WO2007128184A1

WO2007128184A1 - A method and apparatus of calculating each path weighting value and a rake receiver

Info

Publication number: WO2007128184A1
Application number: PCT/CN2007/000072
Authority: WO
Inventors: Haihong Xu; Zhilin Zhao; Zhijian Yu
Original assignee: Huawei Technologies Co., Ltd.
Priority date: 2006-05-10
Filing date: 2007-01-09
Publication date: 2007-11-15
Also published as: CN1983839A; CN100578954C

Abstract

A method of calculating each path weighting value comprises steps of: obtaining a noise correlation matrix by multiple access interference function and noise interference function achieved by calculating; calculating to obtain each path weighting value based on each path fading function and noise correlation matrix achieved. An apparatus and a RAKE receiver are used to calculate each path weighting value based on the above method. It efficiently reduces the computational complexity of calculating each path combining weighting value while it does not affect the performance of the G-RAKE receiver through using the above method. Moreover, users of different spread spectrum factor may commonly use a single RAKE receiver because the calculation of each path weighting value is independent of the spread spectrum factor, and it is not necessary to add the number of RAKE receivers for user of different spread spectrum factor.

Description

Method, device and RAKE receiver for calculating weight value of each path

Technical field

The present invention relates to a wireless communication signal receiving technique, and in particular, to a method, a device and a RAKE receiver for calculating weight values of respective paths. Background of the invention

Unlike conventional modulation techniques, the channel bandwidth in a code division multiplex (CDMA) spread spectrum system is much larger than the flat fading bandwidth of the channel. Therefore, it is required to have a good autocorrelation property of the CDMA spreading code when selecting a signal, so that an equalization algorithm can be used to eliminate inter-symbol interference between adjacent symbols. Thus, if delay spread occurs in the wireless channel, it can be regarded as only the retransmission of the transmitted signal. If the delay between different paths of the multipath signal exceeds one chip period, the signals of the different paths can actually be regarded as being mutually uncorrelated.

For the above reasons, the multipath signal contains information that can be utilized, so the receiver can improve the signal to noise ratio of the received signal by combining the path signals, specifically, according to the input multipath delay information, each path channel. The fading information and other external parameters (including the system Gaussian noise power N. and the chip-level energy E of the receiving user) calculate the weighting factor ω of each path; according to the input multipath delay information and the scrambling code in the received signal, The frequency code information completes the descrambling and despreading operation on the chip-level received signal, and outputs the descrambled and despread symbol-level received signal y; and then performs the combining process by using the formula xy, where [represents the common transposition. A receiver employing this signal processing method is called a RAKE receiver.

In order to meet the increasing rate requirements of users, existing wireless communication systems have introduced various technologies with higher peak rates, such as High Speed Downlink Packet Access (HSDPA) and high-speed uplink packet access ( High Speed Uplink Packet Access, HSUPA X The introduction of these new technologies brings traditional RAKE reception technology Bigger challenge. On the one hand, the traditional RAKE receiver technology cannot meet the throughput requirements of high-speed services; on the other hand, due to the self-interference characteristics of CDMA systems, the interference caused by high-rate service users to other users should be as small as possible, that is, satisfied. Under the premise of high-rate service performance indicators, the required signal-to-noise ratio (SR) is reduced as much as possible. Therefore, it is very necessary to introduce advanced receiving techniques that outperform conventional RAKE receivers.

Receivers with better performance than traditional RAKE receivers include the following categories:

Enhanced RAKE receiver, which is only based on the traditional RAKE receiving technology, and its performance improvement is very limited;

Equilibrium class advanced receiver, although this kind of receiver can theoretically completely eliminate inter-symbol interference (ISI) caused by multipath and multiple access interference (MUI) between different users, so as to obtain good performance, but because of its The structure and the traditional RAKE receiver are very different, and the impact on the existing receiver structure is large;

Universal RAKE (Generalized Rake, G-RAKE) receiver, which fundamentally considers the dominant factor limiting the performance of traditional RAKE receivers, not only achieves the same performance as balanced receivers, but also has traditional RAKE reception. A similar implementation structure.

The G-RAKE receiver needs to calculate the weighting coefficients of each path according to the following formula:

ω = R-' J7 , (1) where,

h = ^∑g _l R _p (d - T _l ) , (2) h is the amount related to the fading information of each path. At present, there is no unified Chinese name in the field. For convenience of description, h is called the path fading function. . - _Z Tr,)x ? 1⁄2 +m7 -rr ₉ ) -ίΤ-τ,) R _p * (d ₂ + mT _c - iT - r _? )[l - S(m)S(i)]

+N _Q R _p {d _l -d ₂ ) o

(3)

In formula (2) or (3), E. The symbol-level energy for the target user, E, is the symbol-level energy of the interfering user, N. For the unilateral power spectral density of the system Gaussian white noise, the symbol-level energy and the unilateral power spectral density of the system Gaussian white noise can be estimated according to the algorithm in which the above quantities are used as known parameters.扩 is the target user's spreading factor, which is the intersymbol interference function, which is the multiple access interference function, 7; is the duration of a spread spectrum chip, T = NT _C , g is the channel fading of the / path and the first path respectively The factors, r, and are the propagation delays of the /path and the first, respectively, L is the total number of effective paths, ^ is the autocorrelation function of the transmit pulse shaping filter, and the calculation expression of the shaping filter used by the transmitting end Pre-obtained; d, ^, and ^ are the multipath delays corresponding to the despreading. Although compared with the traditional RAKE receiver, the G-RAKE receiver can better reduce the impact of multiple access interference or inter-symbol interference on the receiver performance, but the computational overhead required to calculate the path weighting value is used. It is very large, so the implementation complexity is high and the cost is high. SUMMARY OF THE INVENTION In view of this, the embodiment of the present invention provides a method for calculating each path weighting value, which can effectively reduce the calculation amount of calculating each path combining weight. The method comprises the following steps:

Calculating a multiple access interference function and a noise function of the background thermal noise according to each path fading factor, each path position information, and a multipath delay corresponding to the despreading path; The multiple access interference function ?^, and the background thermal noise interference function, respectively, and the sum of the user chip-level energy E _c - _T independent of the spreading factor and the single-sided power spectrum of the system Gaussian white noise The density N _{Q is} multiplied and then added to obtain a noise correlation matrix R _u . The weighted value ω of each path is calculated according to the obtained noise correlation matrix R _u .

The embodiment of the present invention further provides a device for calculating the weighting value of each path, which is located in the RAKE receiver system, and specifically includes:

a multiple access interference function calculation unit, configured to receive each path fading factor from the RAKE receiver, each path position information, and a multipath delay from the multipath search distribution module, calculate a multiple access interference function _w , and calculate The obtained multiple access interference function input noise correlation matrix calculation unit;

The thermal noise interference function calculating unit is configured to receive the multipath delay of the RAKE receiver, and calculate a disturbance function of the background thermal noise according to the multipath delay and the autocorrelation function of the shaping filter of the transmitting end obtained in advance. And the calculated interference function of the background thermal noise

R. Input noise correlation matrix calculation unit;

a noise correlation matrix calculation unit, configured to calculate an interference function of a background thermal noise of the unit according to the multiple access interference function _w and the thermal noise interference function from the multiple access interference function calculation unit

R , , the sum of all user chip-level energies of the receiver system E^ and the unilateral power spectral density N of the system Gaussian white noise. , calculate the noise correlation matrix, and obtain the resulting noise correlation matrix

R _u input weighted value synthesis unit;

a weighting value synthesizing unit configured to synthesize each path weighting value ω according to the received noise correlation matrix, and output the synthesized path weighting values to the RAKE receiver

Moreover, the embodiment of the present invention provides a RAKE receiver, including:

a multipath search allocation module for receiving signals and performing multipath delay acquisition and multipath delay allocation on the received signals; The method is configured to perform descrambling and despreading operations on the chip-level received signal according to the input multipath delay information and the scrambling code and the spreading code information in the received signal, and obtain the descrambling and despreading symbol-level receiving of each path. a descrambling despreading module of signal y;

a channel estimation module for estimating channel fading conditions of each delay according to the input multipath delay information, and outputting channel fading information corresponding to each path delay;

Each path weighting value calculation module is configured to calculate, according to symbol level user energy from the descrambling despreading module, multipath delay from the multipath search allocation module, each path fading factor from the channel estimation module, and each path position information. Address interference function? _MW and background thermal noise interference function. Then according to the sum R'_; obtain each path weighting factor ω, and output the obtained path weighting factor ω to each path combining module;

Each path combining module performs combining processing on the descrambling and despreading symbol-level received signals y according to the input weighting factors ω, and the combined signal Y is outputted to the demodulation decoding module;

The demodulation decoding module performs corresponding demodulation and decoding operations on the received signal Y, and outputs a signal after demodulation and decoding.

It can be seen from the above technical solution that, in the process of calculating the weighting value of each path, the method of the present invention does not need to calculate the inter-symbol interference function, so compared with the G-RAKE receiver, the operation of calculating the combined weights of the paths can be effectively reduced. Quantity, and the quality of the output signal is almost unaffected; and, the calculation of the weight of each path is independent of the spreading factor, users of different spreading factors can share a RAKE receiver without additional additions for users with different spreading factors Number of RAKE receivers. By adopting the scheme of the invention, the calculation overhead and the implementation complexity can be greatly reduced and the cost can be saved under the premise of outputting the same quality signal as the G-RAKE receiver. BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 is a schematic structural view of a RAKE drop machine or a G-RAKE receiver;

^{2 is a} flowchart showing an implementation of each path weighting value calculation module of a RAKE receiver according to an embodiment of the present invention;

FIG. 3 is a structural diagram of each path weighting value calculation module of a RAKE receiver according to an embodiment of the present invention.

Mode for carrying out the invention

In order to make the objects, technical solutions and advantages of the present invention more comprehensible, the present invention will be further described in detail with reference to the accompanying drawings.

According to the simplified formula (4) proposed by the current method for calculating the weight of each path of the G-RAKE receiver:

= R(n _l:> n ₂ ),

Equation (3) can be reduced to:

+N _Q R _p «-d ₂ ),

Where, for / and /3⁄4 related functions,

d -r

n ₂ =d ₂ -T _q xN, ovs

N, ovs

N _ovs is the sampling rate of the G-RAKE receiver module.

As can be seen from equation (3), among the constituent elements of the matrix ^, Ri _Si and ! The expression of i _{MlM differs} only in the sum of i: the former has less summation than /=. It is precisely because of this difference between R _IS! and , that the calculation of 7? _βί in the formula (5) after the chemical cylinder contains the summation operation of the product of 2N-1 terms. If you think of _iS ,=i? _ww in the implementation of G-RAKE, then the formula

(5) can be further reduced to:

1-1 i-1

η" ^η 2, -·^ ^κ — ^τ ! + N ₀ R (d _x -d ₂ ),

/=0 q=Q

Where + is the symbol level energy. Table 1 shows the calculation amount of each time slot of the G-RAKE receiver, where M is the number of iterations for obtaining the weighting value, and generally takes 1~3; J is the total number of paths of the G-RAKE receiver, Generally, the value is J 2J.

Table 1 72 It can be seen from the data in Table 1 that the calculation of _s/ sum takes up a considerable proportion of the total calculation of G-RAKE, so it is considered that R _IS i = R will be valid during the G-RAKE implementation. The computational complexity of the G-RAKE receiver is reduced, and the implementation complexity is reduced.

According to formula (5), since R _IS1 = R, the time and energy E are obtained. Or the matrix of energy multiplication is equal, so only the total symbol level energy _Er is known without distinguishing the symbol level energy of the target user. And interfere with the user's symbol-level energy E, which can reduce the number of parameters required by the G-RAKE receiver, thereby further reducing its implementation complexity.

In addition, the spreading factor 中 can be extracted, so equation (5) can be rewritten as:

1-1 L-1

R{n _x ,n ₂ )--R _p {d _x -T _l ) R {d ₂ -T _q ) + N,R _p (d _x -d ₂ )

1=0 q=0

L-\ L-\

+ N ₀ R _p (d,-d ₂ )

/=0 3⁄4f=0 N That is, the energy weighting factor of R _MU1 changes from ^ to =, that is, the sum of all user chip-level energies independent of the spreading factor, so the size is independent of the spreading factor.

It can be seen from equation (9) that if R _IS! = R _m! is considered in the implementation of the G-RAKE receiver, the transmission environment of the target user is the same, that is, the multipath delay and the channel fading are the same, all of which The user can share the calculation results of a G-RAKE receiver.

According to the above analysis, the G-RAKE receiver of the embodiment of the present invention adopts the structure shown in Fig. 1, and includes the following modules:

The multipath search distribution module 110 is configured to receive signals, complete multipath delay acquisition and multipath delay allocation of the received signals, and output delay information of each multipath to the descrambling and despreading module respectively. 120. The channel estimation module 130 and the path weighting value calculation module 140; the descrambling and despreading module 120, configured to use the multipath delay information and the received signal according to the input The scrambling code and the spreading code information complete the descrambling and despreading operation on the chip-level received signal, and output the descrambled and despread symbol-level received signal y to each path combining module 150;

The channel estimation module 130 estimates the channel fading condition of each delay according to the input multipath delay information, and outputs channel fading information corresponding to each path delay to each path weighting value calculation module 140;

Each path weighting value calculation module 140 calculates a weighting factor of each path according to the input multipath delay information, each channel fading information, and other external parameters, including the system Gaussian noise power N ₀ and the chip level energy E of the receiving user. ω, and the calculated weighting factor ω is output to each path combining module 150;

Each of the path combining modules 150 performs a combining process on the descrambled and despread path data y according to the input path weighting factor ω, and outputs the resultant signal Y to the demodulation decoding module 160. [ ] ^w indicates a total transposition;

The demodulation decoding module 160 performs a corresponding demodulation and decoding operation on the received signal Y, and outputs a signal after demodulation and decoding.

The implementation process of each path weighting value calculation module is specifically as shown in FIG. 1, and includes the following steps:

Step 201: Calculate each path fading function h according to the input path fading factor g, the multipath position r, and the symbol level energy of the target user using equation (2);

Step 202: Find: From formula (8)

(10) This step is to calculate the multiple access interference function according to the input path fading factor g, the multipath position r, the multipath delays ^, and ^ corresponding to the despreading path and the formula (10). Step 303: Calculate: From formula (9)

R _n . = R _p (d “d ₂ ) (11) This step calculates the background heat according to the formula ( 11 ) according to the autocorrelation function of the shaping filter at the transmitting end, _p , the input multipath delay 4 and d ₂ . Interference function of noise;

Step 204: The sum of all user chip-level energies E _{c T} , which is independent of the spreading factor, and the single-sided power spectral density N of the system Gaussian white noise. And the calculation results of step 202 and step 203, the noise correlation matrix ^ is calculated according to formula (9) _;

Step 205: The path fading function h obtained in step 201 and the noise correlation matrix obtained in step 204 are calculated by using formula (1) to obtain the weighted values of the paths of the G-RAKE receiver.

In order to implement the above method, the structure and connection relationship of each part of each path weighting value calculation module 140 of the RAKE receiver in the embodiment of the present invention are as shown in FIG. 3, and the details are as follows:

Each path fading function calculation unit 141 is configured to receive the target user symbol level energy E from the descrambling and despreading module 120. And the multipath fading factor and the multipath position information r of the channel estimation module 130 calculate the path fading function h according to the formula (2) by using the received parameters, and input the calculated path fading function h into the weighting unit. 145;

The multiple access interference function calculation unit 142 is configured to receive the multipath fading factor & and the multipath position information 信道·, from the channel estimation module 130, and the multipath delay and the multipath search allocation module 110, and use the received The above parameters are calculated according to formula (10), the multiple access interference function R read, and the calculated multiple access interference function is input to the noise correlation matrix calculation unit 144;

The thermal noise interference function calculation unit 143 is configured to receive the multipath delay and the sum of the multipath search allocation module 110, and the autocorrelation function of the shaping filter at the transmitting end, and calculate the background according to the formula (11) by using the received parameters. Thermal noise interference function, and will calculate The resulting interference function of the background thermal noise?,, the input noise correlation matrix calculation unit 144;

The noise correlation matrix calculation unit 144 is configured to: according to the multiple access interference function from the multiple access interference function calculation unit 142, the interference function _n of the background thermal noise of the thermal noise interference function calculation unit 143, The sum of the user chip level energy E ^ and the unilateral power spectral density N of the system Gaussian white noise. , the noise correlation matrix is calculated according to the formula (9), and the resulting noise correlation matrix _{Ru is} input to the weighting value synthesizing unit 145;

The weighting value synthesizing unit 145 is configured to synthesize each path weighting value to by using the formula (1) according to the received path fading function h and the noise correlation matrix, and output the synthesized path weighting values ω to the respective paths. Merge module 150.

The simulation analysis of the G-RAKE receiver of the embodiment of the present invention shows that in the implementation process of the G-RAKE receiver, it is considered that R _1SI = R deletion hardly affects the output result of the G-RAKE receiver. Therefore, in the implementation process of the G-RAKE receiver, it is considered that the calculation amount of the G-RAKE receiver can be effectively reduced without reducing the quality of the output signal, and the number of parameters required for the G-RAKE receiver is reduced. In turn, the implementation complexity of the G-RAKE receiver is reduced; and the calculation of the G-RAKE weighting value is independent of the spreading factor. Users of different spreading factors can share a G-RAKE receiver without having to use different spreading factors. The user additionally increases the number of G-RAKE receivers.

Those skilled in the art should recognize that since the difference in structure between the G-RAKE receiver and the RAKE receiver is mainly due to different path weight value calculation modules, the method for calculating each path weight value proposed by the present invention and used to implement the same The apparatus for calculating the method is applicable not only to a G-RAKE receiver but also to a RAKE receiver having the same or similar structure.

The above is only the preferred embodiment of the present invention, and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection of the present invention. Within the scope.

Claims

Claim

A method for calculating a weight value of each path, the method comprising the steps of: calculating a multiple access interference function _ω according to each path fading factor, each path position information, and a multipath delay corresponding to a despreading path; Interference function of background thermal noise;

The interference function of the multiple access interference function and the background thermal noise is respectively equal to the sum of all user chip-level energies E _c — _T independent of the spreading factor and the single-sided power spectral density N of the system Gaussian white noise. By multiplying the sum to obtain the noise correlation matrix R _u. Calculated correlation matrix R _u ω weighted value of each path according to the obtained noise.

2. The method according to claim 1, wherein the interference function for calculating the background thermal noise is:

The autocorrelation function of the shaping filter at the transmitting end is multiplied by the difference of the multipath delay corresponding to the despreading path to obtain the interference function of the background thermal noise.

The method according to claim 1, wherein the calculating the weight value w according to the obtained noise correlation matrix ^ comprises:

Calculating each path fading function h according to each path fading factor, multipath position information, and symbol level energy of the target user, multiplying the path fading function h by the noise correlation matrix „, and the result is weighted by each path .

The method according to claim 3, wherein the calculating each path fading function h is: . For the goal

The symbol-level energy of the user is the channel fading factor of the /path, ^ is the autocorrelation function of the transmit pulse shaping filter, and is the multipath delay corresponding to the despreading path, which is the multipath transmission of the /path Broadcast delay.

The method according to claim 1, wherein said calculating the multiple access interference function ^/ / is:

According to the formula

1⁄2"d) ^x ( "d)- W- ^r ') ^xi? ; - Calculate the multiple access interference function; where N is the spreading factor of the target user, / and the label of the path, & and respectively are the / path And the fading factor of the first path, and the position information of the first/diameter and the first diameter, respectively, L is the total number of effective paths, and ^ is the multipath delay corresponding to the despreading path.

The method according to any one of claims 1 to 5, wherein the calculating the noise correlation matrix ^ is:

According to the formula ∑∑Sig[ -i? ( - r,)x (c/ ₂ -r ) + N _Q R (d _x -d ₂ ), g=0 N

Calculate the noise correlation matrix ^; where N is the spreading factor of the target user, / and are the labels of the path, & is the path fading factor, and is the multipath position information, L is the total number of effective paths, and is the despreading path Corresponding multipath delay, E. _ _{R & lt} spreading factor is independent of all user chip-level energies and for; ¾ and correlation function, where _{_{η = ά -τ ι - xN ovs}} , N. _Ra is the sampling rate,

7. A device for calculating weight values for each path, located in a RAKE receiver system, the device comprising:

a multiple access interference function calculation unit for receiving each path fading from the RAKE receiver a falling factor, each path position information, and a multipath delay from the multipath search distribution module, calculating a multiple access interference function, and inputting the calculated multiple access interference function into the noise correlation matrix calculation unit;

a thermal noise interference function calculating unit, configured to receive a multipath delay of the RAKE receiver, and calculate an interference function of the background thermal noise according to the multipath delay and a pre-determined autocorrelation function of the shaping filter of the transmitting end And the calculated interference function of the background thermal noise

R, input noise correlation matrix calculation unit;

a noise correlation matrix calculation unit, configured to calculate a disturbance function of a background thermal noise of the unit according to the multiple access interference function _ω and the thermal noise interference function from the multiple access interference function calculation unit

R , , the sum of all user chip-level energies of the receiver system E^ and the unilateral power spectral density N of the system Gaussian white noise. Calculating a noise correlation matrix ^, and inputting the obtained noise correlation matrix _Ru into a weighted value synthesis unit;

The weighting value synthesizing unit is configured to synthesize the respective path weighting values w according to the received noise correlation matrix, and output the synthesized respective path weighting values ω to the RAKE receiver.

8. The apparatus according to claim 7, wherein the apparatus further comprises path fading function calculation units for receiving target user symbol level energy from the RAKE receiver, and path fading factors of the channel estimation module And each path position information, calculating each path fading function h, and inputting the calculated path fading function h into the weighting value synthesizing unit;

Then, the weighting value synthesizing unit is configured to synthesize the respective path weighting values w according to the path fading function h and the noise correlation matrix, and output the synthesized path weighting values 6) to the RAKE receiver.

9. Apparatus according to claim 7 or claim 8 wherein said RAKE receiver is a universal RAKE receiver.

10. A RAKE receiver comprising for receiving a signal and performing a plurality of received signals a multipath search allocation module for path delay acquisition and multipath delay allocation, configured to perform descrambling on a chip level received signal according to the input multipath delay information and the scrambling code and the spreading code information in the received signal, Despreading operation, obtaining a descrambling and despreading descrambling and despreading module for each of the symbol-level received signals y, for estimating the channel fading of each delay according to the input multipath delay information, outputting and A channel estimation module for channel fading information corresponding to a path delay, wherein the RAKE receiver further comprises:

Each path weighting value calculation module is configured to calculate, according to symbol level user energy from the descrambling despreading module, multipath delay from the multipath search allocation module, each path fading factor from the channel estimation module, and each path position information. The address interference function and the background thermal noise interference function, according to the ? _MW and R _>; obtaining the path weighting factors ω, and outputting the obtained path weighting factors ω to the path combining modules;

The RAKE receiver according to claim 10, wherein the path weight calculation module comprises:

a multiple access interference function calculation unit, configured to calculate a multiple access interference function according to each path fading factor from the channel estimation module, each path position information, and a multipath delay from the multipath search distribution module, and calculate the multiple access Interference function input noise correlation matrix calculation unit;

A thermal noise interference function calculation unit is configured to calculate a disturbance function of the background thermal noise according to the multipath delay from the multipath search distribution module, and calculate the interference of the background thermal noise Function R". input noise correlation matrix calculation unit;

a noise correlation matrix calculation unit, configured to calculate an interference function of a background thermal noise of the unit according to a multiple access interference function and a thermal noise interference function from the multiple access interference function calculation unit

, the sum of the chip-level energy of all users of the receiver system E^ and the unilateral power spectral density N of the system Gaussian white noise. , calculate the noise correlation matrix ?„, and get the resulting noise correlation matrix

R _u input weighted value synthesis unit;

The weighting value synthesizing unit is configured to synthesize the respective path weighting values ω according to the received noise correlation matrix, and output the synthesized path weighting values to the respective path combining modules.

The RAKE receiver according to claim 10, wherein the each path weighting value calculation module further comprises:

Each path fading function calculation unit is configured to receive target user symbol level energy from the descrambling and despreading module, each path fading factor and each path position information of the channel estimation module, calculate each path fading function h, and calculate each calculated The path fading function h inputs a weighting value synthesizing unit; then the weighting value synthesizing unit is configured to synthesize each path weighting value ω according to the path fading function h and the noise correlation matrix, and output the synthesized each to the RAKE receiver The path weighting value ω.

A RAKE receiver according to claim 10, 11 or 12, wherein said RAKE receiver is a general-purpose RAKE receiver.