CN1571319A - A method and apparatus for enhancing accuracy and precision of searcher - Google Patents

Abstract

The invention relates to a method and device of improving accuracy and precision of searcher in CDMA communication system, and the device mainly includes code filter group module and coherence accumulation module and adopts coherence algorithm, including the steps: a. descrambling and dispreading control channel, obtaining the pilot frequency bit symbol information of the control channel; b. correspondingly multiplying the obtained pilot frequency bit symbol by the known pilot frequency information bit; c. averaging the product with time gap, frame and multiframe as unit, and determining by the current channel environment, service quantity, service quality and other parameters; d. using the square of the average value as pilot frequency energy at the current time to input to a peak value detector and making multipaht identification. Its searching precision is high and its anti-interference ability is strong and its operation quantity change is small.

Description

A kind of method and device that improves searcher accuracy and precision

Technical field

The present invention relates to CDMA (code division multiple access) communication system of communication field, relate in particular to method and device that a kind of CDMA of being used for communication system improves searcher accuracy and precision.

Background technology

CDMA has the advantage of many uniquenesses than FDMA (frequency division multiple access) and TDMA (time division multiple access), comprehensive frequency domain, time domain and sign indicating number territory three-dimensional information are handled, and it is good to have anti-interference, and the anti-multipath decline is strong, the secret and safe height can be done attributes such as trade-off between capacity and the quality.Power system capacity is bigger 20 times than the simulation net in theory.Actual bigger 10 times than the simulation net, than the big 4-5 of GSM doubly.

Extensively adopt the Rake technology to realize Base-Band Processing in the CDMA communication system.The spread-spectrum signal that transmitter sends is subjected to the reflection and the refraction of various disorders things such as different buildings, hummock in transmission course, each path has different delays when arriving receiver, forms multipath signal.If the delay of different path signals surpasses the width of a spreading code chip, receiver can come different path differences, different delay lines is passed through in these different paths respectively, the alignment and combine, reach and turn bane into boon, being originally that interference signals becomes useful signal and combines, the basic principle of RAKE receiver that Here it is.

Different path differentiation are finished by searcher, and searcher is the key component in the Rake receiver.Its performance has determined the Base-Band Processing performance, directly influences link performance, service quality and power system capacity.The index of weighing searcher has searching algorithm complexity, search precision, search accuracy etc.Present searcher all is to use incoherent searching algorithm, and the search accuracy is not high, but computation complexity is not low.

Summary of the invention

The object of the present invention is to provide a kind of search accuracy of searcher and method of precision of improving.

Another purpose of the present invention is to provide a kind of searcher device that is used to realize said method.

The present invention can improve the search accuracy and the precision of searcher effectively under the prerequisite that does not improve algorithm computation complexity and hardware implementation complexity.

The present invention is achieved in that

A kind of search accuracy of searcher and method of precision of improving comprises the steps:

Step 1: control channel is carried out descrambling and de-spreading handle, obtain the pilot frequency bit symbol information of control channel;

Step 2: the pilot frequency bit symbol and the known pilot information bit correspondence that obtain are multiplied each other;

Step 3: product is asked on average, and asking can be that unit carries out with time slot, frame, multiframe on average, is determined by parameters such as current channel circumstance, traffic carrying capacity, service quality;

Step 4:, be input to peak detector and can carry out multipath identification with square the pilot energy of mean value as current time.

Wherein, the baseband signal X that receives of the base station of Rake receiver _I ^(t)And X _Q ^(t), at first enter a yard bank of filters, the scrambler S of sign indicating number filter _CI ^(t)And S _CQ ^(t)It is required to be configured to the user.The sign indicating number filter is finished the input baseband signal is finished the descrambling and de-spreading processing, and the spreading code of pilot channel is fixed.

Wherein the calculating process of descrambling and de-spreading is: at first, the sign indicating number bank of filters is according to scrambler real part S _CI ^(t)With imaginary part S _CQ ^(t)To input baseband signal X _I ^(t)And X _Q ^(t)Finish the descrambling and de-spreading computing.

A wireless multipath channel of WCDMA, the user has the L paths, and then its channel response is:

$h\left(t\right)=\underset{l=1}{\overset{L}{\mathrm{\Σ}}}{\mathrm{\α}}_l\mathrm{\δ}(t-{\mathrm{\τ}}_l)e^{\mathrm{j\φl}}$

In the formula, τ _l, α _l, φ _lBe respectively time delay, decay factor and the phase place of user's l paths.Searcher is exactly to τ in the path delay of time _lEstimate.

The signal that antenna is received obtains later on through LPF:

$X_I\left(t\right)=\frac{\mathrm{\α}}{2}[I(t-\mathrm{\τ})\cos \mathrm{\φ}+Q(t-\mathrm{\τ})\sin \mathrm{\φ}]$

$X_Q\left(t\right)=\frac{\mathrm{\α}}{2}[-I(t-\mathrm{\τ})\sin \mathrm{\φ}+Q(t-\mathrm{\τ})\cos \mathrm{\φ}]$

The descrambling and de-spreading algorithm is as follows:

$Z1=\underset{(n-1)T_b}{\overset{{\mathrm{nT}}_b}{\∫}}{X}_I\left(t\right)\·S_{\mathrm{CQ}}\left(t^{\′}\right)\mathrm{dt}=-\frac{\mathrm{\α}{\mathrm{\β}}_c}{2}{p}_n\cos \mathrm{\φ}$

$Z2=\underset{(n-1)T_b}{\overset{{\mathrm{nT}}_b}{\∫}}{X}_Q\left(t\right)\·S_{\mathrm{CI}}\left(t^{\′}\right)\mathrm{dt}=-\frac{\mathrm{\α}{\mathrm{\β}}_c}{2}{p}_n\cos \mathrm{\φ}$

More than utilized the WCDMA control channel to adopt the characteristics of No. 0 ovsf code.

Wherein: T _bBe symbol period, P _nBe the n position of pilot tone, β _cBe power gain factors.

Z12＝Z2-Z1＝αβ _cp _ncosφ

In like manner we can obtain:

$Z3=\underset{(n-1)T_b}{\overset{{\mathrm{nT}}_b}{\∫}}{X}_I\left(t\right)\·S_{\mathrm{CI}}\left(t^{\′}\right)\mathrm{dt}=-\frac{\mathrm{\α}{\mathrm{\β}}_c}{2}{p}_n\sin \mathrm{\φ}$

$Z4=\underset{(n-1)T_b}{\overset{{\mathrm{nT}}_b}{\∫}}{X}_Q\left(t\right)\·S_{\mathrm{CQ}}\left(t^{\′}\right)\mathrm{dt}=-\frac{\mathrm{\α}{\mathrm{\β}}_c}{2}{p}_n\sin \mathrm{\φ}$

Z34＝Z3+Z4＝αβ _cp _nsinφ

Wherein, descrambling and de-spreading obtains two of Z34, Z12, these two is-symbol level data flow.Multiply each other respectively in multiplier with corresponding pilot bit, in integrator, distinguish integration according to pilot bit, in squarer, ask square, in the adder summation, obtain the current search value, wherein at last, integration time window length is relevant with the searcher system configuration that is adopted, the time window length of search accuracy and summation is directly proportional, and time window length is oversize can't in time to reflect the variation of multipath time, simultaneously system's operational capability is required to improve.

The algorithm of relevant accumulative total part is:

$E={(\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{p}_i\·Z12\left(i\right))}^2+{(\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{p}_i\·Z34\left(i\right))}^2$

Wherein, E is the pilot energy under the current time phase place, and N is an integral constant.When N equaled the pilot tone figure place of a time slot, the expression searcher was that unit is accumulated with the time slot.

Wherein, because the multiple access of other channels disturbs and interference of noise may cause peak atenuation even pseudo-peak occur, therefore can adopt the increase integral constant to obtain time gain.

A kind of device that improves searcher accuracy and precision comprises:

Sign indicating number bank of filters module is made up of four sign indicating number filters, the input baseband signal is finished descrambling and de-spreading handle;

The coherent accumulation module comprises two parallel device chains, an adder, is used to finish the coherent superposition of pilot bit;

Described device chain is made of multiplier, integrator, squarer successively, multiplier export to integrator, integrator export to squarer, two parallel device chains are all being exported to adder after the squarer computing, carry out add operation, obtain the current search value;

Described adder export to peak detector.

Described searcher device is the part of Rake receiver.

The present invention adopts coherent approach to realize Energy Estimation, has a higher processing signal to noise ratio than incoherent.Algorithm operation quantity is constant.Adopt the searcher structure of this algorithm identical with routine search device structure, just increased the pilot frequency configuration to searcher, complexity and cost change little, are algorithms very with practical value.The present invention is applicable to multipath is accurately estimated, followed the tracks of.When the input signal-to-noise ratio of received signal hangs down, can embody the superiority of this method.Therefore, can be used in combination with incoherent searcher.

Description of drawings

Fig. 1 provides the Base-Band Processing block diagram of cdma system;

Fig. 2 provides the structured flowchart of searcher of the present invention;

Fig. 3 a and Fig. 3 b provide the present invention and the performance simulation result of routine search device under the Matlab environment.

Embodiment

Below in conjunction with embodiment the present invention is described in further detail:

See also Fig. 1.Radiofrequency signal is received by antenna element (11), through radio-frequency channel (12) vanishing intermediate frequency baseband signal, becomes digital signal through A/D sampling (13) again and enters CDMA baseband processor-Rake receiver (14).In Rake receiver (14), at first obtain multipath energy by searcher (141), multipath distributor (142) obtains the multipath phase place according to allocation strategy from the output of searcher (141), the multipath phase place is assigned to the footpath tracking and the Base-Band Processing in single footpath is finished in descrambling despreader (143), (144), (145), the symbolic information in each footpath of output enters and finishes the multipath merging in channel estimating and the multipath combiner (146), the output baseband signal.

Adopt the theoretical derivation of WCDMA Base-Band Processing of the present invention as follows:

At first, the sign indicating number bank of filters is according to scrambler real part S _CI ^(t)With imaginary part S _CQ ^(t)To input baseband signal X _I ^(t)And X _Q ^(t)Finish the descrambling and de-spreading computing.

A wireless multipath channel of WCDMA, the user has the L paths, and then its channel response is:

$h\left(t\right)=\underset{l=1}{\overset{L}{\mathrm{\Σ}}}{\mathrm{\α}}_l\mathrm{\δ}(t-{\mathrm{\τ}}_l)e^{\mathrm{j\φl}}$

In the formula, τ _l, α _l, φ _lBe respectively time delay, decay factor and the phase place of user's l paths.Searcher is exactly to τ in the path delay of time _lEstimate.

The signal that antenna is received obtains later on through LPF:

$X_I\left(t\right)=\frac{\mathrm{\α}}{2}[I(t-\mathrm{\τ})\cos \mathrm{\φ}+Q(t-\mathrm{\τ})\sin \mathrm{\φ}]$

$X_Q\left(t\right)=\frac{\mathrm{\α}}{2}[-I(t-\mathrm{\τ})\sin \mathrm{\φ}+Q(t-\mathrm{\τ})\cos \mathrm{\φ}]$

The descrambling and de-spreading algorithm is as follows:

$Z1=\underset{(n-1)T_b}{\overset{{\mathrm{nT}}_b}{\∫}}{X}_I\left(t\right)\·S_{\mathrm{CQ}}\left(t^{\′}\right)\mathrm{dt}=-\frac{\mathrm{\α}{\mathrm{\β}}_c}{2}{p}_n\cos \mathrm{\φ}$

$Z2=\underset{(n-1)T_b}{\overset{{\mathrm{nT}}_b}{\∫}}{X}_Q\left(t\right)\·S_{\mathrm{CI}}\left(t^{\′}\right)\mathrm{dt}=-\frac{\mathrm{\α}{\mathrm{\β}}_c}{2}{p}_n\cos \mathrm{\φ}$

More than utilized the WCDMA control channel to adopt the characteristics of No. 0 ovsf code.

Wherein: T _bBe symbol period, p _nBe the n position of pilot tone, β _cBe power gain factors.

Z12＝Z2-Z1＝αβ _cp _ncosφ

In like manner we can obtain:

$Z3=\underset{(n-1)T_b}{\overset{{\mathrm{nT}}_b}{\∫}}{X}_I\left(t\right)\·S_{\mathrm{CI}}\left(t^{\′}\right)\mathrm{dt}=-\frac{\mathrm{\α}{\mathrm{\β}}_c}{2}{p}_n\sin \mathrm{\φ}$

$Z4=\underset{(n-1)T_b}{\overset{{\mathrm{nT}}_b}{\∫}}{X}_Q\left(t\right)\·S_{\mathrm{CQ}}\left(t^{\′}\right)\mathrm{dt}=-\frac{\mathrm{\α}{\mathrm{\β}}_c}{2}{p}_n\sin \mathrm{\φ}$

Z34＝Z3+Z4＝αβ _cp _nsinφ

Secondly, to carry out coherent accumulation:

$E={(\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{p}_i\·Z12\left(i\right))}^2+{(\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{p}_i\·Z34\left(i\right))}^2$

Wherein, E is the pilot energy under the current time phase place, and N is an integral constant.When N equaled the pilot tone figure place of a time slot, the expression searcher was that unit is accumulated with the time slot.Because the multiple access of other channels disturbs and interference of noise may cause peak atenuation even pseudo-peak occur, therefore can adopt the increase integral constant to obtain time gain.

Adopt WCDMA searcher structure of the present invention as shown in Figure 2.Specifically form: sign indicating number bank of filters module (21), coherent accumulation module (22) by two modules.

Sign indicating number bank of filters module is finished descrambling and de-spreading, and the coherent accumulation module is finished the coherent superposition of pilot bit.

The baseband signal X that the base station receives _I ^(t)And X _Q ^(t), at first enter a yard bank of filters (21), the scrambler S of sign indicating number filter (211,212,213,214) _CI ^(t)And S _CQ ^(t)It is required to be configured to the user.Sign indicating number filter (211,212,213,214) is finished the input baseband signal is finished the descrambling and de-spreading processing.The spreading code of pilot channel is fixed.

Secondly descrambling and de-spreading obtains two of Z34, Z12, these two is-symbol level data flow.Multiply each other respectively in multiplier (2211,2212) with corresponding pilot bit, in integrator (2221,2222),, in (2231,2232), ask square, in adder (224) summation, obtain the current search value at last according to pilot bit integration respectively.Integration time window length is relevant with the searcher system configuration that is adopted.The time window length of search accuracy and summation is directly proportional.Time window length is oversize can't in time to reflect the variation of multipath time, simultaneously system's operational capability is required to improve.

Fig. 3 a and Fig. 3 b have shown and have adopted searcher of the present invention and the routine search device performance simulation result under the Matlab environment.Wherein Fig. 3 a is incoherent searcher output, and Fig. 3 b is the searcher output of this method.

Simulated conditions is as follows:

The data source that meets 3Gpp, 20 users, 4 multipaths of every user.

Desirable multidiameter delay value is 3,28,48,54, the chip of unit.

Desirable multipath fading value is 1.0000,0.5828,0.4154,0.6833.

The integral constant of two kinds of searchers is a frame.

Simulation result is as follows:

1) Fig. 3 a time delay estimated value is 3.0000,54.1250,64.6250,82.7500.Fig. 3 b time delay estimated value is 2.8750,28.0000,54.0000,149.6250.As can be seen from the results, Fig. 3 a has two pseudo-footpaths, and Fig. 3 b has a pseudo-footpath.Search precision is improved.

2) see from Fig. 3 a that the energy in two maximum footpaths much at one.By the multipath fading that provides in the simulated conditions as can be seen, the energy in maximum diameter and time big footpath should be (1/0.6833) than ideal ²=2.14.Promptly the energy approximation in time big footpath is 1/2 of a maximum diameter energy.Therefore, can conclude that time big footpath among Fig. 3 a has been subjected to serious interference.

3) see from Fig. 3 b that the proportionate relationship between maximum diameter and inferior big footpath, the 3rd footpath is approximate and simulated conditions is identical.

Claims (6)

1, a kind of method that improves searcher accuracy and precision comprises the steps:

Step 1: control channel is carried out descrambling and de-spreading handle, obtain the pilot frequency bit symbol information of control channel;

Step 2: the pilot frequency bit symbol and the known pilot information bit correspondence that obtain are multiplied each other;

Step 3: product is asked on average, and asking is that unit carries out with time slot, frame, multiframe on average, is determined by parameters such as current channel circumstance, traffic carrying capacity, service quality;

Step 4:, be input to peak detector and carry out multipath identification with square the pilot energy of mean value as current time.

2, improve the method for searcher accuracy and precision according to claim 1, it is characterized in that the processing of descrambling and de-spreading described in the step 1 further comprises:

Baseband signal X _I(t) and X _Q(t), enter a yard bank of filters, with the scrambler S of sign indicating number filter _CI(t) and S _CQ(t) it is required to be configured to the user, and the spreading code of pilot channel is fixed;

The WCDMA control channel adopts No. 0 ovsf code;

Carry out descrambling and de-spreading:

$Z1=\underset{(n-1)T_b}{\overset{{\mathrm{nT}}_b}{\∫}}{X}_I\left(t\right)\·S_{\mathrm{CQ}}\left(t^{\′}\right)\mathrm{dt}=-\frac{{\mathrm{\α\β}}_c}{2}{p}_n\cos \mathrm{\φ}$

$Z2=\underset{(n-1)T_b}{\overset{{\mathrm{nT}}_b}{\∫}}{X}_Q\left(t\right)\·S_{\mathrm{CI}}\left(t^{\′}\right)\mathrm{dt}=\frac{{\mathrm{\α\β}}_c}{2}{p}_n\cos \mathrm{\φ}$

Wherein: T _bBe symbol period, p _nBe the n position of pilot tone, β _cBe power gain factors;

Z12＝Z2-Z1＝αβ _cp _ncosφ

$Z3=\underset{(n-1)T_b}{\overset{n{T}_b}{\∫}}{X}_I\left(t\right)\·S_{\mathrm{CI}}\left(t^{\′}\right)\mathrm{dt}=\frac{{\mathrm{\α\β}}_c}{2}{p}_n\sin \mathrm{\φ}$

$Z4=\underset{(n-1)T_b}{\overset{{\mathrm{nT}}_b}{\∫}}{X}_Q\left(t\right)\·S_{\mathrm{CQ}}\left(t^{\′}\right)\mathrm{dt}=\frac{{\mathrm{\α\β}}_c}{2}{p}_n\sin \mathrm{\φ}$

Z34＝Z3+Z4＝αβ _cp _nsinφ。

3, improve the method for searcher accuracy and precision as claimed in claim 1 or 2, it is characterized in that:

The described processing through descrambling and de-spreading obtains two sign level data streams of Z34, Z12, multiplies each other respectively in multiplier with corresponding pilot bit, according to pilot bit integration respectively, asks in squarer square in integrator, sues for peace in adder, obtains the current search value:

$E={(\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{p}_i\·Z12\left(i\right))}^2+{(\frac{1}{N}\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{p}_i\·Z34\left(i\right))}^2$

Wherein, E is the pilot energy under the current time phase place, and N is an integral constant;

When N equaled the pilot tone figure place of a time slot, the expression searcher was that unit is accumulated with the time slot;

Integration time window length is relevant with the searcher system configuration that is adopted, and the time window length of search accuracy and summation is directly proportional.

4, as the method for raising searcher accuracy as described in the claim 2 and precision, it is characterized in that: increase integral constant and obtain time gain by adopting.

5, a kind of device that improves searcher accuracy and precision, it is characterized in that: described device comprises:

Sign indicating number bank of filters module is made up of four sign indicating number filters, the input baseband signal is finished descrambling and de-spreading handle;

The coherent accumulation module comprises two parallel device chains, an adder, is used to finish the coherent superposition of pilot bit;

Described device chain is made of multiplier, integrator, squarer successively, multiplier export to integrator, integrator export to squarer, two parallel device chains are all being exported to adder after the squarer computing, carry out add operation, obtain the current search value;

Described adder export to peak detector.

6, as the device of raising searcher accuracy as described in the claim 5 and precision, it is characterized in that:

Baseband signal X _I(t) and X _Q(t) enter a yard bank of filters, the scrambler S of sign indicating number filter _CI(t) and S _CQ(t) it is required to be configured to the user;

The spreading code of pilot channel is fixed.

Images