CN103929177B - A/D quantization bit conversion system and method in GNSS receiver - Google Patents

Abstract

The invention provides an A/D quantization bit conversion system and method in a GNSS receiver. The system comprises a comparison module and a threshold adjustment module, wherein the comparison module and the threshold adjustment module work simultaneously, high-order quantized signals are input to the first input end of the comparison module, the comparison module converts the input high-order quantized signals into low-order quantized signals according to threshold values, and then the low-order quantized signals are output from the first low-order output end of the comparison module. The second low-order output end of the comparison module is connected with the input end of the threshold adjustment module, the second input end is connected with the output end of the threshold adjustment module, the threshold adjustment module is used for adjusting threshold values so that the low-order quantized signals output by the comparison module can conform to the corresponding best probability distribution. According to the A/D quantization bit conversion system and method, on the premise that complexity and cost of an existing GNSS receiver are not increased, an anti-interference technique can be applied to the existing low-order quantization GNSS receiver.

Description

A/D quantization digit converting system and method in GNSS receiver

Technical field

The present invention relates to electronic communication and digital signal processing technique field, particularly to A/D in a kind of GNSS receiver Quantization digit converting system and method.

Background technology

Increasingly complicated with electromagnetic environment, GNSS receiver is interfered more and more.Satellite-signal travels to earth table Typically very faint during face, it is highly prone to various artificial or unartificial interference.When anti-interference, A/D quantifies to need using high How position, in order to strengthen output signal-to-noise ratio and suppression interference, retain enough useful informations using more than 8 quantization digits. When research shows to exist strong jamming, the more big required quantization digit of certain limit internal interference is more, and interference free performance is better.

When noiseless, A/D quantization digit is limited to GNSS receiver performance impact, can't be serious using low bit quantification device Deteriorate signal to noise ratio, excellent reception performance can be reached.Research and analyse when 2bit quantifies and select suitable quantization threshold, just have preferably Acquisition performance.Existing GNSS receiver typically adopts low level A/D to quantify, and such as 2bit quantifies.Due to anti-interference when A/D quantify adopt With a high position, and in existing GNSS receiver adopt low level A/D to quantify, for reduces cost with realize Anti-Jamming Technique existing more Application in receiver, needs to realize the high-order conversion quantifying to low bit quantification.At present by quantization digit from high ordertransfer to low Position most implementation methods be：Only retain the high-order former quantized result quantifying, the loss of the method Signal-to-Noise is larger.

Content of the invention

In view of the shortcomings of the prior art, the invention provides A/ in a kind of less GNSS receiver of snr loss D quantization digit converting system and method, for being converted to low bit quantification signal by high-order quantized signal.

In order to solve above-mentioned technical problem, the present invention adopts the following technical scheme that：

First, A/D quantization digit conversion method in GNSS receiver, including step：

Step 1, carries out re-quantization, encoded acquisition according to positive and negative threshold value and amplitude threshold to high-order quantized signal Low bit quantification signal, the initial value of positive and negative threshold value and amplitude threshold is rule of thumb artificially given；

Step 2, whether the low bit quantification signal that analytical procedure 1 obtains meets corresponding optimal probability distribution, if meeting, should Low bit quantification signal is final transformation result, terminates；Otherwise, execution step 3；

Step 3, adjustment positive and negative threshold value and/or amplitude threshold, based on the positive and negative threshold value after adjustment and/or amplitude door Limit value re-executes step 1；

Described re-quantization is carried out to high-order quantized signal according to positive and negative threshold value and amplitude threshold it is specially：

It is set to forward quantizations signal more than the high-order quantized signal of positive and negative threshold value, less than the high position quantization of positive and negative threshold value Signal is set to negative quantified signal；Amplitude interval is divided according to amplitude threshold, each amplitude interval corresponding default quantization knot respectively Really, the quantized result of high-order quantized signal is the interval corresponding default quantized result of the high-order affiliated amplitude of quantized signal amplitude.

The low bit quantification signal that analytical procedure 1 described in step 2 obtains whether meet corresponding optimal probability be distributed into One step includes sub-step：

2.1 arrange the interval amplitude count values of forward quantizations signal-count values, negative quantity signal-count value and each amplitude so that just The ratio of quantized signal count value and negative quantified signal count value meet low bit quantification signal optimal probability distribution, simultaneously so that The ratio of the interval amplitude count value of each amplitude also complies with the optimal probability distribution of low bit quantification signal；

2.2 often a low bit quantification signal, if forward quantizations signal, forward quantizations signal-count value subtracts 1；If negative quantity Change signal, negative quantified signal count value subtracts 1；Meanwhile, by the amplitude count value in this low bit quantification signal amplitude affiliated amplitude interval Subtract 1；

If 2.3 forward quantizations signal-count values and negative quantified signal count value are kept to 0 simultaneously, and each amplitude interval is corresponding Amplitude count value is also kept to 0 simultaneously, then low bit quantification signal meets corresponding optimal probability distribution；Otherwise, low bit quantification signal Do not meet corresponding optimal probability distribution.

Adjustment positive and negative threshold value described in step 3 and/or amplitude threshold are specially：

Plus or minus by positive and negative threshold value plus or minus default positive negative regulator unit and/or preset amplitude threshold Amplitude adjusted unit.

2nd, A/D quantization digit converting system in GNSS receiver, adjusts mould including the comparison module working and thresholding simultaneously Block, high-order quantized signal inputs the first input end of comparison module, and the high-order quantization inputting is believed by comparison module according to threshold value Number be converted to low bit quantification signal, and the first low level outfan output from comparison module；Second low level output of comparison module End connects thresholding adjusting module input, and the second input connects thresholding adjusting module outfan；Thresholding adjusting module is used for adjusting Whole threshold value is so that the low bit quantification signal of comparison module output meets corresponding optimal probability distribution.

Above-mentioned thresholding adjusting module further includes positive and negative threshold adjusting module and amplitude threshold adjusting module, positive and negative threshold Adjusting module and amplitude threshold adjusting module synchronous working, the low bit quantification that positive and negative threshold adjusting module inputs according to comparison module Signal adjusts positive and negative threshold value, and amplitude threshold adjusting module adjusts amplitude threshold according to the low bit quantification signal that comparison module inputs Value.

The invention provides a kind of A/D quantization digit converting system for GNSS receiver and method, using the present invention, The quantization digit in GNSS receiver need not be changed and can achieve application in existing GNSS receiver for the Anti-Jamming Technique.

Compared to the prior art, the present invention has advantages below and effect：

1st, provide A/D quantization digit converting system in GNSS receiver, performance is good, easily realize, and snr loss is little.

2nd, use the present invention, on the premise of not changing existing GNSS receiver system, Anti-Jamming Technique can be applied to The GNSS receiver of existing employing low bit quantification.

Brief description

Fig. 1 is present system structure chart；

Fig. 2 is A_gσ and snr lossGraph of a relation；

Fig. 3 is the flow chart that in specific embodiment, positive and negative threshold value adjusts process；

Fig. 4 is the flow chart that in specific embodiment, amplitude threshold adjusts process.

Specific embodiment

Below in conjunction with accompanying drawing, the present invention is described in detail.

See Fig. 1, present system includes the comparison module simultaneously working and thresholding adjusting module, high-order quantized signal input The first input end of comparison module, high-order quantized signal is converted to low bit quantification signal through comparison module, and from comparison module First low level outfan output.Comparison module also includes the second input and the second low level outfan, its second low level outfan Connect thresholding adjusting module input, the second input connects thresholding adjusting module outfan.

Comparison module realizes high-order quantized signal to the conversion of low bit quantification signal according to threshold value, and threshold value includes 1 Positive and negative threshold value and (2^B- 2) individual amplitude threshold, B is low bit quantification signal digit, and positive and negative threshold value is used for judging quantized signal Positive and negative, amplitude threshold is used for judging to quantify signal amplitude size.Just it is more than the quantized signal of positive and negative threshold value, less than positive and negative The quantized signal of threshold value is negative；Amplitude interval is divided according to amplitude threshold, interval according to the affiliated amplitude of quantized signal amplitude Re-quantization is carried out to quantized signal, the signal after re-quantization is carried out encode i.e. acquisition low bit quantification signal.Re-quantization It is specially：The interval corresponding quantized result of each amplitude is it is assumed that the amplitude interval corresponding quantized result of i is A_i, then amplitude is belonged to The quantized signal of amplitude interval i is quantified as A_i.

The setting of threshold value refers to the optimal probability of the low bit quantification signal of comparison module the second low level outfan output Distribution, therefore needs to obtain the optimal probability distribution of low bit quantification signal, the obtaining step of optimal probability distribution is as follows first：

Step 1, when low bit quantification signal digit is B, tries to achieve snr loss

${\mathrm{SNR}}_1=\frac{A^2}{{\mathrm{\σ}}^2}---\left(1\right)$

${\mathrm{SNR}}_2=\frac{E{\left\{r_B\right[n\left]\right\}}^2}{\mathrm{Var}\left\{r_B\right[n\left]\right\}}=\frac{E{\left\{r_B\right[n\left]\right\}}^2}{\left|E\right\{r_B^2\left[n\right]\}-E{\left\{r_B\right[n\left]\right\}}^2|}---\left(2\right)$

$\frac{{\mathrm{SNR}}_2}{{\mathrm{SNR}}_1}=\frac{2}{\mathrm{\π}}\frac{[1+2\exp \{-\frac{{(1/A_g)}^2}{{2\mathrm{\σ}}^2}\left\}\right]}{1+8\mathrm{erfc}\left(\frac{1/A_g}{\sqrt{2}\mathrm{\σ}}\right)}---\left(3\right)$

In formula (1)～(3)：

SNR₁For the signal to noise ratio of the analogue signal of input, A is satellite-signal amplitude, σ²For receiving the thermal noise power of channel；

SNR₂For the signal to noise ratio of the digital signal of output, input signal r [n] inclusion satellite-signal and Gaussian noise, Gauss Noise obeys N (0, σ²) distribution, the quantized result of input analogue signal r [n] is r_B[n]；E{r_B[n]}²Represent quantized signal r_B [n] power, Var { r_B[n] } represent Gaussian noise power；

A_gThe gain providing for automatic growth control (AGC),For error function.

A under different B values is drawn according to formula (3)_gσ and snr lossRelation curve, see Fig. 2, in Fig. 2 Relation curve includes A when B is respectively 1bit, 2bit, 3bit, 4bit, 5bit_gσ and snr lossRelation curve.

Step 2, according to A_gThe relation curve of σ and snr loss determines corresponding optimal AGC gain A of B value_g.

Optimal AGC gain A_gI.e. snr lossCorresponding A when minimum_gValue, is originally embodied as middle determination Good AGC gain A_gValue is shown in Table 1.

Optimal AGC gain A of table 1_gValue

Step 3, the anti-optimal probability distribution pushing away low bit quantification signal.

By optimal AGC gain A_gValue substitutes into the anti-optimal probability distribution pushing away B position low bit quantification signal of new probability formula (4)：

$p\left(r_B\right[n]=2i+1)=\left\{\begin{array}{cc}\frac{1}{\sqrt{{2\mathrm{\&pi;\&sigma;}}^2}}{\&Integral;}_{(2^{B-1}-1)/A_g-y\left[n\right]}^{\&infin;}\exp \{-\frac{z^2}{{\mathrm{\&sigma;}}^2}\}\mathrm{dz}& i=2^{B-1}-1\\ \frac{1}{\sqrt{{2\mathrm{\&pi;\&sigma;}}^2}}{\&Integral;}_{-\&infin;}^{-(2^{B-1}-1)/A_g-y\left[n\right]}\exp \{-\frac{z^2}{{\mathrm{\&sigma;}}^2}\}\mathrm{dz}& i={-2}^{B-1}\\ \frac{1}{\sqrt{{2\mathrm{\&pi;\&sigma;}}^2}}{\&Integral;}_{i/A_g-y\left[n\right]}^{i+1/A_g-y\left[n\right]}\exp \{-\frac{z^2}{{\mathrm{\&sigma;}}^2}\}\mathrm{dz}& {-2}^{B-1}<i<2^{B-1}-1\end{array}\right.---\left(4\right)$

In formula (4), y [n] represents satellite-signal, r_B[n]={ ± 1, ± 3, ± 5 ... } are quantized result, utilizeCondition, estimates corresponding probit p (r_B[n]=2i+1), obtain optimal probability distribution, be shown in Table 2.

The optimal probability distribution of table 2 low bit quantification signal

Optimal probability distribution adjustment positive and negative threshold value according to low bit quantification signal and amplitude threshold, positive and negative threshold value is adjusted Whole and amplitude threshold adjustment is carried out simultaneously, and the method for adjustment of positive and negative threshold value specifically includes step：

Step 1.1, a bit of in the low bit quantification signal of statistical module output, record this section of low bit quantification signal In forward quantizations signal and negative quantified signal number, and quantized signal is aligned using regressive mode and negative quantified signal number is counted Number.

Count concrete mode be：

Set forward quantizations signal-count value and the initial value of negative quantified signal count value be respectively sign0 and sign1, The ratio of sign0 and sign1 need to meet optimal probability distribution (being shown in Table 2) of low bit quantification signal.Forward quantizations signal and negative quantified Signal often occurs once, and forward quantizations signal-count value and negative quantified signal count value then accordingly subtract 1.

Step 1.2, judges whether the low bit quantification signal of comparison module output meets corresponding optimal probability distribution.

The concrete mode judging as：

Judge whether forward quantizations signal-count value and negative quantified signal count value are kept to 0 simultaneously, be first kept to 0 quantized signal Number is on the high side, then need to adjust current positive and negative threshold value；If be kept to 0 simultaneously, current positive and negative threshold value is final positive and negative Threshold value.

Step 1.3, is adjusted to current positive and negative threshold value.

The specific embodiment of this step is：

By current positive and negative threshold value plus or minus default positive negative regulator unit, the positive and negative threshold value after to adjust is current Positive and negative threshold value, the high-order quantized signal inputting is changed by comparison module according to current positive and negative threshold value and current amplitude threshold value For low bit quantification signal, the then low bit quantification signal execution step 1.1～1.3 to comparison module output, up to comparison module The low bit quantification signal of output makes forward quantizations signal-count value and negative quantified signal count value be kept to 0 simultaneously.Preset positive and negative tune Section unit is bigger, and the adjustment of positive and negative threshold value requires time for shorter；Preset positive negative regulator unit less, the adjustment of positive and negative threshold value needs Time is longer.

The method of adjustment of amplitude threshold specifically includes step：

Step 2.1, divides amplitude interval, the low bit quantification signal of statistical module output according to current amplitude threshold value In a bit of, the interval low bit quantification signal number of the different amplitudes of record, using low interval to different amplitudes of regressive mode Position quantized signal number is counted.

Count concrete mode be：

Set the initial value mag of the interval amplitude count value of different amplitudes_i, i represents the interval numbering of amplitude, i=1,2 ... N, n are amplitude interval quantity, and the initial value ratio of the interval corresponding amplitude count value of each amplitude should meet low bit quantification signal Optimal probability is distributed (being shown in Table 2)；One quantized signal often occurs, the interval amplitude count value of affiliated for this quantized signal amplitude is subtracted 1.

Step 2.2, judges whether the low bit quantification signal of comparison module output meets corresponding optimal probability distribution.

The concrete mode judging as：

Judge whether the interval amplitude count value of each amplitude is kept to 0 simultaneously, be first kept to the quantized signal in 0 amplitude interval Number is on the high side, need to adjust current amplitude threshold value；If be kept to 0 simultaneously, current amplitude threshold value is final amplitude threshold Value.

Step 2.3, is adjusted to current amplitude threshold value.

The specific embodiment of this step is：

Current amplitude threshold value plus or minus predetermined amplitude are adjusted unit, the amplitude threshold after to adjust is current Amplitude threshold, the high-order quantized signal inputting is changed by comparison module according to current positive and negative threshold value and current amplitude threshold value For low bit quantification signal, the then low bit quantification signal execution step 2.1～2.3 to comparison module output, up to comparison module The low bit quantification signal of output makes the interval amplitude count value of each amplitude be kept to 0 simultaneously.Predetermined amplitude regulation unit is bigger, width Degree threshold value adjustment requires time for shorter；Predetermined amplitude adjust unit less, then amplitude threshold adjustment require time for more long.

During the adjustment of above-mentioned positive and negative threshold value and amplitude threshold, current positive and negative threshold value and current amplitude threshold value Initial value rule of thumb set.

To be converted to as a example 2bit quantized signal by 14bit quantized signal below, further illustrate the present invention.

14bit quantized signal inputs comparison module, and comparison module is according to positive and negative threshold value and magnitude threshold initial value to input High-order quantized signal carry out re-quantization and coding, 14bit quantized signal is converted to 2bit quantized signal.Due to low level amount Changing signal digit is 2, therefore needs in the present embodiment to choose 1 positive and negative threshold value v_com and 2 amplitude threshold threshold、-threshold.Positive and negative threshold value v_com is used for judging the positive and negative of quantized signal, signal value is more than v_com's Quantized signal is forward quantizations signal, and the quantized signal that signal value is less than v_com is negative quantified signal.According to magnitude threshold value It is interval that threshold and-threshold divides amplitude：Amplitude more than threshold is interval, the width less than-threshold Degree is interval, [- threshold, 0] and (0, threshold], quantized signal is entered according to the affiliated amplitude of quantized signal amplitude is interval Row quantify, amplitude be more than threshold signal quantization be 3, less than-threshold signal quantization be -3, amplitude belong to [- Threshold, 0] interval signal quantization is -1, and it is 1 that amplitude belongs to [0, threshold] interval signal quantization.

The determination of threshold value, according to the optimal probability distribution of 2bit low bit quantification signal, is known according to table 2：2bit low bit quantification Result be -3 number of signals account for signal sum 15%, 2bit low bit quantification result be -1 number of signals account for signal sum 35%, 2bit low bit quantification result be 1 number of signals account for signal sum 35%, 2bit low bit quantification result be 3 signal Quantity accounts for the 15% of signal sum, and that is, forward quantizations signal number accounts for the 50% of signal sum, and negative quantified signal number accounts for signal sum 50%, [- threshold, threshold] amplitude interval quantization signal number accounts for the 70% of signal sum, [- threshold, Threshold] the interval outer quantized signal number of amplitude accounts for total 30% of signal.

Thresholding adjusting module adjusts door according to the low bit quantification signal adaptive that comparison module the second low level outfan exports Limit value, the adjustment of positive and negative threshold value and amplitude threshold is carried out simultaneously.

In the present embodiment, visible Fig. 3～4 of flow process of threshold value adjustment, comprise the following steps that：

Step 1, initializes positive and negative counting number value and amplitude count value.

Known according to 2bit low bit quantification corresponding optimal probability distribution in table 2：Forward quantizations signal number and negative quantified signal number The ratio accounting for quantized signal sum is respectively 50% and 50%, and that is, the quantity of forward quantizations signal and negative quantified signal ratio is for 1/1.If Fixed forward quantizations signal-count initial value sign0 and negative quantified signal count initial value sign1, sign0/sign1=1/1.

Known according to 2bit low bit quantification corresponding optimal probability distribution in table 2：Quantized result is 1 and -1 number of signals accounts for Quantized signal sum 70%, quantized result be 3 and -3 number of signals account for quantized signal sum 30%.According to amplitude threshold Value threshold division amplitude is interval, interval amplitude counting initial value mag0 and mag1 of each amplitude of setting, mag0 representative [- Threshold, threshold] the interval amplitude counting initial value of amplitude, mag1 representative [- threshold, threshold] width Amplitude outside degree interval counts initial value, mag0/mag1=7/3.

Step 2, judges whether the low bit quantification signal of comparison module output meets corresponding optimal probability distribution.

Output signal according to comparison module is counted in regressive mode.Forward quantizations signal and negative quantified signal often occur Once, forward quantizations signal-count value and negative quantified signal count value accordingly subtract 1；Meanwhile, often a quantized signal occurs, this is believed The interval amplitude count value of number respective amplitude subtracts 1.

Judge 1. whether forward quantizations signal-count value and negative quantified signal count value are kept to 0 simultaneously, judge 2. each width simultaneously Spend whether interval corresponding amplitude count value is kept to 0 simultaneously, if 1. and 2. set up simultaneously, the low bit quantification of comparison module output Signal meets optimal probability distribution, terminates and exports this low bit quantification signal from the first low level outfan of comparison module；No Then, execution step 3.

Step 3, threshold value is adjusted.

If forward quantizations signal-count value and negative quantified signal count value are not kept to 0 simultaneously, adjust positive and negative threshold value.If Sign0 is first kept to 0, illustrates that forward quantizations number of signals is on the high side, v_com plus 1；If sign1 is first kept to 0, negative quantified signal is described Quantity is on the high side, and v_com is subtracted 1.

If the interval corresponding amplitude count value of each amplitude is not kept to 0 simultaneously, adjust magnitude threshold value.If mag0 is first kept to 0, Illustrate that the interval quantized signal quantity of [- threshold, threshold] amplitude is on the high side, threshold is subtracted 1；If mag1 is first It is kept to 0, illustrate that the interval quantized signal quantity of [- threshold, threshold] amplitude is on the low side, threshold is accordingly added 1.

Claims (4)

1. A/D quantization digit conversion method in 1.GNSS receiver is it is characterised in that include step：

   Step 1, carries out re-quantization, encoded acquisition low level according to positive and negative threshold value and amplitude threshold to high-order quantized signal Quantized signal, the initial value of positive and negative threshold value and amplitude threshold is rule of thumb artificially given；

   Step 2, whether the low bit quantification signal that analytical procedure 1 obtains meets corresponding optimal probability distribution, if meeting, this low level Quantized signal is final transformation result, terminates；Otherwise, execution step 3；

   Step 3, adjustment positive and negative threshold value and/or amplitude threshold, based on the positive and negative threshold value after adjustment and/or amplitude threshold Re-execute step 1；

   Described re-quantization is carried out to high-order quantized signal according to positive and negative threshold value and amplitude threshold it is specially：

   It is set to forward quantizations signal more than the high-order quantized signal of positive and negative threshold value, less than the high-order quantized signal of positive and negative threshold value It is set to negative quantified signal；Amplitude interval is divided according to amplitude threshold, each amplitude interval corresponding default quantized result respectively is high The quantized result of position quantized signal is the interval corresponding default quantized result of the high-order affiliated amplitude of quantized signal amplitude.
2. 2. in GNSS receiver as claimed in claim 1 A/D quantization digit conversion method it is characterised in that：

   Whether the low bit quantification signal that the analytical procedure 1 described in step 2 obtains meets corresponding optimal probability distribution further Including sub-step：

   2.1 arrange the interval amplitude count values of forward quantizations signal-count values, negative quantity signal-count value and each amplitude so that forward quantizations The ratio of signal-count value and negative quantified signal count value meets the optimal probability distribution of low bit quantification signal, makes each width simultaneously The ratio of the interval amplitude count value of degree also complies with the optimal probability distribution of low bit quantification signal；

   2.2 often a low bit quantification signal, if forward quantizations signal, forward quantizations signal-count value subtracts 1；If negative quantified letter Number, negative quantified signal count value subtracts 1；Meanwhile, the interval amplitude count value of this affiliated amplitude of low bit quantification signal amplitude is subtracted 1；

   If 2.3 forward quantizations signal-count values and negative quantified signal count value are kept to 0, and the interval corresponding amplitude of each amplitude simultaneously Count value is also kept to 0 simultaneously, then low bit quantification signal meets corresponding optimal probability distribution；Otherwise, low bit quantification signal is not inconsistent Close corresponding optimal probability distribution.
3. 3. in GNSS receiver as claimed in claim 1 A/D quantization digit conversion method it is characterised in that：

   Adjustment positive and negative threshold value described in step 3 and/or amplitude threshold are specially：

   By positive and negative threshold value plus or minus preset positive negative regulator unit and/or by amplitude threshold plus or minus predetermined amplitude Adjust unit.
4. A/D quantization digit converting system in 4.GNSS receiver it is characterised in that：

   Including the comparison module working and thresholding adjusting module simultaneously, high-order quantized signal inputs the first input of comparison module End, the high-order quantized signal inputting is converted to low bit quantification signal according to threshold value by comparison module, and the from comparison module One low level outfan output；Second low level outfan of comparison module connects thresholding adjusting module input, and the second input is even Connect thresholding adjusting module outfan；Thresholding adjusting module be used for adjust threshold value so that comparison module output low bit quantification signal Meet corresponding optimal probability distribution；

   Described thresholding adjusting module further includes positive and negative threshold adjusting module and amplitude threshold adjusting module, and positive and negative threshold is adjusted Mould preparation block and amplitude threshold adjusting module synchronous working, positive and negative threshold adjusting module is believed according to the low bit quantification that comparison module inputs Number adjustment positive and negative threshold value, the low bit quantification signal adjustment amplitude threshold that amplitude threshold adjusting module input according to comparison module Value.