CN112600777B - Frequency offset compensation method and device suitable for 5G-NR receiver - Google Patents
Frequency offset compensation method and device suitable for 5G-NR receiver Download PDFInfo
- Publication number
- CN112600777B CN112600777B CN202011423663.6A CN202011423663A CN112600777B CN 112600777 B CN112600777 B CN 112600777B CN 202011423663 A CN202011423663 A CN 202011423663A CN 112600777 B CN112600777 B CN 112600777B
- Authority
- CN
- China
- Prior art keywords
- value
- frequency offset
- target
- phase
- waveform
- 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.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L27/00—Modulated-carrier systems
- H04L27/0014—Carrier regulation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L27/00—Modulated-carrier systems
- H04L27/0014—Carrier regulation
- H04L2027/0024—Carrier regulation at the receiver end
- H04L2027/0026—Correction of carrier offset
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Circuits Of Receivers In General (AREA)
- Digital Transmission Methods That Use Modulated Carrier Waves (AREA)
Abstract
The invention provides a frequency offset compensation method and a device suitable for a 5G-NR receiver, wherein the method comprises the following steps: the device comprises a subframe-level symbol scheduler, a first register, an initial phase and frequency offset calculator, a second register, a phase accumulator, a table look-up and waveform generator, a CORDIC calibrator and a spectrum calibrator. The invention can realize high-precision frequency offset compensation on the premise of reducing resource occupation.
Description
Technical Field
The present invention relates to the field of data processing technologies, and in particular, to a frequency offset compensation method and apparatus for a 5G-NR receiver.
Background
Wireless communication is divided into two categories: wireless Mobile Communication (Wireless Mobile Communication) and Wireless Local Area Network (LAN) Communication.
In a wireless communication system, frequency offset compensation is often required for a calculated waveform. Assuming that the signal sampling rate is fs, the frequency offset to be compensated is f0, the calculated waveform is x (k) (k =0, … n), and the signal after frequency offset compensation is:
y (k) = x (k) = cos (f 0/fs 2 × pi (k-1)) +1i × sin (f 0/fs 2 × pi (k-1))), k =1, … n. Thereby realizing time domain frequency offset compensation. The existing frequency offset compensation schemes include a table lookup method, a CORDIC implementation method, a secondary table lookup method and the like.
In order to realize high-precision frequency offset compensation, the table lookup method needs to store a relatively large table, and occupies resources.
CORDIC implementations may increase latency, require a large number of iterations to achieve higher accuracy, and may also increase power consumption.
The secondary table lookup method is a better compromise scheme, can better balance storage space and power consumption, but needs to do complex multiplication once after the secondary table lookup, also consumes certain resources, and the performance is also influenced by the size of the secondary table lookup.
Disclosure of Invention
The invention aims to provide a frequency offset compensation method and a frequency offset compensation device suitable for a 5G-NR receiver, which can realize high-precision frequency offset compensation on the premise of reducing resource occupation.
The technical scheme provided by the invention is as follows:
the invention provides a frequency offset compensation method suitable for a 5G-NR receiver, which comprises the following steps:
the sub-frame level symbol scheduler performs sub-frame level scheduling on the input sub-frame data according to the sub-frame parameters obtained from the first register, and outputs the scheduled data and the symbol level data length and the cyclic prefix length parameters thereof;
the initial phase and frequency offset calculator outputs a frequency offset compensation value according to the frequency offset and initial phase parameters obtained from the second register and the refreshing identification bit;
the phase accumulator performs accumulation calculation according to the symbol-level data length, the cyclic prefix length parameter, the frequency offset and initial phase parameter, the refreshing identification bit and the frequency offset compensation value to obtain a phase accumulated value;
the table look-up and waveform generator outputs corresponding first target numerical values, third target numerical values and calculated waveforms according to the phase accumulated values and a preset mapping table;
the CORDIC calibrator calculates and outputs a waveform numerical value after frequency offset compensation according to the calculated waveform, the first target numerical value and the third target numerical value;
and the frequency spectrum calibrator performs complex multiplication calculation according to the waveform numerical value and the scheduled data to obtain frequency offset compensation output data to complete frequency spectrum calibration.
Further, the sub-frame symbol scheduler comprises, before performing sub-frame scheduling on the input sub-frame data according to the sub-frame parameters obtained from the first register, the steps of:
when the radio frequency is restarted or reset and the 5G-NR receiver receives a calculated waveform, the first register updates subframe parameters corresponding to the calculated waveform;
when a subframe data is input to the subframe-level symbol scheduler, the second register updates the self-stored frequency offset, initial phase parameter and refresh identification bit.
Further, the step of outputting the frequency offset compensation value by the initial phase and frequency offset calculator according to the frequency offset and initial phase parameters obtained from the second register and the refreshing identification bit comprises the following steps:
the initial phase and frequency offset calculator judges whether to calculate updated frequency offset and initial phase parameters according to the refreshing identification bits;
when the refresh identification bit is determined to accord with the preset update condition, the initial phase and frequency offset calculator obtains the updated initial phase parameter and substitutes the initial phase parameter into a normalization formula to calculate and obtain a frequency offset compensation value according to the frequency offset parameter;
fn_delta＝round(fn/fs*2^32)
when the refresh identification bit is determined not to be in accordance with the preset update condition, the initial phase and frequency offset calculator does not update the frequency offset and the initial phase parameters;
wherein, fn is the frequency offset value of the current subframe data, fs is the sampling frequency of the current subframe data, and round () is a rounding function.
Further, the step of performing accumulation calculation by the phase accumulator according to the symbol-level data length, the cyclic prefix length parameter, the frequency offset, the initial phase parameter, the refresh identification bit, and the frequency offset compensation value to obtain a phase accumulated value includes:
when the refresh identification bit is determined to meet the preset update condition, the phase accumulator performs accumulation calculation according to the symbol-level data length, the cyclic prefix length parameter, the updated initial phase parameter and the updated frequency offset compensation value to obtain a corresponding phase accumulated value; or the like, or, alternatively,
and when the refresh identification bit is determined not to meet the preset update condition, the phase accumulator performs accumulation calculation according to the symbol-level data length, the cyclic prefix length parameter, the frequency offset and the initial phase parameter to obtain a corresponding phase accumulated value.
Further, the table look-up and waveform generator outputs the corresponding first target value, the third target value and the calculated waveform according to the phase accumulated value and the preset mapping table, and the method comprises the following steps:
the table lookup and waveform generator compares the preset value with the phase accumulated value and updates and obtains a target phase accumulated value according to a comparison result;
the table lookup and waveform generator extracts three high bits from the target phase accumulated value to obtain the first target value;
the table lookup and waveform generator extracts twenty-nine lower bits of the target phase accumulated value to obtain an extracted field, and updates and obtains a target extracted field according to the first target value and the extracted field;
the table look-up and waveform generator extracts seven high bits from the target extraction field to obtain a second target value, and queries a preset mapping table according to the second target value to obtain a corresponding calculated waveform;
and the table look-up and waveform generator extracts seventeenth to twenty-second bits of the target extraction field to obtain the third target value.
Further, the table lookup and waveform generator compares the preset value with the phase accumulated value, and the step of updating and acquiring the target phase accumulated value according to the comparison result comprises the following steps:
if the phase accumulated value is larger than the preset value, the table look-up and waveform generator determines that the target phase accumulated value is equal to the difference value between the phase accumulated value and first preset data;
and if the phase accumulated value is not larger than the preset value, the table look-up and waveform generator determines that the target phase accumulated value is equal to the phase accumulated value.
Further, the table look-up and waveform generator extracts twenty-nine low bits of the target phase accumulated value to obtain an extracted field, and updating and acquiring the target extracted field according to the first target value and the extracted field comprises the following steps:
the table look-up and waveform generator judges whether the first target value is an odd number or not;
if the first target value is an odd number, the table look-up and waveform generator determines that the target extraction field is equal to the difference value between second preset data and the extraction field;
and if the first target value is an even number, the table look-up and waveform generator determines that the target extraction field is equal to the extraction field.
Further, the step of calculating, by the CORDIC calibrator, the waveform value after the output frequency offset compensation according to the calculated waveform, the first target value, and the third target value includes:
the CORDIC calibrator performs iterative computation according to the computed waveform and a third target numerical value to output a calibrated computed waveform;
and the CORDIC calibrator calculates and outputs a waveform numerical value after frequency offset compensation according to the first target numerical value and the calibrated calculated waveform.
The invention also provides a frequency offset compensation device suitable for the 5G-NR receiver, which comprises:
the sub-frame level symbol scheduler is used for performing sub-frame level scheduling on the input sub-frame data according to the sub-frame parameters obtained from the first register and outputting the scheduled data and the symbol level data length and cyclic prefix length parameters thereof;
the initial phase and frequency offset calculator is used for outputting a frequency offset compensation value according to the frequency offset and initial phase parameters obtained from the second register and the refreshing identification bit;
the phase accumulator is used for carrying out accumulation calculation according to the symbol-level data length, the cyclic prefix length parameter, the frequency offset and initial phase parameter, the refreshing identification bit and the frequency offset compensation value to obtain a phase accumulated value;
the table look-up and waveform generator is used for outputting a corresponding first target value, a corresponding third target value and a corresponding calculated waveform according to the phase accumulated value and a preset mapping table;
the CORDIC calibrator is used for calculating and outputting a waveform numerical value after frequency offset compensation according to the calculated waveform, the first target numerical value and the third target numerical value;
and the frequency spectrum calibrator is used for performing complex multiplication calculation according to the waveform numerical value and the scheduled data to obtain frequency offset compensation output data to finish frequency spectrum calibration.
The frequency offset compensation method and the frequency offset compensation device suitable for the 5G-NR receiver can realize high-precision frequency offset compensation on the premise of reducing resource occupation.
Drawings
The above features, technical features, advantages and implementations of a method and apparatus for frequency offset compensation in a 5G-NR receiver will be further described in the following detailed description of preferred embodiments in conjunction with the accompanying drawings.
FIG. 1 is a flow chart of one embodiment of a method of frequency offset compensation for a 5G-NR receiver of the present invention;
FIG. 2 is a schematic diagram of a table lookup and waveform generator of a frequency offset compensation apparatus for a 5G-NR receiver according to the present invention;
FIG. 3 is a schematic diagram of a CORDIC calibrator of a frequency offset compensation apparatus for a 5G-NR receiver according to the present invention;
FIG. 4 is a schematic diagram illustrating the effect of a frequency offset compensation method applied to a 5G-NR receiver according to the present invention;
fig. 5 is a schematic diagram of a frequency offset compensation apparatus suitable for a 5G-NR receiver according to the present invention.
Detailed Description
In the following description, for purposes of explanation and not limitation, specific details are set forth such as particular system structures, techniques, etc. in order to provide a thorough understanding of the embodiments of the present application. It will be apparent, however, to one skilled in the art that the present application may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present application with unnecessary detail.
It will be understood that the terms "comprises" and/or "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
For the sake of simplicity, only the parts relevant to the present invention are schematically shown in the drawings, and they do not represent the actual structure as a product. Moreover, in the interest of brevity and understanding, only one of the components having the same structure or function is illustrated schematically or designated in some of the drawings. In this document, "a" means not only "only one of this but also a case of" more than one ".
It should be further understood that the term "and/or" as used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.
In addition, in the description of the present application, the terms "first", "second", and the like are used only for distinguishing the description, and are not intended to indicate or imply relative importance.
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following description will be made with reference to the accompanying drawings. It is obvious that the drawings in the following description are only some examples of the invention, and that for a person skilled in the art, other drawings and embodiments can be derived from them without inventive effort.
Subcarrier spacing, frame, subframe, slot, symbol of 5 GNR.
In 5GNR, the length of a defined frame (radios frame) is 10ms, each frame is divided into two identical half-frames (half frames), each half-frame is 5ms, and each frame can be further divided into 10 subframes (subframes), each subframe is 1ms.
There are 5 alternative subcarrier spacings in the 5 GNRs, including 15khz, 30khz, 60khz, 120khz, 240khz. The maximum value of the subcarrier spacing is determined by a Cyclic Prefix (CP for short), because the larger the subcarrier spacing is, the shorter the OFDM symbol duration is, the shorter the CP is, one of the functions of the CP is to resist multipath delay, and the CP is greater than the maximum multipath delay, so that the multipath delay cannot be overcome by the excessively short CP.
The length of the slot (slot) in the 5GNR depends on the subcarrier spacing, the wider the subcarrier spacing, the shorter the duration of the slot.
The symbol in 5GNR is the basic unit of modulation.
One embodiment of the present invention, as shown in fig. 1, is a frequency offset compensation method for a 5G-NR receiver, comprising the steps of:
s10, the sub-frame level symbol scheduler performs sub-frame level scheduling on input sub-frame data according to the sub-frame parameters obtained from the first register, and outputs the scheduled data, the symbol level data length of the scheduled data and the cyclic prefix length parameters of the scheduled data;
s20, the initial phase and frequency offset calculator outputs a frequency offset compensation value according to the frequency offset and initial phase parameters obtained from the second register and the refreshing identification bit;
s30, the phase accumulator carries out accumulation calculation according to the symbol-level data length, the cyclic prefix length parameter, the frequency offset and initial phase parameter, the refreshing identification bit and the frequency offset compensation value to obtain a phase accumulated value;
s40, looking up a table and outputting a corresponding first target value, a corresponding third target value and a corresponding calculated waveform by a waveform generator according to the phase accumulated value and a preset mapping table;
the S50CORDIC calibrator calculates and outputs a waveform numerical value after frequency offset compensation according to the calculated waveform, the first target numerical value and the third target numerical value;
and S60, the frequency spectrum calibrator performs complex multiplication calculation according to the waveform numerical value and the scheduled data to obtain frequency offset compensation output data to complete frequency spectrum calibration.
In the embodiment, the time domain frequency offset compensation is performed by combining the table lookup method and the CORDIC, so that higher compensation precision can be realized by using smaller storage resources and operation iteration times, and the requirement on a multiplier is reduced. And selecting a corresponding subframe structure according to the current subcarrier interval by pre-storing a 5G-NR subframe structure and symbol scheduling specified by a protocol, and realizing frequency offset compensation only on useful data by scheduling so as to avoid waste of power consumption. By mapping 0-2 × pi to 0-pi × 4, the operation amount is not required to be increased by applying a trigonometric function transformation formula, and only a table with 1/8 of the original size needs to be stored, so that the storage size required by the table lookup method is greatly simplified.
Based on the foregoing embodiment, before performing the sub-frame level scheduling on the input sub-frame data according to the sub-frame parameter obtained from the first register, the S10 sub-frame level symbol scheduler includes the steps of:
s01, after the radio frequency is restarted or reset, when the 5G-NR receiver receives the calculated waveform, the first register updates the subframe parameter corresponding to the calculated waveform;
s02, when a subframe data is input to the subframe-level symbol scheduler, the second register updates the self-stored frequency offset, initial phase parameters and the refresh identification bit.
Based on the foregoing embodiment, the step of S20, wherein the step of outputting the frequency offset compensation value by the initial phase and frequency offset calculator according to the frequency offset and initial phase parameters obtained from the second register and the refresh flag bit includes:
s21, the initial phase and frequency offset calculator judges whether to calculate updated frequency offset and initial phase parameters according to the refreshing identification bits;
s22, when the refreshing identification bit is determined to accord with the preset updating condition, the initial phase and frequency offset calculator obtains the updated initial phase parameter and substitutes the initial phase parameter into the normalization formula according to the frequency offset parameter to calculate and obtain a frequency offset compensation value;
fn_delta＝round(fn/fs*2^32)
s23, when the refreshing identification bit is determined not to be in accordance with the preset updating condition, the initial phase and frequency offset calculator does not update the frequency offset and the initial phase parameters;
wherein, fn is the frequency offset value of the current sub-frame data, fs is the sampling frequency of the current sub-frame data, round () is the integer function,/represents the division sign,/represents the multiplication sign, and ^ represents the 32 th power of 2.
Based on the foregoing embodiment, the step of performing accumulation calculation by the phase accumulator according to the symbol-level data length, the cyclic prefix length parameter, the frequency offset, the initial phase parameter, the refresh flag, and the frequency offset compensation value to obtain the phase accumulated value includes:
when the refresh identification bit is determined to meet the preset update condition, the phase accumulator performs accumulation calculation according to the symbol-level data length, the cyclic prefix length parameter, the updated initial phase parameter and the updated frequency offset compensation value to obtain a corresponding phase accumulated value; or the like, or a combination thereof,
and when the refresh identification bit is determined not to be in accordance with the preset update condition, the phase accumulator performs accumulation calculation according to the symbol-level data length, the cyclic prefix length parameter, the frequency offset and the initial phase parameter to obtain a corresponding phase accumulated value.
Based on the aforementioned embodiment, the table lookup and waveform generator outputting the corresponding first target value, the third target value and the calculated waveform according to the phase accumulated value and the preset mapping table includes the steps of:
the table lookup and waveform generator compares the preset value with the phase accumulated value and updates and obtains a target phase accumulated value according to a comparison result;
looking up a table and extracting high three bits from the target phase accumulated value by a waveform generator to obtain a first target value;
the table lookup and waveform generator extracts twenty-nine lower bits of the target phase accumulated value to obtain an extracted field, and the target extracted field is obtained according to the first target value and the extracted field;
the table look-up and waveform generator extracts seven high bits from the target extraction field to obtain a second target value, and the preset mapping table is queried according to the second target value to obtain a corresponding calculated waveform;
and the table look-up and waveform generator extracts seventeenth to twenty-second bits from the target extraction field to obtain a third target value.
Based on the embodiment, the table lookup and waveform generator compares the preset value with the phase accumulated value, and updates and acquires the target phase accumulated value according to the comparison result, which comprises the following steps:
if the phase accumulated value is larger than the preset value, the table look-up and waveform generator determines that the target phase accumulated value is equal to the difference value between the phase accumulated value and the first preset data;
if the phase accumulated value is not greater than the preset value, the table look-up and waveform generator determines that the target phase accumulated value is equal to the phase accumulated value.
Based on the aforementioned embodiment, the table lookup and waveform generator extracts twenty-nine low bits from the target phase accumulated value to obtain the extracted field, and the step of updating and acquiring the target extracted field according to the first target value and the extracted field includes:
the table look-up and waveform generator judges whether the first target value is an odd number;
if the first target value is an odd number, determining that the target extraction field is equal to the difference value of the second preset data and the extraction field by the table look-up and waveform generator;
if the first target value is an even number, the table look-up and waveform generator determines that the target extraction field is equal to the extraction field.
Based on the foregoing embodiment, the calculating, by the CORDIC calibrator, the waveform value after the frequency offset compensation is output according to the calculated waveform, the first target value and the third target value includes:
the CORDIC calibrator performs iterative computation according to the computed waveform and a third target value to output a calibrated computed waveform;
and the CORDIC calibrator calculates and outputs a waveform numerical value after frequency offset compensation according to the first target numerical value and the calibrated calculation waveform.
Specifically, the frequency offset compensation method applicable to the 5G-NR receiver in the invention is mainly used for the frequency offset compensation of the 5G-NR receiver, and comprises the following parts: the method comprises the steps of subframe and symbol scheduling, initial phase and frequency offset calculation, phase accumulation, table lookup and waveform generation, CORDIC calibration and frequency offset calibration. The method comprises the following specific steps:
s1, after the radio frequency is restarted, refreshing subframe-level symbol scheduling parameters and corresponding subframe structure parameters in a register 1.
And S2, obtaining a sub-frame-level symbol scheduling parameter according to the S1, inputting the sub-frame-level symbol scheduler, performing sub-frame-level scheduling on data according to the input parameter, and outputting the symbol-level data length of the useful part, the CP length parameter and the scheduled data. And inputting the frequency offset, the initial phase parameter and the FreshFlag parameter in the pre-refresh register 2 by each subframe data.
S3, inputting the initial phase and frequency offset parameters and the FreshFlag parameters in the register 2 into an initial phase and frequency offset calculator:
s301, judging whether the input initial phase and frequency offset parameters are utilized or not according to the freshflag, calculating to obtain and output a U32 initial phase and normalized U32 frequency offset compensation value. If not, no calculation is performed. The normalization mode is as follows: frequency offset compensation value (f 0_ delta = round (f 0/fs 2^ 32)).
And S4, inputting the data length and CP length parameters of each symbol of the current subframe in S2, the initial phase, the freshflag and the frequency offset value (f 0_ delta) output in S301 into a phase accumulator. The phase accumulator is a register that accumulates the phase linearly.
S401, judging whether to refresh the frequency offset value and the initial phase output by the S3 according to the freshflag. If not, the frequency offset value and the initial phase are not changed.
And S402, accumulating and calculating a phase accumulated value according to the data length of each symbol of the subframe, the CP length parameter, the frequency offset value output by the S3 or updated by the S401 and the initial phase.
Each time the phase accumulated value is accumulated, when the accumulated value reaches the first data position of the next symbol when each symbol boundary, the length of the symbol CP is accumulated for +1 times, and then the accumulated value is output again.
And S5, inputting the accumulated value delta output in the S4 into a table look-up and waveform generator to output Findex (namely the first target value of the invention), index1, index2 and waveform x, wherein the specific process is shown in FIG. 2.
S501, if delta is larger than 2^32-1, then delta = delta-2^32.
S502, findex = (30 th to 32 th bits of delta) is taken.
S503, taking delta _ n (namely the extraction field) = (1 to 29 bits of delta), and if Findex (three high bits) is an odd number, refreshing the delta _ n =2^29-1-delta _ n;
s504, take index1 (i.e., the second target value of the present invention) = 23 th to 29 th bits of delta _ n, take index2 (i.e., the third target value of the present invention) = 17 th to 22 th bits of delta _ n.
S505, look up the LUT1 using index1 (i.e., the second target value of the present invention), and obtain waveform data x.
Inputting x obtained in S505, index2 obtained in S504 (i.e. the third target value of the present invention) and Findex calculated in S502 into the CORDIC calibrator for iterative calculation, where the CORDIC iterative block diagram is shown in fig. 3:
s610, performing CORDIC calculation on x and index2 (i.e. the third target value of the present invention), calibrating x, and outputting the calibrated waveform x.
Initializing z (1) = index2 (i.e., third target value of the present invention), x1 (1) = real (x), x2 (1) = imag (x), antantext is a stored atan table.
Where atantext = round (atan (2 ^ - (0)))/(pi) × 2^ 31), cordic iterates 6 times.
S610 specifically includes the steps of:
s611, ifz (i) = =0, x1 (i + 1) = x1 (i), and x2 (i + 1) = x2 (i) are output, and the iteration is skipped. Otherwise d = sign (z (i)).
S612、x1(i+1)＝x1(i)-d*(x2(i)>>7+i)；
x2(i+1)＝x2(i)+d*(x1(i)>>7+i)；
z(i+1)＝z(i)-d*atantext(i)；
When the iteration is performed i times, after the iteration is completed, updating the calibrated x = x1 (i + 1) +1i × x2 (i + 1);
and S620, judging an output waveform y according to Findex and x after calibration in S610.
The real () is a real part evaluation function, the imag () is an imaginary part evaluation function, and the i is an iterative sequence number value.
The invention realizes the time domain frequency offset compensation of the 5G receiver mainly by adding a 5GNR symbol scheduling, simplifying a table look-up method and a CORDIC (Coordinate Rotation Digital Computer) algorithm, namely a Coordinate Rotation Digital calculation method, and replacing multiplication operation by basic addition and shift operation, so that functions such as trigonometric function, multiplication, evolution, inverse trigonometry, exponent and the like are not needed for vector Rotation and orientation calculation.
In a wireless communication system, it is often necessary to compensate a received signal for a frequency offset. If the signal sampling rate is fs, the frequency offset to be compensated is f0, the received signal is x (k) (k =0, … n), the signal after frequency offset compensation is y (k), and x and y are complex numbers, then
y(k)＝x(k)*{cos(f0/fs*2*pi*(k-1))+1i*sin(f0/fs*2*pi*(k-1))}，k＝1，…n
The equivalence is regarded as:
y (k) = x (k) { cos (delta) +1i × sin (delta) }, k =1, … n. delta is between 0 and 2 xpi.
Thereby realizing frequency offset compensation.
The invention splits the process into two processes, and splits delta into: delta = delta1+ delta2. Wherein delta1= m × 2 × pi/1024, wherein m is a non-negative integer.
The above formula may be equivalent to:
y(k)＝x(k)*{cos(delta)+1i*sin(delta)}
＝x(k)*{cos(delta1+delta2)+1i*sin(delta1+delta2)}
＝x(k)*{cos(delta1)+1i*sin(delta1))*(cos(delta2)+1i*sin(delta2)}
＝x(k)*z
z1＝(cos(delta1)+1i*sin(delta1))；
z2＝(cos(delta2)+1i*sin(delta2))；
z＝(cos(delta)+1i*sin(delta))＝z1*z2；
z can be obtained by multiplying z1 and z2 by calculating z1 and then z 2.
Since delta is an integer multiple of 2 xpi/1024, it can be obtained by table lookup. z1 x z2 can be obtained by CORDIC rotating z1 by delta2 angle. The CORDIC is added with an early exit mechanism, so that invalid iteration is avoided, and power consumption is wasted.
For any angle a,0< = a <2 x pi, which can be written as a = b + m x pi/4, where 0= < b < pi/4,m is a number between 0 and 7. Then the
cos(a)＝cos(b+m*pi/4)；
For m =0; cos (a) = cos (b), m =1, cos (a) = sin (pi/4-b);
by analogy, cos (a) can always be expressed as the result of taking the positive cosine values of two angles b, pi/4-b and the negative sign of the positive cosine values. Sin (a) may also be so expressed, as well. We can calculate the angular sine-cosine values in the range of 0-2 x pi and calculate the results in the range of 0-pi/4. Meanwhile, only a cos table and a sin table need to be stored, and then the results of b and pi/4-b can be taken out in a positive and negative sequence number taking mode.
The invention relates to a frequency offset compensation device suitable for a 5G-NR receiver, which comprises a subframe-level symbol scheduler, an initial phase and frequency offset calculator, a phase accumulator, a table look-up and waveform generator, a CORDIC calibrator, a frequency offset calibrator, a register 1 and a register 2. The two registers can be configured with frequency offset, initial phase and subframe structures, and are used for scheduling of the subframe level symbol scheduler and calculation of the initial phase and frequency offset compensator. And the sub-frame level symbol scheduler schedules the data according to sub-frames and symbols so that the scheduled data is output as each symbol data with CP removed. And the initial phase and frequency offset calculator calculates a normalized frequency offset compensation value according to the frequency offset and the sampling rate after the register is refreshed each time. And the phase accumulator is used for accumulating phases according to the length of each symbol in the input subframe and the length parameter of the CP, and controlling the output of the accumulator. And the table look-up and waveform generator is used for carrying out initial table look-up calculation according to the output of the accumulator to obtain an initial waveform. And the CORDIC calibrator calibrates the initial waveform according to the input initial waveform, index2 (namely the third target value of the invention) and Findex to obtain an accurate waveform. And the frequency offset calibrator performs complex multiplication according to the input data and the corresponding waveform thereof to finish frequency offset calibration.
The invention adopts the mode of combining the table look-up method and the CORDIC to carry out time domain frequency offset compensation, can use smaller storage resources and operation iteration times, realizes higher compensation precision and reduces the requirement on a multiplier. CORDIC iterates 6 times, can reach the performance of looking up the 64 size tables again for the second time, and has omitted the complex multiplier after the second time of looking up the tables. By pre-storing the 5G-NR subframe structure and symbol scheduling specified by the protocol, the corresponding subframe structure is selected according to the current subcarrier interval, as shown in FIG. 4, the CP field is screened out, and the scheduling realizes frequency offset compensation only on useful data, thereby avoiding the waste of power consumption. By mapping 0-2 × pi to 0-pi/4, the operation amount is not increased by applying the trigonometric function transformation formula, and only a table with 1/8 of the original size needs to be stored, thereby greatly simplifying the storage size required by the table look-up method. The CORDIC is additionally provided with an early exit mechanism, so that invalid iteration is avoided, and power consumption is wasted.
An embodiment of the present invention, as shown in fig. 5, is a frequency offset compensation apparatus for a 5G-NR receiver, including:
the subframe-level symbol scheduler is used for performing subframe-level scheduling on input subframe data according to the subframe parameters acquired from the first register and outputting the scheduled data and the symbol-level data length and cyclic prefix length parameters thereof;
the initial phase and frequency offset calculator is used for outputting a frequency offset compensation value according to the frequency offset and initial phase parameters obtained from the second register and the refreshing identification bit;
the phase accumulator is used for carrying out accumulation calculation according to the symbol-level data length, the cyclic prefix length parameter, the frequency offset and initial phase parameter, the refreshing identification bit and the frequency offset compensation value to obtain a phase accumulated value;
the table look-up and waveform generator is used for outputting a corresponding first target value, a corresponding third target value and a corresponding calculated waveform according to the phase accumulated value and a preset mapping table;
the CORDIC calibrator is used for calculating and outputting a waveform numerical value after frequency offset compensation according to the calculated waveform, the first target numerical value and the third target numerical value;
and the frequency spectrum calibrator is used for performing complex multiplication calculation according to the waveform numerical value and the scheduled data to obtain frequency offset compensation output data to complete frequency spectrum calibration.
Specifically, this embodiment is a device embodiment corresponding to the method embodiment, and specific effects refer to the method embodiment, which is not described in detail herein.
It will be apparent to those skilled in the art that, for convenience and simplicity of description, the above division of the program modules is merely used as an example, and in practical applications, the above distribution of functions may be performed by different program modules according to needs, that is, the internal structure of the apparatus may be divided into different program units or modules to perform all or part of the above-described functions. Each program module in the embodiments may be integrated in one processing unit, or each unit may exist alone physically, or two or more units are integrated in one processing unit, and the integrated unit may be implemented in a form of hardware, or may be implemented in a form of software program unit. In addition, the specific names of the program modules are only used for distinguishing the program modules from one another, and are not used for limiting the protection scope of the application.
It should be noted that the above embodiments can be freely combined as necessary. The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.
Claims (7)
1. A method of frequency offset compensation for a 5G-NR receiver, comprising the steps of:
the sub-frame level symbol scheduler performs sub-frame level scheduling on the input sub-frame data according to the sub-frame parameters obtained from the first register, and outputs the scheduled data and the symbol level data length and the cyclic prefix length parameters thereof;
the initial phase and frequency offset calculator outputs a frequency offset compensation value according to the frequency offset and initial phase parameters obtained from the second register and the refreshing identification bit;
the phase accumulator performs accumulation calculation according to the symbol-level data length, the cyclic prefix length parameter, the frequency offset and initial phase parameter, the refreshing identification bit and the frequency offset compensation value to obtain a phase accumulated value;
the table look-up and waveform generator outputs corresponding first target numerical values, third target numerical values and calculated waveforms according to the phase accumulated values and a preset mapping table;
the CORDIC calibrator calculates and outputs a waveform numerical value after frequency offset compensation according to the calculated waveform, the first target numerical value and the third target numerical value;
the frequency spectrum calibrator performs complex multiplication calculation according to the waveform numerical value and the scheduled data to obtain frequency offset compensation output data to complete frequency spectrum calibration;
the table lookup and waveform generator outputs corresponding first target values, third target values and calculated waveforms according to the phase accumulated values and a preset mapping table, and the method comprises the following steps:
the table lookup and waveform generator compares the preset value with the phase accumulated value and updates and obtains a target phase accumulated value according to a comparison result;
the table lookup and waveform generator extracts three high bits from the target phase accumulated value to obtain the first target value;
the table look-up and waveform generator extracts twenty-nine low bits of the target phase accumulated value to obtain an extraction field, and updates and obtains a target extraction field according to the first target value and the extraction field;
the table look-up and waveform generator extracts seven high bits from the target extraction field to obtain a second target value, and queries a preset mapping table according to the second target value to obtain a corresponding calculated waveform;
the table look-up and waveform generator extracts seventeenth to twenty-second bits from the target extraction field to obtain a third target value;
the CORDIC calibrator calculates and outputs a waveform numerical value after frequency offset compensation according to the calculated waveform, the first target numerical value and the third target numerical value, and comprises the following steps:
the CORDIC calibrator performs iterative computation according to the computed waveform and a third target numerical value to output a calibrated computed waveform;
and the CORDIC calibrator calculates and outputs a waveform numerical value after frequency offset compensation according to the first target numerical value and the calibrated calculated waveform.
2. The method of claim 1, wherein the sub-frame symbol scheduler comprises, before performing sub-frame level scheduling on the input sub-frame data according to the sub-frame parameters obtained from the first register, the steps of:
when the radio frequency is restarted or reset, when the 5G-NR receiver receives a calculated waveform, the first register updates subframe parameters corresponding to the calculated waveform;
when a subframe data is input to the subframe level symbol scheduler, the second register updates the self-stored frequency offset, initial phase parameter and refresh identification bit.
3. The method of claim 1, wherein the initial phase and frequency offset calculator outputs the frequency offset compensation value according to the updated flag bit and the frequency offset and initial phase parameters obtained from the second register, comprising:
the initial phase and frequency offset calculator judges whether to calculate updated frequency offset and initial phase parameters according to the refreshing identification bits;
when the refreshing identification bit is determined to accord with a preset updating condition, the initial phase and frequency offset calculator acquires an updated initial phase parameter and substitutes the initial phase parameter into a normalization formula to calculate a frequency offset compensation value according to the frequency offset parameter;
fn_delta＝round(fn/fs*2^32)
when the refresh identification bit is determined not to be in accordance with the preset update condition, the initial phase and frequency offset calculator does not update the frequency offset and the initial phase parameters;
wherein, fn is the frequency offset value of the current subframe data, fs is the sampling frequency of the current subframe data, and round () is a rounding function.
4. The method of claim 1, wherein the phase accumulator performs an accumulation calculation according to the symbol-level data length, the cyclic prefix length parameter, the frequency offset and initial phase parameter, the refresh flag bit, and the frequency offset compensation value to obtain a phase accumulation value, and comprises:
when the refresh identification bit is determined to meet the preset update condition, the phase accumulator performs accumulation calculation according to the symbol-level data length, the cyclic prefix length parameter, the updated initial phase parameter and the updated frequency offset compensation value to obtain a corresponding phase accumulated value; or the like, or, alternatively,
and when the refresh identification bit is determined not to meet the preset update condition, the phase accumulator performs accumulation calculation according to the symbol-level data length, the cyclic prefix length parameter, the frequency offset and the initial phase parameter to obtain a corresponding phase accumulated value.
5. The method of claim 1, wherein the table lookup and waveform generator compares the phase accumulation value with a predetermined value, and the step of obtaining the target phase accumulation value according to the comparison result comprises:
if the phase accumulated value is larger than the preset value, the table look-up and waveform generator determines that the target phase accumulated value is equal to the difference value between the phase accumulated value and first preset data;
and if the phase accumulated value is not larger than the preset value, the table look-up and waveform generator determines that the target phase accumulated value is equal to the phase accumulated value.
6. The method of claim 1, wherein the table lookup and waveform generator extracts twenty-nine lower bits of the target phase accumulation value to obtain an extracted field, and updating the extracted field according to the first target value and the extracted field comprises:
the table lookup and waveform generator judges whether the first target value is an odd number;
and if the first target value is an odd number, the table look-up and waveform generator determines that the target extraction field is equal to the difference value between second preset data and the extraction field.
7. A frequency offset compensation apparatus suitable for a 5G-NR receiver, characterized in that, when applied to the frequency offset compensation method suitable for a 5G-NR receiver in any one of claims 1 to 6, the method comprises:
the subframe-level symbol scheduler is used for performing subframe-level scheduling on input subframe data according to the subframe parameters acquired from the first register and outputting the scheduled data and the symbol-level data length and cyclic prefix length parameters thereof;
the initial phase and frequency offset calculator is used for outputting a frequency offset compensation value according to the frequency offset and initial phase parameters obtained from the second register and the refreshing identification bit;
the phase accumulator is used for carrying out accumulation calculation according to the symbol-level data length, the cyclic prefix length parameter, the frequency offset and initial phase parameter, the refreshing identification bit and the frequency offset compensation value to obtain a phase accumulated value;
the table look-up and waveform generator is used for outputting a corresponding first target value, a corresponding third target value and a corresponding calculated waveform according to the phase accumulated value and a preset mapping table; the table lookup and waveform generator compares the preset value with the phase accumulated value and updates and obtains a target phase accumulated value according to a comparison result; the table lookup and waveform generator extracts three high bits from the target phase accumulated value to obtain the first target value; the table lookup and waveform generator extracts twenty-nine lower bits of the target phase accumulated value to obtain an extracted field, and updates and obtains a target extracted field according to the first target value and the extracted field; the table look-up and waveform generator extracts seven high bits from the target extraction field to obtain a second target value, and queries a preset mapping table according to the second target value to obtain a corresponding calculation waveform; the table look-up and waveform generator extracts seventeenth to twenty-second bits from the target extraction field to obtain a third target value;
the CORDIC calibrator is used for calculating and outputting a waveform numerical value after frequency offset compensation according to the calculated waveform, the first target numerical value and the third target numerical value; the CORDIC calibrator performs iterative computation according to the computed waveform and a third target numerical value to output a calibrated computed waveform; the CORDIC calibrator calculates and outputs a waveform numerical value after frequency offset compensation according to the first target numerical value and the calibrated calculation waveform;
and the frequency spectrum calibrator is used for performing complex multiplication calculation according to the waveform numerical value and the scheduled data to obtain frequency offset compensation output data to finish frequency spectrum calibration.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011423663.6A CN112600777B (en) | 2020-12-08 | 2020-12-08 | Frequency offset compensation method and device suitable for 5G-NR receiver |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011423663.6A CN112600777B (en) | 2020-12-08 | 2020-12-08 | Frequency offset compensation method and device suitable for 5G-NR receiver |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112600777A CN112600777A (en) | 2021-04-02 |
| CN112600777B true CN112600777B (en) | 2022-11-22 |
Family
ID=75188876
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011423663.6A Active CN112600777B (en) | 2020-12-08 | 2020-12-08 | Frequency offset compensation method and device suitable for 5G-NR receiver |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112600777B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101039292A (en) * | 2006-03-16 | 2007-09-19 | 中国科学院上海微系统与信息技术研究所 | Method and apparatus for correcting sampling frequency deviation in OFDM system |
| CN104243378A (en) * | 2014-09-28 | 2014-12-24 | 武汉光迅科技股份有限公司 | Frequency-offset compensation module and method applied to DQPSK (differential quadrature reference phase shift keying) system |
| WO2017000406A1 (en) * | 2015-06-30 | 2017-01-05 | 深圳市中兴微电子技术有限公司 | Frequency shift and phase shift compensation method and device, and storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI302066B (en) * | 2006-04-26 | 2008-10-11 | Ind Tech Res Inst | Frequency acquisition and tracking circuit |
-
2020
- 2020-12-08 CN CN202011423663.6A patent/CN112600777B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101039292A (en) * | 2006-03-16 | 2007-09-19 | 中国科学院上海微系统与信息技术研究所 | Method and apparatus for correcting sampling frequency deviation in OFDM system |
| CN104243378A (en) * | 2014-09-28 | 2014-12-24 | 武汉光迅科技股份有限公司 | Frequency-offset compensation module and method applied to DQPSK (differential quadrature reference phase shift keying) system |
| WO2017000406A1 (en) * | 2015-06-30 | 2017-01-05 | 深圳市中兴微电子技术有限公司 | Frequency shift and phase shift compensation method and device, and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112600777A (en) | 2021-04-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9001905B2 (en) | Distance estimation | |
| CN109561495B (en) | Time-frequency tracking method, user equipment and computer readable medium | |
| CN109219948B (en) | Signal processing method and device | |
| CN105992385B (en) | Physical random access channel signal generation method | |
| CN109655847B (en) | Fast capturing method suitable for dynamic signal | |
| US20100182899A1 (en) | OFDM Time Basis Matching With Pre-FFT Cyclic Shift | |
| CN111404850B (en) | Uplink channel estimation method and device based on 5G system | |
| CN102932123B (en) | Method and system for detecting sounding reference signal (SRS) | |
| CN112600777B (en) | Frequency offset compensation method and device suitable for 5G-NR receiver | |
| CN114938255A (en) | High-speed time domain parallel timing synchronization method based on O & M algorithm | |
| KR100754840B1 (en) | Method for Frequency Synchronization of Mobile Telecommunication System Using OFDM | |
| CN112600776B (en) | Frequency offset compensation method and device suitable for OFDM system | |
| CN112311707B (en) | Direct current offset estimation method, device, equipment and storage medium based on UWB system | |
| CN104506271A (en) | Method and device for generating ZC (Zadoff-Chu) sequence | |
| CN102611665B (en) | Integer frequency bias in CMMB system and thin regularly combined estimation method and device | |
| CN113132284A (en) | Carrier phase tracking method and device | |
| CN106161296B (en) | The method of channel estimation and base station | |
| CN115051899A (en) | Frequency offset estimation method and device and computer readable storage medium | |
| KR100457925B1 (en) | A frequency offset calculation method using log transform and linear approximation | |
| CN109729034B (en) | Timing method, device and terminal | |
| He et al. | An efficient implementation of PRACH generator in LTE UE transmitters | |
| JP5515910B2 (en) | Sequence generation device, wireless communication device, sequence generation method, and sequence generation program | |
| Mar et al. | FPGA implementation of SDR based CFO estimation and compensation circuit for OFDM system | |
| WO2013100891A1 (en) | Fractional symbol based phase noise mitigation | |
| CN112887232B (en) | TRS-based PDP estimation method and system in NR |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |