# US 9,882,581 B2 · Compressive sensing-based signal processing method and apparatus

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 9,882,581 B2
- Document ID: US9882581B2
- Status: Granted
- Kind: B2
- Application number: 15/270,911
- Filed: 2016-09-20
- Published: 2017-01-12
- Granted: 2018-01-30
- Earliest claimed: 2014-03-20
- Assignees: Huawei Technologies Co., Ltd.
- Applicants: Huawei Technologies Co., Ltd.
- Inventors: Xiangming Kong; Hufei Zhu
- Web page: https://patentgrep.com/patent/US9882581B2

## CPC

- H03M 7/3062: {Compressive sampling or sensing}
- H04B 1/66: for reducing bandwidth of signals; for improving efficiency of transmission (H04B1/68 takes precedence)
- H03M 1/1255: {Synchronisation of the sampling frequency or phase to the input frequency or phase}
- H03M: CODING; DECODING; CODE CONVERSION IN GENERAL (using fluidic means F15C4/00; optical analogue/digital converters G02F7/00; coding, decoding or code conversion, specially adapted for particular applications, see the relevant subclasses, e.g. G01D, G01R, G06F, G06T, G09G, G10L, G11B, G11C, H04B, H04L, H04M, H04N; ciphering or deciphering for cryptography or other purposes involving the need for secrecy G09C)
- H04B: TRANSMISSION

## Abstract

The present invention relates to a compressive sensing-based signal processing method and apparatus. The method includes: determining distribution, of signal components of an input signal on which frequency mixing has been performed, in a baseband spectrum according to spectral distribution of the input signal; dividing the baseband spectrum according to different signal aliasing patterns formed by the distribution of the signal components, to form multiple frequency bands; determining lowest sampling frequency of the input signal according to the signal aliasing patterns in the frequency bands, and performing sampling on the input signal at sampling frequency greater than the lowest sampling frequency, to obtain a sampled signal; and separately restoring a corresponding signal component in each frequency band of the multiple frequency bands according to the sampled signal, and splicing the signal components restored in the frequency bands, to complete restoration of the input signal.

## Claims

### Claim 1

A compressive sensing-based signal processing method, comprising:
determining distribution, of signal components of an input signal on which frequency mixing has been performed, in a baseband spectrum according to spectral distribution of the input signal;
dividing the baseband spectrum according to different signal aliasing patterns formed by the distribution of the signal components, to form multiple frequency bands, wherein each of the frequency bands corresponds to one signal aliasing pattern;
determining lowest sampling frequency of the input signal according to the signal aliasing patterns in the frequency bands, and performing sampling on the input signal at sampling frequency greater than the lowest sampling frequency, to obtain a sampled signal; and
separately restoring a corresponding signal component in each frequency band of the multiple frequency bands according to the sampled signal, and splicing the signal components restored in the frequency bands, to complete restoration of the input signal.

### Claim 2 (depends on 1)

The method according to claim 1, wherein the determining lowest sampling frequency of the input signal according to the signal aliasing patterns in the frequency bands comprises:
determining a frequency band comprising a largest quantity of aliased signals among the frequency bands, and using lowest sampling frequency required by the frequency band comprising the largest quantity of aliased signals as the lowest sampling frequency of the input signal.

### Claim 3 (depends on 1)

The method according to claim 1, wherein the performing sampling on the input signal at sampling frequency greater than the lowest sampling frequency, to obtain a sampled signal is specifically:
performing sampling on the input signal at sampling frequency fs, to obtain the sampled signal, wherein
f
s
≥
f
s_min
=
Q
×
f
p
m
,
fs_min is the lowest sampling frequency, Q is a maximum quantity of aliased signals in the frequency band, fp is a subband bandwidth, and m is a channel quantity of a system.

### Claim 4 (depends on 1)

The method according to claim 1, wherein after the multiple frequency bands are formed, the method further comprises:
determining a submatrix ASi, corresponding to an observation matrix A for compressive sensing sampling, in each frequency band of the multiple frequency bands according to the signal aliasing pattern in each frequency band of the multiple frequency bands, wherein i=1, . . . , K, wherein
the separately restoring a corresponding signal component in each frequency band of the multiple frequency bands according to the sampled signal, and splicing the signal components restored in the frequency bands, to complete restoration of the entire input signal specifically comprises:
separately restoring a baseband spectrum of the ith (i=1, . . . , K) frequency band according to a signal restoration algorithm formula Zsi(f)=ASi+y(f), wherein ASi+ is a Moore Penrose inverse matrix of a submatrix ASi of the ith frequency band, ZSi(f) is the restored baseband spectrum of the ith frequency band, y(f) is a discrete Fourier transformation of the sampled signal, and K is a total quantity of the frequency bands that are formed on the baseband spectrum; and
splicing the restored baseband spectrums of the frequency bands, to obtain the entire baseband spectrum of the input signal, and restoring the input signal according to the baseband spectrum obtained through splicing.

### Claim 5 (depends on 1)

The signal processing method according to claim 1, wherein before the determining distribution, of signal components of an input signal on which frequency mixing has been performed, in a baseband spectrum according to spectral distribution of the input signal, the method further comprises:
detecting a previously received input signal, to obtain the spectral distribution of the input signal, and performing system configuration according to the spectral distribution of the input signal, wherein
the performing system configuration according to the spectral distribution of the input signal specifically comprises one of or a combination of more of the following:
configuring and adjusting a subband bandwidth fp of the system; or
dynamically adjusting the subband bandwidth fp of the system; or
adjusting a sampling frequency fs of a sampler; or
adjusting a passband bandwidth of a filter; or
configuring a channel quantity m.

### Claim 6 (depends on 2)

The method according to claim 2, wherein the performing sampling on the input signal at sampling frequency greater than the lowest sampling frequency, to obtain a sampled signal is specifically:
performing sampling on the input signal at sampling frequency fs, to obtain the sampled signal, wherein
f
s
≥
f
s_min
=
Q
×
f
p
m
,
fs_min is the lowest sampling frequency, Q is a maximum quantity of aliased signals in the frequency band, fp is a subband bandwidth, and m is a channel quantity of a system.

### Claim 7 (depends on 2)

The signal processing method according to claim 2, wherein before the determining distribution, of signal components of an input signal on which frequency mixing has been performed, in a baseband spectrum according to spectral distribution of the input signal, the method further comprises:
detecting a previously received input signal, to obtain the spectral distribution of the input signal, and performing system configuration according to the spectral distribution of the input signal, wherein
the performing system configuration according to the spectral distribution of the input signal specifically comprises one of or a combination of more of the following:
configuring and adjusting a subband bandwidth fp of the system; or
dynamically adjusting the subband bandwidth fp of the system; or
adjusting a sampling frequency fs of a sampler; or
adjusting a passband bandwidth of a filter; or
configuring a channel quantity m.

### Claim 8 (depends on 3)

The signal processing method according to claim 3, wherein before the determining distribution, of signal components of an input signal on which frequency mixing has been performed, in a baseband spectrum according to spectral distribution of the input signal, the method further comprises:
detecting a previously received input signal, to obtain the spectral distribution of the input signal, and performing system configuration according to the spectral distribution of the input signal, wherein
the performing system configuration according to the spectral distribution of the input signal specifically comprises one of or a combination of more of the following:
configuring and adjusting the subband bandwidth fp of the system; or
dynamically adjusting the subband bandwidth fp of the system; or
adjusting the sampling frequency fs of a sampler; or
adjusting a passband bandwidth of a filter; or
configuring the channel quantity m.

### Claim 9

A compressive sensing-based signal processing apparatus, wherein the apparatus comprises:
a transceiver configured to receive an input signal;
at least one processor; and
a memory coupled to the at least one processor and configured to store instructions which, when executed by the at least one processor, causes the at least one processor to:
determine distribution, of signal components of the input signal on which frequency mixing has been performed, in a baseband spectrum according to spectral distribution of the input signal;
divide the baseband spectrum according to different signal aliasing patterns formed by the distribution of the signal components, to form multiple frequency bands, wherein each of the frequency bands corresponds to one signal aliasing pattern;
determine lowest sampling frequency of the input signal according to the signal aliasing patterns in the frequency bands, and performing sampling on the input signal at sampling frequency greater than the lowest sampling frequency, to obtain a sampled signal; and
separately restore a corresponding signal component in each frequency band of the multiple frequency bands according to the sampled signal, and splicing the signal components restored in the frequency bands, to complete restoration of the input signal.

### Claim 10 (depends on 9)

The apparatus according to claim 9, wherein that the processor is operable to execute the instructions to:
determine a frequency band comprising a largest quantity of aliased signals among the frequency bands, and use lowest sampling frequency required by the frequency band comprising the largest quantity of aliased signals as the lowest sampling frequency of the input signal.

### Claim 11 (depends on 9)

The apparatus according to claim 9, wherein that the processor is configured to perform sampling on the input signal at sampling frequency greater than the lowest sampling frequency, to obtain a sampled signal comprises:
the processor is configured to:
perform sampling on the input signal at sampling frequency fs, to obtain the sampled signal, wherein
f
s
≥
f
s_min
=
Q
×
f
p
m
,
fs_min is the lowest sampling frequency, Q is a maximum quantity of aliased signals in the frequency band, fp is a subband bandwidth, and m is a channel quantity of a system.

### Claim 12 (depends on 9)

The apparatus according to claim 9, wherein after the multiple frequency bands are formed, the processor is further configured to: determine a submatrix ASi, corresponding to an observation matrix A for compressive sensing sampling, in each frequency band of the multiple frequency bands according to the signal aliasing pattern in each frequency band of the multiple frequency bands, wherein i=1, . . . , K, wherein
that the processor is configured to separately restore a corresponding signal component in each frequency band of the multiple frequency bands according to the sampled signal, and splice the signal components restored in the frequency bands, to complete restoration of the entire input signal comprises:
the processor is configured to:
separately restore a baseband spectrum of the ith (i=1, . . . , K) frequency band according to a signal restoration algorithm formula ZSi(f)=ASi+y(f), wherein ASi+ is a Moore Penrose inverse matrix of a submatrix ASi+ of the ith frequency band, ZSi(f) is the restored baseband spectrum of the ith frequency band, y(f) is a discrete Fourier transformation of the sampled signal, and K is a total quantity of the frequency bands that are formed on the baseband spectrum; and
splice the restored baseband spectrums of the frequency bands, to obtain the entire baseband spectrum of the input signal, and restore the input signal according to the baseband spectrum obtained through splicing.

### Claim 13 (depends on 9)

The apparatus according to claim 9, wherein before determining the distribution, of the signal components of the input signal on which frequency mixing has been performed, in the baseband spectrum according to the spectral distribution of the input signal, the processor is further configured to:
detect a previously received input signal, to obtain the spectral distribution of the input signal, and perform system configuration according to the spectral distribution of the input signal, wherein
the performing system configuration according to the spectral distribution of the input signal specifically comprises one of or a combination of more of the following:
configuring and adjusting a subband bandwidth fp of the system; or
dynamically adjusting the subband bandwidth fp of the system; or
adjusting a sampling frequency fs of a sampler; or
adjusting a passband bandwidth of a filter; or
configuring a channel quantity m.

### Claim 14 (depends on 9)

The apparatus according to claim 9, wherein the compressive sensing-based signal processing apparatus is a base station or a terminal.