Adaptive MIMO detection method based on channel condition number

Abstract

The invention provides a self-adaptive MIMO (Multiple Input Multiple Output) detection method based on a channel condition number, and belongs to a wireless communication technology. The method comprises the following steps of: selecting a corresponding condition number threshold value from a candidate threshold table according to an actually used MIMO (Multiple Input Multiple Output) specification; calculating an average value according to each subcarrier channel matrix output by the channel estimation module, and then solving a channel condition number; and comparing the condition number with a threshold value, and selecting a linear detection method or a nonlinear detection method to carry out subsequent signal processing. Performance of the receiver can be effectively improved or power consumption can be effectively reduced.

Description

Technical Field

[0001] The invention relates to wireless communication technology, and in particular to an adaptive MIMO detection method based on a channel condition number. Background Art

[0002] With the increasing demand for access to the Internet anytime and anywhere, wireless communication technology has been receiving extensive attention and research. The MIMO (Multiple Input Multiple Output) system uses multiple antennas at the transmitter and receiver to transmit and receive information simultaneously, which can effectively improve the system capacity and communication reliability without increasing the use of additional spectrum resources.

[0003] MIMO detection is a key and challenging technology in MIMO systems. Its main purpose is to accurately estimate the transmitted data from the received multi-antenna data. Common MIMO detection methods include linear detection and nonlinear detection. Among them, the linear detection method uses the linear combination of the received signals as the detection output by calculating the equalization matrix. Common linear detection methods include zero forcing (ZF) detection and linear minimum mean square error (LMMSE) detection. These two methods have similar performance under high signal-to-noise ratio, but under low signal-to-noise ratio, zero forcing detection has the disadvantage of amplifying noise, while the linear minimum mean square error detection method has better performance because it can better suppress noise amplification. Nonlinear detection mainly includes maximum likelihood detection and simplified maximum likelihood methods such as QRM. Its basic idea is to search for the most likely transmitted signal from the space of all possible transmitted signals by calculating the Euclidean distance when the channel estimation result is obtained. This type of detection method can achieve better performance than linear detection, but the complexity increases rapidly with the increase of signal modulation order and number of streams. Therefore, its use in MIMO systems with a large number of antennas and high-order modulation will be greatly restricted.

[0004] Compared with wired communication, the uncertainty of wireless channels is an important challenge faced by wireless communication. Therefore, the research, modeling and analysis of channels are important parts of wireless communication system design and optimization. In MIMO channels, the spatial correlation between channels of different transmit and receive antenna pairs will greatly affect the difficulty of signal extraction and separation by the receiver. Under medium and high correlation channel conditions, the MIMO detection module can obtain a significant performance gain of more than 3dB compared with linear detection by using nonlinear detection methods. Under low correlation channel conditions, there is no significant difference in the performance of the two detection methods, and the use of low-complexity linear detection methods means a significant reduction in system power consumption. Modern wireless communication system receivers often integrate both detection modules at the same time, but the switching between the two detection methods requires manual configuration through parameter delivery, making it difficult to make the best choice in real time.

[0005] Based on the above analysis, how to adaptively select the most appropriate MIMO detection method according to the channel conditions has important practical significance for balancing the performance and power consumption of the receiver. According to the relevant theory of matrices, when the MIMO channel correlation is strong, the channel matrix is ​​close to singularity, and the condition number of the matrix will increase. Therefore, the condition number of the channel matrix can be used as a basis for judging the selection of MIMO detection methods. Summary of the invention

[0006] In view of the defects in the prior art, the purpose of the present invention is to provide an adaptive MIMO detection method based on the channel condition number. The method of the present invention calculates the condition number of the MIMO channel, compares it with the preset threshold value, and selects the appropriate MIMO detection method in real time, which can effectively improve the performance of the receiver or reduce power consumption.

[0007] The technical solution adopted by the present invention is:

[0008] An adaptive MIMO detection method based on a channel condition number comprises the following steps:

[0009] Step 1: According to the actual MIMO specification, select the corresponding condition number threshold value κ from the candidate threshold table th ;

[0010] Step 2: Based on the channel matrix H output by the channel estimation module, k , using a group of physical resource blocks using the same precoding as the granularity, calculate the average value

[0011]

[0012] Where N is the number of subcarriers in a group of physical resource blocks using the same precoding;

[0013] Then, find the channel condition number:

[0014]

[0015] in, Express Matrix inversion, || || means finding the bi-norm of the matrix;

[0016] Step 3: The calculated channel condition number κ(H) and threshold value κ th For comparison, if κ(H)<κ th , the MIMO detection module selects the linear detection method for subsequent signal processing; if κ(H)>κ th , the MIMO detection module selects a nonlinear detection method for subsequent signal processing.

[0017] Furthermore, in step 1, the candidate threshold table is obtained by:

[0018] For a MIMO specification system, the detection method is set to use a fixed nonlinear detection method, and the system throughput is obtained by simulation as the optimal throughput of the specification system;

[0019] Set the search interval and step, and search for the threshold value in the search interval in a decreasing manner from the largest value; perform simulation at each threshold value, and realize the adaptive selection of the detection method through steps 2 and 3; obtain the air interface throughput and the number of calls of the two detection methods;

[0020] When the throughput of the simulation result is lower than 90% of the optimal throughput, the search is stopped, and the threshold value with the largest number of calls to the linear detection method is selected and used as the condition number threshold value under the current MIMO specification.

[0021] Compared with the prior art, the present invention has the following beneficial effects:

[0022] 1. Reduce bit error rate and improve system throughput.

[0023] The wireless communication environment is complex and changeable, and low, medium, and high channel correlations may occur. The method of the present invention can quickly adapt to changes in the wireless communication environment. Compared with the method of fixed setting for linear detection, when medium and high correlation channels appear, it can effectively reduce the bit error rate, improve the reliability and throughput of the system, and improve the user experience.

[0024] 2. Reduce receiver power consumption.

[0025] Although setting the MIMO detection module fixedly to nonlinear detection can achieve better performance, it also means a lot of energy waste. The method of the present invention can significantly reduce power consumption by reducing the call of complex nonlinear detection without reducing system performance and throughput. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 The flowchart of the adaptive MIMO detection method based on the channel condition number is shown in FIG.

[0027] Figure 2 This is a throughput simulation comparison chart of different MIMO detection methods in 5G NR system. DETAILED DESCRIPTION

[0028] The present invention is further described in detail below in conjunction with the accompanying drawings.

[0029] An adaptive MIMO detection method based on channel condition number, based on the output result of the channel estimation module, obtains the condition number of the channel matrix, and adaptively selects to use a low-complexity linear MIMO detection method or a high-performance nonlinear detection method by comparing it with a preset threshold value. Figure 1 As shown, the specific steps are as follows:

[0030] Step 1: Use the wireless communication system link-level simulation platform to search for systems with different MIMO specifications and obtain candidate threshold values. Select the corresponding condition number threshold value κ from the candidate threshold values ​​according to the actual MIMO specifications used. th .

[0031] The candidate threshold value is obtained as follows:

[0032] For a MIMO specification system, the detection method is set to use a fixed nonlinear detection method, and the system throughput is obtained by simulation as the optimal throughput of the specification system;

[0033] Set the search interval and step, and search the threshold value in the search interval in descending order from the largest value. In this example, the threshold value is searched as [100:-5:0], that is, the threshold value is first set to 100, and then reduced by 5 each time, with the lower limit being 0. Simulate at each threshold value, and realize the adaptive selection of the detection method through steps 2 and 3. Obtain the air interface throughput and the number of calls of the two detection methods.

[0034] When the throughput of the simulation result is lower than 90% of the optimal throughput, the search is stopped, and the threshold value with the largest number of calls to the linear detection method is selected and used as the condition number threshold value under the current MIMO specification.

[0035] Step 2: Based on the channel matrix H output by the channel estimation module, k , the average value is calculated with PRG (a group of physical resource blocks using the same precoding) as the granularity:

[0036]

[0037] Wherein, N is the number of subcarriers in a PRG;

[0038] Then, find the condition number of the matrix:

[0039]

[0040] in, Express Matrix inversion, |||| means finding the bi-norm of the matrix.

[0041] Step 3: Substitute the calculated channel condition number κ(H) and threshold value κ th For comparison. If κ(H)<K th , the MIMO detection module selects linear detection methods such as zero-forcing detection or linear minimum mean square error for subsequent signal processing; if κ(H)>K th , the MIMO detection module selects nonlinear detection methods such as QRM or MFCSO for subsequent signal processing.

[0042] Step 4: In order to verify the effect of this method, the performance of the adaptive method can be evaluated based on the air interface throughput, and the power consumption can be evaluated based on the statistical results of the number of calls of the two detection methods.

[0043] Here is a more specific example:

[0044] For the downlink PDSCH reception process of 5G NR, the specific steps are as follows:

[0045] Step 1: Set the bandwidth, number of antennas, number of streams, channels and other related simulation parameters as shown in the following table:

[0046] Frame rate 100 BWP size 106RB Number of transmitting antennas 4 Number of receiving antennas 4 Number of streams 3 Modulation order QPSK Bitrate 490 / 1024 Channel Type R.PDSCH.2-2.3TDD, Medium A

[0047] Step 2: Set the condition number threshold κ th is 15;

[0048] Step 3: The receiving end sets three MIMO detection modes: linear MMSE detection, nonlinear MFCSO detection, and adaptive detection based on channel condition number, and performs simulation;

[0049] Step 4: Analyze and count the throughput rates achieved by the above three detection methods, as well as the number of calls of MMSE and MFCSO respectively when the adaptive detection method is used.

[0050] Figure 2 The figure shows the throughput simulation comparison of the three detection methods in the 5G NR system. Among them, when using the adaptive detection method, MMSE calls account for 14.7% and MFCSO calls account for 85.3%.

[0051] From the above examples, it can be found that the adaptive MIMO detection method of the present invention has a maximum throughput improvement of 21.65% compared with fixed use of MMSE. Compared with fixed use of MFCSO, the performance is similar, but the number of calls to MMSE accounts for 14.7%, which greatly reduces the power consumption of the receiver.

Claims

1. An adaptive MIMO detection method based on channel condition number, characterized in that: The following steps are involved: Step 1: According to the actual MIMO specification, select the corresponding condition number threshold value κ from the candidate threshold table th ; Step 2: Based on the channel matrix H output by the channel estimation module, k , using a group of physical resource blocks using the same precoding as the granularity, calculate the average value Where N is the number of subcarriers in a group of physical resource blocks using the same precoding; Then, find the channel condition number: in, Express Matrix inversion, || || means finding the bi-norm of the matrix; Step 3: The calculated channel condition number κ(H) and threshold value κ th For comparison, if κ(H)<κ th , the MIMO detection module selects the linear detection method for subsequent signal processing; if κ(H)>κ th , the MIMO detection module selects a nonlinear detection method for subsequent signal processing.

2. The adaptive MIMO detection method based on channel condition number according to claim 1, characterized in that: In step 1, the candidate threshold table is obtained in the following way: For a MIMO specification system, the detection method is set to use a fixed nonlinear detection method, and the system throughput is obtained by simulation as the optimal throughput of the specification system; Set the search interval and step, and search for the threshold value in the search interval in a decreasing manner from the largest value; perform simulation at each threshold value, and realize the adaptive selection of the detection method through steps 2 and 3; obtain the air interface throughput and the number of calls of the two detection methods; When the throughput of the simulation result is lower than 90% of the optimal throughput, the search is stopped, and the threshold value with the largest number of calls to the linear detection method is selected and used as the condition number threshold value under the current MIMO specification.

Images