CN103326743A - Method and device for acquiring merging weight - Google Patents

Abstract

The present invention provides a method and device for acquiring combined weights, wherein the method includes: selecting a time period for calculating channel estimation of the channel; calculating the channel estimation of the channel in the time period; An interpolation operation is performed on the channel estimate, and a combination weight of the receiver is obtained according to the channel estimate obtained after the interpolation operation. Adopting the above-mentioned technical solution provided by the present invention solves the technical problems in related technologies such as the slow updating speed of combining weights and the inability to accurately reflect channel changes, improves the updating speed of combining weights, and makes the granularity of combining weights smaller, thereby achieving In order to resist the fast time-varying fading channel and a certain degree of residual frequency offset, the bit error rate of the receiver is reduced.

Description

Method and device for acquiring combined weight

technical field

The present invention relates to the field of communication, in particular, to a method and device for acquiring combining weights.

Background technique

In CDMA systems, Rake receivers are widely used. The Rake receiver performs descrambling and despreading on the searched multipath, and then performs maximum ratio combination with a certain weight. The usual combination weight is the conjugate of the channel estimate of a certain multipath. With the development of advanced receiver technology, combined multipath is no longer limited to the searched energy path, but also has a certain number of noise paths. At this time, the combined weight is no longer channel estimation, but obtained based on a certain algorithm, such as minimum mean square error (MMSE) or advanced Rake receiver (Arake). The Arake receiver is an extension of the Rake receiver. The number of multipaths it combines is more than the number of real multipaths, and the extra part is called the noise path. Of course, the pooling weights are still based on channel estimates. Whether it is a Rake receiver or an advanced receiver, the calculation of combining weights is very important, which determines the final bit error rate.

In order to combat noise, the channel estimation is generally the result of averaging over a period of time, and the averaged channel estimation will be applied to this period of time. When the channel is rapidly time-varying, or has a large residual frequency offset, the channel response in the average time period is also time-varying. Applying a constant channel estimate to this time period is inappropriate and does not accurately reflect channel variation. The bit error rate will increase when the channel is rapidly time-varying, such as in-vehicle mobile stations. When there is a large residual frequency offset, the rotation of the constellation diagram is caused.

For the above-mentioned problems in related technologies, no effective solution has been proposed yet.

Contents of the invention

Aiming at the technical problems existing in related technologies such as slow update speed of combining weights and inability to accurately reflect channel changes, the present invention provides a method and device for acquiring combining weights to at least solve the above technical problems.

According to one aspect of the present invention, there is provided a method for acquiring combined weights, including: selecting a time period for calculating the channel estimate of the channel; calculating the channel estimate of the channel in the time period; An interpolation operation is performed on the estimate, and a combination weight of the receiver is obtained according to the channel estimate obtained after the interpolation operation.

The above-mentioned receiver is a Rake receiver; the combination weight of the receiver is obtained according to the channel estimate obtained after the interpolation operation, including: performing a conjugate operation on the channel estimate obtained after the interpolation operation to obtain the Rake receiver The combined weight of .

The above-mentioned receiver is an advanced Rake Arake receiver; before performing an interpolation operation on the channel estimate, the above-mentioned method further includes: obtaining an initial combination weight of the Arake receiver;

Obtaining the combination weight of the receiver according to the channel estimate obtained after the interpolation operation includes: using the initial combination weight as an initial value, and using the channel estimate obtained after the interpolation operation as the input of the adaptive filter, and performing the above-mentioned The initial combination weight is adjusted to obtain the combination weight of the Arake receiver after the interpolation operation.

Obtain the combined weight of the Arake receiver after the interpolation operation through the following iterative process, including:

w(0)

e(n)=d(n)-w ^H (n)u(n)

w(n+1)=w(n)+k(n)e ^* (n), where w(0) is the initial combined weight, w(n) is the combined weight after the nth iteration, and w ^H (n) represents the conjugate transpose of w(n), u(n) is the channel estimate obtained after the interpolation operation, d(n) is the expected value of the adaptive filter output, and e(n) is The difference between the adaptive filter output and the expected value is an error, e(n) represents the conjugate of e(n), and k(n) is a weight update vector.

During the iterative process, the method further includes: when the error value in the current iterative process is greater than the error value in the previous iterative process, regenerating the weight update vector so that the iterative process tends to converge.

When the iterative process satisfies one of the following conditions, exit the iterative process: the number of iterations of the iterative process reaches a preset threshold value; the error value satisfies a predetermined condition for exiting the iterative process.

The above channel changes linearly within the selected time period.

Performing an interpolation operation on the channel estimate includes: performing an interpolation operation on the channel estimate by using a linear spline interpolation algorithm.

According to another aspect of the present invention, there is provided a combination weight acquisition device, which is located in the receiver and includes: a selection module, used to select a time period for calculating the channel estimate of the channel; a calculation module, used to calculate the The channel estimate of the channel in the time period; the interpolation module is used to interpolate the channel estimate; the acquisition module is used to obtain the combination weight of the receiver according to the channel estimate obtained after the interpolation operation

The acquisition module is further configured to perform a conjugate operation on the channel estimate obtained after the interpolation operation to obtain the combination weight of the Rake receiver.

The above acquisition module includes: a first acquisition unit, used to acquire the initial combination weight of the advanced Rake Arake receiver before performing interpolation operation on the channel estimate; a second acquisition unit, used to use the initial combination weight as the initial value, the channel estimate obtained after the interpolation operation is used as the input of the adaptive filter, and the initial combination weight is adjusted to obtain the combination weight of the Arake receiver after the interpolation operation.

Through the present invention, the technical solution of obtaining combining weights based on channel estimation after interpolation is adopted, which solves the technical problems of slow updating speed of combining weights and inability to accurately reflect channel changes in related technologies, and improves the updating speed of combining weights, and The granularity of combining weights is made smaller, so as to be able to resist fast time-varying fading channels and a certain degree of residual frequency offset, and reduce the bit error rate of the receiver.

Description of drawings

The accompanying drawings described here are used to provide a further understanding of the present invention and constitute a part of the application. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations to the present invention. In the attached picture:

FIG. 1 is a flowchart of a method for obtaining a combination weight according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of dividing an average interval for a fast time-varying channel according to a preferred embodiment of the present invention;

Fig. 3 is a schematic diagram of obtaining an average channel estimate in an average interval according to a preferred embodiment of the present invention;

FIG. 4 is a schematic diagram of the principle of adaptive filtering according to a preferred embodiment of the present invention;

5 is a schematic diagram of interpolation using a linear spline interpolation algorithm according to a preferred embodiment of the present invention;

FIG. 6 is a schematic diagram of the principle of optimized adaptive filtering based on the embodiment shown in FIG. 4;

FIG. 7 is a structural block diagram of a device for obtaining a combination weight according to an embodiment of the present invention;

Fig. 8 is a structural block diagram based on the acquisition module of the embodiment shown in Fig. 7;

FIG. 9 is a schematic structural diagram of a device for combining weights with a maximum ratio according to a preferred embodiment of the present invention;

Fig. 10 is a schematic diagram of the principle of implementing adaptive filtering by using LMS according to a preferred embodiment of the present invention.

Detailed ways

Hereinafter, the present invention will be described in detail with reference to the drawings and examples. It should be noted that, in the case of no conflict, the embodiments in the present application and the features in the embodiments can be combined with each other.

FIG. 1 is a flow chart of a method for acquiring a combination weight according to an embodiment of the present invention. As shown in Figure 1, the method includes:

Step S102, selecting a time period for calculating the channel estimate of the channel;

Step S104, calculating the channel estimate of the channel in the time period;

Step S106, performing an interpolation operation on the channel estimate, and obtaining a combination weight of the receiver according to the channel estimate obtained after the interpolation operation.

Regardless of whether the channel is fast time-varying or has a large residual frequency offset, the above processing procedure can be used to obtain the combination weight of the receiver. Especially when the channel is fast time-varying or has a large residual frequency offset, it can solve the technical problem that the combination weight cannot accurately reflect the channel change due to the slow update speed of the combination weight, so that the combination weight (such as the maximum ratio combination weight) can be updated faster and more granular. Small, can resist fast time-varying fading channels and a certain degree of residual frequency offset, which reduces the bit error rate of the receiver.

In step S102 and step S104, the following process can be adopted: when the channel is time-varying, as shown in FIG. 2, a certain average interval (ie time period) is selected so that the channel variation in this interval is approximately linear. At this time, the averaged channel estimate reflects the channel response at the center of the average time slot. Based on the assumption of approximate linearity, interpolation is performed on the channel estimate (for example, interpolation is performed by using a linear spline interpolation algorithm) to reduce the granularity of the channel estimate.

In step S106, the channel estimate may be interpolated using the following process:

As shown in Figure 3, assuming that the average window time of the channel estimation is T, then the average result reflects the channel response at the center of the window, so that the center of the window corresponds to the indices n1, n2, n3, and the corresponding channel estimation is w _n1 , w _n2 , w _n3 . Use the linear spline algorithm for interpolation, and let the interpolation position index be n.

If the windows corresponding to n1 and n2 are the first and last averaging windows respectively, then for n<n1 or n>n3, the interpolation belongs to extrapolation, and for n1≤n≤n3, the interpolation belongs to interpolation. Extrapolation belongs to prediction, so the extrapolation method must conform to the known trend, otherwise the error will be relatively large. The known trend is: frequency offset or fading will cause the rotation of channel estimation, but the rotation will not be too large in the average window, and can be approximated by a linear curve.

In step S106, the acquisition schemes of combining weights are different for the Rake receiver (or called the common Rake receiver) and the advanced Rake receiver. in particular,

When the receiver is a Rake receiver, the combination weight may be obtained through the following process: performing a conjugate operation on the channel estimate obtained after the interpolation operation to obtain the combination weight of the Rake receiver. Moreover, since the channel estimate is (approximately) linearly changing, the combining weights of the Rake receiver also satisfy a linear relationship, so that the channel estimate after interpolation can be directly applied to the maximum ratio combining.

When the receiver is an advanced Rake Arake receiver, the above-mentioned combined weights can be obtained through the following process: before interpolation operation is performed on the channel estimate, the initial combined weights of the Arake receiver are obtained; The weight is an initial value, and the channel estimate obtained after the interpolation operation is used as the input of the adaptive filter, and the initial combination weight is adjusted to obtain the combination weight of the Arake receiver after the interpolation operation.

Wherein, the combination weight of the Arake receiver after the interpolation operation can be obtained through the following iterative process, including:

w(0)

e(n)=d(n)-w ^H (n)u(n)

w(n+1)=w(n)+k(n)e ^* (n)

Among them, w(0) is the initial combined weight, w(n) is the combined weight after the nth iteration, w ^H (n) represents the conjugate transpose of w(n), and u(n) is the interpolation The channel estimate obtained after the operation, d(n) is the expected value of the adaptive filter output, and e(n) is the difference between the adaptive filter output and the expected value, which is the error, and e(n) represents The conjugate of e(n), k(n) is the weight update vector. Different adaptive algorithms have different manifestations.

In general, as shown in Figure 4, the input to the adaptive filter is the received training sequence, since the expected value of the training sequence is known. After a certain length of training sequence sliding, the coefficients of the transversal filter tend to be stable. In the above iterative process, the input of the adaptive filter is the channel estimate after interpolation, which is actually the channel impulse response, that is, the training sequence is equivalent to a unit impulse function. After channel fading, the received signal u (that is, the channel estimate obtained after interpolation) is no longer an impulse function, but it is expected to be restored to a unit impulse function after transversal filtering. Therefore, if the power is normalized, the output of w ^Hu u should be equal to 1, and if the power is not normalized, at least the imaginary part of the output value is zero. Therefore, the expected value of the filter output is the real part of w ^H u, and the error value is:

e(n)＝-i*imag(w ^H u)

Where imag represents the imaginary part, that is, the imaginary part in a complex number. "i" is the imaginary unit, which is a square root of "-1".

In the above iterative process, since the weight update vector determines the convergence process of the above iterative process, in order to ensure the convergence speed, the generation of the weight update vector needs to be considered, and the generation of the weight update vector needs to consider the convergence process of the error value. The generation of the initial weight update vector is based on the principle of maximizing the convergence speed. Once the error value in a certain iteration process (or in the current iteration process) is found to be greater than the error value of the previous iteration, that is, it does not converge, it is necessary to regenerate the weight update vector to make the iterative process tend to converge.

In order to control the computational complexity, the iterative process needs to be ended when the output result of the adaptive filter (the channel estimate after interpolation) meets the requirements, and the above iterative process can jump out of the iterative process when one of the following conditions is met: the The number of iterations of the iterative process reaches a preset threshold; the error value satisfies a predetermined condition for exiting the iterative process.

In step S102, in order to make the combination weight also show a linear change, the channel characteristics can change linearly within the selected time period, and the linear change here can be a linear change within a certain allowable offset range, and is not limited to Absolutely linear change.

A variety of interpolation algorithms may be used to perform the interpolation operation on the channel estimate. In a preferred embodiment of the present invention, a linear spline interpolation algorithm is preferably used to perform the interpolation operation on the channel estimate.

As shown in Figure 5, the difference between linear spline interpolation and linear fitting is that the former connects the sampling points into line segments to obtain interpolation points, and extends the line segments outward to obtain extrapolation points. The interpolation algorithm is:

${\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}}{\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; \&Element;</span>\left\{\begin{array}{cccc}\mathrm{<span\; class="notranslate">\; 1</span>}& \mathrm{<span\; class="notranslate">\; 2</span>}& <span\; class="notranslate">\; \&Center\; Dot;</span><span\; class="notranslate">\; \&Center\; Dot;</span><span\; class="notranslate">\; \&CenterDot;</span>& \mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 2</span>}\end{array}\right\}$

${\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 2</span>}}}{\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 3</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; \&Element;</span>\left\{\begin{array}{cc}\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}& <span\; class="notranslate">\; \&CenterDot;</span><span\; class="notranslate">\; \&CenterDot;</span><span\; class="notranslate">\; \&CenterDot;</span>\end{array}\right\}$

In order to better understand the adaptive filtering process (iterative process) in the above embodiment, the principle of the process is briefly described below.

As shown in Figure 6, the optimized adaptive filtering principle is as follows. In this optimization scheme, an adaptive iterative exit mechanism and a weight update vector generation mechanism are added. details as follows:

Step (1): Interpolating the channel estimates, each group of interpolated channel estimates corresponds to an adaptive filter;

Step (2): Use the interpolated channel estimate to calculate the initial weight update vector;

Step (3): Transverse filtering is performed on channel estimates u of all paths at a certain moment, and the filter coefficient is the current combined weight w_now, that is, sum(u.*conj(w_now)); where sum is the sum and conj is the total The yoke operation, ".*" is the dot multiplication operation, that is, the corresponding elements of two vectors are multiplied.

Step (4): get the imaginary part of the result of transverse filtering as error, satisfy e(n)=-i*imag(w ^H u);

Step (5): If the maximum number of iterations or the jumping condition is met (that is, the error satisfies the condition), then jump out of the iteration;

Step (6): If the exit condition is satisfied when exiting the iteration, the final weight value is the weight of the last iteration, otherwise it is the weight value of the previous iteration;

Step (7): If you continue to iterate, then judge whether this iteration diverges. If it diverges, regenerate the weight update vector and ignore this iteration; otherwise, update the weight value;

Step (8): The update of the current combined weight value satisfies the following formula:

w(0)

e(n)=d(n)-w ^H (n)u(n)

w(n+1)=w(n)+k(n)e ^* (n);

Step (9): Repeat steps (3) to (8) if the iteration continues.

In this embodiment, there is also provided a combination weight acquisition device, which is located in the receiver, and is used to implement the above-mentioned embodiments and preferred implementation modes. What has been explained will not be described in detail. to the module for clarification. As used below, the term "module" may be a combination of software and/or hardware that realizes a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementations in hardware, or a combination of software and hardware are also possible and contemplated. Fig. 7 is a structural block diagram of an apparatus for acquiring combination weights according to an embodiment of the present invention. As shown in Figure 7, the device includes:

A selection module 70, connected to the calculation module, for selecting a time period for calculating the channel estimate of the channel;

Calculation module 72, connected to the interpolation module 74, for calculating the channel estimate of the channel in the time period;

An interpolation module 74, connected to the acquisition module 76, for performing an interpolation operation on the channel estimate;

The obtaining module 76 is configured to obtain the combination weight of the receiver according to the channel estimate obtained after the interpolation operation.

As described in the foregoing method embodiments, the acquisition schemes of combining weights are different for a Rake receiver (or called a common Rake receiver) and an advanced Rake receiver. For the Rake receiver, the following solution can be adopted: the acquisition module 76 is also configured to perform a conjugate operation on the channel estimate obtained after the interpolation operation to obtain the combination weight of the Rake receiver.

In order to make the combination weight also show a linear change, within the above-mentioned time period selected by the selection module 70, the channel characteristics can change linearly, and the linear change here can be a linear change within a certain allowable offset range, and is not limited to absolute linear change.

For advanced Rick (Arake) receiver, can adopt following scheme to realize: As shown in Figure 8, above-mentioned acquisition module 76 comprises:

The first obtaining unit 760 is configured to obtain the initial combination weight of the advanced Rake Arake receiver before interpolating the channel estimate;

The second obtaining unit 762 is configured to use the initial combining weight as an initial value, and use the channel estimate obtained after interpolation as the input of the adaptive filter, adjust the initial combining weight, and obtain the interpolated channel estimate The combining weights of the Arake receiver

In order to better understand the above embodiments, detailed descriptions will be given below in conjunction with preferred embodiments and related drawings. It should be noted that, in the following embodiments, it mainly relates to wideband code division multiple access (WCDMA for short) baseband receiver technology, and is particularly related to channel estimation and equalization in a Rake receiver. The following embodiment describes in detail an interpolation scheme of maximum ratio combining weights.

As shown in Figure 9, the device for maximally combining weights includes the following modules:

Control channel descrambling and despreading module 90: used to perform descrambling and despreading processing on the control channel, and each multipath has one descrambling and despreading resource.

Data channel descrambling and despreading module 92: used to perform descrambling and despreading processing on the data channel, each multipath has one descrambling and despreading resource, including energy path and noise path.

Channel estimation module 94: for estimating the channel response of each multipath.

Advanced receiver combination weight calculation module 96: used to calculate the combination weight of energy path and noise path in Arake, and used to calculate all tap coefficients of transversal filter in MMSE receiver. This generally refers to all combined weight calculations different from the Rake receiver, and the calculated combined weight is not equal to the conjugate of the channel estimate.

Maximum ratio combining module 98: performing maximum ratio combining on the multipath symbol-level data that needs to be combined.

Demodulation and decoding module 100: demodulate and decode the maximum ratio combination result to obtain information bits at the sending end.

Channel estimation linear interpolation module 102: Interpolate the averaged channel estimation to obtain channel estimation with smaller granularity.

Adaptive filtering interpolation module 104: take the result of the combined weight calculation of the advanced receiver as the initial value, take the interpolated channel estimate as the input of the adaptive filter, adjust the combined weight through adaptive iteration, and obtain the combined Weights.

The descrambling and despreading module of the control and data channels, the channel estimation and advanced receiver combination weight calculation module, the maximum ratio combination module, and the demodulation and decoding module constitute the necessary modules for the entire channel estimation and equalization, and the channel estimation interpolation and Adaptive filter interpolation is the core module of combined weight interpolation, which is used to improve system performance and resist fast time-varying channels and residual frequency offset.

It should be noted that the channel estimation linear interpolation module 102 in this embodiment is equivalent to the selection module 70, the calculation module 72 and the interpolation module 74 in the above embodiment; the adaptive filter interpolation module 104 in this embodiment is equivalent to The acquisition module 76 in the above embodiment.

The following is based on the principle in the embodiment shown in FIG. 9 to illustrate the processing flow of the maximum ratio combining weight:

Step 1: Descramble and despread the control channel and data channel for all multipaths to obtain symbol-level data.

Step 2: Process the symbol data of the control channel to obtain channel estimates for each path and average them within the averaging window.

Step 3: Calculate the combining weights of the advanced receiver by using the channel statistics obtained in the second step, such as the channel estimate.

Step 4: Perform linear spline interpolation on the channel estimate obtained in the second step to obtain the combined weight of the Rake receiver after interpolation, which satisfies the following formula:

${\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}}{\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; \&Element;</span>\left\{\begin{array}{cccc}\mathrm{<span\; class="notranslate">\; 1</span>}& \mathrm{<span\; class="notranslate">\; 2</span>}& <span\; class="notranslate">\; \&CenterDot;</span><span\; class="notranslate">\; \&CenterDot;</span><span\; class="notranslate">\; \&CenterDot;</span>& \mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 2</span>}\end{array}\right\}$

${\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 2</span>}}}{\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 3</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; w</span>}}_{\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; \&Element;</span>\left\{\begin{array}{cc}\mathrm{<span\; class="notranslate">\; no</span>}\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}& <span\; class="notranslate">\; \&CenterDot;</span><span\; class="notranslate">\; \&CenterDot;</span><span\; class="notranslate">\; \&CenterDot;</span>\end{array}\right\}$

Step 5: Take the interpolated channel estimate obtained in the fourth step as input, and use the combined weight of the advanced receiver obtained in the third step as the initial value, perform adaptive filtering interpolation, and obtain the combined weight of the advanced receiver after interpolation .

Step 6: Perform maximum ratio combining on the symbol data of the data channel that needs to combine multipaths, and the combining weight comes from the fourth or fifth step.

Step 7: Demodulate and decode the combined data to obtain information bits.

The fifth step is further divided into the following steps, as shown in Figure 10, the adaptive filtering is implemented by LMS, and the iterative formula of the LMS adaptive filtering is:

w(0)

e(n)=d(n)-w ^H (n)u(n)

w(n+1)=w(n)+μu(n)e ^* (n)

μ is the iteration step size, and μu(n) constitutes the weight update vector k(n).

Step 1: Interpolate the channel estimates, copy the advanced receiver combining weights to the interpolated granularity, and each group of interpolated channel estimates constitutes an LMS adaptive filter.

Step 2: Use the interpolated channel estimate u to calculate the initial step size, satisfying

$\mathrm{<span\; class="notranslate">\; \&mu;</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; NS</span>}}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; N</span>}<span\; class="notranslate">\; *</span>\mathrm{<span\; class="notranslate">\; mean</span>}<span\; class="notranslate">\; \{</span>{<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; u</span>}<span\; class="notranslate">\; |</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; \}</span><span\; class="notranslate">\; *</span>\mathrm{<span\; class="notranslate">\; 3</span>}}$

Among them, S is the maximum value of the power spectral density of the input u, N is the order of the filter, mean means to take the mean value, and use mean{|u| ² }*3 to approximate S.

Step 3: Transversely filter the channel estimates u of all paths at a certain moment, and the filter coefficient is the current combined weight

Step 4: Take the imaginary part of the result of transversal filtering as the error, satisfying e(n)=-i*imag(w ^H u)

Step 5: If the maximum number of iterations or the exit condition is met, exit the LMS iteration, and the exit condition is

Where th is the threshold and e is the error value. Where real is the real part of the complex number, and imag is the imaginary part.

Step 6: If the exit condition is met when exiting the iteration, the final weight value is the weight of the last iteration, otherwise it is the weight value of the previous iteration

Step 7: If you continue to iterate, then judge whether this iteration diverges, that is, satisfy

|e(n)|＞|e(n-1)|

If it diverges, the step size is divided by 2, and the weight update vector is regenerated; otherwise, the previous weight value and the previous iteration error are updated, that is, the previous weight is equal to the current weight, and the previous iteration error is equal to the current error.

Step 8: The update of the current combined weight value satisfies the following formula:

w(0)

e(n)=d(n)-w ^H (n)u(n)

w(n+1)=w(n)+μu(n)e ^* (n).

Step 9: Repeat steps 3 to 8 if iteration continues.

In another embodiment, software is also provided, and the software is used to implement the technical solutions described in the above embodiments and preferred implementation manners.

In another embodiment, there is also provided a storage medium, in which the software is stored, the storage medium includes but not limited to: optical discs, floppy disks, hard disks, rewritable memories, and the like.

Obviously, those skilled in the art should understand that each module or each step of the above-mentioned present invention can be realized by a general-purpose computing device, and they can be concentrated on a single computing device, or distributed in a network formed by multiple computing devices Alternatively, they may be implemented in program code executable by a computing device so that they may be stored in a storage device to be executed by a computing device, and in some cases in an order different from that shown here The steps shown or described are carried out, or they are separately fabricated into individual integrated circuit modules, or multiple modules or steps among them are fabricated into a single integrated circuit module for implementation. As such, the present invention is not limited to any specific combination of hardware and software.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims (11)

1. A method for obtaining a combined weight, characterized in that, comprising; selecting a time period for calculating a channel estimate for the channel; calculating a channel estimate for the channel during the time period; An interpolation operation is performed on the channel estimate, and a combination weight of the receiver is obtained according to the channel estimate obtained after the interpolation operation. 2. The method according to claim 1, wherein the receiver is a Rake receiver; the combination weight of the receiver is obtained according to the channel estimate obtained after the interpolation operation, comprising: performing a conjugate operation on the channel estimate obtained after the interpolation operation to obtain the combination weight of the Rake receiver. 3. The method according to claim 1, wherein the receiver is an advanced Rake Arake receiver; Before performing an interpolation operation on the channel estimate, the method further includes: obtaining an initial combining weight of the Arake receiver; Obtaining the combination weight of the receiver according to the channel estimate obtained after the interpolation operation includes: using the initial combination weight as an initial value, and using the channel estimate obtained after the interpolation operation as the input of the adaptive filter, and performing the above-mentioned The initial combination weight is adjusted to obtain the combination weight of the Arake receiver after the interpolation operation. 4. method according to claim 3, is characterized in that, obtains the combined weight of described Arake receiver after interpolation operation by following iterative process, comprising: w(0) e(n)=d(n)-w ^H (n)u(n) w(n+1)=w(n)+k(n)e ^* (n) Among them, w(0) is the initial combined weight, w(n) is the combined weight after the nth iteration, w ^H (n) represents the conjugate transpose of w(n), and u(n) is the interpolation The channel estimate that obtains after operation, d (n) is the expected value that described adaptive filter outputs, and e (n) is the error that is the difference between described adaptive filter output and expected value, e ^* (n) Represents the conjugate of e(n), and k(n) is the weight update vector. 5. method according to claim 4, is characterized in that, in described iterative process, described method also comprises: When the error value in the current iteration process is greater than the error value in the previous iteration, regenerate the weight update vector so that the iteration process tends to converge. 6. The method according to claim 4, wherein the iterative process jumps out of the iterative process when one of the following conditions is met: The number of iterations of the iterative process reaches a preset threshold; The error value satisfies a predetermined condition for jumping out of the iterative process. 7. The method according to claim 1, wherein the channel changes linearly within the selected time period. 8. The method according to any one of claims 1 to 7, wherein performing an interpolation operation on the channel estimate comprises: An interpolation operation is performed on the channel estimate by using a linear spline interpolation algorithm. 9. An acquisition device for combining weights, located in a receiver, characterized in that it comprises; A selection module, configured to select a time period for calculating a channel estimate of the channel; a calculation module, configured to calculate a channel estimate of the channel during the time period; an interpolation module, configured to perform an interpolation operation on the channel estimate; The obtaining module is used to obtain the combination weight of the receiver according to the channel estimate obtained after the interpolation operation. 10. The device according to claim 9, wherein the acquisition module is further configured to perform a conjugate operation on the channel estimate obtained after the interpolation operation to obtain a combination weight of the Rake receiver. 11. The device according to claim 9, wherein the acquiring module comprises: The first obtaining unit is used to obtain the initial combination weight of the advanced Rake Arake receiver before performing interpolation operation on the channel estimate; The second acquisition unit is configured to use the initial combination weight as an initial value, and use the channel estimate obtained after the interpolation operation as the input of the adaptive filter, adjust the initial combination weight, and obtain the interpolation operation. Combining weights for the Arake receiver described above.

Images