CN107453732B - Signal sampling rate conversion method and device - Google Patents
Signal sampling rate conversion method and device Download PDFInfo
- Publication number
- CN107453732B CN107453732B CN201610371661.4A CN201610371661A CN107453732B CN 107453732 B CN107453732 B CN 107453732B CN 201610371661 A CN201610371661 A CN 201610371661A CN 107453732 B CN107453732 B CN 107453732B
- Authority
- CN
- China
- Prior art keywords
- domain signal
- signal sequence
- time domain
- sampling rate
- transform 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.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03H—IMPEDANCE NETWORKS, e.g. RESONANT CIRCUITS; RESONATORS
- H03H17/00—Networks using digital techniques
- H03H17/02—Frequency selective networks
- H03H17/06—Non-recursive filters
- H03H17/0621—Non-recursive filters with input-sampling frequency and output-delivery frequency which differ, e.g. extrapolation; Anti-aliasing
- H03H17/0635—Non-recursive filters with input-sampling frequency and output-delivery frequency which differ, e.g. extrapolation; Anti-aliasing characterized by the ratio between the input-sampling and output-delivery frequencies
- H03H17/0685—Non-recursive filters with input-sampling frequency and output-delivery frequency which differ, e.g. extrapolation; Anti-aliasing characterized by the ratio between the input-sampling and output-delivery frequencies the ratio being rational
Abstract
The invention discloses a signal sampling rate conversion method and a signal sampling rate conversion device. In the invention, after a first time domain signal sequence of N sampling points is obtained, the first time domain signal sequence is converted into a first transform domain signal sequence, the first transform domain signal sequence is processed according to the ratio of a second sampling rate to the first sampling rate to obtain a second transform domain signal sequence, the number of the sampling points contained in the second transform domain signal sequence is the product of N and the ratio, then the second transform domain signal sequence is converted into a second time domain signal sequence, and the sampling rate of the second time domain signal sequence is the second sampling rate, thereby realizing the conversion from the first sampling rate to the second sampling rate. Because the invention carries out signal sampling rate conversion based on the transform domain, compared with the signal sampling rate conversion carried out in the time domain, the invention can reduce the processing overhead, reduce the realization complexity and improve the processing precision.
Description
Technical Field
The present invention relates to the field of digital signal processing technologies, and in particular, to a signal sampling rate conversion method and apparatus.
Background
Signal sampling, also called signal sampling, is a discretization of the signal over time, i.e. the sampling of the instantaneous value point by point on the analog signal x (t) at a certain time interval Δ t, which is a prerequisite for digital signal processing.
The sampling rate of the signal is various, for example, there are 9 sampling rates of the audio signal: 8kHz, 11.025kHz, 12kHz, 16kHz, 22.05kHz, 24kHz, 32kHz, 44.1kHz and 48 kHz. Sample Rate Conversion (SRC) is a problem often encountered in signal processing, especially audio signal processing, and a common processing method is to cascade a Sample Rate expander for an upsampling function, a low-pass filter, and a Sample Rate compressor for a downsampling function in a time domain to achieve a specific required SRC task.
Considering the operational stability, causality and physical realizability of the low-pass filter in an actual operating environment, a high-order Finite Impulse Response (FIR) low-pass filter is usually designed in advance according to specific requirements, and thus the computational complexity and the processing delay of the signal sampling rate conversion technology are large.
Therefore, how to perform the signal sampling frequency conversion with low processing overhead and complexity is a problem to be solved at present.
Disclosure of Invention
The embodiment of the invention provides a signal sampling rate conversion method and a signal sampling rate conversion device, which are used for realizing signal sampling frequency conversion with lower processing overhead and complexity.
The signal sampling rate conversion method provided by the embodiment of the invention comprises the following steps:
acquiring a first time domain signal sequence, wherein the sampling rate of the first time domain signal sequence is a first sampling rate, the first time domain signal sequence comprises N sampling points, and N is a positive integer;
transforming the first time domain signal sequence into a first transform domain signal sequence;
processing the first transform domain signal sequence according to a ratio of a second sampling rate to the first sampling rate to obtain a second transform domain signal sequence, wherein the number of sampling points contained in the second transform domain signal sequence is the product of N and the ratio;
and transforming the second transform domain signal sequence into a second time domain signal sequence, wherein the sampling rate of the second time domain signal sequence is a second sampling rate.
Optionally, under the condition that the transform domain is a frequency domain, the first transform domain signal sequence includes a first interval and a second interval, where the second interval is a conjugate mirror image of the first interval;
the second transform domain signal sequence comprises a first interval and a second interval, wherein the second interval is a conjugate mirror image of the first interval; and the sampling points in the first interval in the second transform domain signal sequence are obtained by gating the sampling points in the first interval in the first transform domain signal sequence based on a rectangular function.
Optionally, acquiring a first time domain signal sequence includes:
segmenting the time domain signal sequences of more than N sampling points to obtain P first time domain signal sequences containing N sampling points, wherein the last 2Q sampling points of the previous time domain signal sequence in two adjacent first time domain signal sequences are the same as the first 2Q sampling points of the next time domain signal sequence, N is more than 4Q, and P and Q are both positive integers;
transforming the first time domain signal sequence into a first transform domain signal sequence, comprising:
converting the P first time domain signal sequences into first transform domain signal sequences, and processing the first transform domain signal sequences according to a ratio of a second sampling rate to the first sampling rate to obtain P second transform domain signal sequences;
after transforming the second transform domain signal sequence into a second time domain signal sequence to obtain P second time domain signal sequences, the method further includes:
for the obtained P second time domain signal sequences, deleting the front Q1 sampling points and the rear Q1 sampling points in each second time domain signal sequence, wherein Q1 is the product of Q and the ratio;
and splicing the P second time domain signal sequences after the sampling points are deleted to obtain a time domain signal sequence with a second sampling rate.
The signal sampling rate conversion device provided by the embodiment of the invention comprises:
the device comprises an acquisition module, a processing module and a processing module, wherein the acquisition module is used for acquiring a first time domain signal sequence, the sampling rate of the first time domain signal sequence is a first sampling rate, the first time domain signal sequence comprises N sampling points, and N is an integer greater than 1;
a first transform module for transforming the first time domain signal sequence into a first transform domain signal sequence;
the first processing module is used for processing the first transform domain signal sequence according to a ratio of a second sampling rate to the first sampling rate to obtain a second transform domain signal sequence, wherein the number of sampling points contained in the second transform domain signal sequence is the product of N and the ratio;
and the second conversion module is used for converting the second conversion domain signal sequence into a second time domain signal sequence, and the sampling rate of the second time domain signal sequence is a second sampling rate.
Optionally, under the condition that the transform domain is a frequency domain, the first transform domain signal sequence includes a first interval and a second interval, where the second interval is a conjugate mirror image of the first interval;
the second transform domain signal sequence comprises a first interval and a second interval, wherein the second interval is a conjugate mirror image of the first interval; and the sampling points in the first interval in the second transform domain signal sequence are obtained by gating the sampling points in the first interval in the first transform domain signal sequence based on a rectangular function.
Optionally, the obtaining module is specifically configured to:
segmenting the time domain signal sequences of more than N sampling points to obtain P first time domain signal sequences containing N sampling points, wherein the last 2Q sampling points of the previous time domain signal sequence in two adjacent first time domain signal sequences are the same as the first 2Q sampling points of the next time domain signal sequence, N is more than 4Q, and P and Q are both positive integers;
the first transformation module is specifically configured to: converting the P first time domain signal sequences into first transform domain signal sequences, and processing the first transform domain signal sequences according to a ratio of a second sampling rate to the first sampling rate to obtain P second transform domain signal sequences;
the device further comprises:
a second processing module, configured to transform the second transform domain signal sequence into a second time domain signal sequence, and after P second time domain signal sequences are obtained, delete the first Q1 sampling points and the last Q1 sampling points in each second transform domain signal sequence for the P second time domain signal sequences, where Q1 is a product of Q and the ratio; and splicing the P second time domain signal sequences after the sampling points are deleted to obtain a time domain signal sequence with a second sampling rate.
Another embodiment of the present invention provides a signal sampling rate conversion apparatus, including: a processor, a memory, and a bus interface;
the processor is used for reading the program in the memory and executing the following processes:
acquiring a first time domain signal sequence, wherein the sampling rate of the first time domain signal sequence is a first sampling rate, the first time domain signal sequence comprises N sampling points, and N is a positive integer;
transforming the first time domain signal sequence into a first transform domain signal sequence;
processing the first transform domain signal sequence according to a ratio of a second sampling rate to the first sampling rate to obtain a second transform domain signal sequence, wherein the number of sampling points contained in the second transform domain signal sequence is the product of N and the ratio;
and transforming the second transform domain signal sequence into a second time domain signal sequence, wherein the sampling rate of the second time domain signal sequence is a second sampling rate.
Optionally, under the condition that the transform domain is a frequency domain, the first transform domain signal sequence includes a first interval and a second interval, where the second interval is a conjugate mirror image of the first interval;
the second transform domain signal sequence comprises a first interval and a second interval, wherein the second interval is a conjugate mirror image of the first interval; and the sampling points in the first interval in the second transform domain signal sequence are obtained by gating the sampling points in the first interval in the first transform domain signal sequence based on a rectangular function.
In the above embodiment of the present invention, after the first time domain signal sequence is obtained, the first time domain signal sequence is converted into the first transform domain signal sequence, the first transform domain signal sequence is processed according to the ratio of the second sampling rate to the first sampling rate to obtain the second transform domain signal sequence, the number of sampling points included in the second transform domain signal sequence is the product of N and the ratio, and then the second transform domain signal sequence is converted into the second time domain signal sequence, where the sampling rate of the second time domain signal sequence is the second sampling rate, so as to implement the conversion from the first sampling rate to the second sampling rate. Because the embodiment of the invention carries out signal sampling rate conversion based on the transform domain, compared with the signal sampling rate conversion carried out in the time domain, the embodiment of the invention can reduce the processing overhead, realize the complexity and improve the processing precision.
Drawings
FIG. 1 shows a prior art sample rate of FsConversion to FsA system block diagram of xL/M;
FIG. 2 is a block diagram of a system for implementing a sample rate expander and a low pass filter based on a polyphase filter structure in the second prior art;
FIG. 3 is a block diagram of a system for implementing a low pass filter and a sample rate compressor based on a polyphase filter structure in the prior art III;
FIG. 4 is a block diagram of a system for L times integer upsampling rate conversion in an embodiment of the present invention;
fig. 5 is a schematic diagram of an original signal spectrum and a signal spectrum after the original signal spectrum passes through a sampling frequency expander when L is 3 in the embodiment of the present invention;
FIG. 6 is a block diagram of a system for M times integer downsampling rate conversion in an embodiment of the present invention;
fig. 7 is a schematic diagram of an original signal spectrum and a signal spectrum after the original signal spectrum passes through a low-pass filtering and sampling frequency compressor when M is 4 in the embodiment of the present invention;
FIG. 8 is a schematic diagram illustrating a signal sampling rate conversion process according to an embodiment of the present invention;
FIG. 9 is a schematic diagram of a method for real-time processing of a long-term input signal sequence by a sample rate conversion algorithm according to an embodiment of the present invention;
FIG. 10 is a second schematic diagram illustrating a signal sampling rate conversion process according to an embodiment of the present invention;
fig. 11 is a schematic structural diagram of a signal sampling rate conversion apparatus according to an embodiment of the present invention;
fig. 12 is a schematic structural diagram of a signal sampling rate apparatus according to another embodiment of the present invention.
Detailed Description
Fig. 1 shows a block diagram of a prior art method of signal sample rate conversion. The sampling rate of time-domain digital signal x (n) is FsNow it needs to be converted into a sampling rate Fs’＝FsTime domain digital signal y (M) of xL/M (L and M are any positive integer) and the principle is shown in FIG. 1, wherein, signal x (n) generates sampling rate F after passing through a sampling rate expander1＝L×FsThe intermediate transition signal z (q) is processed by a low-pass filter h (q) to generate another intermediateA transition signal v (q), the intermediate transition signal v (q) is processed by a sampling rate compressor to generate a sampling rate F2 ═ F1/M＝FsTarget signal y (M) of XL/M. Here, the functions of the sampling frequency expander are: interpolating (L-1) zero sample points between each pair of adjacent sample points of the input signal and outputting the zero sample points to enable the output sampling rate to be L times of the input sampling rate; the function of the sample rate compressor is: sampling points are taken once every (M-1) sampling points for an input signal and output, so that the input sampling rate is M times of the output sampling rate; the low-pass filter has a working sampling frequency of F1＝L×FsThe frequency response function is determined by the following formula:
wherein the content of the first and second substances,
is the cut-off frequency of the low-pass filter,
the function is as follows: the method can suppress the image interference introduced by the front-end sampling rate expander in the signal z (q), and prevent the aliasing possibly caused by the back-end sampling rate compressor in the signal y (m).
To further improve the processing efficiency of the system shown in fig. 1, another prior art proposes an implementation technique based on a "poly-phase" filter structure. A sampling rate conversion scheme using a polyphase filter structure to implement the sampling rate expander and the low-pass filter function in fig. 1 is shown in fig. 2, where the filter impulse response pr (n) of the r-th branch is:
pr(n)＝h(nL+r),r＝0,1,…,L-1………………(2)
the time domain input signal x (n) at the time n simultaneously generates L response output sampling points y0(n), y1(n), … and yL-1(n), the selection switch sequentially gates y0(n), y1(n), … and yL-1(n) in a counterclockwise direction, so as to generate L sampling points { v (q) }, q ═ 0,1, …, L-1}, and the signal v (q) is processed by the over-sampling rate compressor and then converted into a time domain signal y (m) with a required sampling rate.
Another sampling rate conversion scheme using a polyphase filter structure to implement the low pass filter and the subsequent sampling rate compressor function in fig. 1 is shown in fig. 3, where the filter impulse response pr (m) of the r-th branch is:
pr(m)＝h(mM+r),r＝0,1,…,M-1………………(3)
the time domain input signal x (n) firstly generates a signal z (q) through a sampling rate expander, a gating switch rotates anticlockwise to sequentially gate and connect a 0 th branch, an M-1 th branch, an M-2 th branch, … and a 1 st branch, and the sum of the filter outputs of each branch is the time domain signal y (M) with the required sampling rate. The input zr (m) of the r-th branch in the figure is:
zr(m)＝z(mM-r),r＝0,1,…,M-1………………(4)
as can be seen from the above description, the existing signal sampling rate conversion technical solution is implemented in the time domain, and the drawbacks thereof mainly include:
the low pass filter h (q) satisfying the condition (as shown in formula 1) needs to be designed in advance. In order to make the low-pass filter work stably and have a linear phase, the low-pass filter is generally implemented by a FIR filter with a symmetrical structure; the passband ripple factor of the FIR low-pass filter needs to be sufficiently small (generally r) in view of effectively suppressing the image interference caused by interpolation and at the same time ensuring the minimum distortion of the filtered signalp<0.01dB) and the stopband attenuation must be sufficiently large (typically r)s<-90 to-80 dB), which results in a sampling rate of (L x F)s) Order N of time domain FIR low pass filter designed under the conditionlpIs very large, so that the processing time delay is very long, and the computation complexity is correspondingly large. Although the low-pass FIR order of each stage design can be relatively effectively reduced by adopting a polyphase filter structure and a multi-stage step processing technology, the FIR order of at least one stage is still very high, thereby bringing great processing delay and computational complexity.
In the frequency domain, an ideal low-pass filter response corresponds to a rectangular window from a distant point of the frequency to its cut-off frequency, and then the frequency spectrum of the input signal of the low-pass filter is directly subjected to the gate weighting processing in the frequency domain, so that the frequency spectrum of the desired target signal can be obtained.
Based on this, the embodiment of the present invention obtains the frequency spectrum of the conversion target signal by performing the gating extraction processing on the frequency spectrum of the input signal to be subjected to the sampling rate conversion on the basis of the ideal frequency response curve (i.e., the rectangular gating window) corresponding to the low-pass filter in the conventional method in the frequency domain, and further obtains the time domain target signal by the conversion from the frequency domain to the time domain.
The embodiment of the invention can be suitable for the signal sampling rate conversion process in the communication field, and is particularly suitable for the audio signal sampling rate conversion process.
In the following, first, the principle on which the signal sampling rate is increased, the principle on which the signal sampling rate is decreased, and the principle on which arbitrary sampling rate conversion is based in the embodiment of the present invention will be explained.
(one) L times integer up-sampling conversion case (i.e. sampling rate from F)sIncrease to L x FsAnd L is a positive integer
Let a time domain signal x (N) comprising N (N is an even number) samples with a sampling rate FsIf necessary, it is converted to a sampling rate of F1＝L×FsThen the time domain signal y (m) will contain N1 ═ L × N sampling points, and the time domain implementation block diagram can be as shown in fig. 4, where the frequency response of the low pass filter is:
wherein the content of the first and second substances,
is the cut-off frequency of the low-pass filter,
according to fig. 4, a signal sequence z (m) containing N1 sampling points obtained by processing a time-domain signal x (N) by a sampling rate expander, and performing N1-point Discrete Fourier Transform (DFT) on z (m) to obtain:
wherein, the operator "·" represents the multiplication operation, which is the same below and is not described separately; mod (k, N) is the remainder of the variable k modulo N, and mod (k, N) is ∈ {0,1,2, …, N-1 }; and X (-) is an N-point DFT of X (N), namely:
formula (6) indicates that: the spectrum of the signal z (m) is formed by the spectrum of the signal x (n) and its mirror spectrum (i.e. mirror interference), as shown in fig. 5, where fig. 5 shows the original signal spectrum and its signal spectrum after passing through the sampling frequency spreader, taking L as an example 3. It follows that the window H (k) is selected in the DFT domain by an ideal low pass filter, acting on Z (e)jω′) Thus, the following desired signal spectrum can be chosen:
the operator "" indicates a complex conjugate operation, and is the same as below, and is not described.
Obtaining y (m) by performing inverse transform of DFT at N1 point for y (k):
(two) M times integer downsampling conversion case (i.e. sampling rate from F)sIs reduced to Fs/M，MIs a positive integer)
Let a time domain signal x (N) comprising N sampling points with a sampling rate FsIf necessary, it is converted to a sampling rate of FsThe time domain signal y (M) of/M, the time domain implementation block diagram of which is shown in fig. 6, wherein the frequency response of the low-pass filter is:
wherein the content of the first and second substances,
the cut-off frequency of the low-pass filter has the effect of avoiding spectral dips caused by subsequent sample rate compressors.
Signal v1(n) is defined as:
then v1(n) can be expressed as:
from fig. 6, it is apparent that:
y(m)＝v1(M·m)＝v(M·m)…………………………………(14)
then z of y (m) is transformed into:
wherein V (z) is a z-transform of v (n).
The z-transforms denoted x (n) and h (n) are X (z) and H (z), respectively, then
V(z)＝X(z)H(z)……………………………(16)
Substituting formula (16) for formula (15) to obtain:
the frequency spectrum Y (e) of the signal Y (m)jω) Comprises the following steps:
wherein the content of the first and second substances,
since the filter H (e) of the formula (11) is usedjω) The signal x (n) is low-pass filtered, so that the output signal of the low-pass filter is at the cut-off frequency
The above frequency components are negligible, so there are
A diagrammatic example of the above process can be seen in fig. 7. Fig. 7 shows an original signal spectrum and a signal spectrum thereof after passing through a low-pass filter and a sampling frequency compressor, taking M-4 as an example.
It can be seen that there is a time domain signal x (N) of N sampling points with a sampling rate FsIf it is to be converted to a sampling rate of Fsand/M (where M is a positive integer), the time domain signal y (M) will contain N1 equal to N/M sampling points, where N1 is an even number, and N is also an even number. The method for realizing integer M times of downsampling rate conversion based on the DFT technology comprises the following steps: calculating x (k) of N-point DFT of time-domain signal x (N) according to equation (7), according to the illustrated example shown in fig. 7, N1-point DFT of time-domain signal y (m) can be given by the following equation:
y (m) is obtained by inverse transformation of DFT at N1 points for y (k) according to equation (10).
(III) sample rate conversion case where arbitrary rational numbers can be expressed (i.e. sample rate from F)sConversion to FsXL/M, L and M are positive integers, and L and M have no common divisor)
Is provided with a time domain signal x (n) with a sampling rate of FsIf it is to be converted to a sampling rate of FsTime domain signal y (M) of x L/M, the time domain implementation block diagram of which is shown in fig. 1, wherein the frequency response of the low pass filter is determined by equation (1). If the time domain signal x (N) includes N samples, the number of samples of the corresponding time domain y (m) signal is N1:
without loss of generality, assuming that N1 and N are both even numbers, it is easy to understand from fig. 1 that this sampling rate conversion can be decomposed into a cascade of L times integer up-sampling conversion and M times integer down-sampling conversion, and then from the foregoing description for these two cases, it can be inferred that the following relationship exists between the DFT (i.e., y (k)) of time-domain signal y (M) and the DFT (i.e., x (k)) of time-domain signal x (N):
the conversion of the upsampling, i.e.,
then there are:
the down-sampling conversion, i.e.,
then there are:
then, the time domain signal y (m) after sampling rate conversion is obtained by solving the IDFT at N1 points of Y (k) according to the formula (10).
Based on the above derived conclusions, fig. 8 shows a flow chart of a general signal sampling rate conversion method provided by the embodiment of the invention. The process may be performed by a signal sampling rate conversion apparatus, and is described by taking as an example a first time domain signal sequence of a first sampling rate converted into a second time domain signal sequence of a second sampling rate. The "first time domain signal sequence" and the "second time domain signal sequence" are not specific to a specific time domain signal sequence, but for clarity of description, the input time domain signal sequence is referred to as the "first time domain signal sequence", and the time domain signal sequence output after the sampling rate conversion is referred to as the "second time domain signal sequence".
As shown in fig. 8, the process may include the following steps:
step 801: the method comprises the steps of obtaining a first time domain signal sequence, wherein the sampling rate of the first time domain signal sequence is a first sampling rate, the first time domain signal sequence comprises N sampling points, and N is a positive integer.
Step 802: the first time domain signal sequence is transformed into a first transform domain signal sequence.
In this step, taking DFT algorithm as an example, the first time domain signal may be subjected to N-point DFT to obtain a first frequency domain signal sequence.
Step 803: and processing the first transform domain signal sequence according to the ratio of the second sampling rate to the first sampling rate to obtain a second transform domain signal sequence, wherein the number of sampling points contained in the second transform domain signal sequence is the product of N and the ratio.
In step 803, taking the transform domain as the frequency domain as an example, the first frequency domain signal sequence may be processed using an ideal low-pass filter or a rectangular function capable of implementing the ideal low-pass filter, so as to obtain a second frequency domain signal sequence. In particular, if the first frequency-domain signal sequence is described as comprising a first interval and a second interval, wherein the second interval is a conjugate mirror of the first interval, then the second frequency-domain signal sequence may also be described as comprising a first interval and a second interval, wherein the second interval is a conjugate mirror of the first interval. The sampling points in the first interval in the second frequency domain signal sequence are obtained by gating the sampling points in the first interval in the first frequency domain signal sequence based on a rectangular function (the rectangular function can realize the function of an ideal low-pass filter) or the ideal low-pass filter. See in particular fig. 5 or fig. 7 in the previous embodiments.
Specifically, to
(L and M are positive integers) represents the ratio of the second sampling rate to the first sampling rate, and the number of sampling points contained in the second frequency domain signal sequence is
If it is
The first frequency domain signal sequence may be processed according to the above equations (23) and (24) to obtain a second frequency domain signal sequence. If it is
The first frequency domain signal sequence may be processed according to the above equation (25) and equation (26) to obtain a second frequency domain signal sequence.
Note that, when the signal sampling rate conversion is performed using the formula (23) and the formula (24) or using the formula (25) and the formula (26), N and N1 both take a positive even number without loss of generality. For example, for
If N is guaranteed to be a positive even number, N1 is a positive even number; in that
In the case of (1), can be according to
The values of N and N1 are determined to ensure that N and N1 are both positive and even numbers. Of course, the present invention is not limited thereto, and if N or N1 is odd, the above formula can be modified
Or
And (6) carrying out rounding.
Step 804: and transforming the second transform domain signal sequence into a second time domain signal sequence, wherein the sampling rate of the second time domain signal sequence is a second sampling rate.
In the above flow, in step 802, an orthogonal transform algorithm or other reversible transform algorithm may be used to transform the time domain into the transform domain, and correspondingly, in step 804, a corresponding inverse transform algorithm may be used to implement the transform from the transform domain into the time domain. Taking the orthogonal transform algorithm as an example, in step 802, the corresponding coefficients in the transform domain of the construction target signal (i.e. the second time domain signal) are extracted by calculating the corresponding coefficients of the input signal in the defined orthogonal transform domain; in step 804, the inverse orthogonal transform is applied to obtain the desired target time domain signal. Orthogonal transformation algorithms include, but are not limited to: DFT algorithm, Wavelet (Wavelet) algorithm, Discrete Cosine Transform (DCT) algorithm, Walsh Transform algorithm, Hartley Transform algorithm, and the like.
Further, the flow shown in fig. 9 is processed for a short-time input sequence signal having N sampling points. For a long-time input sequence signal, it can be divided into a plurality of short-time sequence segments (for example, a short-time sequence containing N sampling points) to be processed respectively, and then the processing results of each short-time sequence segment are combined to form the required output.
Further, considering that the processing error in the transform domain will affect both ends of the time-domain output sequence segment corresponding to the transform domain, in order to further improve and enhance the processing performance of the sampling rate conversion algorithm, the embodiment of the present invention performs the segmentation processing on the input long-term signal sequence by using the "overlap" technique, as shown in fig. 9. Each section of input data comprises N sampling points, 2Q sampling points are overlapped between every two adjacent sections (the sampling points are overlapped, namely the sampling points are the same), for example, the p-th section and the p + 1-th section, the p + 1-th section and the p + 2-th section are overlapped by 2Q sampling points; according to the input data of the nth sampling point in the p-th segment, corresponding N1 point data (where N1 is N × L/M) will be generated, and the 2Q1 output sampling points corresponding to the 2Q input sampling points where the p-th segment and the p + 1-th segment overlap (where Q1 is Q × L/M) will be located at the rear end (i.e., the right end in fig. 9) of the output data in the p-th segment and the front end (i.e., the left end in fig. 9) of the output data in the p + 1-th segment, respectively, so that the last Q1 sampling points at the rear end in the output data in the p-th segment and the first Q1 sampling points at the front end in the output data in the p + 1-th segment (as shown in fig. 9) are deleted; similarly, the first Q1 sampling points at the front end in the p-th output data and the last Q1 sampling points at the rear end in the p + 1-th output data are deleted. It can be seen that, in each piece (N1 points) of data outputted, only the (N1-2Q1) sample point data in the middle thereof is reserved for actual output.
Alternatively, the above-described procedure may be applied to a sampling rate conversion process of an audio signal.
In order to more clearly understand the above embodiment of the present invention, the signal sampling rate conversion flow shown in fig. 8 will be described in detail below with reference to fig. 10.
As shown in fig. 10, an initialization step 1001 may be performed first.
The initialization process may include the following operations:
(1) calculating an up-sampling factor L and a down-sampling factor M according to the input sampling rate and the output sampling rate;
(2) setting the length N of the input buffer x and the length 2Q of the overlapped part of the input data blocks, and calculating the length of the output buffer y
And parameters of the target output buffer z
The input buffer x is used for storing time domain signal sequence segments (the number of sampling points is N) to be subjected to sampling rate conversion, the output buffer y is used for storing the time domain signal sequence segments (the number of sampling points is N1) subjected to sampling rate conversion, and the target output buffer is used for storing signal sequences (the number of sampling points is N1-2Q1) obtained by deleting 2Q1 sampling points which are the first Q1 sampling points and the last Q1 sampling points in the time domain signal sequence segments subjected to sampling rate conversion;
(3) resetting the input buffer, the output buffer and the target output buffer, i.e. zeroing all units of each buffer, wherein [0: N-1] represents the 0 th unit to the N-1 th unit of the buffer, and the others are similar:
x[0:N-1]<—0；
y[0:N1-1]<—0；
z[0:N1-2Q1]<—0；
after the initialization is completed, the following signal sampling rate conversion process can be performed:
step 1002: data overlap blocking step
In step 1002, an (N-2Q) input time domain signal sequence newData [0: N-2Q-1] is read from an input signal sequence segment containing N sampling points, the input buffer x is updated, that is, the 2 nd Q unit to the (N-1) th unit of the input buffer x are set as the read newData [0: N-2Q-1], and the rest of the units of the input buffer x are set as the rest of the input signal sequence segment containing N sampling points except for the newData [0: N-2Q-1], which can be specifically expressed as:
x[0:2Q-1]<—x[N-2Q:N-1]；
x[2Q:N-1]<—newData[0:N-2Q-1]；
Step 1003: signal sample rate conversion for each block of data
In step 1003, the following operations are performed:
(1) performing DFT operation on data in an input buffer x to obtain a frequency domain signal sequence segment X (k) of a signal sequence segment x [0: N-1], wherein 0< ═ k < (N-1);
(2) calculating the frequency domain signal sequence segment Y (k), 0 output by sampling rate conversion<＝k<N1-1. Wherein, if
Calculating according to the formula (23) and the formula (24) to obtain a signal sequence segment Y (k); if it is
The signal sequence segment y (k) is calculated according to the above equation (25) and equation (26).
(3) And calculating a time domain signal sequence segment y [0: N1-1] corresponding to the frequency domain signal sequence segment Y (k) by using DFT inverse transformation.
Step 1004: output of results of signal sample rate conversion
In step 1004, the following operations are performed:
(1) reading (N1-2Q1) sample point data from the output buffer y into the output destination buffer z, as follows:
z[0:N1-2Q1-1]<—y[Q1:N1-Q1-1]；
(2) outputting (N1-2Q1) sample point data: z [0: N1-2Q1-1], and obtaining a time domain signal sequence segment containing N1-2Q1 sampling points.
And repeating the steps 1002 to 1004 until all the time domain signal sequence segments to be subjected to the sampling rate conversion are completely converted.
As can be seen from the foregoing description, in the above embodiments of the present invention, after the first time domain signal sequence is obtained, the first time domain signal sequence is converted into the first transform domain signal sequence, the first transform domain signal sequence is processed according to the ratio of the second sampling rate to the first sampling rate, so as to obtain the second transform domain signal sequence, where the number of sampling points included in the second transform domain signal sequence is the product of N and the ratio, and then the second transform domain signal sequence is converted into the second time domain signal sequence, where the sampling rate of the second time domain signal sequence is the second sampling rate, so as to implement conversion from the first sampling rate to the second sampling rate. Because the embodiment of the invention carries out signal sampling rate conversion based on the transform domain, compared with the signal sampling rate conversion carried out in the time domain, the embodiment of the invention can reduce the processing overhead, reduce the implementation complexity and improve the processing precision.
Specifically, under the condition that the transform domain is a frequency domain, the embodiment of the present invention obtains the frequency spectrum of the transform target signal by performing a gating extraction process on the frequency spectrum of the signal to be transformed by using an ideal frequency response curve (i.e., a rectangular gating window) corresponding to a low-pass filter in the conventional method in the frequency domain, and further obtains the time domain target signal by transforming the frequency domain into the time domain. Because the method equivalently adopts the frequency response curve of the ideal low-pass filter with 0dB passband ripple, - ∞ dB stopband attenuation and 0Hz transition bandwidth in the frequency domain, the method has higher conversion precision than the prior time domain SRC technology; in addition, because some embodiments adopt time domain to frequency domain conversion techniques such as FFT fast conversion, the computational complexity of the method is much smaller than that of the current time domain SRC technique.
Based on the same technical concept, the embodiment of the present invention further provides a signal sampling rate conversion apparatus, which can implement the signal sampling rate conversion process described in the foregoing embodiment.
Referring to fig. 11, a schematic structural diagram of a signal sampling rate conversion apparatus according to an embodiment of the present invention is provided, where the apparatus may include: the obtaining module 1101, the first transforming module 1102, the first processing module 1103, and the second transforming module 1104 may further include a second processing module 1105, where:
an obtaining module 1101, configured to obtain a first time domain signal sequence, where a sampling rate of the first time domain signal sequence is a first sampling rate, the first time domain signal sequence includes N sampling points, and N is an integer greater than 1;
a first transform module 1102 for transforming the first time domain signal sequence into a first transform domain signal sequence;
a first processing module 1103, configured to process the first transform domain signal sequence according to a ratio of a second sampling rate to the first sampling rate to obtain a second transform domain signal sequence, where the number of sampling points included in the second transform domain signal sequence is a product of N and the ratio;
a second transforming module 1104, configured to transform the second transform domain signal sequence into a second time domain signal sequence, where a sampling rate of the second time domain signal sequence is a second sampling rate.
Optionally, under the condition that the transform domain is a frequency domain, the first transform domain signal sequence includes a first interval and a second interval, where the second interval is a conjugate mirror image of the first interval; the second transform domain signal sequence comprises a first interval and a second interval, wherein the second interval is a conjugate mirror image of the first interval; and the sampling points in the first interval in the second transform domain signal sequence are obtained by gating the sampling points in the first interval in the first transform domain signal sequence based on a rectangular function.
Optionally, on the condition that the transform domain is a frequency domain, the first processing module 1103 may be specifically configured to: and when the ratio is larger than 1, processing the first transform domain signal sequence according to the formula (23) and the formula (24) to obtain a second transform domain signal sequence.
Optionally, on the condition that the transform domain is a frequency domain, the first processing module 1103 may be specifically configured to: and when the ratio is smaller than 1, processing the first transform domain signal sequence according to the formula (25) and the formula (26) to obtain a second transform domain signal sequence.
Optionally, the obtaining module 1101 may be specifically configured to: the time domain signal sequence which is larger than N sampling points is segmented to obtain P first time domain signal sequences which comprise N sampling points, the last 2Q sampling points of the previous time domain signal sequence in the two adjacent first time domain signal sequences are the same as the first 2Q sampling points of the next time domain signal sequence, N is larger than 4Q, and P and Q are both positive integers. Accordingly, the first transformation module 1102 may be specifically configured to: converting the P first time domain signal sequences into first transform domain signal sequences, and processing the first transform domain signal sequences according to a ratio of a second sampling rate to the first sampling rate to obtain P second transform domain signal sequences; the second processing module 1105 may be for: after the second transform domain signal sequence is transformed into a second time domain signal sequence to obtain P second time domain signal sequences, deleting front Q1 sampling points and rear Q1 sampling points in each second time domain signal sequence for the P second time domain signal sequences, wherein Q1 is the product of Q and the ratio; and splicing the P second time domain signal sequences after the sampling points are deleted to obtain a time domain signal sequence with a second sampling rate.
Alternatively, the signal processed by the signal sampling rate conversion device may be, but is not limited to, an audio signal.
Referring to fig. 12, a schematic structural diagram of a signal sampling rate conversion apparatus according to another embodiment of the present invention is provided, which can implement the signal sampling rate conversion process described in the foregoing embodiment. As shown in fig. 12, the apparatus may include: a processor 1201, a memory 1202, and a bus interface.
The processor 1201 is responsible for managing a bus architecture and general processing, and the memory 1202 may store data used by the processor 1201 in performing operations.
The bus architecture may include any number of interconnected buses and bridges, with one or more processors, represented by the processor 1201, and various circuits, represented by the memory 1202, being linked together. The bus architecture may also link together various other circuits such as peripherals, voltage regulators, power management circuits, and the like, which are well known in the art, and therefore, will not be described any further herein. The bus interface provides an interface. The processor 1201 is responsible for managing a bus architecture and general processing, and the memory 1202 may store data used by the processor 1201 in performing operations.
The process disclosed by the embodiment of the invention can be applied to the processor 1201, or can be implemented by the processor 1201. In implementation, the steps of the signal processing flow may be implemented by integrated logic circuits of hardware or instructions in the form of software in the processor 1201. The processor 1201 may be a general purpose processor, a digital signal processor, an application specific integrated circuit, a field programmable gate array or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or the like that may implement or perform the methods, steps, and logic blocks disclosed in embodiments of the present invention. A general purpose processor may be a microprocessor or any conventional processor or the like. The steps of a method disclosed in connection with the embodiments of the present invention may be directly implemented by a hardware processor, or may be implemented by a combination of hardware and software modules in the processor. The software module may be located in ram, flash memory, rom, prom, or eprom, registers, etc. storage media as is well known in the art. The storage medium is located in the memory 1202, and the processor 1201 reads information in the memory 1202 and completes the steps of the signal sampling rate conversion process in combination with hardware thereof.
Specifically, the processor 1201, configured to read a program in the memory 1202, executes the following processes:
acquiring a first time domain signal sequence, wherein the sampling rate of the first time domain signal sequence is a first sampling rate, the first time domain signal sequence comprises N sampling points, and N is a positive integer;
transforming the first time domain signal sequence into a first transform domain signal sequence;
processing the first transform domain signal sequence according to the ratio of the second sampling rate to the first sampling rate to obtain a second transform domain signal sequence, wherein the number of sampling points contained in the second transform domain signal sequence is the product of N and the ratio;
and transforming the second transform domain signal sequence into a second time domain signal sequence, wherein the sampling rate of the second time domain signal sequence is a second sampling rate.
Optionally, under the condition that the transform domain is a frequency domain, the first transform domain signal sequence includes a first interval and a second interval, where the second interval is a conjugate mirror image of the first interval; the second transform domain signal sequence comprises a first interval and a second interval, wherein the second interval is a conjugate mirror image of the first interval; and the sampling points in the first interval in the second transform domain signal sequence are obtained by gating the sampling points in the first interval in the first transform domain signal sequence based on a rectangular function.
The specific implementation process of the above flow can be referred to the foregoing embodiments, and is not repeated here.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (8)
1. A method of signal sample rate conversion, comprising:
acquiring a first time domain signal sequence, wherein the sampling rate of the first time domain signal sequence is a first sampling rate, the first time domain signal sequence comprises N sampling points, and N is a positive integer;
transforming the first time domain signal sequence into a first transform domain signal sequence;
processing the first transform domain signal sequence according to a ratio of a second sampling rate to the first sampling rate to obtain a second transform domain signal sequence, wherein the number of sampling points contained in the second transform domain signal sequence is the product of N and the ratio;
when the ratio is greater than 1, processing the first transform domain signal sequence according to the following formula to obtain a second transform domain signal sequence:
wherein Y (k) represents the kth sample point in the second transform domain signal sequence, 0 ≦ k ≦ N1-1,
representing the ratio of the second sampling rate to the first sampling rate,
n and N1 are both even numbers;
when the ratio is smaller than 1, processing the first transform domain signal sequence according to the following formula to obtain a second transform domain signal sequence:
wherein Y (k) represents the kth sample point in the second transform domain signal sequence, 0 ≦ k ≦ N1-1,
representing the ratio of the second sampling rate to the first sampling rate,
n and N1 are both even numbers;
transforming the second transform domain signal sequence into a second time domain signal sequence, wherein the sampling rate of the second time domain signal sequence is a second sampling rate;
specifically, the acquiring the first time domain signal sequence includes:
segmenting the time domain signal sequences of more than N sampling points to obtain P first time domain signal sequences containing N sampling points, wherein the last 2Q sampling points of the previous time domain signal sequence in two adjacent first time domain signal sequences are the same as the first 2Q sampling points of the next time domain signal sequence, N is more than 4Q, and P and Q are both positive integers;
the transforming the first time domain signal sequence into a first transform domain signal sequence, and processing the first transform domain signal sequence according to a ratio of a second sampling rate to the first sampling rate to obtain a second transform domain signal sequence, includes:
converting the P first time domain signal sequences into first transform domain signal sequences, and processing the first transform domain signal sequences according to a ratio of a second sampling rate to the first sampling rate to obtain P second transform domain signal sequences;
after transforming the second transform domain signal sequence into a second time domain signal sequence to obtain P second time domain signal sequences, the method further includes:
for the P second time domain signal sequences, deleting the front Q1 sampling points and the rear Q1 sampling points in each second time domain signal sequence, wherein Q1 is the product of Q and the ratio, and Q1 is a positive integer;
and splicing the P second time domain signal sequences after the sampling points are deleted to obtain a time domain signal sequence with a second sampling rate.
2. The method of claim 1, wherein the first transform domain signal sequence comprises a first interval and a second interval under the condition that the transform domain is a frequency domain, wherein the second interval is a conjugate mirror image of the first interval;
the second transform domain signal sequence comprises a first interval and a second interval, wherein the second interval is a conjugate mirror image of the first interval; and the sampling points in the first interval in the second transform domain signal sequence are obtained by gating the sampling points in the first interval in the first transform domain signal sequence based on a rectangular function.
3. The method of any of claims 1-2, wherein the signal comprises an audio signal.
4. A signal sample rate conversion device, comprising:
the device comprises an acquisition module, a processing module and a processing module, wherein the acquisition module is used for acquiring a first time domain signal sequence, the sampling rate of the first time domain signal sequence is a first sampling rate, the first time domain signal sequence comprises N sampling points, and N is an integer greater than 1;
a first transform module for transforming the first time domain signal sequence into a first transform domain signal sequence;
the first processing module is used for processing the first transform domain signal sequence according to a ratio of a second sampling rate to the first sampling rate to obtain a second transform domain signal sequence, wherein the number of sampling points contained in the second transform domain signal sequence is the product of N and the ratio;
when the ratio is greater than 1, processing the first transform domain signal sequence according to the following formula to obtain a second transform domain signal sequence:
wherein Y (k) represents the kth sample point in the second transform domain signal sequence, 0 ≦ k ≦ N1-1,
representing the ratio of the second sampling rate to the first sampling rate,
n and N1 are both even numbers;
when the ratio is smaller than 1, processing the first transform domain signal sequence according to the following formula to obtain a second transform domain signal sequence:
wherein Y (k) represents the kth sample point in the second transform domain signal sequence, 0 ≦ k ≦ N1-1,
representing the ratio of the second sampling rate to the first sampling rate,
n and N1 are both even numbers;
a second transform module, configured to transform the second transform domain signal sequence into a second time domain signal sequence, where a sampling rate of the second time domain signal sequence is a second sampling rate;
specifically, the obtaining module is specifically configured to:
segmenting the time domain signal sequences of more than N sampling points to obtain P first time domain signal sequences containing N sampling points, wherein the last 2Q sampling points of the previous time domain signal sequence in two adjacent first time domain signal sequences are the same as the first 2Q sampling points of the next time domain signal sequence, N is more than 4Q, and P and Q are both positive integers;
the first transformation module is specifically configured to: transforming the P first time-domain signal sequences into a first transform-domain signal sequence;
the first processing module is configured to: processing the first transform domain signal sequence according to the ratio of a second sampling rate to the first sampling rate to obtain P second transform domain signal sequences;
the device further comprises:
a second processing module, configured to transform the second transform domain signal sequence into a second time domain signal sequence, and after P second time domain signal sequences are obtained, delete the first Q1 sampling points and the last Q1 sampling points in each second transform domain signal sequence for the P obtained second time domain signal sequences, where Q1 is a product of Q and the ratio, and Q1 is a positive integer; and splicing the P second time domain signal sequences after the sampling points are deleted to obtain a time domain signal sequence with a second sampling rate.
5. The apparatus of claim 4, wherein the first transform domain signal sequence comprises a first interval and a second interval under the condition that the transform domain is a frequency domain, wherein the second interval is a conjugate mirror of the first interval;
the second transform domain signal sequence comprises a first interval and a second interval, wherein the second interval is a conjugate mirror image of the first interval; and the sampling points in the first interval in the second transform domain signal sequence are obtained by gating the sampling points in the first interval in the first transform domain signal sequence based on a rectangular function.
6. The apparatus of any of claims 4 to 5, wherein the signal comprises an audio signal.
7. A signal sample rate conversion device, comprising: a processor, a memory, and a bus interface;
the processor is used for reading the program in the memory and executing the following processes:
acquiring a first time domain signal sequence, wherein the sampling rate of the first time domain signal sequence is a first sampling rate, the first time domain signal sequence comprises N sampling points, and N is a positive integer;
transforming the first time domain signal sequence into a first transform domain signal sequence;
processing the first transform domain signal sequence according to a ratio of a second sampling rate to the first sampling rate to obtain a second transform domain signal sequence, wherein the number of sampling points contained in the second transform domain signal sequence is the product of N and the ratio;
when the ratio is greater than 1, processing the first transform domain signal sequence according to the following formula to obtain a second transform domain signal sequence:
wherein Y (k) represents the kth sample point in the second transform domain signal sequence, 0 ≦ k ≦ N1-1,
representing the ratio of the second sampling rate to the first sampling rate,
n and N1 are both even numbers;
when the ratio is smaller than 1, processing the first transform domain signal sequence according to the following formula to obtain a second transform domain signal sequence:
wherein Y (k) represents the second transform domain signalThe k-th sampling point in the sequence is more than or equal to 0 and less than or equal to N1-1,
representing the ratio of the second sampling rate to the first sampling rate,
n and N1 are both even numbers;
transforming the second transform domain signal sequence into a second time domain signal sequence, wherein the sampling rate of the second time domain signal sequence is a second sampling rate;
specifically, the acquiring the first time domain signal sequence includes:
segmenting the time domain signal sequences of more than N sampling points to obtain P first time domain signal sequences containing N sampling points, wherein the last 2Q sampling points of the previous time domain signal sequence in two adjacent first time domain signal sequences are the same as the first 2Q sampling points of the next time domain signal sequence, N is more than 4Q, and P and Q are both positive integers;
the transforming the first time domain signal sequence into a first transform domain signal sequence, and processing the first transform domain signal sequence according to a ratio of a second sampling rate to the first sampling rate to obtain a second transform domain signal sequence, includes:
converting the P first time domain signal sequences into first transform domain signal sequences, and processing the first transform domain signal sequences according to a ratio of a second sampling rate to the first sampling rate to obtain P second transform domain signal sequences;
after transforming the second transform domain signal sequence into a second time domain signal sequence to obtain P second time domain signal sequences, the method further includes:
for the P second time domain signal sequences, deleting the front Q1 sampling points and the rear Q1 sampling points in each second time domain signal sequence, wherein Q1 is the product of Q and the ratio, and Q1 is a positive integer;
and splicing the P second time domain signal sequences after the sampling points are deleted to obtain a time domain signal sequence with a second sampling rate.
8. The apparatus of claim 7, wherein the first transform domain signal sequence comprises a first interval and a second interval under the condition that the transform domain is a frequency domain, wherein the second interval is a conjugate mirror of the first interval;
the second transform domain signal sequence comprises a first interval and a second interval, wherein the second interval is a conjugate mirror image of the first interval; and the sampling points in the first interval in the second transform domain signal sequence are obtained by gating the sampling points in the first interval in the first transform domain signal sequence based on a rectangular function.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610371661.4A CN107453732B (en) | 2016-05-30 | 2016-05-30 | Signal sampling rate conversion method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610371661.4A CN107453732B (en) | 2016-05-30 | 2016-05-30 | Signal sampling rate conversion method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107453732A CN107453732A (en) | 2017-12-08 |
| CN107453732B true CN107453732B (en) | 2021-09-24 |
Family
ID=60484903
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610371661.4A Active CN107453732B (en) | 2016-05-30 | 2016-05-30 | Signal sampling rate conversion method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107453732B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5495432A (en) * | 1994-01-03 | 1996-02-27 | Industrial Technology Research Institute | Apparatus and method for sampling rate conversion |
| CN101083639A (en) * | 2006-06-01 | 2007-12-05 | 上海无线通信研究中心 | Fraction time sampling based filter group and filtering method |
| JP2014127804A (en) * | 2012-12-26 | 2014-07-07 | Onkyo Corp | Sampling rate conversion device and program |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TW351885B (en) * | 1994-04-29 | 1999-02-01 | Ind Tech Res Inst | Method and apparatus for sampling rate conversion |
| EP0957579A1 (en) * | 1998-05-15 | 1999-11-17 | Deutsche Thomson-Brandt Gmbh | Method and apparatus for sampling-rate conversion of audio signals |
| US8369973B2 (en) * | 2008-06-19 | 2013-02-05 | Texas Instruments Incorporated | Efficient asynchronous sample rate conversion |
| US20110182582A1 (en) * | 2009-04-15 | 2011-07-28 | Nec Laboratories America Inc. | Joint sample rate conversion and cd compensation in frequency domain for coherent polmux |
-
2016
- 2016-05-30 CN CN201610371661.4A patent/CN107453732B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5495432A (en) * | 1994-01-03 | 1996-02-27 | Industrial Technology Research Institute | Apparatus and method for sampling rate conversion |
| CN101083639A (en) * | 2006-06-01 | 2007-12-05 | 上海无线通信研究中心 | Fraction time sampling based filter group and filtering method |
| JP2014127804A (en) * | 2012-12-26 | 2014-07-07 | Onkyo Corp | Sampling rate conversion device and program |
Non-Patent Citations (2)
| Title |
|---|
| 《A Parallel Architecture for Sampling Rate Conversion Based on Pseudo-DFT》;Qiang Yuan等;《2008 4th International Conference on Wireless Communications, Networking and Mobile Computing》;20081231;1-4 * |
| 《通信信号变采样频域实现方法》;盛广铭等;《通信技术》;20100228;第43卷(第218期);191-194 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107453732A (en) | 2017-12-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| AU2013223996B2 (en) | Low delay real-to-complex conversion in overlapping filter banks for partially complex processing | |
| US20070094317A1 (en) | Method and system for B-spline interpolation of a one-dimensional signal using a fractional interpolation ratio | |
| SE519885C2 (en) | Method and apparatus for converting sampling frequency | |
| Grossmann et al. | An $ L_1 $-method for the design of linear-phase FIR digital filters | |
| Budunova et al. | Digital frequency-selective filters based on spectra of atomic functions | |
| Bi et al. | Sampling rate conversion based on DFT and DCT | |
| CN107453732B (en) | Signal sampling rate conversion method and device | |
| EP2730026B1 (en) | Low-delay filtering | |
| Kumar et al. | Near perfect reconstruction quadrature mirror filter | |
| Russell et al. | Efficient arbitrary sampling rate conversion with recursive calculation of coefficients | |
| WO2016105574A1 (en) | High order b-spline sampling rate conversion (src) | |
| Pei et al. | A two-channel nonuniform perfect reconstruction filter bank with irrational down-sampling factors | |
| WO2019152101A1 (en) | Sample rate conversion with pitch-based interpolation filters | |
| Babic | Windowing design method for polynomial-based interpolation filters | |
| Yang et al. | Model-Based Design Methodology for Sampling Rate Converter | |
| Xu et al. | Modified polyphase filter for arbitrary sampling rate conversion | |
| Sundararajan et al. | Multirate Digital Signal Processing | |
| Zuluaga et al. | Poly-phase filter-bank realization for the simulation of electric-network transients | |
| Giron-Sierra et al. | Digital filters | |
| Kumar et al. | Delay efficient generalized rational sampling rate conversion polyphase FIR filter | |
| Prajapati et al. | Extension of the ERDMOSM1 and ERDMOSM2 Reduced Group Delay Finite Impulse Response (FIR) in Digital Filtering | |
| Bhandari et al. | Fractional delay filters based on generalized cardinal exponential splines | |
| Sozański et al. | Selected Methods of Signal Filtration and Separation and Their Implementation | |
| CN117833867A (en) | Non-integer multiple downsampling CIC filter | |
| Feldbauer et al. | Multirate signal processing |
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 |