CN1134923C - Time-space joined multi-path search method for CDMA communication system - Google Patents

Abstract

The present invention relates to time space combined multi-path searching method for a CDMA communication system. The method comprises: direction vectors a(theta1) in LTotal_Beam directions theta1 of arrays are determined according to the modes of the arrays; an inverse matrix Rxx<-1> of a correlation matrix is obtained from the correlation matrix Rxx of input signals of the arrays, and further, the weighting vector W1 of LTotal_Beam arrays is obtained by calculating according to a(theta1) and Rxx<-1>; LBeam odd numbers of W1 and a time-domain correlated vector Rxd(tau) are used for carrying out correlated calculation to obtain time-domain correlated coefficients; according to a usual multi-path searching method, effective positive integers LFinger of multi-path time delay is determined from all the time-demain correlated coefficients; according to a corresponding number 1 of the time delay, the spatial domain information of a multi-path signal can be known, namely that the direction of the multi-path signal is near theta1, and thus, the weighting vector W1 of the multi-path arrays can be output; an initial position of a search time window TSearch and the initial direction of spatial domain search are adjusted according to the time delay information and the spatial domain information of a path of signals having the strongest of multiple paths, and the minimal angle intervals of LBeam odd numbers of a(theta1) spatial domain searchers in spatial directions reach minimal space angle intervals deltathetastep.

Description

Time-space joined multi-path search method in a kind of code division multiple address communication system

(1) technical field

The present invention relates to the method for reseptance of the array acceptor of code division multiple access (CDMA) multi-media radio communication system, specifically, relate to a kind of time-space joined multi-path search method.

(2) background technology

For a long time, wireless communication system is faced with the contradiction between limited frequency spectrum resources and the constantly quick user who increases all the time.Although FDMA frequency division multiple access, TDMA time division multiple access and CDMA CDMA (Code Division Multiple Access) have improved power system capacity to a certain extent, this can't satisfy the demand of growing wireless traffic.Therefore people begin to utilize the spatial domain characteristic of channel, as methods such as employing adaptive antenna arrays.Airspace filter characteristic by utilizing array has improved capacity at the communication quality that has improved wireless communication system in varying degrees.

Multipath searching problem in cdma system is a key technology of cdma system, after adopting the adaptive antenna technology, power system capacity will enlarge, the Signal to Interference plus Noise Ratio of receiver is reduced, how at adaptive array, under low Signal to Interference plus Noise Ratio condition, finish Multipath searching and array received becomes array antenna key in application in cdma system.In seeing this Study on Technology result, mainly adopt multi-beam to cooperate the method for Multipath searching at present, carry out wave beam earlier and form, in each wave beam, carry out Multipath searching again.

The existing multi-beam that adopts cooperates multipath searching method, (user is a lot of under the white noise environment of space, the user evenly distributes in the space, and type of service is similar) be effective, but number of users is few in system, inhomogeneous or customer service type (does not have multimedia service) to the user not simultaneously in spatial distribution, and the space just coloured noise may occur.Because the main foundation that multi-beam is selected is the watt level of signal, when coloured noise occurring, wave beam just may select mistake, thereby reduces time domain and volume-search coverage performance.

(3) summary of the invention

The object of the present invention is to provide a kind of time-space joined multi-path search method that has in the code division multiple address communication system that suppresses the System in Spatial Colored Noise ability.

Time-space joined multi-path search method in a kind of code division multiple address communication system provided by the present invention comprises the following steps:

1) from the correlation matrix of array input signal X (t), tries to achieve inverse matrix

, and, determine that array is at L according to array format _{Total_Beam}Individual direction θ _lOn direction vector a (θ _l), l=1 ..., L _{Total_Beam}, L _{Total_Beam}Be natural number;

Even domain search when 2) carrying out is T _SearchBe window search time, τ ∈ T _Search, calculate time domain associated vector R _Xd(τ) $R_{\mathrm{xd}}\left(\mathrm{\&tau;}\right)=\frac{1}{T}{\&Integral;}_0^{T}X\left(t\right)S_{\mathrm{Pilot}}^{*}(t-\mathrm{\&tau;})\mathrm{dt}$ Wherein, S _Piot(t-τ) postpones the signal that τ obtains later in time for pilot signal, the conjugation of " * " expression plural number;

3) carry out volume-search coverage, with array at direction θ _lOn direction vector a (θ _l) as the volume-search coverage vector, l=1 ..., L _{Total_Beam}, therefrom select L _BeamOdd number a (θ _l), l=1,3 ..., L _Beam, L _BeamFor more than or equal to 1 odd number, calculate the array weight vectors W of corresponding l direction vector _l: $W_l=R_{\mathrm{xx}}^{-1}a\left({\mathrm{\&theta;}}_1\right)l=\mathrm{1,2},\&CenterDot;\&CenterDot;\&CenterDot;,L_{\mathrm{Beam}}$

And computing array weighing vector W _lWith time domain associated vector R _XdCoefficient correlation (τ)

, coefficient correlation In the n time search of subscript " n " expression, it is the array weight vectors W of l direction that subscript " l " is represented _LWith time domain associated vector R _Xd(τ) carry out related operation;

4) from all

(τ ∈ T _Search, l=1,2 ..., L _Beam) in, determine effective positive integer L according to the method for common Multipath searching _FingerIndividual multidiameter delay can be known the spatial information (si) of multipath signal according to the numbering 1 of this time delay correspondence, and promptly the direction of multipath signal is at θ _lNear, can export multipath array weight vectors W thus _l

5) according to the strongest footpath time delay information of signal of power in the multipath and spatial information (si) to window T search time _SearchInitial position and the inceptive direction of volume-search coverage adjust, the initial position of adjusting window search time is in the center of search window to guarantee main footpath time location, adjusts direction vector a (θ according to main diameter direction principle placed in the middle _l) sensing, and satisfy L arranged all the time _BeamOdd number a (θ _l) vector carries out volume-search coverage, in the spatial domain scope of being searched for, is the center with the main diameter direction, with minimum space angle intervals Δ θ _StepPursue domain search device direction for step-length, finally make L _BeamOdd number a (θ _l) minimum angles of volume-search coverage device on direction in space reach minimum space angle intervals Δ θ at interval _Step

6) repeat said process, constantly export L _FingerIndividual multidiameter delay and multipath array weight vectors W _l

Time-space joined multi-path search method in the above-mentioned code division multiple address communication system wherein, is asked for inverse matrix in the step 1) from the correlation matrix of array input signal X (t)

Can adopt alternative manner to calculate: $R_{\mathrm{xx}}^{-1}\left(n\right)=R_{\mathrm{xx}}^{-1}(n-1)-\frac{R_{\mathrm{xx}}^{-1}(n-1)X\left(n\right)X^H\left(n\right)R_{\mathrm{xx}}^{-1}(n-1)}{1+X\left(n\right)R_{\mathrm{xx}}^{-1}\left(n\right)X\left(n\right)}$ $R_{\mathrm{xx}}^{-1}\left(0\right)=\mathrm{\&mu;I}$

In following formula, Represent the inverse matrix that the n time snap obtains, I representation unit matrix, μ are less than a constant of 1 greater than 0.

Above-mentioned technical scheme, be used for CDMA multi-media radio communication system, can finish the Syndicating search of time domain and volume-search coverage effectively, the time delay and the directional information of multipath signal are provided, the array weight vectors that suppresses System in Spatial Colored Noise can be provided simultaneously, thereby solve in the cdma system than Multipath searching under low signal interference noise ratio, the fading channel and array received problem.

(4) description of drawings

Fig. 1 is the structured flowchart that the present invention uses in cdma system;

Fig. 2 is the schematic flow sheet of time-space joined multi-path search method of the present invention.

(5) embodiment

As shown in Figure 1, the system configuration that is applied in the cdma system of the present invention is described as follows:

Array received module 101 forms the digital baseband array signal with the radiofrequency signal that receives on the array antenna by methods such as low noise amplification, down-conversion and A/D conversion.

After space-time Multipath searching module 102 received digital baseband signal, the space-time searching algorithm that adopts the present invention to propose obtained the time delay and the array weight of each multipath branch of received signal.The multidiameter delay control signal of space-time Multipath searching module 102 outputs is used to control multipath separation module 103 received signal is separated into several roads multipath signal.Space-time Multipath searching module 102 output array weighted signals make 104 pairs of each road multipath signals of array signal synthesis module, and weighting is synthetic respectively.The merging control signal control channel of space-time Multipath searching module 102 outputs is estimated and high specific merging module 105 only effective multipath signal is carried out channel estimating and high specific merges.

The space-time Multipath searching algorithm of space-time Multipath searching module 102 is described below:

As shown in Figure 2, for the signal s (t) that the user sends, the signal of array received can be expressed as behind the process multipath transmisstion: $X\left(t\right)=\underset{i=1}{\overset{L}{\mathrm{\&Sigma;}}}a\left({\mathrm{\&theta;}}_l\right)h_l\left(t\right)s(t-{\mathrm{\&tau;}}_1)+n\left(t\right)---\left(1\right)$

A (θ wherein _l) the expression array is to from direction θ ₁The response of l bar multipath signal; h _l(t) the multiple decline of expression l bar multipath signal; τ _lThe time delay of representing l bar multipath signal; N (t) represents other subscriber signal, its corresponding multipath signal and thermal noise.Abundant when number of users, user's wireless traffic is similar, and the user is when evenly distributing on the locus, and in cdma system, n (t) can be similar to and think white Gaussian noise.Less when number of users, during skewness, n (t) subscriber signal s (t) relatively is exactly a kind of System in Spatial Colored Noise on the locus for user's wireless traffic type difference (for example high speed wireless data business in 3-G (Generation Three mobile communication system)) or user.

Handle in order to carry out space-time, we carry out following processing to the array signal that receives:

The first step: by the inverse matrix of autocorrelation matrix of step 201 computing array input signal

Because the output of array received step is digital baseband signal, we adopt alternative manner to calculate at digital baseband $R_{\mathrm{xx}}^{-1}\left(n\right)=R_{\mathrm{xx}}^{-1}(n-1)-\frac{R_{\mathrm{xx}}^{-1}(n-1)X\left(n\right)X^H\left(n\right)R_{\mathrm{xx}}^{-1}(n-1)}{1+X\left(n\right)R_{\mathrm{xx}}^{-1}\left(n\right)X\left(n\right)}---\left(2\right)$ $R_{\mathrm{xx}}^{-1}\left(0\right)=\mathrm{\&mu;I}---\left(3\right)$

In formula (2) and (3), Represent the inverse matrix that the n time snap obtains, I representation unit matrix, μ are less than a constant of 1 greater than 0.It is to be noted the inverse matrix that formula (2) and formula (3) calculate

With effective per family, promptly all are identical with using per family in the Multipath searching process to all

, and the snap cycle can be greater than spreading rate, so calculated load is little.

Second step: according to array format, with minimum space angle delta θ _StepBe the interval, and determine that according to the form of array array is at L _{Total_Beam}Individual direction θ _lOn direction vector a (θ _l) l=1,2 ..., L _{Total_Beam}L _{Total_Beam}For greater than 1 integer.If use a ^H(θ _l) to array weight, then can be at direction in space θ _lForm 1 wave beam.Wherein subscript " H " is represented conjugate transpose

The 3rd step: domain search when time domain displacement correlation step 202 is carried out, even T _SearchBe window search time, τ ∈ T _Search, calculate time domain associated vector R _Xd(τ) $R_{\mathrm{xd}}\left(\mathrm{\&tau;}\right)=\frac{1}{T}{\&Integral;}_0^{T}X\left(t\right)S_{\mathrm{Pilot}}^{*}(t-\mathrm{\&tau;})\mathrm{dt}$ Wherein, S _Piot(t-τ) postpones the signal that τ obtains later in time for pilot signal, the conjugation of " * " expression plural number.In general cdma system, pilot signal is all arranged in order to Multipath searching and channel estimating.Therefore this signal is known.

The 4th step: some volume-search coverage devices 203 carry out volume-search coverage, with Δ θ _InitialBe interval (initial space angle intervals Δ θ _InitialBe minimum space angle intervals Δ θ _StepIntegral multiple), from direction vector a (θ _l) l=1,2 ..., L _{Total_Beam}In select L _BeamOdd number a (θ _l) l=1,2 ..., L _Beam, L _BeamFor more than or equal to 1 odd number, calculate array weight vectors W _l: $W_l=R_{\mathrm{xx}}^{-1}a\left({\mathrm{\&theta;}}_1\right)l=1,2,...,L_{\mathrm{Beam}}---\left(5\right)$

And calculating W _lWith Multipath searching matched filter output vector R _XdCoefficient correlation (τ), that is: $C_l^n\left(\mathrm{\&tau;}\right)=(1-a)\&times;{\left|\right|W}_l^H{R}_{\mathrm{xd}}\left(\mathrm{\&tau;}\right){\left|\right|}^2+a\&times;C_l^{n-1}\left(\mathrm{\&tau;}\right)---\left(6\right)$

The 5th step: make T _SearchBe window search time, from all $C_l^n\left(\mathrm{\&tau;}\right)\mathrm{\&tau;}\&Element;T_{\mathrm{Bearch}}l=\mathrm{1,2},...,L_{\mathrm{Beam}}$ In, determine effective L according to the method for common Multipath searching _FingerIndividual multidiameter delay and corresponding volume-search coverage device numbering l thereof, L _FingerBe positive integer, l=1,2 ..., L _Beam, and according to volume-search coverage device numbering 1 output multipath array weight vectors W _l

The 6th step: time delay information and volume-search coverage device according to most powerful path in the multipath are numbered window T search time _SearchInitial position and the inceptive direction of volume-search coverage device adjust.The initial position of adjusting window search time is in the center of search window to guarantee main footpath time location.The adjustment of volume-search coverage device is the L that has all the time that carry out according to main diameter direction principle placed in the middle and satisfied _BeamOdd number a (θ _l) vector carries out volume-search coverage.In the spatial domain scope of being searched for, be the center with the main diameter direction, with minimum space angle intervals Δ θ _StepAdjust volume-search coverage device direction gradually for step-length, finally make L _BeamOdd number a (θ _l) minimum angles of volume-search coverage device on direction in space reach minimum space angle intervals Δ θ at interval _Step

The 7th step:, repeat said process from the first step.

As can be seen, when obtaining multidiameter delay information, array weight information, can also obtain user's spatial information from said process, this also is very favourable to the position of determining the mobile subscriber.

Claims (2)

1. the time-space joined multi-path search method in the code division multiple address communication system comprises the following steps:

1) from the correlation matrix of array input signal X (t), tries to achieve inverse matrix

, and, determine that array is at L according to array format _{Total_Beam}Individual direction θ _lOn direction vector a (θ _l), l=1 ..., L _{Total_Beam}, L _{Total_Beam}Be natural number;

Even domain search when 2) carrying out is T _SearchBe window search time, τ ∈ T _Search, calculate time domain associated vector R _Xd(τ) $R_{\mathrm{xd}}\left(\mathrm{\&tau;}\right)=\frac{1}{T}{\&Integral;}_0^{T}X\left(t\right)S_{\mathrm{Pilot}}^{*}(t-\mathrm{\&tau;})\mathrm{dt}$ Wherein, S _Piot(t-τ) postpones the signal that τ obtains later in time for pilot signal, the conjugation of " * " expression plural number;

3) carry out volume-search coverage, with array at direction θ _lOn direction vector a (θ _l) as the volume-search coverage vector, l=1 ..., L _{Total_Beam}, therefrom select L _BeamOdd number a (θ _l), l=1,3 ..., L _Beam, L _BeamFor more than or equal to 1 odd number, calculate the array weight vectors W of corresponding l direction vector _l: $W_l=R_{\mathrm{xx}}^{-1}a\left({\mathrm{\&theta;}}_1\right)l=\mathrm{1,2},\&CenterDot;\&CenterDot;\&CenterDot;,L_{\mathrm{Beam}}$

And computing array weighing vector W _lWith time domain associated vector R _XdCoefficient correlation (τ)

, coefficient correlation

In the n time search of subscript " n " expression, it is the array weight vectors W of l direction that subscript " l " is represented _lWith time domain associated vector R _Xd(τ) carry out related operation;

4) from all

(τ ∈ T _Search, l=1,2 ..., L _Beam) in, determine effective positive integer L according to the method for common Multipath searching _FingerIndividual multidiameter delay can be known the spatial information (si) of multipath signal according to the numbering 1 of this time delay correspondence, and promptly the direction of multipath signal is at θ _lNear, can export multipath array weight vectors W thus _l

5) according to the strongest footpath time delay information of signal of power in the multipath and spatial information (si) to window T search time _SearchInitial position and the inceptive direction of volume-search coverage adjust, the initial position of adjusting window search time is in the center of search window to guarantee main footpath time location, adjusts direction vector a (θ according to main diameter direction principle placed in the middle _l) sensing, and satisfy L arranged all the time _BeamOdd number a (θ _l) vector carries out volume-search coverage, in the spatial domain scope of being searched for, is the center with the main diameter direction, with minimum space angle intervals Δ θ _StepPursue domain search device direction for step-length, finally make L _BeamOdd number a (θ _l) minimum angles of volume-search coverage device on direction in space reach minimum space angle intervals Δ θ at interval _Step

6) repeat said process, constantly export L _FingerIndividual multidiameter delay and multipath array weight vectors W _l

2. the time-space joined multi-path search method in a kind of code division multiple address communication system according to claim 1 is characterized in that: ask for inverse matrix in the described step 1) from the correlation matrix of array input signal X (t) Can adopt alternative manner to calculate: $R_{\mathrm{xx}}^{-1}\left(n\right)=R_{\mathrm{xx}}^{-1}(n-1)-\frac{R_{\mathrm{xx}}^{-1}(n-1)X\left(n\right)X^H\left(n\right)R_{\mathrm{xx}}^{-1}(n-1)}{1+X\left(n\right)R_{\mathrm{xx}}^{-1}\left(n\right)X\left(n\right)}$ $R_{\mathrm{xx}}^{-1}\left(0\right)=\mathrm{\&mu;I}$

In following formula, Represent the inverse matrix that the n time snap obtains, I representation unit matrix, μ are less than a constant of 1 greater than 0.