CN102118333B - Method and device for canceling DC offsets - Google Patents

Abstract

The embodiment of the invention discloses a method and a device for canceling DC (Direct Current) offsets. The method comprises the following steps: carrying out least-squares DC estimation on input signals, obtaining an estimation value of DC offsets, and obtaining compensation signals of input signals according to the estimation value of the DC offsets; unwinding the compensation signals to obtain first unwinding signals, and unwinding the input signals to obtain second unwinding signals; and comparing an SNR (Signal to Noise Ratio) estimation value of the first unwinding signals with that of the second unwinding signals; and determining the DC offsets and canceling output signals according to comparison results. Compared with the traditional technical scheme for canceling DC offsets, the technical scheme provided by the embodiment of the invention has the following advantages: output signals can be dynamically selected according to the SNR state, so that the suitability to various application scenes can be achieved and the receiving performance can be improved under various application scenes.

Description

A kind of method for eliminating dc bias and device

Technical field

The present invention relates to communication technical field, particularly relate to a kind of method for eliminating dc bias and device.

Background technology

In communication technical field, radio circuit mainly can be divided into three parts, transmitter, receiver, local oscillation circuit according to function.For receiver, mainly comprise the types such as superheterodyne receiver, zero intermediate frequency reciver (homodyne receiver) and nearly zero intermediate frequency reciver.Zero intermediate frequency reciver refers to radiofrequency signal is directly become to the receiver that centre frequency is zero baseband signal by mixing, zero intermediate frequency reciver structure is high owing to having chip integration, required external device is few, cost is low, advantage low in energy consumption, be widely applied, the radio-frequency transmitter for example adopting in GSM (Global System for Mobile communication, global system for mobile communications) is generally zero intermediate frequency reciver.But in the course of work of zero intermediate frequency reciver, the direct voltage being produced in base band by local oscillator self-mixing and high reject signal self-mixing can worsen received signal, needs DC-offset correction so receive the accuracy of signal in order to improve.

A kind of scheme of prior art is the average direct current removing method of time slot: method that the sampled point of the reception signal in same time slot is averaging obtains dc-bias, then DC-offset correction to received signal.But, the impact of being launched symbol, multipath, noise etc., the determined direct current biasing error of the method is larger, has a strong impact on the accuracy that receives signal.

The another kind of scheme of prior art is LS (Least Square, least square) direct current method of estimation: channel impulse response and direct current are combined to estimation, thereby obtain dc-bias according to joint estimate, then DC-offset correction to received signal.But, known according to simulation result, in some scene, for example low SNR (Signal to Noise Ratio, signal to noise ratio) under scene or HT100 channel, the determined direct current biasing error of the method is larger, may not improve with respect to the average direct current removing method of time slot, even may cause performance loss.

Visible, under different application scenarioss, all may there is larger error in existing two kinds of determined direct current biasings of method for eliminating dc bias, cause DC-offset correction exactly, thereby affect receptivity.

Summary of the invention

For solving the problems of the technologies described above, the embodiment of the present invention provides a kind of method for eliminating dc bias and device, and to improve the receptivity under various application scenarioss, technical scheme is as follows:

A kind of method for eliminating dc bias, comprising:

Input signal is carried out to the estimation of least square direct current, obtain direct current biasing estimated value, according to described direct current biasing estimated value, obtain the compensating signal of described input signal;

Described compensating signal is untwisted, obtain the first de-rotated signal, and described input signal is untwisted, obtain the second de-rotated signal;

Signal-to-noise ratio (SNR) estimation value to described the first de-rotated signal and the signal-to-noise ratio (SNR) estimation value of described the second de-rotated signal compare, and determine direct current biasing elimination output signal according to comparative result.

The embodiment of the present invention also provides a kind of direct current biasing cancellation element, comprising:

Direct current biasing determination module, for input signal being carried out to the estimation of least square direct current, obtains direct current biasing estimated value;

Compensating signal determination module, for the direct current biasing estimated value definite according to described direct current biasing determination module, obtains the compensating signal of described input signal;

The module of untwisting, for described compensating signal is untwisted, obtains the first de-rotated signal, and described input signal is untwisted, and obtains the second de-rotated signal;

Output signal determination module, compares for the signal-to-noise ratio (SNR) estimation value of the signal-to-noise ratio (SNR) estimation value to described the first de-rotated signal and described the second de-rotated signal, determines direct current biasing elimination output signal according to comparative result.

According to embodiment of the present invention scheme, first input signal is carried out to least square direct current and estimate to be compensated signal; Before output direct current biasing is eliminated result, respectively the original input signal untwisting and the signal to noise ratio of the compensating signal untwisting are carried out to valuation and comparison, then according to comparative result, a selection signal-to-noise ratio (SNR) estimation value preferably road signal is eliminated output signal as final direct current biasing, compared with existing direct current biasing cancellation scheme, the scheme that the embodiment of the present invention provides can be according to signal to noise ratio Dynamic Selection output signal, thereby adaptation plurality of application scenes, improves the receptivity under various application scenarioss.

Accompanying drawing explanation

In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, to the accompanying drawing of required use in embodiment or description of the Prior Art be briefly described below, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skills, do not paying under the prerequisite of creative work, can also obtain according to these accompanying drawings other accompanying drawing.

The flow chart of a kind of method for eliminating dc bias that Fig. 1 provides for the embodiment of the present invention;

The signal processing mode schematic diagram of a kind of method for eliminating dc bias that Fig. 2 provides for the embodiment of the present invention;

The second flow chart of a kind of method for eliminating dc bias that Fig. 3 provides for the embodiment of the present invention;

The third flow chart of a kind of method for eliminating dc bias that Fig. 4 provides for the embodiment of the present invention;

The 4th kind of flow chart of a kind of method for eliminating dc bias that Fig. 5 provides for the embodiment of the present invention;

The second signal processing mode schematic diagram of a kind of method for eliminating dc bias that Fig. 6 provides for the embodiment of the present invention;

The 5th kind of flow chart of a kind of method for eliminating dc bias that Fig. 7 provides for the embodiment of the present invention;

The third signal processing mode schematic diagram of a kind of method for eliminating dc bias that Fig. 8 provides for the embodiment of the present invention;

The 4th kind of signal processing mode schematic diagram of a kind of method for eliminating dc bias that Fig. 9 provides for the embodiment of the present invention;

The structural representation of a kind of direct current biasing cancellation element that Figure 10 provides for the embodiment of the present invention;

The second structural representation of a kind of direct current biasing cancellation element that Figure 11 provides for the embodiment of the present invention;

The third structural representation of a kind of direct current biasing cancellation element that Figure 12 provides for the embodiment of the present invention.

Embodiment

In prior art, the average direct current removing method of time slot and least square direct current method of estimation are two kinds of method for eliminating dc bias, but under some application scenarios, all may there is larger error in these two kinds of determined direct current biasings of method, cause DC-offset correction exactly, thereby affect receptivity.A kind of method for eliminating dc bias and device that the embodiment of the present invention provides, can improve the receptivity under various application scenarioss effectively.

First a kind of method for eliminating dc bias embodiment of the present invention being provided describes, and the method comprises:

Input signal is carried out to the estimation of least square direct current, obtain direct current biasing estimated value, according to described direct current biasing estimated value, obtain the compensating signal of described input signal;

Described compensating signal is untwisted, obtain the first de-rotated signal, and described input signal is untwisted, obtain the second de-rotated signal;

Signal-to-noise ratio (SNR) estimation value to described the first de-rotated signal and the signal-to-noise ratio (SNR) estimation value of described the second de-rotated signal compare, and determine direct current biasing elimination output signal according to comparative result.

According to embodiment of the present invention scheme, first input signal is carried out to least square direct current and estimate to be compensated signal; Before output direct current biasing is eliminated result, respectively the original input signal untwisting and the signal to noise ratio of the compensating signal untwisting are carried out to valuation and comparison, then according to comparative result, a selection signal-to-noise ratio (SNR) estimation value preferably road signal is eliminated output signal as final direct current biasing, compared with existing direct current biasing cancellation scheme, the scheme that the embodiment of the present invention provides can be according to signal to noise ratio Dynamic Selection output signal, thereby adaptation plurality of application scenes, improves the receptivity under various application scenarioss.

Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described, obviously, described embodiment is only the present invention's part embodiment, rather than whole embodiment.Based on the embodiment in the present invention, those of ordinary skills, not making the every other embodiment obtaining under creative work prerequisite, belong to the scope of protection of the invention.

Shown in Fig. 1, Fig. 2, a kind of method for eliminating dc bias that the embodiment of the present invention provides, can comprise the following steps:

S101, carries out the estimation of least square direct current to input signal, obtains direct current biasing estimated value;

In the present embodiment, input signal is carried out before direct current biasing elimination, can directly carry out the estimation of least square direct current to input signal, thereby obtain direct current biasing estimated value corresponding to input signal, carry out subsequent treatment.

Wherein, input signal is carried out to the estimation of least square direct current, as shown in Figure 3, specifically can comprise the following steps:

S101a, carries out Blind Modulation detection to input signal, obtains the modulation type of input signal;

Input signal is carried out to Blind Modulation and detect the modulation type that can obtain current input signal, for example, for gsm system, can learn that current input signal is through GMSK modulation or 8PSK modulation, in subsequent step, by according to the input signal of different modulation types, carrying out least square direct current corresponding different training sequence symbols while estimating.

S101b, according to the modulation type of input signal, determines the training sequence symbols of least square direct current estimation function;

Still describe as an example of gsm system example:

If current input signal, through GMSK modulation, is carried out training sequence symbols

rotation;

If current input signal, through 8PSK modulation, is carried out training sequence symbols

rotation.

S101c, carries out the estimation of least square direct current according to determined training sequence symbols to input signal.

The definite training sequence symbols difference of input signal of different modulating type, so when having determined after training sequence symbols, the corresponding least square direct current of current input signal estimation function can uniquely be determined, and then input signal is carried out to the estimation of least square direct current.

Wherein, the computational process that least square direct current is estimated is as follows:

Least square direct current method of estimation is by combining estimation to channel impulse response and direct current biasing.Wherein, being modeled as of the previous input signal that untwists receiving:

$r(61+\frac{L-1}{2}+n)=\underset{k=0}{\overset{L-1}{\mathrm{\Σ}}}{\mathrm{ts}}^{\′}(L-1-k+n)\·h^{\′}\left(k\right)+(d_I+j{d}_Q)+n_0\left(n\right)$

n＝0，1，2......26-L

Wherein, d _i+ jd _qthe direct current biasing estimated value d on WeiI/Q road _i, d _qplural form; Ts ' is (n) postrotational training sequence symbols, and L is channel impulse response length; H ' is (k) channel estimating; n ₀(n) be noise.

With matrix notation input signal be:

R＝SH+N

Matrix corresponding to training sequence symbols is:

Channel impulse response and direct current biasing confederate matrix are:

$H={\left[\begin{array}{c}{h}^{\′}\left(0\right)\\ h^{\′}\left(1\right)\\ .\\ .\\ .\\ h^{\′}(L-1)\\ (d_I+j{d}_Q)\end{array}\right]}_{(L+1)\×1}$

Input signal homography is:

$R={\left[\begin{array}{c}r(61+\frac{L-1}{2})\\ r(62+\frac{L-1}{2})\\ .\\ .\\ .\\ r(87-L+\frac{L-1}{2})\end{array}\right]}_{(26-L+1)\×1}$

Noise homography is:

$N={\left[\begin{array}{c}{n}_0\left(0\right)\\ n_0\left(1\right)\\ .\\ .\\ .\\ n_0(26-L+1)\end{array}\right]}_{(26-L+1)\×1}$

Therefore, the LS solution of H is:

H＝(S ^HS) ^-1S ^HR

Solve after H, can obtain input signal I/Q road direct current biasing estimated value d _i, d _q.

S102, according to described direct current biasing estimated value, obtains the compensating signal of described input signal;

Utilize least square direct current method of estimation, obtain the direct current biasing estimated value d on input signal I/Q road _i, d _q; When obtaining after direct current biasing estimated value, can directly from input signal, deduct direct current biasing estimated value, be compensated signal.

In order further to improve the accuracy of compensating signal, can also revise to a certain extent direct current biasing estimated value:

S102a, multiplies each other direct current biasing estimated value and default gain factor a, the direct current biasing estimated value after being gained;

S102b deducts the direct current biasing estimated value after gain from input signal, is compensated signal.

Wherein, the value of gain factor a is near of numerical value 1, and it can dynamically obtain according to simulation result, or comes based on experience value to determine.For different input signals, the value of gain factor a may be different, for example, in the time that direct current biasing estimated value is bigger than normal with respect to default value, can choose the gain factor that is less than 1; And direct current biasing estimated value is time less than normal with respect to default value, can choose the gain factor that is greater than 1.By using suitable gain factor, can make direct current biasing estimated value be revised, thereby the compensating signal that order finally obtains is more accurate.

S103, untwists to described compensating signal, obtains the first de-rotated signal, and described input signal is untwisted, and obtains the second de-rotated signal;

Determining after compensating signal according to direct current biasing estimated value, need to untwist to compensating signal, obtaining the first de-rotated signal; Input signal is untwisted, obtain the second de-rotated signal.From subsequent step, the first de-rotated signal or the second de-rotated signal will be eliminated output signal as direct current biasing.

About the realization of the mode of specifically untwisting, can be referring to the relevant regulations of 3GPP, the embodiment of the present invention does not elaborate to this.

S104, the signal-to-noise ratio (SNR) estimation value to described the first de-rotated signal and the signal-to-noise ratio (SNR) estimation value of described the second de-rotated signal compare, and determine direct current biasing elimination output signal according to comparative result.

In the present embodiment, the input signal of conciliating supination to separating the compensating signal of supination carries out the contrast of signal to noise ratio, and according to comparative result, a selection signal-to-noise ratio (SNR) estimation value preferably road signal is eliminated output signal as final direct current biasing.Wherein, determine direct current biasing elimination output signal according to the comparing result of the signal-to-noise ratio (SNR) estimation value of the signal-to-noise ratio (SNR) estimation value of the first de-rotated signal and the second de-rotated signal, as shown in Figure 4, can comprise:

S104a, carries out signal-to-noise ratio (SNR) estimation to the first de-rotated signal and the second de-rotated signal respectively, corresponding the first signal-to-noise ratio (SNR) estimation value SNR1 and the second signal-to-noise ratio (SNR) estimation value SNR2 of obtaining;

Wherein, the computational process of signal-to-noise ratio (SNR) estimation is as follows:

The principle of signal-to-noise ratio (SNR) estimation is to obtain noise signal by the training sequence part receiving in signal with rebuilding reception signal subtraction, thereby obtains noise power.On the other hand, calculate available signal power with channel impulse response.According to available signal power and noise power, obtain the signal to noise ratio of signal.

Noise signal:

$d\left(n\right)=r(61+\frac{L-1}{2}+n)-\underset{k=0}{\overset{L-1}{\mathrm{\Σ}}}\mathrm{ts}(L-1-k+n)\·h\left(k\right)$

n＝0...26-L

Noise power:

${\widehat{\mathrm{\σ}}}_n^2=\frac{1}{26-L+1}\·\underset{n=0}{\overset{26-L}{\mathrm{\Σ}}}{\left|d\left(n\right)\right|}^2$

Available signal power:

$E_{\mathrm{useful}}=\underset{k=0}{\overset{L-1}{\mathrm{\Σ}}}{\left|h\left(k\right)\right|}^2$

Signal-to-noise ratio (SNR) estimation value:

$\mathrm{SNR}=\frac{E_{\mathrm{useful}}}{{\widehat{\mathrm{\σ}}}_n^2}$

S104b, judges whether the first signal-to-noise ratio (SNR) estimation value SNR1 is greater than the second signal-to-noise ratio (SNR) estimation value SNR2, if so, performs step S104c, otherwise, execution step S104d;

S104c, is defined as direct current biasing by the first de-rotated signal and eliminates output signal;

S104d, is defined as direct current biasing by the second de-rotated signal and eliminates output signal.

Shown in Fig. 2, in the time that the first signal-to-noise ratio (SNR) estimation value SNR1 is less than the second signal-to-noise ratio (SNR) estimation value SNR2, show the real direct current biasing of excess input signal of error that direct current biasing is estimated, cause signal to noise ratio through the compensating signal of DC compensation with respect to the input signal variation of not passing through DC compensation, now the second de-rotated signal corresponding to input signal through DC compensation not should be defined as to direct current biasing and eliminate output signal.Otherwise, the first larger signal-to-noise ratio (SNR) estimation value de-rotated signal is defined as to direct current biasing and eliminates output signal.

Be understandable that, in S104a, the first de-rotated signal and the second de-rotated signal carried out to signal-to-noise ratio (SNR) estimation, can not have sequencing not.

In the present embodiment, first input signal is carried out to least square direct current and estimate to be compensated signal; Before output direct current biasing is eliminated result, respectively the original input signal untwisting and the signal to noise ratio of the compensating signal untwisting are carried out to valuation and comparison, then according to comparative result, a selection signal-to-noise ratio (SNR) estimation value preferably road signal is eliminated output signal as final direct current biasing, compared with existing direct current biasing cancellation scheme, the scheme that the embodiment of the present invention provides can be according to signal to noise ratio Dynamic Selection output signal, thereby adaptation plurality of application scenes, improves the receptivity under various application scenarioss.

In said method embodiment, directly utilize least square direct current method of estimation to carry out the estimation of direct current biasing to input signal, thereby be compensated signal, carry out follow-up processing.In another embodiment of the present invention, can also be before input signal be carried out to the estimation of least square direct current, by certain mode, first the direct current biasing of input signal is carried out to a rough estimate, to eliminate the impact of excessive direct current biasing on dynamic range of signals, another kind of method for eliminating dc bias shown in Figure 5, the embodiment of the present invention provides, can comprise the following steps:

S201, carries out the average direct current of time slot to input signal and eliminates;

Wherein, the computational process of the average direct current removing method of time slot is as follows:

The average direct current method of estimation of time slot adopts the method that the sampled point of the input signal in same time slot is averaging to obtain direct current biasing estimated value, and formula is as follows:

$\hat{m}=\frac{1}{N}\underset{i=0}{\overset{N-1}{\mathrm{\Σ}}}r\left(i\right)$

Wherein, r (i), i=0,1,2......N-1 is the sampled point of the input signal in same time slot, wherein each symbol can be chosen a sampled point or multiple sampled point; N is sampled point number;

for direct current biasing estimated value.

Input signal is carried out to direct current biasing compensation, and formula is as follows:

$\tilde{r}\left(i\right)=r\left(i\right)-\hat{m}$

for the signal after DC-offset correction.

S202, carries out DAGC processing to the signal carrying out after the average direct current of time slot is eliminated.

The signal of eliminating after excessive direct current biasing is carried out to DAGC (Digital Auto Gain Control, digital Auto Gain control) processing;

S203, carries out the estimation of least square direct current to the input signal of processing through DAGC, obtains direct current biasing estimated value;

S204, obtains the compensating signal of input signal according to direct current biasing estimated value;

S205, untwists to described compensating signal, obtains the first de-rotated signal, and described input signal is untwisted, and obtains the second de-rotated signal;

S206, the signal-to-noise ratio (SNR) estimation value to the first de-rotated signal and the signal-to-noise ratio (SNR) estimation value of the second de-rotated signal compare, and determine direct current biasing elimination output signal according to comparative result;

Wherein step S203-S206 is identical with the step S101-S104 of the last embodiment of the present invention, no longer describes in detail at this.

Shown in Figure 6, in the present embodiment, eliminate by advance input signal being carried out to an average direct current of time slot, direct current biasing in input signal is carried out to rough estimate to eliminate excessive direct current biasing, the signal of eliminating after excessive direct current biasing is carried out to DAGC processing, wherein, the average direct current of time slot is eliminated and can be eliminated the impact of excessive direct current biasing on dynamic range of signals, carrying out DAGC signal after treatment eliminates subsystem by input direct-current biasing and does further fine tuning and process to obtain final direct current biasing and eliminate output signal, thereby improve the accuracy of compensating signal, further improve the receptivity under various application scenarioss.Wherein, in Fig. 6, the operation principle that direct current biasing is eliminated part can, referring to shown in the dotted line frame part of Fig. 2, no longer describe in detail here.

In another embodiment of the present invention, realize direct current biasing eliminate by the mode that least square direct current is estimated and signal-to-noise ratio (SNR) estimation Dynamic Selection combines of iteration repeatedly, make the more accurate of direct current biasing elimination, receptivity is better improved.Shown in Figure 7, a kind of method for eliminating dc bias that the present embodiment provides can comprise the following steps:

S301, carries out the average direct current of time slot to input signal and eliminates;

S302, carries out DACG processing to the signal carrying out after the average direct current of time slot is eliminated;

S303, carries out the estimation of least square direct current to the input signal of processing through DAGC, obtains direct current biasing estimated value;

S304, obtains the compensating signal of input signal according to direct current biasing estimated value;

S305, untwists to described compensating signal, obtains the first de-rotated signal, and described input signal is untwisted, and obtains the second de-rotated signal;

Because the implementation of the step S301-S305 of the present embodiment and the step S201-S205 of last embodiment is basic identical, introduce no longer in detail at this.

S306, respectively to the first de-rotated signal and the second de-rotated signal signal-to-noise ratio (SNR) estimation, corresponding the first signal-to-noise ratio (SNR) estimation value SNR1 and the second signal-to-noise ratio (SNR) estimation value SNR2 of obtaining;

S307, judge whether the first signal-to-noise ratio (SNR) estimation value SNR1 is greater than the second signal-to-noise ratio (SNR) estimation value SNR2, and the current least square direct current having carried out estimates whether number of times t is less than default iterations T, if,, using compensating signal as input signal, carry out S303, carry out iterative processing, otherwise, execution step S308;

In the present embodiment, can, according to the default iterations T of actual conditions, for example, can disturb power to determine according to interchannel noise.In the time that the least square direct current having carried out estimates that number of times t is less than default iterations T, show also to need to carry out iterative processing, execution step S303; Otherwise iteration finishes, directly enter step S308, export final direct current biasing and eliminate output signal.

In addition, in the situation that meeting t and being less than T, if the first signal to noise ratio snr 1 is less than the second signal to noise ratio snr 2, show that the error of direct current estimation has exceeded the true direct current biasing of signal, cause like this signal to noise ratio of signal after DC compensation poor on the contrary, due to direct current biasing eliminate output signal should be signal to noise ratio compared with great mono-tunnel, so now should finishing iteration, the second higher signal to noise ratio de-rotated signal is directly defined as to direct current biasing erasure signal.

In sum, in the time that t equals T or SNR1 and is less than SNR2, iteration finishes, and now directly performs step S308, exports final direct current biasing and eliminates output signal; And in the time that t is less than T and SNR1 and is greater than these two conditions of SNR2 and meets, using the compensating signal not untwisting as input signal, execution step S303, carries out next iteration processing simultaneously.

S308, determines direct current biasing elimination output signal according to the signal-to-noise ratio (SNR) estimation result of this iterative processing.If the first signal-to-noise ratio (SNR) estimation value SNR1 is greater than the second signal-to-noise ratio (SNR) estimation value SNR2, the first de-rotated signal is defined as to direct current biasing and eliminates output signal; Otherwise, the second de-rotated signal is defined as to direct current biasing and eliminates output signal.

Shown in Fig. 8 and Fig. 9, in the present embodiment, by the mode that least square direct current is estimated and signal-to-noise ratio (SNR) estimation Dynamic Selection combines of iteration repeatedly, input signal is carried out to the elimination of direct current biasing.Iterative processing repeatedly makes the elimination of direct current biasing more accurate, and receptivity is better improved.

Be understandable that, in the time of iterations T=1, be equivalent to only carry out an iterative processing, this situation is equivalent to the corresponding scheme of an embodiment in the present invention.

In addition, in another embodiment of the invention, also can not carry out rough estimate to the direct current biasing in input signal, only implement above-mentioned steps S303-S308 input signal is carried out to least square direct current elimination iterative processing repeatedly.

By the description of above embodiment of the method, those skilled in the art can be well understood to the mode that the present invention can add essential general hardware platform by software and realize, can certainly pass through hardware, but in a lot of situation, the former is better execution mode.Based on such understanding, the part that technical scheme of the present invention contributes to prior art in essence in other words can embody with the form of software product, this computer software product is stored in a storage medium, comprise that some instructions (can be personal computers in order to make a computer equipment, server, or the network equipment etc.) carry out all or part of step of method described in each embodiment of the present invention.And aforesaid storage medium comprises: various media that can be program code stored such as read-only memory (ROM), random access memory (RAM), magnetic disc or CDs.

Corresponding to embodiment of the method above, the embodiment of the present invention also provides a kind of direct current biasing cancellation element, and as shown in figure 10, this device comprises:

Direct current biasing determination module 110, for input signal being carried out to the estimation of least square direct current, obtains direct current biasing estimated value;

Compensating signal determination module 120, for the direct current biasing estimated value definite according to described direct current biasing determination module 110, obtains the compensating signal of described input signal;

The module 130 of untwisting, for described compensating signal is untwisted, obtains the first de-rotated signal, and described input signal is untwisted, and obtains the second de-rotated signal;

Output signal determination module 140, compares for the signal-to-noise ratio (SNR) estimation value of the signal-to-noise ratio (SNR) estimation value to described the first de-rotated signal and described the second de-rotated signal, determines direct current biasing elimination output signal according to comparative result.

As shown in figure 11, wherein, direct current biasing determination module 110 can comprise:

Blind Modulation detecting unit 111, for described input signal is carried out to Blind Modulation detection, obtains the modulation type of described input signal;

Training sequence determining unit 112, for the modulation type definite according to Blind Modulation detecting unit, determines the training sequence symbols of least square direct current estimation function;

Direct current biasing obtains unit 113, according to the determined training sequence symbols of training sequence determining unit, input signal is carried out to the estimation of least square direct current, obtains direct current biasing estimated value.

As shown in figure 12, wherein, output signal determination module 140 can comprise:

Signal-to-noise ratio (SNR) estimation unit 141, for respectively the first de-rotated signal and the second de-rotated signal being carried out to signal-to-noise ratio (SNR) estimation, corresponding the first signal-to-noise ratio (SNR) estimation value and the second signal-to-noise ratio (SNR) estimation value of obtaining;

Output signal determining unit 142, for being greater than in the first signal-to-noise ratio (SNR) estimation value in the second signal-to-noise ratio (SNR) estimation value situation, is defined as direct current biasing by the first de-rotated signal and eliminates output signal; Otherwise, the second de-rotated signal is defined as to direct current biasing and eliminates output signal.

In another embodiment of the present invention, can also be before input signal be carried out to the estimation of least square direct current, by certain mode, first the direct current biasing of input signal is carried out to a rough estimate, to eliminate the impact of excessive direct current biasing on dynamic range of signals, accordingly, this device can also comprise the average direct current cancellation module of time slot and gain process module:

The average direct current cancellation module of time slot, eliminates for described input signal being carried out to the average direct current of time slot;

Gain process module, for the definite signal of the average direct current cancellation module of time slot is carried out to digital Auto Gain control processing, and exports result to described direct current biasing determination module 110.

In another embodiment of the present invention, more accurate for what direct current biasing was eliminated, receptivity is better improved, can be by repeatedly the least square direct current estimation of iteration and the mode that signal-to-noise ratio (SNR) estimation Dynamic Selection combines realize direct current biasing elimination, output signal determination module 140 so, can comprise:

Signal-to-noise ratio (SNR) estimation unit, for respectively the first de-rotated signal and the second de-rotated signal being carried out to signal-to-noise ratio (SNR) estimation, corresponding the first signal-to-noise ratio (SNR) estimation value and the second signal-to-noise ratio (SNR) estimation value of obtaining;

Output signal determining unit, estimate that for be greater than the second signal-to-noise ratio (SNR) estimation value and the current least square direct current that carried out in the first signal-to-noise ratio (SNR) estimation value number of times is less than default iterations, using compensating signal as input signal input direct-current, biasing determination module 110, carries out iterative processing; Otherwise

Determine direct current biasing elimination output signal according to the signal-to-noise ratio (SNR) estimation result of this iterative processing:

If the first signal-to-noise ratio (SNR) estimation value is greater than the second signal-to-noise ratio (SNR) estimation value, the first de-rotated signal is defined as to direct current biasing and eliminates output signal; Otherwise, the second de-rotated signal is defined as to direct current biasing and eliminates output signal.

In order further to improve the accuracy of compensating signal, can also revise to a certain extent direct current biasing estimated value, so, described compensating signal determination module 120, can comprise:

Gain unit, for direct current biasing estimated value definite described direct current biasing determination module and default gain factor are multiplied each other, the direct current biasing estimated value after being gained;

Compensating signal determining unit, for deduct the definite direct current biasing estimated value of described gain unit from input signal, obtains compensating signal.

For device embodiment, because it is substantially corresponding to embodiment of the method, so relevant part is referring to the part explanation of embodiment of the method.Device embodiment described above is only schematic, the wherein said unit as separating component explanation can or can not be also physically to separate, the parts that show as unit can be or can not be also physical locations, can be positioned at a place, or also can be distributed in multiple network element.Can select according to the actual needs some or all of module wherein to realize the object of the present embodiment scheme.Those of ordinary skills, in the situation that not paying creative work, are appreciated that and implement.

In several embodiment provided by the present invention, should be understood that disclosed apparatus and method, not exceeding in the application's spirit and scope, can realize in other way.Current embodiment is a kind of exemplary example, should not serve as restriction, and given particular content should in no way limit the application's object.For example, the division of described unit or subelement, is only that a kind of logic function is divided, and when actual realization, can have other dividing mode, and for example multiple unit or multiple subelement combine.In addition, multiple unit can or assembly can in conjunction with or can be integrated into another system, or some features can ignore, or do not carry out.

In addition, the schematic diagram of institute's tracing device and method and different embodiment, not exceeding in the application's scope, can with other system, module, technology or method in conjunction with or integrated.Another point, shown or discussed coupling each other or direct-coupling or communication connection can be by some interfaces, indirect coupling or the communication connection of device or unit can be electrically, machinery or other form.

The above is only the specific embodiment of the present invention; it should be pointed out that for those skilled in the art, under the premise without departing from the principles of the invention; can also make some improvements and modifications, these improvements and modifications also should be considered as protection scope of the present invention.

Claims (8)

1. a method for eliminating dc bias, is characterized in that, comprising:

Steps A, carries out the estimation of least square direct current to input signal, obtains direct current biasing estimated value, according to described direct current biasing estimated value, obtains the compensating signal of described input signal; Wherein, described to input signal carry out least square direct current estimate comprise: described input signal is carried out to Blind Modulation detection, obtains the modulation type of described input signal; According to the modulation type of described input signal, determine the training sequence symbols of least square direct current estimation function; According to determined training sequence symbols, input signal is carried out to the estimation of least square direct current; Described according to described direct current biasing estimated value, the compensating signal that obtains described input signal comprises: direct current biasing estimated value and default gain factor are multiplied each other, and the direct current biasing estimated value after being gained; From input signal, deduct the direct current biasing estimated value after gain, obtain compensating signal;

Step B, untwists to described compensating signal, obtains the first de-rotated signal, and described input signal is untwisted, and obtains the second de-rotated signal;

Step C, the signal-to-noise ratio (SNR) estimation value to described the first de-rotated signal and the signal-to-noise ratio (SNR) estimation value of described the second de-rotated signal compare, and according to comparative result, de-rotated signal large signal-to-noise ratio (SNR) estimation value is defined as to direct current biasing and eliminates output signal.

2. method according to claim 1, is characterized in that, before the estimation of least square direct current, also comprises input signal is carried out:

Input signal is carried out to the average direct current of time slot to be eliminated;

The signal carrying out after the average direct current of time slot is eliminated is carried out to digital Auto Gain control processing.

3. method according to claim 1, it is characterized in that, described signal-to-noise ratio (SNR) estimation value to described the first de-rotated signal and the signal-to-noise ratio (SNR) estimation value of described the second de-rotated signal compare, according to comparative result, de-rotated signal large signal-to-noise ratio (SNR) estimation value is defined as to direct current biasing and eliminates output signal, comprising:

Respectively the first de-rotated signal and the second de-rotated signal are carried out to signal-to-noise ratio (SNR) estimation, corresponding the first signal-to-noise ratio (SNR) estimation value and the second signal-to-noise ratio (SNR) estimation value of obtaining;

If the first signal-to-noise ratio (SNR) estimation value is greater than the second signal-to-noise ratio (SNR) estimation value, the first de-rotated signal is defined as to direct current biasing and eliminates output signal; Otherwise, the second de-rotated signal is defined as to direct current biasing and eliminates output signal.

4. a method for eliminating dc bias, is characterized in that, comprising:

Steps A, carries out the estimation of least square direct current to input signal, obtains direct current biasing estimated value, according to described direct current biasing estimated value, obtains the compensating signal of described input signal; Wherein, described to input signal carry out least square direct current estimate comprise: described input signal is carried out to Blind Modulation detection, obtains the modulation type of described input signal; According to the modulation type of described input signal, determine the training sequence symbols of least square direct current estimation function; According to determined training sequence symbols, input signal is carried out to the estimation of least square direct current; Described according to described direct current biasing estimated value, the compensating signal that obtains described input signal comprises: direct current biasing estimated value and default gain factor are multiplied each other, and the direct current biasing estimated value after being gained; From input signal, deduct the direct current biasing estimated value after gain, obtain compensating signal;

Step B, untwists to described compensating signal, obtains the first de-rotated signal, and described input signal is untwisted, and obtains the second de-rotated signal;

Step C, carries out signal-to-noise ratio (SNR) estimation to the first de-rotated signal and the second de-rotated signal respectively, corresponding the first signal-to-noise ratio (SNR) estimation value and the second signal-to-noise ratio (SNR) estimation value of obtaining;

Be greater than the second signal-to-noise ratio (SNR) estimation value and the current least square direct current that carried out estimates that number of times is less than default iterations in the first signal-to-noise ratio (SNR) estimation value, using compensating signal as input signal, return to repeated execution of steps A, from steps A, carry out iterative processing; Otherwise

Determine direct current biasing elimination output signal according to the signal-to-noise ratio (SNR) estimation result of this iterative processing:

If the first signal-to-noise ratio (SNR) estimation value is greater than the second signal-to-noise ratio (SNR) estimation value, the first de-rotated signal is defined as to direct current biasing and eliminates output signal; Otherwise, the second de-rotated signal is defined as to direct current biasing and eliminates output signal.

5. a direct current biasing cancellation element, is characterized in that, comprising:

Direct current biasing determination module, for input signal being carried out to the estimation of least square direct current, obtains direct current biasing estimated value; Described direct current biasing determination module comprises that Blind Modulation detecting unit, training sequence determining unit and direct current biasing obtain unit; Described Blind Modulation detecting unit, for described input signal is carried out to Blind Modulation detection, obtains the modulation type of described input signal; Described training sequence determining unit, for the modulation type definite according to Blind Modulation detecting unit, is determined the training sequence symbols of least square direct current estimation function; Described direct current biasing obtains unit for according to the determined training sequence symbols of training sequence determining unit, input signal being carried out to the estimation of least square direct current, obtains direct current biasing estimated value;

Compensating signal determination module, for the direct current biasing estimated value definite according to described direct current biasing determination module, obtains the compensating signal of described input signal; Described compensating signal determination module comprises gain unit and compensating signal determining unit; Described gain unit is used for direct current biasing estimated value definite described direct current biasing determination module and default gain factor to multiply each other, the direct current biasing estimated value after being gained; Described compensating signal determining unit, for deducting the definite direct current biasing estimated value of described gain unit from input signal, obtains compensating signal;

The module of untwisting, for described compensating signal is untwisted, obtains the first de-rotated signal, and described input signal is untwisted, and obtains the second de-rotated signal;

Output signal determination module, compare for the signal-to-noise ratio (SNR) estimation value to described the first de-rotated signal and the signal-to-noise ratio (SNR) estimation value of described the second de-rotated signal, according to comparative result, de-rotated signal large signal-to-noise ratio (SNR) estimation value is defined as to direct current biasing and eliminates output signal.

6. device according to claim 5, is characterized in that, also comprises: the average direct current cancellation module of time slot and gain process module:

The average direct current cancellation module of time slot, eliminates for described input signal being carried out to the average direct current of time slot;

Gain process module, for the definite signal of the average direct current cancellation module of time slot is carried out to digital Auto Gain control processing, and exports result to described direct current biasing determination module.

7. device according to claim 5, is characterized in that, output signal determination module, comprising:

Signal-to-noise ratio (SNR) estimation unit, for respectively the first de-rotated signal and the second de-rotated signal being carried out to signal-to-noise ratio (SNR) estimation, corresponding the first signal-to-noise ratio (SNR) estimation value and the second signal-to-noise ratio (SNR) estimation value of obtaining;

Output signal determining unit, for being greater than in the first signal-to-noise ratio (SNR) estimation value in the second signal-to-noise ratio (SNR) estimation value situation, is defined as direct current biasing by the first de-rotated signal and eliminates output signal; Otherwise, the second de-rotated signal is defined as to direct current biasing and eliminates output signal.

8. a direct current biasing cancellation element, is characterized in that, comprising:

Direct current biasing determination module, for input signal being carried out to the estimation of least square direct current, obtains direct current biasing estimated value; Described direct current biasing determination module comprises that Blind Modulation detecting unit, training sequence determining unit and direct current biasing obtain unit; Described Blind Modulation detecting unit, for described input signal is carried out to Blind Modulation detection, obtains the modulation type of described input signal; Described training sequence determining unit, for the modulation type definite according to Blind Modulation detecting unit, is determined the training sequence symbols of least square direct current estimation function; Described direct current biasing obtains unit for according to the determined training sequence symbols of training sequence determining unit, input signal being carried out to the estimation of least square direct current, obtains direct current biasing estimated value;

Compensating signal determination module, for the direct current biasing estimated value definite according to described direct current biasing determination module, obtains the compensating signal of described input signal; Described compensating signal determination module comprises gain unit and compensating signal determining unit; Described gain unit is used for direct current biasing estimated value definite described direct current biasing determination module and default gain factor to multiply each other, the direct current biasing estimated value after being gained; Described compensating signal determining unit, for deducting the definite direct current biasing estimated value of described gain unit from input signal, obtains compensating signal;

The module of untwisting, for described compensating signal is untwisted, obtains the first de-rotated signal, and described input signal is untwisted, and obtains the second de-rotated signal;

Signal-to-noise ratio (SNR) estimation unit, for respectively the first de-rotated signal and the second de-rotated signal being carried out to signal-to-noise ratio (SNR) estimation, corresponding the first signal-to-noise ratio (SNR) estimation value and the second signal-to-noise ratio (SNR) estimation value of obtaining;

Output signal determining unit, estimate that for be greater than the second signal-to-noise ratio (SNR) estimation value and the current least square direct current that carried out in the first signal-to-noise ratio (SNR) estimation value number of times is less than default iterations, using compensating signal as input signal input direct-current, biasing determination module, carries out iterative processing; Otherwise

Determine direct current biasing elimination output signal according to the signal-to-noise ratio (SNR) estimation result of this iterative processing:

If the first signal-to-noise ratio (SNR) estimation value is greater than the second signal-to-noise ratio (SNR) estimation value, the first de-rotated signal is defined as to direct current biasing and eliminates output signal; Otherwise, the second de-rotated signal is defined as to direct current biasing and eliminates output signal.

Images