CN100578954C - Method for receiving multi-path signal, device for calculating each path weighted value and PAKE receiver - Google Patents

Abstract

The invention is concerned with a method about counting multi-path weighted values for RAKE receiver or universal RAKE (G-RAKE) receiver, it is: counts multi-address disturbing function RMUI and disturbing function Rn of background thermal noise to get yawp correlation matrix Ru. According to the multi-path fading function h and yawp correlation matrix Ru, count multi-path weighted values Omega. The invention is also concerned with a device to count multi-path weighted values and RAKE receiver to count multi-path weighted values using the said method. This invention reduces the work amount of multi-path weighted values without the influence of G-RAKE receiver's capability and the amount of multi-path weighted values has nothing to do with enlarging frequency gene. The user having different enlarging frequency gene can use one RAKE receiver without adding additional RAKE receiver.

Description

Device and RAKE receiver of the method for receiving multipath signals, each footpath weighted value of calculating

Technical field

The present invention relates to the wireless communication signals reception technique, particularly device and RAKE receiver of a kind of method of receiving multipath signals, each footpath weighted value of calculating.

Background technology

In code division multiplexing (CDMA) spread spectrum system, channel width is far longer than the flat fading bandwidth of channel.Being different from traditional modulation technique need eliminate intersymbol interference between adjacent-symbol with equalization algorithm, and the CDMA spreading code just requires it that good autocorrelation performance is arranged when selecting signal.Like this, the time delay expansion that occurs in wireless channel just can be counted as just being transmitted the transmission once more of signal.If when the mutual time delay of multipath signal surpassed a chip period, in fact multipath signal can be counted as incoherent mutually.

Owing to contain utilizable information in multipath signal, receiver can adopt the receiver of sort signal processing mode to be called as RAKE receiver by merging the signal to noise ratio that multipath signal improves received signal.Figure 1 shows that the structural representation of RAKE receiver, specifically comprise as lower module:

Multipath searching distribution module 110, be used for received signal, and finish that to received signal multidiameter delay is caught and multidiameter delay distributes, export the delayed data that obtains each bar multipath after handling to descrambling and de-spreading module 120, channel estimation module 130 and each footpath weighted value computing module 140 respectively;

Descrambling and de-spreading module 120, be used for finishing the descrambling of chip-level received signal, de-spreading operation according to the multipath delay information of input and scrambler, the spreading code information of received signal, the symbol level received signal y after output descrambling, the despreading to each directly merge module 150;

Channel estimation module 130 is estimated the channel fading situation of each time-delay according to the multipath delay information of input, and the channel fading information of output and each multipath delay correspondence is to each footpath weighted value computing module 140;

Each footpath weighted value computing module 140, multipath delay information, each footpath channel fading information and other external parameters according to input comprise the Gaussian noise power N of system ₀With the chip-level ENERGY E that receives the user _iCalculate the weighted factor ω in each footpath, and export the weighted factor ω that is calculated to each footpath and merge module 150;

Each footpath merges module 150, according to each the footpath weighted factor ω that imports, to each the footpath data y formula Y=ω after descrambling, the despreading ^H* y merges processing, merges processing back gained signal Y and exports demodulation coding module 160 to; Wherein [] ^HTransposition is gripped in expression altogether;

Demodulation coding module 160 is separated the mediation decoded operation accordingly to the signal Y that is received, and the signal after the output demodulation coding.

In order to satisfy the ever-increasing rate requirement of user, existing wireless communication system has been introduced the more and more higher various technology of peak rate, insert (High Speed Downlink PacketAccess as high speed downlink packet, HSDPA) and high speed uplink packet insert (High Speed Uplink Packet Access, HSUPA).The introducing of these new technologies has brought bigger challenge to traditional RAKE reception technique.On the one hand, traditional RAKE receiver technology can't satisfy the throughput demand of high speed business; On the other hand because the self-interference characteristic of cdma system, the interference that requires the high-rate service user that other users are produced should be as far as possible little, that is to say and reduce its needed signal to noise ratio (snr) as much as possible under the prerequisite that satisfies the high-rate service performance index.Therefore, very be necessary to introduce the advanced reception technique that performance is better than traditional RAKE receiver.

The receiver that existing performance is better than traditional RAKE receiver comprises following a few class:

The enhancement mode RAKE receiver, this receiver just improves on the basis of traditional RAKE reception technique, and its performance improves very limited;

Balanced class advanced receivers, though the multiple access that this receiver can be eliminated between caused intersymbol interference of multipath (ISI) and the different user theoretically fully disturbs (MUI), thereby obtain good performance, but, impact bigger to existing receiver structure because there are very big-difference in its structure and traditional RAKE receiver;

Generalized RAKE (Generalized Rake, G-RAKE) receiver, this receiver has fundamentally considered to limit the leading factor of traditional RAKE receiver performance, not only can reach and the identical performance of balanced class receiver, and have the implementation structure similar with traditional RAKE receiver.

The structural similarity of the structure of G-RAKE receiver and traditional RAKE receiver shown in Figure 1, just wherein the concrete calculation processes of each footpath weighted value computing module 140 and the calculation processes of traditional this module of RAKE receiver there are differences.

Each footpath weighted value computing module of G-RAKE receiver needs to calculate

$\mathrm{\ω}=R_u^{-1}h,---\left(1\right)$

Wherein,

$h=\sqrt{E_0}\underset{l=0}{\overset{L-1}{\mathrm{\Σ}}}{g}_l{R}_p(d-{\mathrm{\τ}}_l),---\left(2\right)$

H is and each information-related amount that directly declines that this area does not have unified Chinese as yet at present, for ease of describing, claims that h is each footpath decay function;

$R_u=E_0{R}_{\mathrm{ISI}}+E_I{R}_{\mathrm{MUI}}+N_0{R}_{n^{\′}}$

$=\frac{E_0}{N^2}\underset{I=0}{\overset{L-1}{\mathrm{\Σ}}}\underset{q=0}{\overset{L-1}{\mathrm{\Σ}}}\underset{\underset{i\≠0}{i=-\∞}}{\overset{+\∞}{\mathrm{\Σ}}}{g}_l{g}_q^{*}\underset{m=1-N}{\overset{N-1}{\mathrm{\Σ}}}(N-|m\left|\right)\×R_p(d_1+{\mathrm{mT}}_c-\mathrm{iT}-{\mathrm{\τ}}_l)\×R_p^{*}(d_2+{\mathrm{mT}}_c-\mathrm{iT}{-\mathrm{\τ}}_q)---\left(3\right)$

$+\frac{E_I}{N^2}\underset{I=0}{\overset{L-1}{\mathrm{\Σ}}}\underset{q=0}{\overset{L-1}{\mathrm{\Σ}}}\underset{i=-\∞}{\overset{+\∞}{\mathrm{\Σ}}}{g}_l{g}_q^{*}\underset{m=1-N}{\overset{N-1}{\mathrm{\Σ}}}(N-|m\left|\right)\×R_p(d_1+{\mathrm{mT}}_c-\mathrm{iT}-{\mathrm{\τ}}_l)\×R_p^{*}(d_2+{\mathrm{mT}}_c-\mathrm{iT}-{\mathrm{\τ}}_q)[1-\mathrm{\δ}\left(m\right)\mathrm{\δ}\left(i\right)]$

$+N_0{R}_p(d_1-d_2).$

In formula (2) or (3), E ₀Be targeted customer's symbol level energy, E _IBe the symbol level energy of interference user, N ₀Be the one-sided power spectrum density of system's white Gaussian noise, the symbol level energy can estimate that with the one-sided power spectrum density of system's white Gaussian noise in receiver system, above-mentioned amount is used as known parameters according to relevant algorithm; N is targeted customer's a spreading factor, R _ISIBe intersymbol interference function, R _MUIBe multiple access interference function, T _cBe the duration of a spread-spectrum code chip, T=NT _c, g _l, g _qBe respectively the channel fading factor in l footpath and q footpath, τ _l, τ _qBe the multipath transmisstion time delay in l footpath and q footpath, L is the sum of effective diameter, R _pBe to send the auto-correlation function of pulse shaping filter, obtain in advance according to the calculation expression of the employed forming filter of system; D, d ₁And d ₂For separating the pairing multidiameter delay of hole enlargement.

According to the aforementioned calculation method, the calculating of each footpath weighted value of G-RAKE receiver comprises the steps: as shown in Figure 2

Step 201: according to each footpath fading factor g of input _l, multi-path location τ _lSymbol level ENERGY E with the targeted customer ₀Utilize formula (2) to calculate h;

Step 202: ask for R _ISI: from formula (3) as can be known:

$R_{\mathrm{ISI}}=\frac{1}{N^2}\underset{l=0}{\overset{L-1}{\mathrm{\Σ}}}\underset{q=0}{\overset{L-1}{\mathrm{\Σ}}}\underset{\underset{i\≠0}{i=-\∞}}{\overset{+\∞}{\mathrm{\Σ}}}{g}_l{g}_q^{*}\underset{m=1-N}{\overset{N-1}{\mathrm{\Σ}}}(N-|m\left|\right)\×R_p(d_1+{\mathrm{mT}}_c-\mathrm{iT}-{\mathrm{\τ}}_l)\×R_p^{*}(d_2+{\mathrm{mT}}_c-\mathrm{iT}-{\mathrm{\τ}}_q)---\left(4\right)$

This step is exactly each the footpath fading factor g according to input _l, multi-path location τ _l, separate the pairing multidiameter delay d of hole enlargement, d ₁And d ₂And interference function R between formula (4) compute sign _ISI

Step 203: ask for R _MUI: from formula (3) as can be known:

$R_{\mathrm{MUI}}=\frac{1}{N^2}\underset{l=0}{\overset{L-1}{\mathrm{\Σ}}}\underset{q=0}{\overset{L-1}{\mathrm{\Σ}}}\underset{i=-\∞}{\overset{+\∞}{\mathrm{\Σ}}}{g}_l{g}_q^{*}\underset{m=1-N}{\overset{N-1}{\mathrm{\Σ}}}(N-|m\left|\right)\×R_p(d_1+{\mathrm{mT}}_c-\mathrm{iT}-{\mathrm{\τ}}_l)\×R_p^{*}(d_2+{\mathrm{mT}}_c-\mathrm{iT}-{\mathrm{\τ}}_q)[1-\mathrm{\δ}\left(m\right)\mathrm{\δ}\left(i\right)]---\left(5\right)$

This step is exactly each the footpath fading factor g according to input _l, multi-path location τ _lSeparate the pairing multidiameter delay d of hole enlargement, d ₁And d ₂And formula (4) calculates multiple access interference function R _MUI

Step 204: ask for R _{N '}: from formula (3) as can be known:

R _n′＝R _p(d ₁-d ₂) (6)

This step is according to the auto-correlation function R of forming filter _p, the input multidiameter delay d ₁And d ₂, utilize formula (6) to calculate the interference function R of background thermal noise _{N '}

Step 205: according to the targeted customer's who imports spreading factor N, targeted customer's symbol level ENERGY E ₀, interference user the symbol level ENERGY E _I, system's white Gaussian noise one-sided power spectrum density N ₀With the result of calculation of step 202, calculate noise correlation matrix R to step 204 _u

Step 206: according to the noise correlation matrix R of step 205 gained _uEach footpath that the h that obtains with step 201 calculates the G-RAKE receiver merges weighted factor.

But, comparing with traditional RAKE receiver, the computing cost that the processing method of each footpath weighted value of the calculating that the G-RAKE receiver is adopted needs is very big, thereby implementation complexity is bigger, and cost is also higher.

Summary of the invention

In view of this, the objective of the invention is to, propose a kind of method of receiving multipath signals, can reduce to calculate the operand that each footpath merges weights effectively.This method comprises the steps:

A, calculate each footpath decay function h according to each footpath fading factor, multi-path location information of input and targeted customer's symbol level energy;

B, according to each footpath fading factor of input, multi-path location information is separated the pairing multidiameter delay of hole enlargement and is calculated multiple access interference function R _MUI

C, according to the auto-correlation function R of forming filter _p, input multidiameter delay, calculate the interference function R of background thermal noise _{N '}

D, according to input with all irrelevant user code chip level energy sum E of spreading factor _{C_T}, system's white Gaussian noise one-sided power spectrum density N ₀And the result of calculation of step B and step C, according to formula R _u=E _{C_T}R _MUI+ N ₀R _{N '}Calculate noise correlation matrix R _u

E, according to the noise correlation matrix R of each footpath decay function h and step D gained of steps A gained _u, calculate each footpath weighted value ω; According to each footpath weighted value ω, to the formula Y=ω of the multipath signal y after descrambling, the despreading ^H* y merges processing, and gained signal Y separates the mediation decoded operation after being combined processing, and the signal after the output demodulation coding.

Each footpath decay function h of the described calculating of steps A is:

According to formula $h=\sqrt{E_0}\underset{l=0}{\overset{L-1}{\mathrm{\Σ}}}{g}_l{R}_p(d-{\mathrm{\τ}}_l)$ Calculate each footpath decay function h, wherein, E ₀Be targeted customer's symbol level energy, g _lBe the channel fading factor in l footpath, R _pFor sending the auto-correlation function of pulse shaping filter, d is for separating the pairing multidiameter delay of hole enlargement, τ _lIt is the multipath transmisstion time delay in l footpath.

The described calculating multiple access of step B interference function R _MUIFor:

According to formula

$R_{\mathrm{MUI}}=\frac{1}{N}\underset{l=0}{\overset{L-1}{\mathrm{\Σ}}}\underset{q=0}{\overset{L-1}{\mathrm{\Σ}}}{g}_l{g}_q^{*}[\underset{i=-\∞}{\overset{+\∞}{\mathrm{\Σ}}}{R}_p(d_1+{\mathrm{iT}}_c-{\mathrm{\τ}}_l)\×R_p^{*}(d_2+{\mathrm{iT}}_c-{\mathrm{\τ}}_q)-R_p(d_1-{\mathrm{\τ}}_l)\×R_p^{*}(d_2-{\mathrm{\τ}}_q)]$ Calculate multiple access interference function R _MUIWherein N is targeted customer's a spreading factor, and l and q are the label in footpath, g _lAnd g _qBe each footpath fading factor, τ _lAnd τ _qBe multi-path location information, L is the effective diameter sum, d ₁And d ₂For separating the pairing multidiameter delay of hole enlargement.

The interference function R of the described calculating background of step C thermal noise _{N '}For: according to formula R _{N '}=R _p(d ₁-d ₂) calculate the interference function R of background thermal noise _{N '}, wherein, R _pFor sending the auto-correlation function of pulse shaping filter, d ₁And d ₂For separating the pairing multidiameter delay of hole enlargement.

The described noise correlation matrix R of step D _uFor:

According to formula

$R_u=(E_0+E_I)R_{\mathrm{MUI}}+N_0{R}_{n^{\′}}$

$=E_{c\_T}\underset{l=0}{\overset{L-1}{\mathrm{\Σ}}}\underset{q=0}{\overset{L-1}{\mathrm{\Σ}}}{g}_l{g}_q^{*}[\frac{R(n_1,n_2)}{N}-R_p(d_1-{\mathrm{\τ}}_l)\×R_p^{*}(d_2-{\mathrm{\τ}}_q)]+N_0{R}_p(d_1-d_2),$

Noise correlation matrix R _uWherein N is targeted customer's a spreading factor, and l and q are the label in footpath, g _lAnd g _qBe each footpath fading factor, τ _lAnd τ _qBe multi-path location information, L is the effective diameter sum, d ₁And d ₂For separating the pairing multidiameter delay of hole enlargement, E _{C_T}For with all irrelevant user code chip level energy sums of spreading factor, R (n ₁, n ₂) be n ₁And n ₂Correlation function, wherein

N _OVSBe sample rate.

Described each the footpath weighted value ω that calculates of step e is: according to formula $\mathrm{\ω}=R_u^{-1}h$ Obtain each footpath weighted value ω.

The present invention also aims to, propose a kind of device that calculates each footpath weighted value, be arranged in the RAKE receiver system, specifically comprise:

Each footpath decay function computing unit, be used to receive targeted customer's symbol level energy from the descrambling and de-spreading module of RAKE receiver, the multipath fading factor of channel estimation module and multi-path location information, calculate each footpath decay function h, and each footpath decay function h input weighted value synthesis unit that will calculate;

Multiple access interference function computing unit is used to receive the multipath fading factor, multi-path location information from the channel estimation module of RAKE receiver, and from Multipath searching distribution module multidiameter delay, calculates multiple access interference function R _MUI, and with the multiple access interference function R that calculates _MUIInput noise correlation matrix computing unit;

Thermal noise interference function computing unit is used to receive the multidiameter delay of the Multipath searching distribution module of RAKE receiver, and the auto-correlation function of forming filter, calculates the interference function R of background thermal noise _{N '}, and with the interference function R of the background thermal noise that calculates _{N '}Input noise correlation matrix computing unit;

The noise correlation matrix computing unit is used to receive the multiple access interference function R from multiple access interference function computing unit _MUI, thermal noise interference function computing unit the interference function R of background thermal noise _{N '}Make intersymbol interference function multiple access interference function R _ISIEqual multiple access interference function R _MUI, and according to the intersymbol interference function R _ISI, multiple access interference function R _MUI, the background thermal noise interference function R _{N '}, all user code chip level energy sum E of receiver system _{C_T}And the one-sided power spectrum density N of system's white Gaussian noise ₀, calculate noise correlation matrix R _u, and with gained noise correlation matrix R _uInput weighted value synthesis unit;

The weighted value synthesis unit is used for according to each footpath decay function h and noise correlation matrix R of being received _u, synthetic each footpath weighted value ω, and each the weighted value ω directly that merges directly that module output synthesize to each.

Described RAKE receiver is a generalized RAKE receiver.

Another object of the present invention is to, a kind of RAKE receiver of utilizing said method to calculate each footpath weighted value is proposed, comprise that Multipath searching distribution module, descrambling and de-spreading module, channel estimation module, each footpath weighted value computing module, each footpath merge module and demodulation decoding module, it is characterized in that described each footpath weighted value computing module comprises:

Each footpath decay function computing unit, be used to receive targeted customer's symbol level energy from the descrambling and de-spreading module, each footpath decay function h of the multipath fading factor of channel estimation module and multi-path location information calculations, and each footpath decay function h input weighted value synthesis unit that will calculate;

Multiple access interference function computing unit is used to receive the multipath fading factor, multi-path location information from channel estimation module, and from Multipath searching distribution module multidiameter delay, calculates multiple access interference function R _MUI, and with the multiple access interference function R that calculates _MUIInput noise correlation matrix computing unit;

Thermal noise interference function computing unit is used to receive the multidiameter delay of Multipath searching distribution module, and the auto-correlation function of forming filter, calculates the interference function R of background thermal noise _{N '}, and with the interference function R of the background thermal noise that calculates _{N '}Input noise correlation matrix computing unit;

The noise correlation matrix computing unit is used to receive the multiple access interference function R from multiple access interference function computing unit _MUI, thermal noise interference function computing unit the interference function R of background thermal noise _{N '}Make intersymbol interference function multiple access interference function R _ISIEqual multiple access interference function R _MUI, and according to the intersymbol interference function R _ISI, multiple access interference function R _MUI, the background thermal noise interference function R _{N '}, all user code chip level energy sum E of receiver system _{C_T}And the one-sided power spectrum density N of system's white Gaussian noise ₀, calculate noise correlation matrix R _u, and with gained noise correlation matrix R _uInput weighted value synthesis unit;

The weighted value synthesis unit is used for according to each footpath decay function h and noise correlation matrix R of being received _u, synthetic each footpath weighted value ω, and each the weighted value ω directly that merges directly that module output synthesize to each.

Described RAKE receiver is a generalized RAKE receiver.

As can be seen from the above technical solutions, the inventive method is calculated in the process of each footpath weighted value, need not interference function between compute sign, so compare with the G-RAKE receiver, can reduce effectively to calculate the operand that each footpath merges weights, and the quality of output signal is influenced hardly; And calculating and the spreading factor of each footpath weighted value have nothing to do, and the user of different spreading factors can a shared RAKE receiver, and need not additionally to increase the RAKE receiver number for the user of different spreading factors.Adopt the present invention program, can under the prerequisite of output and G-RAKE receiver equal quality signal, reduce the complexity of computing cost and realization greatly, save cost.

Description of drawings

Fig. 1 is the structural representation of RAKE receiver or G-RAKE receiver;

Fig. 2 is each footpath weighted value computing module realization flow figure of prior art G-RAKE receiver;

Fig. 3 is each footpath weighted value computing module realization flow figure of embodiment of the invention RAKE receiver;

Fig. 4 is the structure chart of each footpath weighted value computing module of embodiment of the invention RAKE receiver.

Embodiment

For making the purpose, technical solutions and advantages of the present invention clearer, the present invention is further elaborated below in conjunction with accompanying drawing.

The formula of reduction (7) that proposes according to each the footpath weighted value computational methods that has the people to the G-RAKE receiver at present:

$\frac{1}{N^2}\underset{i=-\∞}{\overset{+\∞}{\mathrm{\Σ}}}\underset{m=1-N}{\overset{N-1}{\mathrm{\Σ}}}(N-|m\left|\right)\×R_p(d_1+{\mathrm{mT}}_c-\mathrm{iT}-{\mathrm{\τ}}_l)\×R_p^{*}(d_2+{\mathrm{mT}}_c-\mathrm{iT}-{\mathrm{\τ}}_q)$

$=\frac{1}{N}\underset{i=-\∞}{\overset{+\∞}{\mathrm{\Σ}}}{R}_p(d_1+{\mathrm{iT}}_c-{\mathrm{\τ}}_l)\×R_p^{*}(d_2+{\mathrm{iT}}_c-{\mathrm{\τ}}_q)---\left(7\right)$

$=R(n_1,n_2),$

Can be with formula (3) abbreviation:

$R_u=E_0{R}_{\mathrm{ISI}}+E_1{R}_{\mathrm{MUI}}+N_0{R}_{n^{\′}}$

$=E_0\underset{l=0}{\overset{L-1}{\mathrm{\Σ}}}\underset{q=0}{\overset{L-1}{\mathrm{\Σ}}}{g}_l{g}_q^{*}[R(n_1,n_2)-\frac{1}{N^2}\underset{m=1-N}{\overset{N-1}{\mathrm{\Σ}}}{R}_p(d_1+{\mathrm{mT}}_c-{\mathrm{\τ}}_l)\×R_p^{*}(d_2+{\mathrm{mT}}_c-{\mathrm{\τ}}_q)]$

(8)

$+E_I\underset{l=0}{\overset{L-1}{\mathrm{\Σ}}}\underset{q=0}{\overset{L-1}{\mathrm{\Σ}}}{g}_l{g}_q^{*}[R(n_1,n_2)-\frac{1}{N}{R}_p(d_1-{\mathrm{\τ}}_l)\×R_p^{*}(d_2-{\mathrm{\τ}}_q)]$

$+N_0{R}_p(d_1-d_2),$

Wherein, R (n ₁, n ₂) be n ₁And n ₂Correlation function,

N _OVSSample rate for the G-RAKE receiver module.

From formula (3) matrix R as can be seen _uComponent in, R _ISIAnd R _MUIExpression formula only to the summation of i the time, have any different: the sum operation when the former has lacked i=0 than the latter.Just because of R _ISIAnd R _MUIThis difference, cause in the formula (8) behind the abbreviation R _ISICalculating comprised the sum operation of 2N-1 item product.If in the implementation procedure of G-RAKE, think R _ISI=R _MUI, then formula (8) further abbreviation be:

$R_u=(E_0+E_I)R_{\mathrm{MUI}}+N_0{R}_{n^{\′}}$

$=(E_0+E_I)\underset{l=0}{\overset{L-1}{\mathrm{\Σ}}}\underset{q=0}{\overset{L-1}{\mathrm{\Σ}}}{g}_l{g}_q^{*}[R(n_1,n_2)-\frac{1}{N}{R}_p(d_1-{\mathrm{\τ}}_l)\×R_p^{*}(d_2-{\mathrm{\τ}}_q)]$

(11)

$+N_0{R}_p(d_1-d_2)$

$=E_T\underset{l=0}{\overset{L-1}{\mathrm{\Σ}}}\underset{q=0}{\overset{L-1}{\mathrm{\Σ}}}{g}_l{g}_q^{*}[R(n_1,n_2)-\frac{1}{N}{R}_p(d_1-{\mathrm{\τ}}_l)\×R_p^{*}(d_2-{\mathrm{\τ}}_q)]+N_0{R}_p(d_1-d_2),$

E wherein _T=E ₀+ E _I, be the symbol level energy.Table 1 shows the operand of each time slot of G-RAKE receiver, and wherein, M is an iterations of asking for weighted value ω, and general value is 1～3; J is total footpath number of G-RAKE receiver, and general value is J ≈ 2L.

Table 1

From the data of table 1 as can be seen, ask for R _ISIAnd R _MUIAmount of calculation occupy quite great proportion in the amount of calculation of G-RAKE, therefore in the G-RAKE implementation procedure, think R _ISI=R _MUIWith effectively reduce the G-RAKE receiver operand, reduce its implementation complexity.

According to formula (8), because R _ISI=R _MUI, ask for R _uThe time and ENERGY E ₀, E _IThe matrix that multiplies each other equates, therefore just only need know total symbol level ENERGY E _TAnd need not therefrom to distinguish E ₀, E _I, and then can reduce the required number of parameters of G-RAKE receiver, further reduce its implementation complexity.

In addition, from formula (4) and formula (5) as can be seen, R _MUIIn spreading factor

Can extract, so formula (8) can be rewritten as:

$R_u=E_T\underset{l=0}{\overset{L-1}{\mathrm{\Σ}}}\underset{q=0}{\overset{L-1}{\mathrm{\Σ}}}{g}_l{g}_q^{*}[R(n_1,n_2)-\frac{1}{N}{R}_p(d_1-{\mathrm{\τ}}_l)\×R_p^{*}(d_2-{\mathrm{\τ}}_q)]+N_0{R}_p(d_1-d_2)$

(12)

$=E_{c\_T}\underset{l=0}{\overset{L-1}{\mathrm{\Σ}}}\underset{q=0}{\overset{L-1}{\mathrm{\Σ}}}{g}_l{g}_q^{*}[\frac{R(n_1,n_2)}{N}-R_p(d_1-{\mathrm{\τ}}_l)\×R_p^{*}(d_2-{\mathrm{\τ}}_q)]+N_0{R}_p(d_1-d_2)$

Be R _MUIThe energy weighted factor from E _TBecome $\frac{E_T}{N}=E_{c\_T},$ Just with all irrelevant user code chip level energy sums of spreading factor, so R _uSize and spreading factor irrelevant.If therefore in the G-RAKE implementation procedure, think R _ISI=R _MUIAs long as it is identical to carry out user's transmission environment that G-RAKE receives, promptly multidiameter delay is identical with channel fading, the result of calculation that all these users just can a shared G-RAKE receiver.

According to above analysis, the G-RAKE receiver of the embodiment of the invention adopts structure shown in Figure 1, it is the structure of existing G-RAKE receiver or existing RAKE receiver, be wherein each footpath add+implementation procedure of weights computing module adopts the present invention program, specifically as shown in Figure 3, comprise the steps:

Step 301: according to each footpath fading factor g of input _l, multi-path location τ _lSymbol level ENERGY E with the targeted customer ₀Utilize formula (2) to calculate each footpath decay function h; Step 201 in the weights computing module implementation procedure of this step and existing G-RAKE receiver is identical;

Step 302: ask for R _MUI: from formula (12) as can be known

$R_{\mathrm{MUI}}=\frac{1}{N}\underset{l=0}{\overset{L-1}{\mathrm{\Σ}}}\underset{q=0}{\overset{L-1}{\mathrm{\Σ}}}{g}_l{g}_q^{*}[\underset{i=-\∞}{\overset{+\∞}{\mathrm{\Σ}}}{R}_p(d_1+{\mathrm{iT}}_c-{\mathrm{\τ}}_l)\×R_p^{*}(d_2+{\mathrm{iT}}_c-{\mathrm{\τ}}_q)-R_p(d_1-{\mathrm{\τ}}_l)\×R_p^{*}(d_2-{\mathrm{\τ}}_q)],---\left(13\right)$

This step is exactly each the footpath fading factor g according to input _l, multi-path location τ _lSeparate the pairing multidiameter delay d of hole enlargement, d ₁And d ₂And formula (13) calculates multiple access interference function R _MUI

Step 303: ask for R _{N '}: from formula (12) as can be known

R _n′＝R _p(d ₁-d ₂) (14)

This step is according to the auto-correlation function R of forming filter _p, the input multidiameter delay d ₁And d ₂, according to the interference function R of formula (14) calculating background thermal noise _{N '}Formula (14) is identical with formula (6), so the step 204 in the weights computing module implementation procedure of this step and existing G-RAKE receiver is identical;

Step 304: according to input with all irrelevant user code chip level energy sum E of spreading factor _{C_T}, system's white Gaussian noise one-sided power spectrum density N ₀And the result of calculation of step 302 and step 303, calculate noise correlation matrix R according to formula (12) _u

Step 305: according to each the footpath decay function h of step 301 gained and the noise correlation matrix R of step 304 gained _u, utilize formula (1) to calculate each footpath weighted value ω of G-RAKE receiver.

For realizing said method, the each several part structure of each footpath weighted value computing module 140 of embodiment of the invention RAKE receiver and annexation are as shown in Figure 4, and be specific as follows:

Each footpath decay function computing unit 141 is used to receive the targeted customer's symbol level ENERGY E from descrambling and de-spreading module 120 ₀, and the multipath fading factor g of channel estimation module 130 _lWith multi-path location information τ _l, utilize the above-mentioned parameter that is received to calculate each footpath decay function h according to formula (2), and each footpath decay function h input weighted value synthesis unit 145 that will calculate;

Multiple access interference function computing unit 142 is used to receive the multipath fading factor g from channel estimation module 130 _lWith multi-path location information τ _l, and multidiameter delay d, the d of Multipath searching distribution module 110 ₁And d ₂, utilize the above-mentioned parameter that is received to calculate multiple access interference function R according to formula (13) _MUI, and with the multiple access interference function R that calculates _MUIInput noise correlation matrix computing unit 144;

Thermal noise interference function computing unit 143 is used to receive the multidiameter delay d of Multipath searching distribution module 110 ₁And d ₂, and the auto-correlation function R of forming filter _p, utilize the above-mentioned parameter that is received to calculate the interference function R of background thermal noise according to formula (14) _{N '}, and with the interference function R of the background thermal noise that calculates _{N '}Input noise correlation matrix computing unit 144;

Noise correlation matrix computing unit 144 is used for according to the multiple access interference function R from multiple access interference function computing unit 142 _MUI, thermal noise interference function computing unit 143 the interference function R of background thermal noise _{N '}, all user code chip level energy sum E of receiver system _{C_T}And the one-sided power spectrum density N of system's white Gaussian noise ₀, calculate noise correlation matrix R according to formula (12) _u, and with gained noise correlation matrix R _uInput weighted value synthesis unit 145;

Weighted value synthesis unit 145 is used for according to each footpath decay function h and noise correlation matrix R of being received _u, utilize synthetic each the footpath weighted value ω of formula (1), and export each the footpath weighted value ω that synthesize to each and directly merge module 150.

Each footpath weighted value computing module of the embodiment of the invention shown in Figure 4, with each footpath weighted value computing module of the G-RAKE receiver of prior art relatively, reduced an intersymbol interference function calculation unit and the input and output relevant and connected with this unit; On the processing method, also correspondingly saved the expense of interference function between compute sign.

With each footpath weighted value computing module shown in Figure 4 each footpath weighted value computing module 140 as RAKE receiver shown in Figure 1, the structure of RAKE receiver then shown in Figure 1 and each several part annexation are the structure and the each several part annexation of the present invention program's RAKE receiver.

The simulation analysis that the G-RAKE receiver that the present invention program is proposed carries out shows, thinks R in the implementation procedure of G-RAKE receiver _ISI=R _MUICan the output result of G-RAKE receiver be impacted hardly.Therefore, in the implementation procedure of G-RAKE receiver, think R _ISI=R _MUICan under the prerequisite that does not influence quality of output signals, effectively reduce the operand of G-RAKE receiver, reduce the required number of parameters of G-RAKE receiver, and then reduce the implementation complexity of G-RAKE receiver; And making the calculating of G-RAKE weighted value and spreading factor irrelevant, the user of different spreading factors can a shared G-RAKE receiver, and need not additionally to increase G-RAKE receiver number for the user of different spreading factors.

Those skilled in the art will be appreciated that, because the G-RAKE receiver is mainly that with the structural difference of RAKE receiver each footpath weighted value computing module is different, therefore, the method of each footpath weighted value of the calculating that the present invention proposes and the device that is used to realize these computational methods, not only be applied in the G-RAKE receiver, also can be applicable to have in the RAKE receiver of identical or similar structures.

The above only is preferred embodiment of the present invention, not in order to restriction the present invention, all any modifications of being done within the spirit and principles in the present invention, is equal to and replaces and improvement etc., all should be included within protection scope of the present invention.

Claims (10)

1, a kind of method of receiving multipath signals is characterized in that, this method comprises the steps:

A, calculate each footpath decay function h according to each footpath fading factor, multi-path location information of input and targeted customer's symbol level energy;

B, according to each footpath fading factor of input, multi-path location information is separated the pairing multidiameter delay of hole enlargement and is calculated multiple access interference function R _MUI

C, according to the auto-correlation function R of forming filter _p, input multidiameter delay, calculate the interference function R of background thermal noise _{N '}

D, according to input with all irrelevant user code chip level energy sum E of spreading factor _{C_T}, system's white Gaussian noise one-sided power spectrum density N ₀And the result of calculation of step B and step C, according to formula R _u=E _{C_T}R _MUI+ N ₀R _{N '}Calculate noise correlation matrix R _u

E, according to the noise correlation matrix R of each footpath decay function h and step D gained of steps A gained _u, calculate each footpath weighted value ω; According to each footpath weighted value ω, to the formula Y=ω of the multipath signal y after descrambling, the despreading ^H* y merges processing, and gained signal Y separates the mediation decoded operation after being combined processing, and the signal after the output demodulation coding.

2, method according to claim 1 is characterized in that, each footpath decay function h of the described calculating of steps A is:

According to formula $h=\sqrt{E_0}\underset{l=0}{\overset{L-1}{\mathrm{\Σ}}}{g}_l{R}_p(d-{\mathrm{\τ}}_l)$ Calculate each footpath decay function h, wherein, E ₀Be targeted customer's symbol level energy, L is the effective diameter sum, and l is the label in footpath, g _lBe the channel fading factor in l footpath, R _pBe the auto-correlation function of forming filter, d is for separating the pairing multidiameter delay of hole enlargement, τ _lIt is the multipath transmisstion time delay in l footpath.

3, method according to claim 1 is characterized in that, the described calculating multiple access of step B interference function R _MUIFor:

According to formula

$R_{\mathrm{MUI}}=\frac{1}{N}\underset{l=0}{\overset{L-1}{\mathrm{\Σ}}}\underset{q=0}{\overset{L-1}{\mathrm{\Σ}}}{g}_l{g}_q^{*}[\underset{i=-\∞}{\overset{+\∞}{\mathrm{\Σ}}}{R}_p(d_1+{\mathrm{iT}}_c-{\mathrm{\τ}}_l)\×R_p^{*}(d_2+{\mathrm{iT}}_c-{\mathrm{\τ}}_q)-R_p(d_1-{\mathrm{\τ}}_l)\×R_p^{*}(d_2-{\mathrm{\τ}}_q)]$ Calculate multiple access interference function R _MUIWherein N is targeted customer's a spreading factor, and l and q are the label in footpath, g _lAnd g _qBe each footpath fading factor, τ _lAnd τ _qBe multi-path location information, L is the effective diameter sum, d ₁And d ₂For separating the pairing multidiameter delay of hole enlargement, R _pBe the auto-correlation function of forming filter, T _cBe the duration of a spread-spectrum code chip, i is a summing target.

4, method according to claim 1 is characterized in that, the interference function R of the described calculating background of step C thermal noise _{N '}For: according to formula R _{N '}=R _p(d ₁-d ₂) calculate the interference function R of background thermal noise _{N '}, wherein, R _pBe the auto-correlation function of forming filter, d ₁And d ₂For separating the pairing multidiameter delay of hole enlargement.

5, method according to claim 1 is characterized in that, the described noise correlation matrix R of step D _uFor:

According to formula

$R_u=(E_0+E_I)R_{\mathrm{MUI}}+N_0{R}_{n^{\′}}$

$=E_{c\_T}\underset{l=0}{\overset{L-1}{\mathrm{\Σ}}}\underset{q=0}{\overset{L-1}{\mathrm{\Σ}}}{g}_l{g}_q^{*}[\frac{R(n_1,n_2)}{N}-R_p(d_1-{\mathrm{\τ}}_l)\×R_p^{*}(d_2-{\mathrm{\τ}}_q)]+N_0{R}_p(d_1-d_2),$

Noise correlation matrix R _uE wherein ₀Be targeted customer's symbol level energy, E _IBe the symbol level energy of interference user, N ₀Be the one-sided power spectrum density of system's white Gaussian noise, R _{N '}Be the interference function of background thermal noise, N is targeted customer's a spreading factor, and l and q are the label in footpath, g _lAnd g _qBe each footpath fading factor, τ _lAnd τ _qBe multi-path location information, L is the effective diameter sum, d ₁And d ₂For separating the pairing multidiameter delay of hole enlargement, E _{C_T}For with all irrelevant user code chip level energy sums of spreading factor, R (n ₁, n ₂) be n ₁And n ₂Correlation function, wherein

N _OVSBe sample rate.

According to each described method of claim 1 to 5, it is characterized in that 6, described each the footpath weighted value ω that calculates of step e is: according to formula $\mathrm{\ω}=R_u^{-1}h$ Obtain each footpath weighted value ω.

7, a kind of device that calculates each footpath weighted value is arranged in the RAKE receiver system, it is characterized in that this device comprises:

Each footpath decay function computing unit, be used to receive targeted customer's symbol level energy from the descrambling and de-spreading module of RAKE receiver, each footpath decay function h of the multipath fading factor of channel estimation module and multi-path location information calculations, and each footpath decay function h input weighted value synthesis unit that will calculate;

Multiple access interference function computing unit is used to receive the multipath fading factor, multi-path location information from the channel estimation module of RAKE receiver, and from Multipath searching distribution module multidiameter delay, calculates multiple access interference function R _MUI, and with the multiple access interference function R that calculates _MUIInput noise correlation matrix computing unit;

Thermal noise interference function computing unit is used to receive the multidiameter delay of the Multipath searching distribution module of RAKE receiver, and the auto-correlation function of forming filter, calculates the interference function R of background thermal noise _{N '}, and with the interference function R of the background thermal noise that calculates _{N '}Input noise correlation matrix computing unit;

The noise correlation matrix computing unit is used to receive the multiple access interference function R from multiple access interference function computing unit _MUI, thermal noise interference function computing unit the interference function R of background thermal noise _{N '}Make intersymbol interference function multiple access interference function R _ISIEqual multiple access interference function R _MUI, and according to the intersymbol interference function R _ISI, multiple access interference function R _MUI, the background thermal noise interference function R _{N '}, all user code chip level energy sum E of receiver system _{C_T}And the one-sided power spectrum density N of system's white Gaussian noise ₀, calculate noise correlation matrix R _u, and with gained noise correlation matrix R _uInput weighted value synthesis unit;

The weighted value synthesis unit is used for according to each footpath decay function h and noise correlation matrix R of being received _u, synthetic each footpath weighted value ω, and each the weighted value ω directly that merges directly that module output synthesize to each.

8, device according to claim 7 is characterized in that, described RAKE receiver is a generalized RAKE receiver.

9, a kind of RAKE receiver of calculating each footpath weighted value, comprise that Multipath searching distribution module, descrambling and de-spreading module, channel estimation module, each footpath weighted value computing module, each footpath merge module and demodulation decoding module, it is characterized in that described each footpath weighted value computing module comprises:

Each footpath decay function computing unit, be used to receive targeted customer's symbol level energy from the descrambling and de-spreading module, the multipath fading factor of channel estimation module and multi-path location information are calculated each footpath decay function h, and each footpath decay function h input weighted value synthesis unit that will calculate;

Multiple access interference function computing unit is used to receive the multipath fading factor, multi-path location information from channel estimation module, and from Multipath searching distribution module multidiameter delay, calculates multiple access interference function R _MUI, and with the multiple access interference function R that calculates _MUIInput noise correlation matrix computing unit;

Thermal noise interference function computing unit is used to receive the multidiameter delay of Multipath searching distribution module, and the auto-correlation function of forming filter, calculates the interference function R of background thermal noise _{N '}, and with the interference function R of the background thermal noise that calculates _{N '}Input noise correlation matrix computing unit;

The noise correlation matrix computing unit is used to receive the multiple access interference function R from multiple access interference function computing unit _MUI, thermal noise interference function computing unit the interference function R of background thermal noise _{N '}Make intersymbol interference function multiple access interference function R _ISIEqual multiple access interference function R _MUI, and according to the intersymbol interference function R _ISI, multiple access interference function R _MUI, the background thermal noise interference function R _{N '}, all user code chip level energy sum E of receiver system _{C_T}And the one-sided power spectrum density N of system's white Gaussian noise ₀, calculate noise correlation matrix R _u, and with gained noise correlation matrix R _uInput weighted value synthesis unit;

The weighted value synthesis unit is used for according to each footpath decay function h and noise correlation matrix R of being received _u, synthetic each footpath weighted value ω, and each the weighted value ω directly that merges directly that module output synthesize to each.

10, RAKE receiver according to claim 9 is characterized in that, described RAKE receiver is a generalized RAKE receiver.

Images