阅读说明：本技术 一种基于循环的iq不平衡自适应盲补偿方法和系统 (Circulation-based IQ imbalance adaptive blind compensation method and system ) 是由 吕方明 贺小勇 于 2020-03-31 设计创作，主要内容包括：本发明属于通信技术领域,涉及一种基于循环的IQ不平衡自适应盲补偿方法,包括：接收通过I、Q两路已经出现不平衡的信号x(n),并截取一定长度的离散信号；设置步进系数；对接收信号x(n)进行共轭处理,得到x<Sup>*</Sup>(n)；对接收信号x(n)进行处理得到补偿后的输出信号y(n)＝x(n)+w(n)x<Sup>*</Sup>(n),其中w(n)为补偿系数；引入步进系数并对补偿系数w(n)进行平滑处理,通过循环迭代的方式得到系统所需的补偿系数w(n),进而获得补偿后的输出信号y(n)。本发明的IQ不平衡自适应盲补偿方法能够在任何正交接收器设置中运行,无论是单通道还是多通道,并且与理想基带等效信号的任何特定结构或特性无关。本发明还提供一种基于循环的IQ不平衡自适应盲补偿系统。(The invention belongs to the technical field of communication, and relates to an IQ imbalance adaptive blind compensation method based on circulation, which comprises the following steps: receiving I, Q two paths of unbalanced signals x (n) and intercepting a discrete signal with a certain length; setting a stepping coefficient; conjugate processing is carried out on the received signal x (n) to obtain x * (n); processing the received signal x (n) to obtain a compensated output signal y (n) ═ x (n) + w (n) x * (n), wherein w (n) is a compensation factor; introducing a stepping coefficient and smoothing the compensation coefficient w (n), obtaining the compensation coefficient w (n) required by the system in a loop iteration mode, and further obtaining a compensated output signal y (n). The IQ imbalance adaptive blind compensation method of the present invention can operate in any quadrature receiver setup, whether single-channel or multi-channel, and is independent of any particular structure or characteristics of the ideal baseband equivalent signal. The invention also provides an IQ imbalance self-adaptive filter based on circulationA blind compensation system is adapted.)

1. A loop-based IQ imbalance adaptive blind compensation method, comprising:

receiving I, Q two paths of unbalanced signals x (n) and intercepting a discrete signal with a certain length;

setting a stepping coefficient;

conjugate processing is carried out on the received signal x (n) to obtain x^*(n)；

Processing the received signal x (n) to obtain a compensated output signal y (n) ═ x (n) + w (n) x^*(n), wherein w (n) is a compensation factor;

introducing a stepping coefficient and smoothing the compensation coefficient w (n), obtaining the compensation coefficient w (n) required by the system in a loop iteration mode, and further obtaining a compensated output signal y (n).

2. The IQ imbalance adaptive blind compensation method according to claim 1, wherein the step of introducing a step coefficient and smoothing the compensation coefficient w (n), and obtaining the compensation coefficient w (n) required by the system by loop iteration, and further obtaining the compensated output signal y (n) comprises:

introducing a stepping coefficient, and processing the compensation coefficient to obtain: w (n +1) ═ w (n) - μ y (n), where w (n) is a compensation coefficient and μ is a first-order step coefficient;

smoothing the compensation coefficient w (n) to obtain w '(n +1) ═ β w' (n) + (1- β) w (n +1), wherein β is a smoothing coefficient;

continuously taking values of discrete signals, and taking values of the formula y (n) ═ x (n) + w (n) x^*(n), w (n +1) ═ w (n) - μ y (n) is subjected to a cyclic treatment;

when the compensation coefficient w (n) tends to be stable, the compensation coefficient value required by the system is obtained, and the compensated receiving signal y (n) is obtained.

3. The IQ imbalance adaptive blind compensation method according to claim 2, wherein the process of obtaining the compensation coefficients required by the system comprises:

in order to obtain a compensation coefficient w (n) required by the system, the second-order statistical characteristic of the baseband modulation signal is utilized;

according to the second-order statistical characteristic of the general ideal baseband modulation signal, the baseband modulation signal is complementary with an autocorrelation function:

wherein E [. X [ ]]Indicating a desire;

the matrix expression form is as follows:

E[Y(n)y(n)]0, where y (N) ═ y (N) y (N-1) y (N-2) … y (N-N +1)]^T

Wherein: y (N) is a discrete form of the compensator output signal, τ is any integer, and N is the matrix order;

accordingly, a recursion formula of the high-order compensation coefficient w (n) can be obtained:

E[W(n+1)]＝E[W(n)]-ME[Y(n)y(n)]

simplifying expectations, we obtain:

W(n+1)＝W(n)-MY(n)y(n)

wherein, M ═ diag (μ)_C，μ₂，μ₃，…，μ_N) Is a stepping coefficient matrix;

thus, the compensation algorithm can be expressed as:

y(n)＝x(n)+W(n)^TX^*(n)

W(n+1)＝W(n)-MY(n)y(n)

wherein: x^*(n) is the conjugate of the complex signal X (n) of the matrix, X^*(n)＝[x^*(n)x^*(n-1)x^*(n-2)…x^*(n-N+1)]^T；W(n)^TThe order conversion form of the high-order compensation coefficient W (n);

degenerates to first order N-1 and adds a smoothing factor β, the expression becomes:

y(n)＝x(n)+wx^*(n)

w(n+1)＝w(n)-μy(n)y(n)

w′(n+1)＝βw′(n)+(1-β)w(n+1)

wherein: mu is a first-order step coefficient, x (n) is a discrete form of IQ imbalance signal x (t), x^*(n) is the conjugate term of the signal x (n), and w (n) is the compensation factor.

4. The IQ imbalance adaptive blind compensation method according to claim 1, wherein the IQ imbalance adaptive blind compensation method further comprises: a received signal x (n) is fixed-point processed.

5. The IQ imbalance adaptive blind compensation method according to claim 1, wherein setting the step factor comprises: the initial step coefficient set is large, coarse calibration is carried out, and after a certain data volume is processed, a small step coefficient is set, and fine calibration is carried out.

6. A loop-based IQ imbalance adaptive blind compensation system, comprising:

a receiving module: used for receiving I, Q two paths of unbalanced signals x (n) and intercepting a discrete signal with a certain length;

setting a module: for setting a stepping coefficient;

conjugation module: for performing conjugation processing on received signal x (n) to obtain x^*(n)；

A compensation module: output signal y (n) ═ x (n) + w (n) x for processing received signal x (n) to obtain compensation^*(n), wherein w (n) is a compensation coefficient；

A circulation module: the method is used for introducing a stepping coefficient and smoothing the compensation coefficient w (n), obtaining the compensation coefficient w (n) required by the system in a loop iteration mode, and further obtaining a compensated output signal y (n).

7. The IQ imbalance adaptive blind compensation system according to claim 6, wherein the loop module comprises:

the compensation coefficient processing module: the method is used for introducing a stepping coefficient, and processing the compensation coefficient to obtain: w (n +1) ═ w (n) - μ y (n), where w (n) is a compensation coefficient and μ is a first-order step coefficient;

the smoothing processing module is used for smoothing the compensation coefficient w (n) to obtain w '(n +1) ═ β w' (n) + (1- β) w (n +1), and β is a smoothing coefficient;

an iteration module: for continuous evaluation of discrete signals, the formula y (n) ═ x (n) + w (n) x^*(n), w (n +1) ═ w (n) - μ y (n) is subjected to a cyclic treatment; when the compensation coefficient w (n) tends to be stable, the compensation coefficient required by the system is obtained, and a compensated receiving signal y (n) is obtained.

8. The IQ imbalance adaptive blind compensation system according to claim 6, wherein the IQ imbalance adaptive blind compensation system further comprises:

a fixed point processing module: for performing a fixed-point processing on the received signal x (n).

9. The IQ imbalance adaptive blind compensation system according to claim 6, wherein the process of the loop module obtaining the compensation coefficients required by the system comprises:

in order to obtain a compensation coefficient w (n) required by the system, the second-order statistical characteristic of the baseband modulation signal is utilized;

according to the second-order statistical characteristic of the general ideal baseband modulation signal, the baseband modulation signal is complementary with an autocorrelation function:

wherein E [. X [ ]]Indicating a desire;

the matrix expression form is as follows:

E[Y(n)y(n)]0, where y (N) ═ y (N) y (N-1) y (N-2) … y (N-N +1)]^T

Wherein: y (N) is a discrete form of the compensator output signal, τ is any integer, and N is the matrix order;

accordingly, a recursion formula of the high-order compensation coefficient w (n) can be obtained:

E[W(n+1)]＝E[W(n)]-ME[Y(n)y(n)]

simplifying expectations, we obtain:

W(n+1)＝W(n)-MY(n)y(n)

wherein, M ═ diag (μ)_C，μ₂，μ₃，…，μ_N) Is a stepping coefficient matrix;

thus, the compensation algorithm can be expressed as:

y(n)＝x(n)+W(n)^TX^*(n)

W(n+1)＝W(n)-MY(n)y(n)

wherein: x^*(n) is the conjugate of the complex signal X (n) of the matrix, X^*(n)＝[x^*(n)x^*(n-1)x^*(n-2)…x^*(n-N+1)]^T；W(n)^TThe order conversion form of the high-order compensation coefficient W (n);

degenerates to first order (N ═ 1) and adds a smoothing factor β, the expression becomes:

y(n)＝x(n)+wx^*(n)

w(n+1)＝w(n)-μy(n)y(n)

w′(n+1)＝βw′(n)+(1-β)w(n+1)

wherein: mu is a first-order step coefficient, x (n) is a discrete form of IQ imbalance signal x (t), x^*(n) is the conjugate term of the signal x (n), and w (n) is the compensation factor.

10. The IQ imbalance adaptive blind compensation system according to claim 6, wherein the setting module, in setting the step factor: the initial step coefficient set is large, coarse calibration is carried out, and after a certain data volume is processed, a small step coefficient is set, and fine calibration is carried out.

Technical Field

The invention belongs to the technical field of communication, and relates to an IQ imbalance adaptive blind compensation method and system based on circulation.

Background

IQ imbalance refers to the amplitude and phase mismatch between the In-phase (I) and Quadrature-phase (Q) branches of the transmitter and receiver. Ideally, the in-phase and quadrature branches have equal amplitude gain and 90 degree phase offset. However, in an actual communication system, it is generally difficult to achieve the ideal situation, and therefore IQ imbalance occurs. IQ imbalance may occur at the transmitter, non-ideal up-conversion, I and Q branch imbalance filters, digital-to-analog converters, etc. At the receiver, IQ imbalance is caused by non-ideal down-conversion, unbalanced filters, amplification and sampling of the I and Q branches, etc.

A common method for suppressing IQ imbalance starts from hardware, and high-performance analog devices (such as a filter, an amplifier, an analog-to-digital converter, and the like) are adopted, and although the high-performance analog devices can fundamentally suppress the influence of IQ imbalance, the high-performance analog devices generally have larger volume and higher cost, and accordingly, the power consumption and price of the mobile transceiver device are increased. Furthermore, even a high-performance analog device cannot completely suppress IQ imbalance, and its ability to suppress IQ imbalance under different environments (temperature, humidity, and the like) is different, and therefore, it is not practical to suppress IQ imbalance in the analog domain. IQ imbalance cannot be suppressed and compensated in the digital domain by digital signal processing means, but from the analog domain.

However, the existing IQ imbalance compensation algorithm in the digital domain is difficult to consider both high compensation performance and low computation complexity, and it is also difficult to adjust the algorithm structure for different systems.

Disclosure of Invention

Aiming at the defects of the prior art, the invention provides an IQ imbalance adaptive blind compensation method based on circulation. The method is a parameter-adjustable adaptive blind compensation method and can be used for different IQ unbalanced systems.

The invention also provides an IQ imbalance adaptive blind compensation system based on circulation.

The invention discloses a cycle-based IQ imbalance adaptive blind compensation method, which is realized by adopting the following technical scheme:

a loop-based IQ imbalance adaptive blind compensation method, comprising:

receiving I, Q two paths of unbalanced signals x (n) and intercepting a discrete signal with a certain length;

setting a stepping coefficient;

conjugate processing is carried out on the received signal x (n) to obtain x^*(n)；

Processing the received signal x (n) to obtain a compensated output signal y (n) ═ x (n) + w (n) x^*(n), wherein w (n) is a compensation factor;

introducing a stepping coefficient and smoothing the compensation coefficient w (n), obtaining the compensation coefficient w (n) required by the system in a loop iteration mode, and further obtaining a compensated output signal y (n).

Preferably, the step of introducing a step coefficient and smoothing the compensation coefficient w (n), obtaining the compensation coefficient w (n) required by the system in a loop iteration manner, and further obtaining the compensated output signal y (n) includes:

introducing a stepping coefficient, and processing the compensation coefficient to obtain: w (n +1) ═ w (n) - μ y (n), where w (n) is a compensation coefficient and μ is a first-order step coefficient;

smoothing the compensation coefficient w (n) to obtain w '(n +1) ═ β w' (n) + (1- β) w (n +1), wherein β is a smoothing coefficient;

continuously taking values of discrete signals, and taking values of the formula y (n) ═ x (n) + w (n) x^*(n), w (n +1) ═ w (n) - μ y (n) is subjected to a cyclic treatment;

when the compensation coefficient w (n) tends to be stable, the compensation coefficient value required by the system is obtained, and the compensated receiving signal y (n) is obtained.

Preferably, the process of obtaining the compensation coefficient required by the system comprises:

in order to obtain a compensation coefficient w (n) required by the system, the second-order statistical characteristic of the baseband modulation signal is utilized;

according to the second-order statistical characteristic of the general ideal baseband modulation signal, the baseband modulation signal is complementary with an autocorrelation function:

E[y(n)y(n-τ)]＝0，wherein E [. X [ ]]Indicating a desire;

the matrix expression form is as follows:

E[Y(n)y(n)]0, where y (N) is ═ y (N) y (N-1) y (N-2)]^T

Wherein: y (N) is a discrete form of the compensator output signal, τ is any integer, and N is the matrix order;

accordingly, a recursion formula of the high-order compensation coefficient w (n) can be obtained:

E[W(n+1)]＝E[W(n)]-ME[Y(n)y(n)]

simplifying expectations, we obtain:

W(n+1)＝W(n)-MY(n)y(n)

wherein, M ═ diag (μ)₁，μ₂，μ₃，...，μ_N) Is a stepping coefficient matrix;

thus, the compensation algorithm can be expressed as:

y(n)＝x(n)+W(n)^TX^*(n)

W(n+1)＝W(n)-MY(n)y(n)

wherein: x^*(n) is the conjugate of the complex signal X (n) of the matrix, X^*(n)＝[x^*(n)x^*(n-1)x^*(n-2)...x^*(n-N+1)]^T；W(n)^TThe order conversion form of the high-order compensation coefficient W (n);

degenerates to first order N-1 and adds a smoothing factor β, the expression becomes:

y(n)＝x(n)+wx^*(n)

w(n+1)＝w(n)-μy(n)y(n)

w′(n+1)＝βw′(n)+(1-β)w(n+1)

wherein: mu is a first-order step coefficient, x (n) is a discrete form of IQ imbalance signal x (t), x^*(n) is the conjugate term of the signal x (n), and w (n) is the compensation factor.

Preferably, the IQ imbalance adaptive blind compensation method further comprises: a received signal x (n) is fixed-point processed.

Preferably, the setting of the step factor includes: the initial step coefficient set is large, coarse calibration is carried out, and after a certain data volume is processed, a small step coefficient is set, and fine calibration is carried out.

The IQ imbalance self-adaptive blind compensation system based on circulation is realized by adopting the following technical scheme:

a loop-based IQ imbalance adaptive blind compensation system, comprising:

a receiving module: used for receiving I, Q two paths of unbalanced signals x (n) and intercepting a discrete signal with a certain length;

setting a module: for setting a stepping coefficient;

conjugation module: for performing conjugation processing on received signal x (n) to obtain x^*(n)；

A compensation module: output signal y (n) ═ x (n) + w (n) x for processing received signal x (n) to obtain compensation^*(n), wherein w (n) is a compensation factor;

a circulation module: the method is used for introducing a stepping coefficient and smoothing the compensation coefficient w (n), obtaining the compensation coefficient w (n) required by the system in a loop iteration mode, and further obtaining a compensated output signal y (n).

Preferably, the circulation module comprises:

the compensation coefficient processing module: the method is used for introducing a stepping coefficient, and processing the compensation coefficient to obtain: w (n +1) ═ w (n) - μ y (n), where w (n) is a compensation coefficient and μ is a first-order step coefficient;

the smoothing processing module is used for smoothing the compensation coefficient w (n) to obtain w '(n +1) ═ β w' (n) + (1- β) w (n +1), and β is a smoothing coefficient;

an iteration module: for continuous evaluation of discrete signals, the formula y (n) ═ x (n) + w (n) x^*(n), w (n +1) ═ w (n) - μ y (n) is subjected to a cyclic treatment; when the compensation coefficient w (n) tends to be stable, the compensation coefficient required by the system is obtained, and a compensated receiving signal y (n) is obtained.

Preferably, the IQ imbalance adaptive blind compensation system further comprises:

a fixed point processing module: for performing a fixed-point processing on the received signal x (n).

Preferably, the process of obtaining the compensation coefficient required by the system by the circulation module comprises:

in order to obtain a compensation coefficient w (n) required by the system, the second-order statistical characteristic of the baseband modulation signal is utilized;

according to the second-order statistical characteristic of the general ideal baseband modulation signal, the baseband modulation signal is complementary with an autocorrelation function:

E[y(n)y(n-τ)]＝0，wherein E [. X [ ]]Indicating a desire;

the matrix expression form is as follows:

E[Y(n)y(n)]0, where y (N) is ═ y (N) y (N-1) y (N-2)]^T

Wherein: y (N) is a discrete form of the compensator output signal, τ is any integer, and N is the matrix order;

accordingly, a recursion formula of the high-order compensation coefficient w (n) can be obtained:

E[W(n+1)]＝E[W(n)]-ME[Y(n)y(n)]

simplifying expectations, we obtain:

W(n+1)＝W(n)-MY(n)y(n)

wherein, M ═ diag (μ)₁，μ₂，μ₃，...，μ_N) Is a stepping coefficient matrix;

thus, the compensation algorithm can be expressed as:

y(n)＝x(n)+W(n)^TX^*(n)

W(n+1)＝W(n)-MY(n)y(n)

wherein: x^*(n) is the conjugate of the complex signal X (n) of the matrix, X^*(n)＝[x^*(n)x^*(n-1)x^*(n-2)...x^*(n-N+1)]^T；W(n)^TThe order conversion form of the high-order compensation coefficient W (n);

degenerates to first order (N ═ 1) and adds a smoothing factor β, the expression becomes:

y(n)＝x(n)+wx^*(n)

w(n+1)＝w(n)-μy(n)y(n)

w′(n+1)＝βw′(n)+(1-β)w(n+1)

wherein: mu is a first-order step coefficient, x (n) is a discrete form of IQ imbalance signal x (t), x^*(n) is the conjugate term of the signal x (n), and w (n) is the compensation factor.

Preferably, the setting module, in setting the step factor: the initial step coefficient set is large, coarse calibration is carried out, and after a certain data volume is processed, a small step coefficient is set, and fine calibration is carried out.

Compared with the prior art, the invention has at least the following beneficial effects:

1. the invention utilizes the extensive linear filtering of the received baseband signal, utilizes the second-order statistical characteristic of the ideal baseband equivalent signal, introduces the stepping coefficient and the compensation coefficient, and carries out IQ compensation based on the cyclic characteristic.

2. As a parameter-adjustable adaptive blind compensation method, the invention can deal with IQ unbalanced systems under different conditions by adjusting related parameters.

3. The second-order statistical characteristic of the baseband equivalent signal is not influenced by most of non-ideal factors except IQ imbalance, so the method has strong robustness.

4. The invention estimates IQ imbalance in real time in a digital domain, and carries out corresponding compensation aiming at different environments, thereby solving the problem that IQ imbalance degrees are different due to different performances of analog devices under different environments.

5. The method can adapt to the time-varying characteristics of different simulation systems, can adaptively adjust the output along with the performance change of the simulation systems, greatly reduces the compensation cost of IQ imbalance, and continuously improves the performance along with the improvement of chip processes.

Drawings

FIG. 1 is a simplified flowchart of a loop-based IQ imbalance adaptive blind compensation method according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of IQ imbalance of a signal according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of a compensator according to an embodiment of the present invention;

FIG. 4 is a graph illustrating the variation trend of the compensation coefficient w (n) according to an embodiment of the present invention;

fig. 5 is a constellation diagram after IQ imbalance compensation according to an embodiment of the present invention.

Detailed Description

Specific embodiments of the present invention will be described below with reference to the accompanying drawings, but the present invention is not limited thereto.