | A kind of symbol timing synchronization method and system based on the complementary binary sequence pairs of shielding | |
| US20080317184A1 (en) | Method for transmitting synchronization signal in mobile multimedia system | |
| CN112039816B (en) | Downlink synchronization method for narrow-band Internet of things system | |
| CN113315541A (en) | Pseudo-random phase sequence spread spectrum modulation method | |
| CN101330482A (en) | Method for implementing rapid channel estimation using FWT in DMB-T system | |
| US8126103B2 (en) | Frame synchronization using correlation between permuted sequences | |
| US8423598B2 (en) | Circuit arrangement and method for receiving specially designed coded signals | |
| KR100688086B1 (en) | Correlator and its method using symmetry of correlation coefficient | |
| CN101390355B (en) | Efficient bank of correlators for sets of GCL sequences | |
| US20130170576A1 (en) | Assembly and Method for Detecting Multiple Level Signals |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| AD01 | Patent right deemed abandoned |
Effective date of abandoning: 20250318 |
|
| AD01 | Patent right deemed abandoned |