CN112865892A - Adaptive generalized selection diversity combining method and system based on signal-to-noise ratio sequencing - Google Patents

Abstract

The invention discloses an adaptive generalized selection diversity combining method and system based on signal-to-noise ratio sorting. The method includes: receiving multi-channel branch signals; taking one frame of data for each branch signal, and obtaining the current state of each branch signal through analysis Part of the information of the valid data of the frame is used as the pre-judgment information; according to the signal-to-noise ratio sorting result of the branches participating in the merging of the previous frame, several branches are selected as the first branch set, and each branch in the first branch set is calculated in turn according to the pre-judgment information of the current frame. According to the signal-to-noise ratio of these branches, the current frames of the branches that meet the conditions of diversity merging are stored in the set, and the SNR sorting results of the branches participating in the merging of the current frame are recorded; the maximum ratio is used for the data frames in the set. Merge criteria for diversity merging. The invention greatly reduces the processing time required by the system and the complexity and power consumption of the system calculation, is simple and effective in implementation, and has better adaptability to changes in the channel environment.

Description

Adaptive generalized selection diversity combining method and system based on signal-to-noise ratio ranking

technical field

The present invention relates to the field of communication systems, in particular to an adaptive generalized selection diversity combining method and system based on signal-to-noise ratio ranking.

Background technique

Diversity technology can effectively reduce the influence of noise and interference on the signal, thereby improving the signal quality. The schematic diagram of sending and receiving is shown in Figure 1.

Different algorithms of diversity combining techniques that can be used at the receiving end include maximum ratio combining (MRC), equal gain combining (EGC), selective combining (SC), handover resident combining (SSC), generalized selective combining (GSC), adaptive combining Methods such as generalized selection combining (A-GSC).

Maximum Ratio Combining (MRC) is an algorithm for selecting combining weights by maximizing the signal-to-noise ratio of combined signals as a combining criterion. The weight coefficient of Maximum Ratio Combining (MRC) is positively correlated with the signal amplitude and inversely correlated with the noise power. Therefore, the essence of the Maximum Ratio Combining (MRC) is to give a larger weight coefficient to the branch with good channel conditions, and vice versa. Give a smaller weighting factor. Therefore, the maximum ratio combining (MRC) enhances the useful signal and weakens the noise and interference, so that the received signal can be enhanced. Assuming that the noise components are independent of each other, the combined signal-to-noise ratio of maximum ratio combining (MRC) is the sum of the signal-to-noise ratios of all combining branches.

The generalized selection merging is to select the L/ _C branches with the best performance among all the L branches for maximum ratio (MRC) or equal gain merging (EGC), which reduces the system complexity while ensuring a certain performance. The complexity is much lower than maximum ratio combining (MRC). However, in this solution, some weak branches may be merged because the merge branch L _C is too large, or some strong branches may be ignored due to the merge branch L _C being too small, which affects the system performance. The selection of the merging branch _LC can only be set in advance, and cannot be flexibly changed according to the channel conditions, and lacks adaptability.

Adaptive Generalized Selective Merging (A-GSC), which continuously adds the strongest branch to achieve a predetermined communication quality, until it is estimated that adding the next branch will not compensate for the gap from expected performance. Similar to the known generalized selective combining (GSC) reception scheme, this reception scheme periodically inserts short training patterns into the transmit signal. During this time, the receiver performs all necessary operations to achieve an appropriate diversity combining scheme. By employing an adaptive generalized selective combining (A-GSC) scheme, the receiver attempts to increase the combined signal _- to-noise ratio _γc above a threshold _γT , which determines the overall quality of the communication and is usually dynamically adjusted as needed. However, the threshold _γT is not the only criterion for deciding the number of merged branches, once the receiver estimates that adding the next branch does not compensate for the gap from the expected performance, it stops the selection process and follows the maximum ratio merge (MRC) rule Merge the selected branch. However, this scheme needs to estimate the signal-to-noise ratio of all diversity branches in the training mode period of each frame. When the training mode interval is much smaller than the channel variation, it will waste a lot of computing resources and power consumption.

SUMMARY OF THE INVENTION

The purpose of the present invention is to overcome the defects of the prior art, and propose an adaptive diversity combining method and system based on SNR ranking.

In order to achieve the above object, the present invention proposes an adaptive diversity combining method based on signal-to-noise ratio ranking, and the method includes:

receive multiple branch signals;

Take one frame of data for each branch signal, and obtain partial information of the current frame valid data of each branch signal through analysis as pre-judgment information;

According to the SNR sorting result of the branches participating in the merge in the previous frame, several branches are selected as the first branch set, and the SNR of each branch in the first branch set is calculated in turn according to the pre-judgment information of the current frame. ratio, save the current frame of the branch that meets the conditions of diversity merging into the set, and record the SNR sorting result of the branch that the current frame participates in the merging;

Diversity merging is performed on the data frames in the set using the maximum ratio merging criterion.

As an improvement of the above method, the method also includes:

When the received data frame is the first frame, the signal-to-noise ratio of each branch is calculated separately from the pre-judgment information of the frame;

Sort according to the signal-to-noise ratio, select several branches that meet the preset conditions, merge according to the adaptive generalized selection and merge method, and record the signal-to-noise ratio sorting result of the branches participating in the merge in the first frame.

As an improvement of the above method, the signal-to-noise ratio of each branch is calculated separately from the pre-judgment information of the frame; specifically, it includes:

Perform time domain-frequency domain transformation on the pre-judgment information of the first frame of each branch;

The second-order moment M ₂ is calculated for the transformed prediction information as:

Among them, y _n is the pre-judgment information in the frequency domain, n is the length of the pre-judgment information, and E is the operator for obtaining the mean value;

Calculate the fourth moment M ₄ as:

Calculated from the second-order moment M ₂ and the fourth-order moment M ₄ , the estimated value s ₁ of the pre-judgment information power of this branch is:

From the second-order moment M ₂ and the estimated signal power value s ₁ , the estimated value s ₂ of the pre-judgment information noise power of this branch is calculated as:

s ₂ =M ₂ -s ₁

From the estimated value of signal power s ₁ and the estimated value of noise power s ₂ , the estimated value ρ of the pre-judgment information signal-to-noise ratio of the branch is calculated as:

As an improvement of the above method, the sorting according to the signal-to-noise ratio, selecting several branches that meet the preset conditions, merging according to the adaptive generalized selection merging method, and recording the signal-to-noise ratio sorting result of the branches participating in the merge in the first frame ; specifically:

Sort according to the numerical value of the signal-to-noise ratio of each branch to obtain the sorting result of the branch signal-to-noise ratio;

Take each branch in turn from the branch SNR sorting result for judgment, if the branch satisfies the preset conditions and the SNR is less than or equal to the preset threshold γ _T , then the first branch of the branch is selected according to the adaptive generalized selection and merging method. frames are merged;

After each branch in the branch signal-to-noise ratio sorting result is judged, record the branch signal-to-noise ratio sorting result of the first frame participating in the merging.

As an improvement of the above method, according to the SNR sorting result of the branches participating in the merge in the previous frame, several branches are selected as the first branch set, and the pre-judgment information of the current frame is used to calculate the number of each branch in the first branch set in turn. Signal-to-noise ratio, according to the signal-to-noise ratio of these branches, save the current frames of the branches that meet the conditions of diversity merging into the set, and record the SNR sorting results of the branches participating in the merging of the current frame; specifically:

Step 1) according to the branch signal-to-noise ratio sorting result of the previous frame participating in the merge, select several branches as the first branch set;

Step 2) Take out the branch with the largest residual signal-to-noise ratio from the first branch set;

Step 3) calculate the signal-to-noise ratio of the current frame of this branch by the pre-judgment information;

Step 4) judge whether the signal-to-noise ratio of the current frame of this branch meets the preset condition, if it meets, store this frame in the set, and go to step 5); if not, then go to step 6);

Step 5) Sum the signal-to-noise ratio of the current frame of the branch in the set to obtain the output signal-to-noise ratio, and judge whether the output signal-to-noise ratio is greater than the threshold value, if yes, go to step 7), if no, go to step 6) ;

Step 6) judge whether the branch of the first branch set has been traversed, if yes, then go to step 7), if no, then go to step 2);

Step 7) judge whether all branches are screened, if yes, go to step 8); if no, select the remaining branches at random, and go to step 3);

Step 8) Use the maximum ratio merging rule to merge the data frames in the set, and record the branch SNR sorting result of the current frame participating in the merging.

As an improvement of the above method, it is described whether the signal-to-noise ratio of the current frame of the branch meets the preset condition, and if so, the frame is stored in the set; specifically:

Take the signal-to-noise ratio of each branch from the sorting result of the branch signal-to-noise ratio and multiply it by the given μ value, μ∈(0,1), and judge whether the product is greater than γ ₍₁₎ and the signal-to-noise ratio is less than or equal to the set value. Threshold γ _T , if it matches, the frame data is stored in the set; where γ ₍₁₎ is the maximum SNR in the sorted set of branch SNRs participating in the merge of the previous frame.

An adaptive diversity combining system based on signal-to-noise ratio sorting, the system comprises: a receiving module, an acquiring module, a judging module and a diversity combining module; wherein,

The receiving module is used for receiving multiple branch signals;

The obtaining module is used to obtain one frame of data for each branch signal, and obtain part of the information of the current frame valid data of each branch signal through analysis as prejudgment information;

The judging module is used to select several branches as the first branch set according to the SNR sorting result of the branches participating in the merge in the previous frame, and sequentially calculate the SNR of each branch in the first branch set according to the pre-judgment information of the current frame , according to the signal-to-noise ratios of these branches, store the current frames of the branches that meet the conditions of diversity merging into the set, and record the results of the SNR sorting of the branches that the current frame participates in the merging;

The diversity merging module is used to perform diversity merging on the data frames in the set using the maximum ratio merging criterion.

Compared with the prior art, the advantages of the present invention are:

1. While the performance of the present invention is not far from the adaptive generalized selection combining (A-GSC) scheme, it can reduce the unnecessary calculation of the signal-to-noise ratio of the diversity branch, especially when the training mode interval is much smaller than the channel variation. When the system is used, the processing time required by the system and the complexity and power consumption of the system calculation are greatly reduced, the implementation is simple and effective, and it has better adaptability to changes in the channel environment;

2. In practical applications, if the transmitting end or the receiving end is in a state of sometimes moving and sometimes static, the channel is often switched between slow fading and fast fading. The diversity combining method in the present invention is not only suitable for slow fading environments, but also It is suitable for fast fading environment and has better adaptability. Especially in slow fading environment, the training mode interval is much smaller than the change of the channel. Refer to the signal-to-noise ratio of the diversity branch in the previous frame to select this frame to participate in The merged diversity branch prioritizes the stronger branches of the previous frame, which can reduce the consumption of a lot of unnecessary computing resources while having little impact on performance.

Description of drawings

FIG. 1 is a schematic diagram of sending and receiving according to Embodiment 1 of the present invention;

2 is a flowchart of an adaptive diversity combining method according to Embodiment 1 of the present invention;

Fig. 3 is the IEEE 802.11a physical layer protocol data unit (PPDU) frame structure used in Embodiment 1 of the present invention;

FIG. 4 is a composition diagram of an adaptive diversity combining system according to Embodiment 2 of the present invention.

reference number

201, receiving unit 202, acquiring unit 203, calculating unit

204, the first judgment unit 205, the second judgment unit 206, the third judgment unit

207, the fourth judgment unit 208, the diversity merging unit

Detailed ways

The technical solutions of the present invention will be described in detail below with reference to the accompanying drawings and embodiments.

Example 1

As shown in FIG. 2 , Embodiment 1 of the present invention provides an adaptive diversity combining method based on signal-to-noise ratio ranking.

The implementation steps of the adaptive generalized selection diversity combining method are divided into a training mode and a data mode, and the training mode and the data mode are not divided by the processed data but by the processing state. The training mode performs operations such as judgment, signal-to-noise ratio estimation, and screening of antenna branches participating in the merger, while the data mode performs MRC merger according to the information obtained by the training mode. Specific steps are as follows:

Training mode:

(1) Receive multi-channel diversity signals;

(2) The first frame is merged according to the adaptive generalized selection combining (A-GSC) method;

(3) In the training mode of the remaining frames, the diversity branches participating in the merging of the previous frame are sorted from large to small according to the size of the signal-to-noise ratio of the previous frame;

(4) According to the sorting result, calculate the signal-to-noise ratio of the first branch (that is, the strongest branch of the previous frame) in this frame, and judge whether it satisfies the preset condition;

(5) Determine whether the output signal-to-noise ratio is greater than the set threshold γ _T , where the output signal-to-noise ratio is the sum of the signal-to-noise ratios of the diversity branches participating in the merging; The output signal-to-noise ratio is greater than the threshold γ _T ;

(6) If the signal-to-noise ratio is still less than the threshold γ _T , the remaining unscreened diversity branches are randomly selected for estimation and judgment, until the output signal-to-noise ratio is greater than the threshold γ _T or all diversity branches are screened;

Data schema:

(7) Merge the diversity branches that pass the screening according to the maximum ratio combining (MRC) criterion.

Wherein the step of judging whether the diversity branch satisfies the preset condition in (4) comprises:

(a) Set a value of μ, where μ∈(0, 1);

(b) Set the value of γ ₍ 1) as the maximum value of the SNR in the diversity branch participating in the merging in the previous frame, if it is the first frame, then γ ₍₁₎ is the SNR in the diversity branch participating in the merging in the first frame. the maximum value of ;

(c) judging whether the product of the signal-to-noise ratio of the diversity branch and the given μ value is greater than γ ₍₁₎ , if it is greater than the diversity branch satisfies the preset condition;

The first frame is merged according to the Adaptive Generalized Selection Combining (A-GSC) method;

The multipath diversity signal consists of reference signal and non-reference signal. Among them, the reference signal is a sequence known to both the receiving and transmitting ends and used for auxiliary receiving (channel estimation calculation), also known as the training sequence; the non-reference signal is sent by the transmitting end, and requires the receiving end to pass equalization and demodulation. , decoding and other processes to obtain the data sequence.

Perform the following operations in the training mode of the remaining frames; as shown in Figure 2:

S101. Receive a multi-channel diversity signal;

S102, sorting the branches participating in the merging of the previous frame according to their signal-to-noise ratios in the previous frame from large to small;

S103, according to the sorting result, take out the remaining strongest branches that have not been screened;

S104, calculate the signal-to-noise ratio of the branch;

This scheme uses the effective data part in the frame as the pre-judgment information to estimate the signal-to-noise ratio: using the second-order fourth-order moment method to estimate the SNR calculate. First calculate the second moment:

Wherein y _n is the received valid data part with the cyclic prefix removed and after the time domain-frequency domain transformation, the range of n is the length of the valid data part, and E is the mean value.

Then calculate the fourth moment:

The estimated signal power is:

The noise power estimate is:

s ₂ =M ₂ -s ₁

The estimated signal-to-noise ratio is:

S105, determine whether the calculated signal-to-noise ratio of the branch satisfies the preset condition, if yes, go to S106; if not, go to S108;

S106, adding the branch to the diversity merge;

S107, determine whether the output signal-to-noise ratio is greater than the set threshold, if so, go to S111; if not, go to S108;

S108, determine whether all the branches participating in the sorting are all screened, if so, go to S109; if not, go to S103;

S109, determine whether all the branches are screened, if yes, go to S111; if not, go to S110;

S110, randomly select the remaining unscreened branches, and perform S104;

S111. Use the MRC criterion to perform diversity merging;

Among them, the MRC criterion is the maximum ratio combining criterion.

The method of the present invention uses the situation of the diversity branches selected in the previous frame to participate in the merging as a reference to screen the diversity branches of the current frame, prioritizes the stronger branches of the previous frame, adds the branches that meet the preset conditions to the merging, and outputs the When the signal-to-noise ratio exceeds the set threshold, the screening is stopped, and then diversity merging is performed according to the MRC criterion.

Calculate the product of the diversity branch signal-to-noise ratio and the given μ value, if the product value is greater than γ ₍₁₎ , it is judged that the preset condition is met; if the product value is less than or equal to γ ₍₁₎ , it is judged that the pre-determined condition is not met. set conditions;

Specifically, μ is a predetermined value, and generally set μ∈(0,1); γ ₍₁₎ is the maximum value of the signal-to-noise ratio in the diversity branch participating in the merging in the previous frame, if it is the first frame, γ _{( 1)} is the maximum value of the signal-to-noise ratio in the diversity branch that the first frame participates in merging;

Possible alternatives: set a value, and recalculate the signal-to-noise ratio of all diversity branches every time the number of received frames reaches this value to filter the branches participating in the merge, and then continue to follow the method of the scheme in the next frame The value can be dynamically changed according to the actual situation. Other methods can also be used for the calculation of the signal-to-noise ratio.

As shown in FIG. 3 , it is the frame format of IEEE 802.11a, and the data domain part of the frame is selected here as the pre-judgment information, thereby calculating the signal-to-noise ratio.

It should be noted that this embodiment uses the frame format of IEEE 802.11a and the second-order fourth-order moment method for SNR estimation as an example to illustrate the method, but it is not limited to this protocol, nor is it limited to the second-order fourth-order moment method The signal-to-noise ratio estimation method based on the above method satisfies this method as long as it is multi-antenna reception.

Example 2

Based on the above method, Embodiment 2 of the present invention proposes an adaptive diversity combining system based on SNR ranking. The system includes: a receiving module, an acquiring module, a judging module and a diversity combining module; as shown in Figure 4, wherein,

The receiving module is used for receiving multiple branch signals; it is composed of a receiving unit 201;

The acquisition module is used to obtain a frame of data for each branch signal, and obtain part of the information of the current frame valid data of each branch signal through analysis as the pre-judgment information; it is composed of an acquisition unit 202;

The judgment module is used to select a number of branches as the first branch set according to the SNR sorting result of the branches participating in the merge in the previous frame, and from the first branch set, according to the pre-judgment information of the current frame, sequentially calculate the SNR of the corresponding branch , according to the signal-to-noise ratios of these branches, store the current frames of the branches that meet the conditions of diversity merging into the set, and record the branch SNR sorting results of the current frames participating in the merging; specifically including the calculation unit 203, the first judgment unit 204, the first The second judging unit 205, the third judging unit 206 and the fourth judging unit 207:

The calculation unit 203 is used to calculate its signal-to-noise ratio according to the diversity signal; if it is the first frame, calculate the signal-to-noise ratio of all the diversity branches and sort them in descending order;

The first judgment unit 204 is used for judging whether the signal-to-noise ratio of the diversity branch satisfies the preset condition; if the judgment result is yes, then the branch is added to the diversity combining unit and enters the second judgment unit 205; if the judgment result is no, the The branch does not join the diversity combining unit, and enters the third judgment unit 206;

The second judgment unit 205 is used to judge whether the output signal-to-noise ratio exceeds the threshold γ _T , if the judgment result is yes, then enter the diversity combining unit 208; if the judgment result is no, enter the third judgment unit 206;

The third judging unit 206 is used for judging whether all the diversity branches participating in the merging of the previous frame have been screened out, and if the judgment result is yes, the fourth judgment unit is entered; if the judgment result is no, the calculation unit 203 is entered;

The fourth judging unit 207 is used to judge whether all the diversity branches have been screened, and if the judgment result is yes, then enter the diversity merging unit 208; if the judgment result is no, then randomly select a diversity branch that has not been screened to enter the calculation unit 203;

The diversity merging module is used to perform diversity merging on the data frames in the set using the maximum ratio merging criterion, and is composed of a diversity merging unit 208 .

The channel condition may not change much within several frames of signal transmission, and the signal-to-noise ratio of the diversity branch of this frame may be the same or not different from the previous frame, so the diversity branch participating in the merging is also the same as the previous frame. , there is no need to recalculate the signal-to-noise ratio of all diversity branches every frame like the A-GSC scheme, especially when the training mode interval is much smaller than the channel variation, this scheme can achieve a performance that is not much different from the A-GSC scheme. At the same time, the number of branches that need to calculate the signal-to-noise ratio and the processing time of the system in the training mode are significantly reduced, saving the consumption of computing resources; and the use of this merging scheme can eliminate branches that have little gain in system performance, and has better self-adaptation performance, reducing the power consumption of the system.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the embodiments, those of ordinary skill in the art should understand that any modification or equivalent replacement of the technical solutions of the present invention will not depart from the spirit and scope of the technical solutions of the present invention, and should be included in the present invention. within the scope of the claims.

Claims (7)

1. An adaptive generalized selection diversity combining method based on signal-to-noise ratio ranking, the method comprising: receive multiple branch signals; Take one frame of data for each branch signal, and obtain partial information of the current frame valid data of each branch signal through analysis as pre-judgment information; According to the SNR sorting result of the branches participating in the merge in the previous frame, several branches are selected as the first branch set, and the SNR of each branch in the first branch set is calculated in turn according to the pre-judgment information of the current frame. ratio, save the current frame of the branch that meets the conditions of diversity merging into the set, and record the SNR sorting result of the branch that the current frame participates in the merging; Diversity merging is performed on the data frames in the set using the maximum ratio merging criterion. 2. The adaptive generalized selection diversity combining method based on signal-to-noise ratio sorting according to claim 1, wherein the method further comprises: When the received data frame is the first frame, the signal-to-noise ratio of each branch is calculated separately from the pre-judgment information of the frame; Sort according to the signal-to-noise ratio, select several branches that meet the preset conditions, merge according to the adaptive generalized selection and merge method, and record the signal-to-noise ratio sorting result of the branches participating in the merge in the first frame. 3. The adaptive generalized selection diversity combining method based on signal-to-noise ratio sorting according to claim 2, wherein the signal-to-noise ratio of each branch is calculated separately from the pre-judgment information of the frame; specifically comprising: Perform time domain-frequency domain transformation on the pre-judgment information of the first frame of each branch; The second-order moment M ₂ is calculated for the transformed prediction information as:

Among them, y _n is the pre-judgment information in the frequency domain, n is the length of the pre-judgment information, and E is the operator for obtaining the mean value; Calculate the fourth moment M ₄ as:

Calculated from the second-order moment M ₂ and the fourth-order moment M ₄ , the estimated value s ₁ of the pre-judgment information power of this branch is:

From the second-order moment M ₂ and the estimated signal power value s ₁ , the estimated value s ₂ of the pre-judgment information noise power of this branch is calculated as: s ₂ =M ₂ -s ₁ From the estimated value of signal power s ₁ and the estimated value of noise power s ₂ , the estimated value p of the pre-judgment information SNR of this branch is calculated as:

4. The adaptive generalized selection diversity combining method based on signal-to-noise ratio sorting according to claim 3, wherein the sorting is performed according to the signal-to-noise ratio, several branches that meet preset conditions are selected, and the adaptive generalized selection is performed according to the The merging method is used to merge, and record the SNR sorting results of the branches participating in the merging of the first frame; specifically, it includes: Sort according to the numerical value of the signal-to-noise ratio of each branch to obtain the sorting result of the branch signal-to-noise ratio; Take each branch in turn from the branch SNR sorting result for judgment, if the branch satisfies the preset condition and the SNR is less than or equal to the preset threshold γT, then select the first frame of the branch according to the adaptive generalized selection and merging method. to merge; After each branch in the branch signal-to-noise ratio sorting result is judged, record the branch signal-to-noise ratio sorting result of the first frame participating in the merging. 5. The adaptive generalized selection diversity combining method based on signal-to-noise ratio sorting according to claim 1, wherein, according to the signal-to-noise ratio sorting result of the branches participating in the merger of the previous frame, several branches are selected as the first Branch set, calculate the signal-to-noise ratio of each branch in the first branch set in turn according to the prediction information of the current frame. Merged branch SNR sorting results; specifically include: Step 1) according to the branch signal-to-noise ratio sorting result of the previous frame participating in the merge, select several branches as the first branch set; Step 2) Take out the branch with the largest residual signal-to-noise ratio from the first branch set; Step 3) calculate the signal-to-noise ratio of the current frame of this branch by the pre-judgment information; Step 4) judge whether the signal-to-noise ratio of the current frame of this branch meets the preset condition, if it meets, store this frame in the set, and go to step 5); if not, then go to step 6); Step 5) Sum the signal-to-noise ratio of the current frame of the branch in the set to obtain the output signal-to-noise ratio, and judge whether the output signal-to-noise ratio is greater than the threshold value, if yes, go to step 7), if no, go to step 6) ; Step 6) judge whether the branch of the first branch set has been traversed, if yes, then go to step 7), if no, then go to step 2); Step 7) judge whether all branches are screened, if yes, go to step 8); if no, select the remaining branches at random, and go to step 3); Step 8) Use the maximum ratio merging rule to merge the data frames in the set, and record the branch SNR sorting result of the current frame participating in the merging. 6. The adaptive generalized selection diversity combining method based on signal-to-noise ratio sorting according to claim 5, wherein the said judgment is whether the signal-to-noise ratio of the current frame of the branch meets a preset condition, and if so, the frame is Stored in the collection; specifically: Take the signal-to-noise ratio of each branch from the sorting result of the branch signal-to-noise ratio and multiply it by the given μ value, μ∈(0,1), and judge whether the product is greater than γ ₍₁₎ and the signal-to-noise ratio is less than or equal to the set value. Threshold γ _T , if it matches, the frame data is stored in the set; where γ ₍₁₎ is the maximum SNR in the sorted set of branch SNRs participating in the merge of the previous frame. 7. An adaptive generalized selection diversity combining system based on signal-to-noise ratio sorting, characterized in that the system comprises: a receiving module, an acquiring module, a judging module and a diversity combining module; wherein, The receiving module is used for receiving multiple branch signals; The obtaining module is used to obtain one frame of data for each branch signal, and obtain part of the information of the current frame valid data of each branch signal through analysis as prejudgment information; The judging module is used to select several branches as the first branch set according to the SNR sorting result of the branches participating in the merge in the previous frame, and sequentially calculate the SNR of each branch in the first branch set according to the pre-judgment information of the current frame , according to the signal-to-noise ratios of these branches, store the current frames of the branches that meet the conditions of diversity merging into the set, and record the results of the SNR sorting of the branches that the current frame participates in the merging; The diversity merging module is used to perform diversity merging on the data frames in the set using the maximum ratio merging criterion.

Images