Multi-channel self-adaptive rate power multiplication module and multi-channel self-adaptive rate power multiplication method
By designing a multi-channel adaptive rate multi-weight module, using adaptive parameter analysis and module combination, general weighting processing for different scenarios is realized, solving the problems of long design cycles of traditional modules and difficulty in resource reuse, and improving processing efficiency and Convenient configuration.
Patent Information
- Application Number
- CN202411391166.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-08
- Publication Date
- 2025-01-03
AI Technical Summary
The traditional multi-channel multi-weight module design needs to be customized according to different application scenarios, resulting in a long design cycle and the inability to multiplex resources in different modes, making it difficult to adapt to changes in high and low data rates and beam counts.
A multi-channel adaptive rate multiplication module is designed, including adaptive parameter analysis module, weight split cache module, AD data cache module, AD data string conversion module, AD data copy module, channel weighting module and folding addition The module realizes universal multi-channel weighting processing by adaptively calculating the number of weight split groups, the number of data copy copies and the number of folding times.
It realizes universal multi-scenario and simple parameter configuration, and can make full use of hardware resources in different working scenarios to improve weighted processing efficiency.
Smart Images
Figure CN119254286A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the field of computer technology, and in particular relates to a multi-channel adaptive rate weighting module and method. Background Art
[0002] In modern communications, radars and other multi-input and multi-output MIMO systems, the front-end data is sampled and packaged by the analog-to-digital converter (AD), and is generally sent to the back-end through optical fiber for multi-channel data adaptive weighted processing, and the weighted module implements multi-channel weighted processing of AD data. In the back-end design, field programmable gate arrays (FPGAs) are generally used to implement multi-channel weighting.
[0003] Traditional multi-channel weighted modules and methods are customized according to data characteristics such as the data rate, number of input paths, number of single channels, and number of output multi-beams of AD data. Traditional multi-channel weighted modules and methods have the following disadvantages:
[0004] (1) The weighting method needs to be customized for different application scenarios. The weighting method for forming multiple beams for low data rates is not applicable to high data rates, and the weighting method for forming fewer beams for high data rates is not applicable to low data rates. This results in a long design cycle for multi-scenario applications.
[0005] (2) For the same system, if high data rate and low data rate scenarios exist simultaneously in different modes, the number of beams in different scenarios is adjusted accordingly, and traditional processing methods cannot reuse resources and are difficult to implement. Summary of the invention
[0006] In order to overcome the deficiencies in the prior art, the present invention proposes a multi-channel adaptive rate weighting module and method, which designs a universal multi-channel data weighting architecture, and can adaptively weight channels according to the data rate, input number of channels, single channel number and output multi-beam number characteristics of the data. It has the characteristics of being universal in various scenarios, simple parameter configuration and convenient application.
[0007] To achieve the above-mentioned purpose, the multi-channel adaptive rate weighted module of the present invention comprises: an adaptive parameter parsing module (1), a weight splitting cache module (2), an AD data cache module (3), an AD data serial-to-parallel conversion module (4), an AD data replication module (5), a channel weighting module (6) and a folding and adding module (7).
[0008] The adaptive parameter analysis module (1) adaptively calculates the number of weight split groups, the number of data replication parameters and the number of folding additions according to the data rate, the number of input paths, the number of single-path channels and the number of output multi-beams.
[0009] The weight splitting cache module (2) splits the pre-calculated weights of each channel according to the weight splitting group number, and the split weights are cached in RAM for subsequent data multiplication processing.
[0010] The AD data buffer module (3) is used to receive the AD data received by the front end, and use fifo to buffer the data. When the data storage capacity reaches a range gate, it starts to read the data and sends it to the AD data serial-to-parallel conversion module (4).
[0011] The AD data serial-to-parallel conversion module (4) converts the AD data into parallel arrangement output according to the number of single-channel channels.
[0012] The AD data replication module (5) replicates the parallel AD data in a number of copies according to the replication number parameter.
[0013] The channel weighting module (6) groups the parallel channel data and weight split data according to the data copy number parameter analyzed by the adaptive parameter analysis module (1), multiplies the data and weight of each group, and then adds them in parallel to obtain a single channel weighted result.
[0014] The folding and adding module (7) performs folding and adding on the weighted results of the N channels according to the folding and adding times parameter analyzed by the adaptive parameter analyzing module (1) to obtain a system channel weighted result.
[0015] The multi-channel adaptive rate weighting method of the present invention comprises the following specific steps:
[0016] (a) The adaptive parameter analysis module (1) receives the data rate R, the number of input paths N, the number of single-path channels I and the number of output multi-beams B, and calculates the number of weight split groups G (G=I*C), the number of data copies C and the number of folding additions F.
[0017] (b) Sending the serial weights to the weight splitting cache module (2) Splitting the serial input weight data into G groups according to the weight splitting group number G, and caching them separately in RAM.
[0018] (c) Waiting for the arrival of AD data, and putting the received AD data into the FIFO of the AD data buffer module (3).
[0019] (d) Determine whether the number of cached AD data reaches the total number of channels of one range gate. If yes, read out the cached AD data of one range gate and send it to the AD data serial-to-parallel conversion module (4) for serial-to-parallel conversion. Otherwise, continue to cache and wait.
[0020] (e) The parallel AD data after serial-to-parallel conversion is sent to the AD data replication module (5) and replicated as C copies.
[0021] (f) The C pieces of parallel data and the G groups of weights are sent to the channel weighting module (6) for parallel multiplication. After the data and weights of each group are multiplied, they are added in parallel to obtain a single channel weighted result.
[0022] (g) The weighted results of the N channels are input to the folding and adding module (7), and the folding and adding are performed according to the folding and adding times F to ensure that the same beams are added correspondingly, and finally the weighted results of the C system channels are output.
[0023] The beneficial effects of the present invention are
[0024] 1. The multi-channel adaptive rate weighting module adopts a parametric design method, which has a wide range of application scenarios and strong universality. By configuring the input data rate, number of input paths, number of single channels and number of beams, it can adaptively generate combined parameters of multiple internal scenarios and complete adaptive channel weighting processing.
[0025] 2. The multi-channel adaptive rate weighted module and method are designed to reuse hardware resources, which can fully utilize the hardware computing power in different working scenarios and improve the weighted processing efficiency.
[0026] 3. Easy configuration and simple application: This method has simple configuration parameters and does not require understanding of the internal processing flow. It can be used by simply configuring the weighted parameters and serial data interface. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] Figure 1 This is a module block diagram of the present invention.
[0028] Figure 2 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION
[0029] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0030] like Figure 1 As shown, a multi-channel adaptive rate weighted module includes: an adaptive parameter parsing module, a weight splitting cache module, an AD data cache module, an AD data serial-to-parallel conversion module, an AD data replication module, a channel weighting module and a folding and adding module.
[0031] The adaptive parameter parsing module adaptively calculates the number of weight split groups, the number of data replication parameters and the number of folding additions according to the data rate, the number of input paths, the number of single channels and the number of output multi-beams.
[0032] The weight splitting cache module splits the pre-calculated channel weights according to the number of weight splitting groups, and the split weights are cached in RAM for subsequent data multiplication processing.
[0033] The AD data buffer module is used to receive the AD data received by the front end, and uses fifo to buffer the data. When the data storage capacity reaches a range gate, it starts to read the data and sends it to the AD data serial-to-parallel conversion module (4).
[0034] The AD data serial-to-parallel conversion module converts the AD data into parallel arrangement output according to the number of single channels.
[0035] The AD data replication module replicates the parallel AD data in several copies according to the replication number parameter.
[0036] The channel weighting module groups the parallel channel data and weight split data according to the data replication number parameter analyzed by the adaptive parameter analysis module (1), multiplies the data and weight of each group and then adds them in parallel to obtain a single channel weighted result.
[0037] The folding and adding module performs folding and adding on the weighted results of the N channels according to the folding and adding times parameter analyzed by the adaptive parameter analyzing module (1) to obtain the system channel weighted results.
[0038] The method steps of the present invention specifically include:
[0039] (a) The adaptive parameter analysis module (1) receives the data rate R, the number of input paths N, the number of single-path channels I, and the number of output multi-beams B, and calculates the number of weight split groups G (G=I*C), the number of data copies C, and the number of folding additions F.
[0040] (b) Sending the serial weights to the weight splitting cache module (2) Splitting the serial input weight data into G groups according to the weight splitting group number G, and caching them separately in RAM.
[0041] (c) Waiting for the arrival of AD data, and putting the received AD data into the FIFO of the AD data buffer module (3).
[0042] (d) Determine whether the number of cached AD data reaches the total number of channels of one range gate. If yes, read out the cached AD data of one range gate and send it to the AD data serial-to-parallel conversion module (4) for serial-to-parallel conversion. Otherwise, continue to cache and wait.
[0043] (e) The parallel AD data after serial-to-parallel conversion is sent to the AD data replication module (5) and replicated as C copies.
[0044] (f) The C pieces of parallel data and the G groups of weights are sent to the channel weighting module (6) for parallel multiplication. After the data and weights of each group are multiplied, they are added in parallel to obtain a single channel weighted result.
[0045] (g) The weighted results of the N channels are input to the folding and adding module (7), and the folding and adding are performed according to the folding and adding times F to ensure that the same beams are added correspondingly, and finally the weighted results of the C system channels are output.
[0046] Combination Figure 1 , Figure 2 The present invention is described in further detail:
[0047] Figure 2 The workflow diagram of the channel weighting method is described, taking the parameters of data rate 10Mbps, input number 32, single channel number 8, and output multi-beam number 48 as an example.
[0048] The adaptive parameter analysis module calculates the number of multi-beams formed in each internal channel according to the 10Mbps data rate table, which is 24. Combined with the output multi-beam number of 48, it can be obtained that the number of data copies is 48 / 24=2, and the number of weight splitting groups is the number of data copies 2*the number of single channels 8=16. Taking the logarithm of 2 for the input number of 32, it can be obtained that the number of folding additions is 5.
[0049] according to Figure 2 The process splits the weights into 16 channels and caches them in RAM. After waiting for the complete range gate data to arrive, the serial AD data is converted into 8-channel serial-to-parallel and copied twice to form 16-channel parallel data. The 16-channel weights are read out from RAM in sequence and multiplied with the 16-channel data to obtain the data weighted result. The data weighted result is divided into 2 groups and added in parallel to obtain the weighted result of a single channel.
[0050] The 32-way 32*2 group channel weighted results are folded and added 5 times to obtain the final 2 groups of folded and added results, that is, the final channel weighted results.
[0051] The present invention is not limited to the above specific implementations, and the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made to the above implementations based on the technical essence of the present invention shall be included in the protection scope of the present invention.
Claims
1. A multi-channel adaptive rate weighting module, characterized in that: include: Adaptive parameter parsing module (1), weight splitting cache module (2), AD data cache module (3), AD data serial-to-parallel conversion module (4), AD data replication module (5), channel weighting module (6) and folding and adding module (7); The adaptive parameter analysis module (1) adaptively calculates the number of weight split groups, the number of data replication parameters and the number of folding additions according to the data rate, the number of input paths, the number of single-path channels and the number of output multi-beams; The weight splitting cache module (2) splits the pre-calculated weights of each channel according to the weight splitting group number, and the split weights are cached in RAM for subsequent data weight multiplication processing; The AD data buffer module (3) is used to receive the AD data received by the front end, use FIFO to buffer the data, and start reading the data when the data storage capacity reaches a range gate, and send it to the AD data serial-to-parallel conversion module (4); The AD data serial-to-parallel conversion module (4) converts the AD data into parallel arrangement output according to the number of single-channel channels; The AD data replication module (5) replicates the parallel AD data in a number of copies according to the replication number parameter; The channel weighting module (6) groups the parallel channel data and the weight splitting data according to the data copy number parameter analyzed by the adaptive parameter analysis module (1), multiplies the data of each group by the weight, and then adds them in parallel to obtain a single channel weighted result; The folding and adding module (7) performs folding and adding on the weighted results of the N channels according to the folding and adding times parameter analyzed by the adaptive parameter analyzing module (1) to obtain a system channel weighted result.
2. A multi-channel adaptive rate weighting method, applied to a multi-channel adaptive rate weighting module according to claim 1, characterized in that: The following steps are involved: (a) A method for calculating the number of weight split groups G, the number of data copies C, and the number of folding additions F using the received data rate R, the number of input paths N, the number of single-path channels I, and the number of output multi-beams B; (b) sending the serial weights to the weight splitting cache module (2) splitting the serial input weight data into G groups according to the weight splitting group number G, and caching them separately in RAM; (c) sending the parallel AD data after serial-to-parallel conversion to the AD data replication module (5) and replicating C copies; (d) sending the C pieces of parallel data and the G groups of weights to the channel weighting module (6) for parallel multiplication, and adding the multiplication of each group of data and weights in parallel to obtain a single channel weighted result; (e) The weighted results of the N channels are input to the folding and adding module (7), and the folding and adding are performed according to the folding and adding times F to ensure that the same beams are added correspondingly, and finally the weighted results of the C system channels are output.