Automatic gain control method based on signal-to-noise ratio estimation of over-sampled signal
Technical Field
The invention relates to the technical field of digital wireless communication transmission, in particular to an Automatic Gain Control (AGC) method based on signal-to-noise ratio estimation of an oversampling signal.
Background
In a wireless communication system, the signal strength has a large dynamic range along with the difference of factors such as the strength of signal transmitting end power, channel environment, near-far effect, receiving condition and the like. In order to ensure the validity and reliability of the signal before receiving and demodulating, an Automatic Gain Control (AGC) circuit is often designed to adjust the signal.
Generally, the AGC circuit has three implementations of feedforward, feedback and hybrid loop. The feedback loop AGC usually adopts algorithms such as output signal envelope detection method, square detection method or power error detection method to obtain error voltage and control voltage, and then performs linear logarithmic increment gain adjustment, for example, in "Khoury JM, On the Design of Constant setting time AGC Circuits [ J]Signal peak gain control as designed in IEEE Transactions on Circuits and Systems,1998,45(3): 283-; in the Wednesday, Luman Honghua, Huangjian nationalityConstant settling time digital AGC loop design [ J]The constant establishing time AGC method based on the square power error detection method is designed in the aircraft measurement and control bulletin, 2013,32(4): 316-; in "Liu A, An J, Wang A. Performance Analysis of aDigital Feedback AGC with Constant setting Time [ C]IEEE 12th international conference on Communication Technology,2010 "digital AGC loop design for QPSK signals; design and simulation of coherent automatic gain control of Liuxi, Von Wen, Lichul, and Zhunan satellite measurement and control receiver [ J]Information and electronics engineering, 2012,10(6):654 and 658 ", based on the AGC method after coherent demodulation. The signal power estimation is mainly P1~P5:
P1=I2(n)+Q2(n)
P3=|I(n)|+|Q(n)|
These methods often estimate the signal power as the total power of the received signal, i.e., the sum of the power of the desired signal and the power of the noise. In the case of low signal-to-noise ratio, the power of the useful signal after a/D sampling cannot meet the expected requirement even if the signal before a/D sampling is adjusted. Digital coherent AGC methods, data decision-based and data-aided AGC methods can accurately estimate the useful signal power, however, these methods all require demodulation of the signal first, increasing the processing delay.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides an Automatic Gain Control (AGC) method based on the signal-to-noise ratio estimation of an oversampling signal, which can eliminate noise interference, more accurately improve the automatic gain control accuracy of the oversampling signal compared with the traditional method, enable the power consumption of a system to be more efficient, reduce the processing time delay and simultaneously improve the performance of demodulation synchronization.
The above purpose of the invention is mainly realized by the following technical scheme:
an automatic gain control method based on signal-to-noise ratio estimation of an oversampled signal, comprising:
calculating a second moment and a fourth moment of the oversampled signal;
calculating useful signal power estimation of the over-sampled signal according to the second moment and the fourth moment of the over-sampled signal;
obtaining a signal-to-noise ratio estimation of the over-sampled signal according to the useful signal power estimation;
comparing the signal-to-noise ratio estimate of the over-sampled signal with a signal-to-noise ratio threshold, and if the signal-to-noise ratio estimate of the over-sampled signal is greater than or equal to the signal-to-noise ratio threshold, determining the power error of the over-sampled signal as the difference between the second moment of the over-sampled signal and the square of the reference signal level; and if the signal-to-noise ratio estimation of the over-sampled signal is smaller than the signal-to-noise ratio threshold, the power error of the over-sampled signal is the difference between the power estimation of the useful signal and the square of the reference signal level.
In the above automatic gain control method, the oversampled signal is y (n), and the expression is as follows:
wherein: pSFor useful signal power, PNW (n) is additive white gaussian noise, θ is carrier phase offset, and x (n) is the transmit signal.
In the above automatic gain control method, the transmission signal x (n) is expressed as follows:
x(n)=∑ia(i)h(nTs-iT-τ)
h () is the raised cosine filter function with roll-off coefficient α, T is the symbol period, Tsτ is the timing offset for the sampling period, and a (i) is the transmit side signal.
In the above automatic gain control method, the second moment M of the oversampled signal2And fourth order moment M4Respectively, as follows:
wherein: e [ ] is desirable.
In the above automatic gain control method, a specific method for calculating the useful signal power estimation of the oversampled signal from the second moment and the fourth moment of the oversampled signal is as follows:
useful signal power estimation for MPSK and MQAM signals
Comprises the following steps:
wherein: is defined as:
=E[|a(i)|4]R(0)+2λM∑m≥1R(m)
r (m) represents a correlation function, specifically:
wherein: h (v) is a frequency domain function of h (); lambda [ alpha ]MThe parameters are corresponding to different modulation modes, a (i) is a transmitting end signal, α is a roll-off coefficient, and m is an integer greater than or equal to 0.
In the above automatic gain control method, λ is a value for MPSK and MQAM signalsMThe values of (A) are as follows:
in the above automatic gain control method, a specific method for obtaining the snr estimate of the oversampled signal based on the power estimate of the useful signal is as follows:
signal-to-noise ratio estimation of oversampled signals for MPSK and MQAM signals
The method comprises the following specific steps:
wherein:
is an estimate of the noise power.
In the above automatic gain control method, the signal-to-noise ratio estimate of the oversampled signal is compared to a signal-to-noise ratio threshold,
if the following conditions are met:
wherein:
a signal-to-noise ratio estimate for the oversampled signal; rho
thIs the signal-to-noise ratio threshold; e
PIs the power error of the oversampled signal; m
2Is the second moment of the oversampled signal; a. the
refIs a reference signal level;
if the following conditions are met:
wherein:
is a useful signal power estimate.
In the above automatic gain control method, for MPSK and MQAM signals, the signal-to-noise ratio threshold ρthThe value is 10 dB.
Compared with the prior art, the invention has the following beneficial effects:
(1) the invention provides an Automatic Gain Control (AGC) method based on signal-to-noise ratio estimation of an over-sampled signal, which comprises the steps of firstly designing a useful signal power estimation algorithm and a signal-to-noise ratio estimation algorithm of the over-sampled signal, then applying the useful signal power estimation algorithm and the signal-to-noise ratio estimation algorithm to a digital AGC loop, and respectively switching a power error detector in the AGC loop into a new power estimation algorithm and a traditional power estimation algorithm according to SNR estimation values at medium-low signal-to-noise ratio and medium-high signal-to-noise ratio, wherein the method.
(2) The AGC method based on the signal-to-noise ratio estimation of the over-sampled signal is suitable for MPSK and MQAM modulation signals under low signal-to-noise ratio, can eliminate noise interference, and can more accurately improve the accuracy of automatic gain control compared with the traditional method;
(3) the AGC method based on the signal-to-noise ratio estimation of the over-sampled signal switches to adopt a new method and a traditional method to calculate a power error function based on the signal-to-noise ratio estimation of the over-sampled signal, so that the power consumption of the system can be more efficient;
(4) the AGC method based on the signal-to-noise ratio estimation of the over-sampled signal estimates the signal power and the signal-to-noise ratio based on the over-sampled signal, does not need coherent demodulation, reduces the processing time delay and simultaneously improves the demodulation synchronization performance.
Drawings
FIG. 1 is a block diagram of an AGC method based on signal-to-noise ratio estimation of an oversampled signal in accordance with the present invention;
fig. 2 is a diagram illustrating the signal power estimation bias performance of the AGC method based on the signal-to-noise ratio estimation of the oversampled signal in embodiment 1 of the present invention;
fig. 3 shows the amplitude gain convergence performance of the AGC method based on the signal-to-noise ratio estimation of the oversampled signal in embodiment 1 of the present invention and the conventional method.
Detailed Description
The invention is described in further detail below with reference to the following figures and specific examples:
fig. 1 shows a block diagram of an AGC method based on an estimation of a signal-to-noise ratio of an oversampled signal, which shows that the AGC method based on the estimation of the signal-to-noise ratio of the oversampled signal mainly comprises the following steps:
step one, calculating a second moment and a fourth moment of the over-sampling signal y (n). The received signal y (t) is the signal a (i) at the transmitting end after shaping and filtering, and is matched with the received signal at the receiver through an AWGN channel, and after sampling, the signal is expressed as:
wherein, PSFor useful signal power, PNW (n) is additive white gaussian noise with zero mean and variance of 1, θ is carrier phase offset, and x (n) is a transmit signal, expressed as:
x(n)=∑ia(i)h(nTs-iT-τ)
where h (t) is a raised cosine filter function with roll-off factor α, and t is nTs-iT- τ; t is the symbol period, TsIs the sampling period, tau is the timing deviation, and | tau | is less than or equal to T/2.
Second moment M of oversampled signal y (n)2Fourth order moment M4Respectively calculated as:
wherein, E2]N may be averaged by accumulatingPThe data symbols to obtain an approximate result.
Step two, using the second moment and the fourth moment obtained in the step one for estimating the useful signal power of the over-sampled signal, wherein for MPSK and MQAM signals, the useful signal power estimation method comprises the following steps:
wherein, is defined as:
=E[|a(i)|4]R(0)+2λM∑m≥1R(m)
wherein, for MPSK and MQAM signals, lambdaMThe values of (A) are as follows:
r (m) (m is an integer of 0 or more) represents a correlation function, specifically:
wherein: h (v) is a frequency domain function of h (t); lambda [ alpha ]MParameters corresponding to different modulation modes, a (i) is a transmitting end signal, and α is a roll-off coefficient.
When calculating the value, R (m) is accumulated and summed, and the value range of m can be approximated by a finite number. When the roll-off coefficient α is 0.25, 0.35, or 0.5, the values for the BPSK signal are ≈ 1.331, 1.215, or 1.066, respectively; for M-PSK, M is more than or equal to 4, and the values are respectively about 1.106, 1.022 and 0.920; for a 16QAM signal, the values are ≈ 1.316, 1.235, 1.120, respectively.
And (III) using the result of the useful signal power estimation in the step (II) for the signal-to-noise ratio estimation of the over-sampled signal, wherein for the MPSK and MQAM signals, the signal-to-noise ratio estimation method comprises the following steps:
wherein:
is an estimate of the noise power.
And step (IV) comparing the useful signal power estimation result of the over-sampling signal with a reference signal to extract a power error, wherein a power error detection function in the AGC loop can be expressed as:
wherein A isrefIs the reference signal level.
Similarly, when applied to a constant settling time digital AGC loop or a simplified constant settling time digital AGC loop, the power error function is expressed as:
comparing the signal-to-noise ratio estimation of the over-sampled signal with a signal-to-noise ratio threshold, wherein if the signal-to-noise ratio estimation of the over-sampled signal is greater than or equal to the signal-to-noise ratio threshold, the power error of the over-sampled signal is the difference between the second moment of the over-sampled signal and the square of the reference signal level; and if the signal-to-noise ratio estimation of the over-sampled signal is smaller than the signal-to-noise ratio threshold, the power error of the over-sampled signal is the difference between the power estimation of the useful signal and the square of the reference signal level. Namely:
if the following conditions are met:
wherein:
a signal-to-noise ratio estimate for the oversampled signal; rho
thIs the signal-to-noise ratio threshold; e
PFor oversampling the power of the signalAn error; m
2Is the second moment of the oversampled signal; a. the
refIs a reference signal level;
if the following conditions are met:
wherein:
is a useful signal power estimate. For MPSK and MQAM signals, rho in the embodiment of the invention
thThe value is 10 dB.
In this step, that is, the signal-to-noise ratio estimation method in step (three) is combined with the power error detection function in step (four) to obtain an AGC method based on signal-to-noise ratio estimation, which is used to further improve the working efficiency of the communication system, and the error detection function of this method is:
the step switches the power error detector in the AGC loop into a new power estimation algorithm and a traditional power estimation algorithm respectively according to the SNR estimated value in the middle-low signal-to-noise ratio and the middle-high signal-to-noise ratio, and can accurately control the signal gain.
Example 1
The processing results of the prior art timing recovery method and the timing recovery method of the present invention are compared.
Signal power estimation performance
The simulation performance of the signal power estimation adopts QPSK signal, oversampling is 4 times, the roll-off coefficient is α -0.25, and the signal power is PS=1,NPAt 1000, i.e. at 10M symbol rate the smoothing period is 0.1 ms. Fig. 2 shows the signal power estimation bias performance of the AGC method based on the signal-to-noise ratio estimation of the oversampled signal in embodiment 1 of the present invention; from the analysis of FIG. 2, the square ratio of the modulus valueCompared with the (MSL) method, the new algorithm with low signal-to-noise ratio can obtain accurate signal power estimation performance.
(II) AGC amplitude gain convergence performance
The convergence performance of the new algorithm and the AGC loop estimated by adopting the formula (2) signal power, and the normalized bandwidth of the loop is BLT=5×10-3Signal-to-noise ratio SNR of 5dB, reference signal power to useful signal power ratio (P)ref/PS) Set to 3dB and 6dB respectively, namely the signal amplitude ratio is RA=Aref/ASAnd ≈ 1.4 and 0.5, as shown in fig. 3, the amplitude gain convergence performance of the AGC method based on the signal-to-noise ratio estimation of the oversampled signal in embodiment 1 of the present invention and the conventional method is shown, and it can be known from the analysis of fig. 3 that the amplitude gain convergence of the conventional MSL method is about 1.13 and 0.44, but the new algorithm of the present invention can more accurately converge the amplitude gain to 1.4 and 0.5, and the influence of the algorithm on signal amplification is larger than that of signal reduction.
The above description is only for the best mode of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention.
Those skilled in the art will appreciate that the invention may be practiced without these specific details.