阅读说明：本技术 一种自适应模拟信号解调方法 (Self-adaptive analog signal demodulation method ) 是由 吕新 黄均安 于 2020-07-06 设计创作，主要内容包括：本发明公开了一种自适应模拟信号解调方法,包括以下步骤：a)识别器调制识别:先对未知调制信号进行量化处理并进行相移变换处理,使得所述未知调制信号变为复信号；b)解调器调制解调：对所述复信号进行归一化处理并计算归一化幅度方差,得到模拟语音信号；c)对于FM调制信号,先进行反正切运算求出频率值,算出信号频率,然后进过滤波处理解调出频率信号；对于AM调制信号,先进行包络检波提取出幅度数据,然后进行处理采用卷积方式完成低通滤波。d)对解调后的信号进行静噪处理。该方法能够自动识别信号调制类型并进行解调与静噪优化。并且识别参数不涉及频域变换的复杂运算,在保证识别精度的同时大大提高了系统反应时间与运算速度。(The invention discloses a self-adaptive analog signal demodulation method, which comprises the following steps: a) the modulation and identification of the identifier, namely, firstly carrying out quantization processing on an unknown modulation signal and carrying out phase shift transformation processing on the unknown modulation signal so as to change the unknown modulation signal into a complex signal; b) the demodulator modulates and demodulates: carrying out normalization processing on the complex signal and calculating a normalized amplitude variance to obtain a simulated voice signal; c) for FM modulation signals, performing arc tangent operation to obtain a frequency value, calculating signal frequency, and then performing filtering processing to demodulate frequency signals; for AM modulation signals, envelope detection is firstly carried out to extract amplitude data, and then processing is carried out to complete low-pass filtering by adopting a convolution mode. d) And performing squelch processing on the demodulated signal. The method can automatically identify the signal modulation type and carry out demodulation and squelch optimization. And the identification parameters do not relate to the complex operation of frequency domain transformation, so that the identification precision is ensured, and the system reaction time and the operation speed are greatly improved.)

1. A method for demodulating an adaptive analog signal, comprising the steps of:

a) the modulation and identification of the identifier, namely, firstly carrying out quantization processing on an unknown modulation signal and carrying out phase shift transformation processing on the unknown modulation signal so as to change the unknown modulation signal into a complex signal;

b) the demodulator modulates and demodulates: normalizing the complex signal and calculating normalized amplitude variance to obtain analog voice signal for judgment,

if the FM modulation signal is smaller than the threshold value T0, performing FM demodulation to obtain an FM modulation signal;

if the amplitude is larger than the threshold value T0, performing AM demodulation to obtain an AM modulation signal;

c) for FM modulation signals, performing arc tangent operation to obtain a frequency value, calculating signal frequency, and then performing filtering processing to demodulate frequency signals;

for AM modulation signals, envelope detection is firstly carried out to extract amplitude data, and then the amplitude data are processed to finish low-pass filtering in a convolution mode;

d) and performing squelch processing on the demodulated signal.

2. The adaptive analog signal demodulation method according to claim 1, wherein: in the step a), the modulation signal is quantized by converting int16 data of an unknown modulation signal into an amplitude voltage value of a double type.

3. An adaptive analog signal demodulation method according to claim 2, characterized in that: in the step a), the phase shift transformation processing method is to decompose the signal into an in-phase signal i (t) and a quadrature signal q (t) by Hilbert transformation, then calculate a mean square value m0 of the signal amplitude, and perform normalization processing on the signal by using the mean square value m0 to complete the function of a limiter; then, calculating the amplitude of the normalized signal to complete the function of the envelope detector; and finally calculating the variance value zeta of the total envelope.

4. An adaptive analog signal demodulation method according to claim 3, characterized in that: in the step b), for the AM modulation signal, setting a mute signal threshold T1, verifying by a simulation experiment T1< < T0, equally dividing the signal envelope and respectively calculating the variances ζ i, in an AM mute state, making the AM modulation signal envelope constant, firstly making variance decision, filtering the mute signal, making the remaining variances mean again, performing mute processing to improve the recognition accuracy, and completing the modulation recognition of the analog voice signal.

5. The adaptive analog signal demodulation method according to claim 1, wherein: after the step d), performing squelch processing on the demodulated signal, and then performing multi-thread processing.

Technical Field

The invention relates to a self-adaptive analog signal demodulation method.

Background

The demodulation scheme is widely applied to the civil and commercial fields, and single and quick transmission is guaranteed. However, with the emergence of electronic communication and electronic warfare developed in the times, the analog modulation signals with unknown parameters need to be demodulated, and the demodulation processing mode is single demodulation and therefore has no self-adaptive capability. Under the unknown analog signal modulation mode, the signal can not be correctly modulated, and in the fields of military affairs and monitoring, the traditional demodulation mode is completely invalid when the parameter unknown modulation mode is unknown. Under the condition of single-path demodulation, if the data stream of the signal to be demodulated is more and the signal is a multi-path signal, the demodulation performance of one path is greatly reduced.

Disclosure of Invention

The invention aims to provide an adaptive analog signal demodulation method aiming at the defects of the prior art, which firstly identifies the modulation type of a modulated signal with unknown modulation mode and modulation index, correspondingly demodulates the corresponding modulated signal, performs squelch processing, and then stores the demodulated data in a database.

In order to solve the technical problems, the following technical scheme is adopted:

an adaptive analog signal demodulation method, comprising the steps of:

a) the modulation and identification of the identifier, namely, firstly carrying out quantization processing on an unknown modulation signal and carrying out phase shift transformation processing on the unknown modulation signal so as to change the unknown modulation signal into a complex signal;

b) the demodulator modulates and demodulates: normalizing the complex signal and calculating normalized amplitude variance to obtain analog voice signal for judgment,

if the FM modulation signal is smaller than the threshold value T0, performing FM demodulation to obtain an FM modulation signal;

if the amplitude is larger than the threshold value T0, performing AM demodulation to obtain an AM modulation signal;

c) for FM modulation signals, performing arc tangent operation to obtain a frequency value, calculating signal frequency, and then performing filtering processing to demodulate frequency signals;

for AM modulation signals, envelope detection is firstly carried out to extract amplitude data, and then the amplitude data are processed to finish low-pass filtering in a convolution mode;

d) and performing squelch processing on the demodulated signal.

Further, in the step a), the modulation signal is quantized by converting int16 data of an unknown modulation signal into an amplitude voltage value of a double type.

Further, in the step a), the phase shift transformation processing method is to decompose the signal into an in-phase signal i (t) and a quadrature signal q (t) by Hilbert transformation, then calculate a mean square value m0 of the signal amplitude, and perform normalization processing on the signal by using the mean square value m0 to complete the function of a limiter; then, calculating the amplitude of the normalized signal to complete the function of the envelope detector; and finally calculating the variance value zeta of the total envelope.

Further, in the step b), for the AM modulation signal, setting a mute signal threshold T1, verifying by a simulation experiment T1< < T0, equally dividing the signal envelope and respectively calculating variances ζ i, in an AM mute state, making a variance decision on the AM modulation signal envelope which is constant, filtering the mute signal, averaging the remaining variances again, performing mute processing to improve the recognition accuracy, and completing the modulation recognition on the analog voice signal.

Further, after the step d), performing squelch processing on the demodulated signal, and then performing multi-thread processing.

Due to the adoption of the technical scheme, the method has the following beneficial effects:

the invention relates to a self-adaptive analog signal demodulation method, which processes and filters an analog voice signal containing a mute modulation signal, judges a modulation mode by using a mean square value of an envelope variance, and demodulates and identifies 128 paths of parallel calculation. And phase shifting the signal to be demodulated by using Hilbert transform, calculating the mathematical expectation of the signal through the mathematical characteristics of in-phase and quadrature components, normalizing the signal, performing square sum to complete AM envelope extraction, or performing inverse tangent to complete FM frequency extraction, and finally completing the corresponding demodulation target. And filtering the demodulated voice signal for white noise and howling noise, and finally storing the demodulated data in a database.

The method can automatically identify the signal modulation type and carry out demodulation and squelch optimization. And the identification parameters do not relate to the complex operation of frequency domain transformation, so that the identification precision is ensured, and the system reaction time and the operation speed are greatly improved.

Drawings

The invention will be further described with reference to the accompanying drawings in which:

FIG. 1 is a flow chart of a method for demodulating an adaptive analog signal according to the present invention;

FIG. 2 is a graph of threshold-signal-to-noise ratio as a function of the present invention;

FIG. 3 is a flow chart of the recognizer of the present invention;

FIG. 4 is a flow chart of a demodulator according to the present invention;

FIG. 5 is a flow diagram of a multithreading module according to the present invention;

FIG. 6 is a diagram showing CPU usage before and after multi-core parallel processing in the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood, however, that the description herein of specific embodiments is only intended to illustrate the invention and not to limit the scope of the invention. Moreover, in the following description, descriptions of well-known structures and techniques are omitted so as to not unnecessarily obscure the concepts of the present invention.

As shown in fig. 1 to 6, an adaptive analog signal demodulation method includes the following steps:

and carrying out modulation identification on the identifier, preprocessing the system, configuring related CPU core and thread environment, and binding the algorithm program to the corresponding CPU.

Specifically, initialization work is firstly performed for the 64-core 128-path parallel processor and the matlab squelch algorithm is called. Considering that matlab is thread-safe, 64-core 128-path parallel processing is realized by adopting multithreading and a windows library function SetProcesssAffinityMask; in order to make VC + + call the matlab algorithm, the matlab algorithm needs to be packaged into dll to facilitate the call of VC + +.

a) The modulation and identification of the identifier, namely, firstly carrying out quantization processing on an unknown modulation signal and carrying out phase shift transformation processing on the unknown modulation signal so as to change the unknown modulation signal into a complex signal;

specifically, for an unknown modulation signal, processing is performed on quantized data, and int16 data is converted into an amplitude voltage value of a double type.

Then, decomposing the signal into an in-phase signal I (t) and a quadrature signal Q (t) by using Hilbert transform by using a phase shift method, then calculating a mean square value m0 of the signal amplitude, normalizing the signal by using the mean square value, and dividing the data of each sampling point by the mean square value m0 to normalize the signal, thereby completing the function of an amplitude limiter.

c) The demodulator modulates and demodulates: normalizing the complex signal and calculating normalized amplitude variance to obtain analog voice signal for judgment,

if the FM modulation signal is smaller than the threshold value T0, performing FM demodulation to obtain an FM modulation signal;

if the amplitude is larger than the threshold value T0, performing AM demodulation to obtain an AM modulation signal;

for FM modulation signals, performing arc tangent operation to obtain a frequency value, calculating signal frequency, and then performing filtering processing to demodulate frequency signals;

for AM modulation signals, envelope detection is firstly carried out to extract amplitude data, and then processing is carried out to complete low-pass filtering by adopting a convolution mode.

Specifically, the amplitude of the normalized signal is calculated first, and the function of the envelope detector is completed. And finally, calculating a variance value zeta of the total envelope to obtain an analog voice signal.

The modulation identification principle is as follows: the AM signal is an amplitude modulation signal, a signal to be modulated is modulated onto the amplitude of a carrier signal, so that the envelope of the modulated signal is greatly changed, and the FM modulation is to modulate the signal to be modulated onto the frequency of the carrier while the amplitude of the signal is kept unchanged, so that the FM modulation is a constant envelope modulation mode, and the envelope is basically not changed. Setting a certain threshold T0 determines whether it is an amplitude variation signal.

However, since the analog voice does not always have a communication signal, at the time of AM muting, if the muting time is long, the amplitude envelope after AM modulation still does not change, and erroneous recognition may occur. Through setting a mute signal threshold T1 and verifying T1< < T0 by simulation experiments, the signal envelope is equally divided and the variance ζ i is respectively solved, so that in a mute state, the modulation signal envelope is constant, variance judgment is firstly carried out, a mute signal is filtered, the rest variances are solved again, mute processing can be carried out, the identification accuracy is improved, and the modulation identification of the analog voice signal is completed.

For AM modulation signals, only amplitude limiting signals which are subjected to normalization processing in identification need to be subjected to envelope detection to directly extract amplitude data, and finally low-pass filtering is completed in a convolution mode after processing. The envelope signal may be uniformly quantized for subsequent read processing if desired.

For FM modulation signals, Herbert conversion and normalization operation are carried out by adopting signal operation like AM, signal data are normalized, I path signals and Q path signals are divided to obtain arctangent, signal frequency is calculated, and then frequency signals can be demodulated through LPF low-pass filtering.

d) And performing squelch processing on the demodulated signal. There are two types of invalid speech segments in the demodulated analog speech signal: firstly, the original voice signal has white noise or invalid voice segments (such as engine sound or table knocking sound, etc.), and the time of a listener is wasted because such voice segments do not have any valuable information; secondly, some language segments are changed into howling sound through transmission, and the listening experience of a listener is influenced. Therefore, after AM is carried out, sound squelch processing is carried out on the real voice signals after FM signals are demodulated, white noise and harsh howling noise are shielded and filtered, and finally, gentle voice is output.

After the step (d), performing a multi-thread process after performing a squelch process on the demodulated signal. The server obtains the transmitted data and carries out system multithreading processing firstly. Firstly, searching attributes, then acquiring the number of CPU logic cores, then binding a demodulation thread into a corresponding logic core, then starting the work of a demodulator, inputting data into the demodulator with an idle thread, carrying out data processing such as envelope detection and the like, outputting a demodulated voice signal, and then starting squelch processing to filter out noise such as howling and the like in the voice signal. And finally, storing the clear sound signals in a local database SQLite according to the requirement or sending the clear sound signals to a server for transmission and subsequent calling.

The above is only a specific embodiment of the present invention, but the technical features of the present invention are not limited thereto. Any simple changes, equivalent substitutions or modifications made on the basis of the present invention to solve the same technical problems and achieve the same technical effects are all covered in the protection scope of the present invention.