CN108112068A

CN108112068A - A kind of rapid automatic gain control method suitable for ofdm system

Info

Publication number: CN108112068A
Application number: CN201711256629.2A
Authority: CN
Inventors: 杨建平; 王亚茜; 姜永广; 乐强; 肖飞; 隋天宇
Original assignee: CETC 30 Research Institute
Current assignee: CETC 30 Research Institute
Priority date: 2017-12-04
Filing date: 2017-12-04
Publication date: 2018-06-01
Anticipated expiration: 2037-12-04
Also published as: CN108112068B

Abstract

The invention discloses a kind of rapid automatic gain control methods suitable for ofdm system, it is respectively used to calculate absolute error power and access adjust gain by using two kinds of power detection algorithm joint-detection power and by testing result, in addition A/D saturation detections are carried out while power is detected, stops power detection immediately when A/D reaches saturation and is quickly once decayed to the gain of current VGA；If unsaturated, whether restrain determining whether absolute error power is more than error threshold Th according to current AGC₁Or Th₂, and progress Gain tuning is decided whether according to judging result.The method of the present invention solves the problems, such as that traditional AGC technologies cannot take into account convergent speed and stability simultaneously using single power detection algorithm, the setting of double error thresholds efficiently avoids the problem of single thresholding AGC borders concussion, and the decline soon introducing of mechanism of A/D saturation detections further accelerates the convergence rate of AGC.

Description

Fast automatic gain control method suitable for OFDM system

Technical Field

The invention relates to a fast automatic gain control method suitable for an OFDM system.

Background

The Automatic Gain Control (AGC) technique is to automatically adjust the attenuation and gain of the reception path according to the detected magnitude of the received signal power, and adjust the received signal power to the optimal demodulation range of the demodulator, so as to obtain the optimal demodulation performance. The automatic gain control circuit can effectively enlarge the dynamic range of the receiving signal of the receiver and increase the distance of communication transmission, and is an important component of the receiver of a wireless communication system. The current broadband wireless communication system widely adopts Orthogonal Frequency Division Multiplexing (OFDM) technology as a key technical means of wireless transmission, and requires high AGC convergence speed and good stability. However, an obvious feature of the OFDM signal is that the OFDM signal has a high peak-to-average power ratio (PAPR), and the dynamic range of the signal is large, so that it is difficult for the conventional AGC technique using a single power detection algorithm to consider both the convergence speed and stability, and therefore, a targeted improvement is required to adapt to a new requirement.

Although some published documents present some AGC control algorithms for OFDM systems, most of them have high complexity and have certain limitations, and the problem that the convergence speed and stability are contradictory to each other cannot be really solved.

Disclosure of Invention

In order to overcome the above disadvantages of the prior art, the present invention provides a fast automatic gain control method suitable for an OFDM system, which aims to solve the problem that the conventional AGC technique in the OFDM system is difficult to consider both the convergence speed and stability, avoid the influence caused by the high peak-to-average ratio OFDM signal and the bursty interference signal, and greatly improve the convergence speed of the AGC while ensuring the stable convergence of the AGC.

The technical scheme adopted by the invention for solving the technical problems is as follows: a fast automatic gain control method suitable for an OFDM system comprises the following steps:

step one, performing combined power detection and simultaneously performing A/D saturation detection on an ADC;

step two, judging whether A/D saturation is achieved: if not, finishing a detection period and calculating to obtain the average power P of the detection window _SWA (m) and the average power P of the received signal _IIR (m) then entering step three; if yes, immediately stopping power detection and entering a sixth step;

step three, judging whether the current AGC is converged: if not, entering the step four; if yes, entering a fifth step;

step four, utilizing P _IIR (m) calculating the absolute error power and determining whether the absolute error power is greater than an error threshold Th ₁ : if yes, entering a step seven; if not, entering step eight;

step five, utilizing P _IIR (m) calculating absolute error power, and determining whether the absolute error power is greater than an error threshold Th ₂ : if yes, entering a step seven; if not, entering step eight;

step six, quickly attenuating the gain of the current VGA once, setting the convergence mark to be 0 at the same time, and then entering the step nine;

step seven, utilizing P _SWA (m) adjusting the gain, setting the convergence flag to 0, and then entering the ninth step;

step eight, keeping the gain of the path unchanged, setting the convergence mark to be 1, and then entering the step nine;

and step nine, gain configuration is carried out, and then the step one is returned.

Compared with the prior art, the invention has the following positive effects:

the method of the invention adopts two power detection algorithms with complementary characteristics to carry out joint detection, and the detection results are respectively used for error judgment and gain calculation, thereby solving the problems that the traditional AGC technology adopts a single power detection algorithm and can not simultaneously give consideration to the speed and stability of convergence, the setting of double error thresholds effectively avoids the problem of single threshold AGC boundary oscillation, and the introduction of an A/D saturation detection fast attenuation mechanism further accelerates the convergence speed of AGC.

Drawings

The invention will now be described, by way of example, with reference to the accompanying drawings, in which:

FIG. 1 is a control flow diagram of the method of the present invention.

Detailed Description

One of the key technologies of the invention is joint power detection, and the core idea is to distinguish the detection power for error judgment and the detection power for calculating error gain, wherein the former adopts a power detection algorithm with good flexibility and smooth filtering characteristic to perform power detection, and the latter adopts a power detection algorithm with good real-time performance and high accuracy to perform power detection, so that the problem that the convergence speed and the stability of AGC are contradictory can be effectively solved.

The second key technology of the invention is double-error threshold setting, and the idea is to set double-error thresholds and set the threshold of AGC convergence to be strictly wider, thereby effectively solving the problem of single threshold boundary oscillation and ensuring that the stability of AGC is better.

The third key technology of the invention is saturation fast attenuation processing, and the idea is to perform saturation detection on the A/D and perform fast attenuation on the channel gain when the A/D is saturated, so that the problem of inaccurate power detection when the A/D is saturated can be effectively solved, and the convergence rate of AGC is further increased.

The specific contents are as follows:

1. joint power detection

The invention adopts two power detection algorithms to detect the average power of a sampling signal x (n) at the same time:

the first is a Sliding Window Average power detection method (SWA), in which the Average power of a detection Window can be obtained by directly accumulating and averaging the power of sampling points in the detection Window. The calculation formula is as follows:

in the formula, W is the window length of the detection window.

The second method is an Infinite Impulse Response filtering power detection method (IIR), which uses a first-order IIR filter to filter the real-time power of the sampling point and also can calculate the average power of the received signal. The calculation formula is as follows:

P _IIR (m)＝(1-α)P _IIR (m-1)+αx ² (n) (2)

wherein α ∈ (0, 1).

The invention uses the result detected by the IIR detection method to judge whether AGC starts gain adjustment or not, and uses the result detected by the SWA detection method to calculate the size of the adjustment gain.

2. Double error threshold setting

To facilitate threshold decision and gain calculation, all power is converted to dB here. Detecting power P by IIR detection method _IIR (m) and a reference power P _ref Taking the absolute value after subtraction to obtain the absolute error power delta P _IIR ＝|P _IIR (m)-P _ref L. The invention sets double error thresholds Th ₁ And Th ₂ Wherein Th ₁ <Th ₂ . Initially, the AGC is in an out-of-lock state, gain adjustment is needed, and when delta P is reached _IIR <Th ₁ When the AGC is converged, the AGC stops adjusting the gain, and the original state is maintained only at delta P _IIR >Th ₂ When the AGC is considered to be out of lock, the gain adjustment needs to be restarted.

3. Treatment of saturated fast decay

The invention detects the power and detects the saturation of ADC, and carries out special treatment. When the ADC is saturated, the absolute value of the sampling signal is limited to the maximum value which can be represented by the ADC, when the maximum value is detected by the AGC, the power detection is immediately stopped, the gain of the current VGA is quickly attenuated once, then the detection is restarted, and the steps are repeated until no maximum value is detected in a detection period, and a normal AGC control flow is started. The influence of ADC saturation on the convergence speed of AGC can be eliminated to a great extent.

Based on the above, the control flow of the AGC of the present invention is shown in fig. 1. Wherein, G _curr Is the current detection cycle path gain, G _next And the AGC convergence flag is an AGC convergence flag, and the power-on initial is 0, that is, the AGC convergence flag is in an out-of-lock state. The method specifically comprises the following steps:

step two, judging whether A/D saturation is achieved: if not, completing a detection period and calculating to obtain P _SWA (m) and P _IIR (m) then entering step three; if yes, immediately stopping power detection and entering a sixth step;

step three, judging whether the current AGC is converged, namely AGC _ conv _ flag = =1: if not, entering the step four; if yes, entering the step five;

step four, judging whether | P is satisfied _IIR (m)-P _ref |>Th ₁ : if yes, the AGC is considered to be unlocked, and the step seven is entered; if not, the AGC is considered to be converged, and the step eight is entered;

step five, judging whether | P is satisfied _IIR (m)-P _ref |>Th ₂ : if so, determining that the AGC is unlocked, and entering a seventh step; if not, the AGC is considered to be converged, and the step eight is entered;

step six, the current path gain value G _curr Subtracting the fast attenuation value G _fast Obtaining the path gain G of the next detection period _next I.e. G _next ＝G _curr -G _fast Setting the convergence flag to be 0, namely agc _ conv _ flag =0, and then entering step nine;

step seven, the gain value G of the current path is calculated _curr Plus a reference power P _ref And the detected power P _SWA Difference of (m)Taking the value as the gain value G of the next detection cycle path _next I.e. G _next ＝G _curr +P _ref -P _SWA (m) while setting the convergence flag to 0, i.e., agc _ conv _ flag =0, and then proceeding to step nine;

step eight, keeping the path gain unchanged, namely G _next ＝G _curr Meanwhile, the convergence flag is set to 1, that is, agc _ conv _ flag =1: then entering the step nine;

and step nine, gain configuration is carried out, and then the step one is re-entered.

Claims

1. A fast automatic gain control method suitable for an OFDM system is characterized in that: the method comprises the following steps:

step four, utilizing P _IIR (m) calculating the absolute error power and determining whether the absolute error power is greater than an error threshold Th ₁ : if yes, entering a seventh step; if not, entering step eight;

step five, utilizing P _IIR (m) calculating the absolute error power and determining whether the absolute error power is greater than an error threshold Th ₂ : if yes, entering a step seven; if not, entering step eight;

2. The method according to claim 1, wherein the method comprises: step four and step five said utilizing P _IIR (m) the method of calculating the absolute error power is: delta P _IIR ＝|P _IIR (m)-P _ref L, wherein P _ref Is the reference power.

3. The method according to claim 1, wherein the method comprises: step six the method for rapidly attenuating the gain of the current VGA once comprises the following steps: the current path gain value G _curr Subtracting the fast attenuation value G _fast Path gain G as next detection period _next 。

4. The method according to claim 1, wherein the method comprises: the method for adjusting the gain comprises the following steps: the current path gain value G _curr Plus a reference power P _ref And P _SWA (m) as the path gain G of the next detection period _next 。

5. The method of claim 1, wherein the fast automatic gain control is applied to an OFDM system, and the method comprises: p is obtained by calculation according to the following formula _SWA (m)：

Where x (n) is the sampling signal and W is the window length of the detection window.

6. According to the rightThe fast automatic gain control method for the OFDM system according to claim 1, wherein: p is obtained by calculation according to the following formula _IIR (m)：

P _IIR (m)＝(1-α)P _IIR (m-1)+αx ² (n)

Wherein x (n) is the sampling signal, and alpha is epsilon (0, 1).