CN201674495U - Processing device for calculating automatic frequency control parameter by using joint detection - Google Patents

Abstract

The utility model relates to a processing device for calculating automatic frequency control parameters by using joint detection, which comprises a hardware module and a software module and is characterized in that the hardware module comprises a hard decision device, the output end of the hard decision device is connected with the input end of a first multiple correlation device, the output end of the first multiple correlation device is connected with the input end of a second multiple correlation device, the output end of the second multiple correlation device is connected with the input end of a self-adapting bit width adjuster, and the first multiple correlation device and the second multiple correlation device are simultaneously connected with a time schedule controller. Thereby, through the multiple correlation devices, the utility model has the advantages that the joint detection of common-frequency adjacent cells is realized, the received data causing common-frequency cell interference and other user interference in the cell is eliminated, and the estimation of the offset of local frequency and network frequency is more reasonable and accurate.

Description

Processing unit with the joint-detection calculating parameters for automatic frequency control

Technical field

The utility model relates to a kind of processing unit, relates in particular to a kind of processing unit with the joint-detection calculating parameters for automatic frequency control.

Background technology

In cordless communication terminal system, automatic frequency control AFC is an important function very.By automatic control adjustment to local oscillator frequencies so that itself and network side corresponding frequencies are consistent, with satisfy synchronously, the requirement of demodulation etc., the performance that improves terminal system is had crucial effects.

In the TD-SCDMA terminal, it is relevant that the reception result that generally adopts intermediate code Midambl (MA) and descending synchronous code SYNC-DL and corresponding training sequence carry out, and estimates frequency deviation, and then obtain the AFC parameter.Yet when having inter-frequency interference cell, if only co-channel interference being used as common noise handles, owing to there is fixing interference on SYNC-DL or the MA training sequence, in the phase place rotation value that will cause estimating by existing method, have a stationary phase deviation fixing, that can not eliminate, this will cause also existing in the final frequency offset estimating value frequency departure to a certain degree.Conventional device can't overcome the interference that these deviations cause, and can't make the TD-SCDMA terminal applies to more field.

Summary of the invention

The purpose of this utility model is to overcome the above problem that prior art exists, and a kind of processing unit with the joint-detection calculating parameters for automatic frequency control is provided.

For realizing the processing unit of the purpose of this utility model with the joint-detection calculating parameters for automatic frequency control, include hardware module and software module, wherein: described hardware module comprises the hard decision device, the output of hard decision device is connected with the input of first complex correlator, the output of first complex correlator is connected with the input of second complex correlator, the output of second complex correlator connects the input of self adaptation bit wide adjuster, self adaptation bit wide regulator output is connected into software module, is connected with time schedule controller simultaneously on first complex correlator and second complex correlator.

Further, the processing unit of above-mentioned usefulness joint-detection calculating parameters for automatic frequency control, wherein, described software module comprises the phase calculation assembly, self adaptation bit wide regulator output is connected into the input of computation module, and the output of computation module connects the input of filtering and automatic frequency control assembly.

Adopt technical solutions of the utility model,, adopt the method for decision-feedback to estimate frequency deviation by the joint detection results of data division in the TD-SCDMA structure of time slot.Rely on a plurality of complex correlator, the joint-detection that realization is frequently faced the sub-district has together been eliminated the reception data that co-frequency cell interference and other users of this sub-district disturb, and is reasonable more and accurate to the skew estimation of local frequency and network.

The purpose of this utility model, advantage and characteristics will illustrate by the non-limitative illustration of following preferential embodiment and explain that these embodiment only provide as an example with reference to accompanying drawing.

Description of drawings

Fig. 1 is the organigram of this processing unit.

The implication of each Reference numeral is as follows among the figure:

1 hard decision device, 2 first complex correlator

3 second complex correlator, 4 time schedule controllers

5 self adaptation bit wide adjusters, 6 phase calculation assemblies

7 filtering and automatic frequency control

Assembly

Embodiment

The processing unit of usefulness joint-detection calculating parameters for automatic frequency control as shown in Figure 1, it comprises hardware module and software module, unusual part is: described hardware module comprises hard decision device 1, the output of hard decision device 1 is connected with the input of first complex correlator 2, the output of first complex correlator 2 is connected with the input of second complex correlator 3, the output of second complex correlator 3 connects the input of self adaptation bit wide adjuster 5, the output of self adaptation bit wide adjuster 5 is connected into software module, is connected with time schedule controller 4 simultaneously on first complex correlator 2 and second complex correlator 3.Software module comprises phase calculation assembly 6, and the output of self adaptation bit wide adjuster 5 is connected into the input of computation module, and the output of computation module connects filtering and automatic frequency control assembly 7.Software module is read the real part and the imaginary part of correlated results from hardware module in real work, calculate phase place and phase place is carried out filtering, calculates corresponding parameters for automatic frequency control.

Find that in actual use in order to save the hardware module resource, by staggering, first complex correlator 2 can be realized by identical hardware module circuit multiplexer with second complex correlator 3 on sequential.

Work disposal flow process of the present utility model is as follows: at first, choose S the character position among the DATA1 and the data of S character position among the DATA2 and handle.Two parts data midfeather N1 DATA1 symbol and N2 DATA2 symbol.The data of DATA1 and DATA2 have all been passed through joint-detection, the interference that can effectively eliminate all the other code channels and face the sub-district with frequency.

Suppose S character position among the DATA1, on m the code channel in certain Common transport channel, the data (joint-detection demodulation output) of a s symbol are:

I _m，s ¹+j×Q _m，s ¹ (m＝1，2，…，M，s＝1，2，…，S)

S among a DATA2 character position, on m the code channel in certain Common transport channel, the data (joint-detection demodulation output) of a s symbol are:

I _m，s ²+j×Q _m，s ² (m＝1，2，…，M，s＝1，2，…，S)

Then, joint detection results is carried out hard decision, promptly obtain sgn (I respectively _{M, s} ¹), sgn (Q _{M, s} ¹), sgn (I _{M, s} ²) and sgn (Q _{M, s} ²); Wherein, function sgn is for getting symbol manipulation, that is:

$\mathrm{sgn}\left(x\right)=\left\{\begin{array}{cc}1,& x\&GreaterEqual;0\\ -1,& x<0\end{array}\right.$

According to declaring the result firmly, remove the data-modulated information in this two segment data respectively, and do relevant treatment.For example, the last period data correlated results can be calculated as follows:

$\mathrm{Corr}1=\underset{s=1}{\overset{S}{\mathrm{\&Sigma;}}}\{(I_{m,s}^1+j\&times;Q_{m,s}^1)\&times;\mathrm{conj}(\mathrm{sgn}\left(I_{m,s}^1\right)+j\&times;\mathrm{sgn}\left(Q_{m,s}^1\right))\}$

$=\underset{s=1}{\overset{S}{\mathrm{\&Sigma;}}}\left[\right|I_{m,s}^1|+|Q_{m,s}^1\left|\right]+j\&times;[\mathrm{sgn}\left(I_{m,s}^1\right)\&times;\mathrm{sgn}\left(Q_{m,s}^1\right)\&times;\left(\right|Q_{m,s}^1|-|I_{m,s}^1\left|\right)]$

Similarly, can obtain the correlated results of back one piece of data by following formula:

$\mathrm{<figure-callout\; id="2"\; label="Corr"\; filenames="H200920257507XE00011.png"\; state="\{\{state\}\}">Corr</figure-callout>}2=\underset{s=1}{\overset{S}{\mathrm{\&Sigma;}}}\{(I_{m,\mathrm{<figure-callout\; id="2"\; label="s"\; filenames="H200920257507XE00011.png"\; state="\{\{state\}\}">s</figure-callout>}}^2+j\&times;Q_{m,s}^2)\&times;\mathrm{conj}(\mathrm{sgn}\left(I_{m,s}^2\right)+j\&times;\mathrm{sgn}\left(Q_{m,s}^2\right))\}$

$=\underset{s=1}{\overset{S}{\mathrm{\&Sigma;}}}\left[\right|I_{m,\mathrm{<figure-callout\; id="2"\; label="s"\; filenames="H200920257507XE00011.png"\; state="\{\{state\}\}">s</figure-callout>}}^2|+|Q_{m,\mathrm{<figure-callout\; id="2"\; label="s"\; filenames="H200920257507XE00011.png"\; state="\{\{state\}\}">s</figure-callout>}}^2\left|\right]+j\&times;[\mathrm{sgn}\left(I_{m,s}^2\right)\&times;\mathrm{sgn}\left(Q_{m,s}^2\right)\&times;\left(\right|Q_{m,\mathrm{<figure-callout\; id="2"\; label="s"\; filenames="H200920257507XE00011.png"\; state="\{\{state\}\}">s</figure-callout>}}^2|-|I_{m,\mathrm{<figure-callout\; id="2"\; label="s"\; filenames="H200920257507XE00011.png"\; state="\{\{state\}\}">s</figure-callout>}}^2\left|\right)]$

Subsequently, calculating Corr1 is relevant with the conjugation of Corr2

Corr3＝conj(Corr1)×Corr2

Then the phase information of Corr3 is the phase place rotation of this section period.

Then, the self adaptation bit wide is adjusted, and adopts dsp software to read Corr3.During this period, consider that the width of bus is generally 16bit, it need limit real part, the imaginary part bit wide of Corr3.For guaranteeing the precision of subsequent calculations, hardware module output Corr3 result's number of significant digit should be wide as far as possible, and therefore output can not be taked the mode of fixed bit cut position.Thus, the present invention has provided the method that a kind of self adaptation bit wide is adjusted.Specifically, be real part imaginary part 26bit signed number with Corr3 result of calculation, the 16bit bus is output as example, then exports Output real part and imaginary part surely and is the 16bit signed number.

The Output sign bit is following to be got:

sgn(Re{output})＝sgn(Re{Corr3})

sgn(Im{output})＝sgn(Im{Corr3})。

The Output range value is following to be got: first figure place different with sign bit is w1 among the note Re{Corr3} bit from high to low, and first figure place different with sign bit is w2 among the note Im{Corr3} bit from high to low.Get w=max (w1, w2).If w 〉=15, [w:w-14] that get Corr3 real part imaginary part is as the part beyond the Output divided-by symbol position; If [14:0] then got as the part beyond the Output divided-by symbol position in w＜15.

At last, after software module reads Output, try to achieve the phase place deflection of every chip (chip) time correspondence

θ＝angle(Output)/((N1+N2+S)×SF+144)

Wherein SF represents spreading factor, carries out filtering by software module afterwards and asks automatic frequency control (AFC) Control Parameter to get final product.

Again further, (UE) is under the different mode when terminal, can utilize joint-detection (JD) result of different physical channels to carry out frequency offset estimating.Particularly:

(1) when terminal is in Cell searching (Cell Search) state, can utilizes the JD of broadcast channel (BCH) demodulation to export and carry out frequency offset estimating;

(2) be in the free time during (IDLE) pattern when terminal, still can utilize the JD of broadcast channel demodulation to export and carry out frequency offset estimating; Prerequisite is all to be distributed in time slot 0 (TS0) such as Page Indication Channel (PICH), paging channel common signal channels such as (PCH) to go up (note: although need not do decoding this moment to broadcast channel, 1/2 code channel on the TS0 is certain to participate in the JD task);

(3) when terminal is in connection (Connected) pattern, utilizes the JD of the minimum time slot of the Common transport channel (CCTrCH) of downward special physical channel (DPCH), minimum code channel to export and do frequency offset estimating; Why select minimum time slot, minimum code channel for use, unless be because be in discontinuous reception (DTX) state, otherwise certainly exist data.

Adopt technical solutions of the utility model,, adopt the method for decision-feedback to estimate frequency deviation by the joint detection results of data division in the TD-SCDMA structure of time slot.Rely on a plurality of complex correlator, the joint-detection that realization is frequently faced the sub-district has together been eliminated the reception data that co-frequency cell interference and other users of this sub-district disturb, and is reasonable more and accurate to the skew estimation of local frequency and network.

Certainly, more than only be concrete exemplary applications of the present utility model, protection range of the present utility model is not constituted any limitation.In addition to the implementation, the utility model can also have other execution mode.All employings are equal to the technical scheme of replacement or equivalent transformation formation, all drop within the utility model scope required for protection.

Claims (1)

1. use the processing unit of joint-detection calculating parameters for automatic frequency control, include hardware module, it is characterized in that: described hardware module comprises the hard decision device, the output of hard decision device is connected with the input of first complex correlator, the output of first complex correlator is connected with the input of second complex correlator, the output of second complex correlator connects the input of self adaptation bit wide adjuster, is connected with time schedule controller simultaneously on first complex correlator and second complex correlator.

Images