A RIS-assisted communication network CSI estimation prediction system and method

By building a virtual transmission environment in the RIS auxiliary communication system, extracting indoor environment characteristics and predicting channel parameters, the problems of difficulty and low accuracy of channel multiplication fading estimation in the prior art are solved, and high accuracy prediction of indoor CSI and Simplify the calculation process.

CN117221844BActive Publication Date: 2024-11-08HUANENG YIMIN COAL POWER CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202311240017.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-09-25
Publication Date
2024-11-08
Estimated Expiration
2043-09-25

AI Technical Summary

Technical Problem

The prior art is difficult to estimate the multiplicative fading effect in the channel in RIS-assisted communication system, and the existing CSI prediction methods have poor accuracy in indoor scenarios and high computational complexity, so they cannot effectively process the small-scale fading characteristics of the channel, resulting in Information loss.

Method used

The indoor environment features are extracted by the environmental information acquisition and preprocessing module, and the virtual transmission environment is constructed. The large-scale and small-scale fading parameters of the channel are obtained through the feature extraction module. The CSI is predicted by using the ray tracing method to reduce the calculation complexity and improve the estimation accuracy.

Benefits of technology

It realizes accurate prediction and estimation of CSI of RIS-assisted communication networks in indoor scenarios, reduces the computational complexity, improves the accuracy of estimation and prediction, simplifies the process, and makes it more reliable.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117221844B_ABST
    Figure CN117221844B_ABST
Patent Text Reader

Abstract

A RIS-assisted communication network CSI estimation prediction system and method, relating to the field of communication technology, the method comprising: obtaining indoor environment information through an indoor access point or base station, taking the obtained indoor environment information as input, and preprocessing the indoor environment information using a data preprocessing technology; extracting features that have a significant impact on the indoor communication transmission environment from the indoor environment information using a feature extraction method; constructing a virtual transmission environment using the extracted features, obtaining channel large-scale fading parameters and channel small-scale fading parameters through indoor access points to fit the channel propagation characteristic curve, thereby obtaining an accurate indoor communication transmission environment, and based on the obtained indoor communication transmission environment, predicting and estimating the channel state information CSI of the indoor communication transmission environment in the virtual transmission environment based on a ray tracing method. The present invention improves the accuracy of CSI estimation and prediction, and the implementation process is simpler and more reliable.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a RIS-assisted communication network CSI estimation and prediction system and method. Background Art

[0002] Reconfigurable Intelligent Surface (RIS) consists of small, low-power and reconfigurable antennas or passive components that can intelligently reflect, refract or scatter incident electromagnetic waves. RIS is deployed in environments such as walls or surfaces and can be independently programmed to adjust its phase, amplitude or polarization, enabling it to actively modify the propagation characteristics of the incident electromagnetic wave and intelligently manipulate the phase of the incident electromagnetic wave. RIS optimizes the wireless link between the transmitter and the receiver by actively adjusting the reflected signal, thereby achieving multiple functions including beamforming, signal focusing, interference elimination and enhanced signal strength. At the same time, RIS is a passive component that is usually deployed as an additional layer in a wireless communication system without transmission or reception functions, and changes the propagation environment by reflecting the incident signal. The main purpose of RIS is to enhance the performance of wireless communication systems by optimizing the propagation environment. It can improve signal strength, expand coverage, mitigate interference and achieve better spectrum efficiency by adjusting the reflected signal.

[0003] Channel state information (CSI) is key information in wireless communication systems. It provides knowledge about the characteristics and quality of the communication channel between the transmitter and the receiver, and is an important means of measuring the characteristics of the communication channel. In wireless communication systems, signals propagate in the air and encounter various obstacles and environmental conditions, which cause the communication channel to introduce losses such as fading, attenuation, interference, and noise. These effects may change with time and frequency, making the channel a dynamic and challenging component of the communication link.

[0004] The prior art closest to the present invention is the traditional method for predicting CSI, including the following cases:

[0005] (1) Pilot-based channel estimation method: One of the most common methods for predicting CSI is pilot-based channel estimation. In this method, the transmitter sends a known pilot symbol or sequence during a specific time slot or frequency subcarrier. The receiver measures the received pilot symbols and estimates the channel response based on the known transmitted pilot symbols. This estimated channel response provides the CSI required by the receiver to decode the data symbols.

[0006] (2) Feedback from the receiver: In some systems, the receiver can estimate the channel conditions locally and then feed this information back to the transmitter. This approach is common in feedback-based MIMO systems and certain multi-user communication scenarios.

[0007] (3) Channel prediction methods based on statistics or machine learning: In a dynamic channel environment where the channel changes rapidly, channel prediction techniques can be used. The receiver observes the past channel behavior and predicts the future channel conditions based on statistics or machine learning methods.

[0008] Since RIS is a passive component, it is usually deployed as an additional layer in wireless communication systems without transmission or reception functions. It mainly changes the propagation environment by reflecting the incident signal. According to this feature of RIS, the above three methods, namely the pilot-based channel estimation method, the feedback method from the receiver, and the channel prediction method based on statistics or machine learning, which are based on calculation and signal processing, are not feasible in RIS-assisted communication systems. The reason is that the RIS channel is a cascade channel with multiplicative fading. The existing CSI estimation and prediction methods are difficult to estimate the impact of multiplicative fading in the RIS channel.

[0009] In addition, the existing method for obtaining CSI of RIS-assisted communication system is to perform three-dimensional tensor operation at the receiving end. By increasing the dimension, the complex RIS-assisted CSI prediction problem is converted into a tensor solution problem. However, most receiving devices (including a large number of IoT devices) have limited computing power and cannot complete tensor calculation tasks; at the same time, the computational complexity of tensor operation is much higher than the vector calculation in the traditional CSI prediction method, which greatly increases the requirements for computing devices. At the same time, this method has the problem of poor CSI prediction accuracy in indoor scenarios, which is mainly reflected in the difficulty of this method in obtaining small-scale fading characteristics of the channel and the serious information loss problem in feature acquisition. Summary of the invention

[0010] In order to solve the problems existing in the prior art, the present invention provides a RIS-assisted communication network CSI estimation and prediction system and method.

[0011] The technical solution adopted by the present invention to solve the technical problem is as follows:

[0012] A RIS-assisted communication network CSI estimation and prediction system of the present invention comprises:

[0013] The environmental information acquisition and preprocessing module is used to acquire indoor environmental information and preprocess the indoor environmental information using data preprocessing technology;

[0014] A feature extraction module, used to extract features that have a significant impact on the indoor communication transmission environment from the indoor environmental information preprocessed by the environmental information acquisition and preprocessing module 201;

[0015] The CSI prediction module is used to construct a virtual transmission environment, obtain the large-scale fading parameters and small-scale fading parameters of the channel, and predict the channel state information CSI of the RIS-assisted indoor communication transmission environment from the features extracted by the feature extraction module that have a significant impact on the indoor communication transmission environment.

[0016] Furthermore, the feature extraction module takes into account prior information in the field of channel modeling when extracting features that have a significant impact on the indoor communication transmission environment.

[0017] A RIS-assisted communication network CSI estimation and prediction method of the present invention is implemented by using the above-mentioned RIS-assisted communication network CSI estimation and prediction system, and the method comprises the following steps:

[0018] Step S101: Acquire indoor environment information through an indoor access point or base station, use the acquired indoor environment information as input, and pre-process the indoor environment information using a data pre-processing technology;

[0019] Step S102: using a feature extraction method, based on the actual communication scenario, extracting features that have a significant impact on the indoor communication transmission environment from the indoor environment information;

[0020] Step S103: construct a virtual transmission environment using the features extracted in step S102, and obtain channel large-scale fading parameters and channel small-scale fading parameters from the virtual transmission environment; obtain channel large-scale fading parameters and channel small-scale fading parameters through indoor access points to fit the channel propagation characteristic curve, so as to obtain an accurate indoor communication transmission environment, and based on the obtained indoor communication transmission environment, predict and estimate the channel state information CSI of the indoor communication transmission environment in the virtual transmission environment based on the ray tracing method.

[0021] Furthermore, in step S101, the indoor environment information is an indoor environment picture.

[0022] Furthermore, in step S101, the indoor environment information is acquired using a depth camera or a camera.

[0023] Furthermore, in step S101, before obtaining the indoor environment information, an update time for obtaining the indoor environment information is preset.

[0024] Furthermore, in step S101, the indoor environment information is preprocessed using data preprocessing technology, and the specific operation steps are as follows:

[0025] (1) Outlier detection and deletion: Identify pixels in the indoor environment information that do not conform to the overall environmental characteristics and may be erroneous, and repair the erroneous pixels;

[0026] (2) Data conversion: The indoor environment information is converted into binary vector information after outlier detection and removal.

[0027] Furthermore, in step S102, the features that have a significant impact on the indoor communication transmission environment include the size of the indoor space, the placement of indoor objects, the distance between indoor objects, the shape of indoor objects, the material of indoor objects, and the speed and direction of indoor moving objects.

[0028] Furthermore, in step S102, the feature extraction method is implemented based on an LSTM neural network or a CNN neural network.

[0029] Furthermore, in step S103, a virtual transmission environment is constructed using the Deep MIMO data set.

[0030] The beneficial effects of the present invention are:

[0031] The present invention provides a RIS-assisted communication network CSI estimation and prediction system and method, for indoor scenarios, by acquiring key features that have a significant impact on the indoor communication transmission environment, and after virtually modeling the indoor communication transmission environment, accurately predicting and estimating the channel state information CSI of the RIS-assisted indoor communication network, reducing the complexity of CSI estimation and prediction through information related to the indoor communication transmission environment, and improving the accuracy of estimation and prediction.

[0032] Compared with the existing solutions, the RIS-assisted communication network CSI estimation and prediction system and method of the present invention ensures the accuracy of RIS-assisted communication network CSI estimation and prediction while making the implementation process simpler and more reliable. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 A structural block diagram of a RIS-assisted communication network CSI estimation and prediction system provided by the present invention.

[0034] Figure 2 A flow chart of a RIS-assisted communication network CSI estimation and prediction method provided by the present invention. DETAILED DESCRIPTION

[0035] A RIS-assisted communication network CSI estimation and prediction system and method provided by the present invention will be specifically described below with reference to the accompanying drawings.

[0036] In a first aspect, the present invention provides a RIS-assisted communication network CSI estimation and prediction system.

[0037] See also Figure 1 To illustrate, a RIS-assisted communication network CSI estimation and prediction system of the present invention mainly includes: an environment information acquisition and preprocessing module 201 , a feature extraction module 202 and a CSI prediction module 203 .

[0038] The environment information acquisition and preprocessing module 201 is used to acquire indoor environment information that can help model the indoor environment, such as indoor environment pictures, and use the acquired indoor environment information as input, while using data preprocessing technology to perform preprocessing such as noise reduction on the indoor environment information.

[0039] The indoor environment information mentioned above is mainly an indoor environment picture, which may be a point cloud picture, etc., for showing an overview of the indoor environment.

[0040] Among them, the feature extraction module 202 is used to extract features that have a significant impact on the indoor communication transmission environment from the indoor environmental information preprocessed by the environmental information acquisition and preprocessing module 201. These features include the size of the indoor space, the placement of indoor objects, the distance between indoor objects, the shape of indoor objects, the material of indoor objects, and the speed and direction of indoor moving objects.

[0041] Meanwhile, for the feature extraction module 202, some prior information in the field of channel modeling is taken into consideration in this specific implementation, for example, the signal absorption of metal is much stronger than that of materials such as concrete.

[0042] Among them, the CSI prediction module 203 is used to construct a virtual transmission environment, obtain channel large-scale fading parameters and channel small-scale fading parameters, and predict the channel state information CSI of the RIS-assisted indoor communication transmission environment from the features extracted by the feature extraction module 202 that have a significant impact on the indoor communication transmission environment.

[0043] In a second aspect, the present invention provides a RIS-assisted communication network CSI estimation and prediction method, which is implemented using the above-mentioned RIS-assisted communication network CSI estimation and prediction system.

[0044] See also Figure 2 To illustrate, a RIS-assisted communication network CSI estimation and prediction method of the present invention specifically includes the following steps:

[0045] Step S101: Indoor environment information that can help model the indoor environment is obtained through an indoor access point (AP) or base station, such as an indoor environment picture, and the obtained indoor environment information is used as input. At the same time, data preprocessing technology is used to perform preprocessing such as noise reduction on the indoor environment information to prepare for the next step of feature extraction.

[0046] When obtaining indoor environment information that can help model the indoor environment through an indoor access point or base station, the update time for obtaining the indoor environment information may be preset first.

[0047] The updating time for obtaining the indoor environment information can be set according to the requirements of the actual application, and the updating time for obtaining the indoor environment information can be any time such as 1 minute or 30 seconds. The present invention does not make any specific limitation on this.

[0048] Furthermore, the indoor access point or base station may use a depth camera, a camera, or other methods to obtain indoor environment pictures and other indoor environment information to help model the indoor environment. The present invention does not make any specific limitation on this.

[0049] Specifically, the indoor environment information is preprocessed by using data preprocessing technology to reduce noise. The specific operation steps of the preprocessing are as follows:

[0050] First, outlier detection and deletion. For example, identify pixels in indoor environment images that do not conform to the overall environmental characteristics and may be erroneous, and repair the erroneous pixels based on the indoor environment information presented in the indoor environment images.

[0051] Second, data conversion: After outlier detection and removal, the erroneous pixels in the indoor environment image are repaired, and then the indoor environment image is converted into binary vector information.

[0052] The specific algorithms for outlier detection and deletion and data conversion can be set according to the requirements of actual applications, and the present invention does not make any specific limitation on this.

[0053] Step S102: Using a feature extraction method, based on the actual communication scenario required, extract features that have a significant impact on the indoor communication transmission environment from the indoor environment information, such as the size of the indoor space, the placement of indoor objects, the distance between indoor objects, the shape of indoor objects, the material of indoor objects, and the speed and direction of indoor moving objects.

[0054] The feature extraction method used can be implemented based on neural networks such as LSTM and CNN, and can be implemented according to the actual application scenario and requirements. The present invention does not make specific limitations on this.

[0055] For the feature extraction process, some prior information in the field of channel modeling is considered in this specific implementation, for example, metal absorbs signals much more strongly than materials such as concrete.

[0056] Characteristics such as the position, shape, and material of objects in the real physical environment play a vital role in channel modeling of wireless communication systems. Understanding these aspects is crucial because the wireless channel is significantly affected by the signal propagation environment. Considering characteristics such as the position, shape, and material of objects in the real physical environment can enable more accurate and realistic channel modeling, and also help engineers design and optimize wireless communication systems so that they can adapt to the specific characteristics of the signal propagation environment and provide better performance and coverage.

[0057] Among them, the placement of indoor objects and the distance between indoor objects directly affect path loss and signal attenuation. Signal strength weakens when propagating in space, and objects in the path may cause additional attenuation. Therefore, knowing the location and distance of these objects helps to accurately estimate path loss and predict signal strength at different points in the coverage area. At the same time, objects may act as interference sources or generate noise, thus affecting the overall channel performance. Knowing the location and distance of these objects can help identify potential interference and noise sources in the communication system.

[0058] At the same time, the shape and position of indoor objects also affect multipath propagation. When signals encounter obstacles, they reflect, diffract, and scatter, causing multiple signal paths to arrive at the receiver at different times. Multipath fading effects can cause constructive or destructive interference, affecting the quality and reliability of the signal. Therefore, understanding the shape and position of these objects helps to accurately construct a virtual transmission environment and fit the channel propagation characteristic curve. And the presence of large objects or structures indoors may cause shadow effects, where certain areas have low signal strength due to obstacles. These shadow areas may cause signal changes within the coverage area. Therefore, understanding the shape of indoor objects is crucial for channel modeling.

[0059] In addition, different materials have different degrees of reflection, absorption and transmission characteristics for radio waves. The material properties of indoor objects affect the way signals interact with them, thereby affecting signal propagation and scattering behavior. Therefore, understanding the material properties of these objects helps to accurately construct a virtual transmission environment and fit the channel propagation characteristic curve.

[0060] Among them, in mobility scenarios, it will cause Doppler shift in the received signal. Therefore, knowing the speed and direction of indoor moving objects is crucial to accurately model the Doppler spectrum of the channel.

[0061] Step S103: construct a virtual transmission environment using the features extracted in step S102, and directly obtain large-scale channel fading parameters such as path loss and signal attenuation factor, as well as small-scale channel fading parameters such as multipath fading, channel Doppler spectrum, antenna placement and radiation pattern from the virtual transmission environment. Therefore, large-scale channel fading parameters and small-scale channel fading parameters can be directly obtained through indoor access points, and the channel propagation characteristic curve can be directly fitted to obtain an accurate indoor communication transmission environment, and based on the obtained indoor communication transmission environment, the channel state information CSI of the indoor communication transmission environment is predicted and estimated in the virtual transmission environment based on the ray tracing method.

[0062] Large-scale fading, also known as path loss or macroscopic fading, refers to the long-term variation of signal strength over a relatively large distance in a wireless communication link. It is mainly affected by the distance between the transmitter and the receiver and the overall geometry of the propagation environment. Compared with small-scale fading, large-scale fading is usually considered to be a slowly changing phenomenon.

[0063] Small-scale fading, also called fast fading or Rayleigh fading, refers to the rapid fluctuation of signal strength in a short period of time due to constructive and destructive interference caused by multipath propagation. It is a fast-changing phenomenon that occurs over short distances or short time intervals (such as a fraction of a wavelength or a few symbol durations).

[0064] The antenna positions on both sides of the transmitter and receiver are very important in determining the spatial characteristics of the channel. The antenna radiation pattern and its interaction with nearby objects affect the signal's directivity, gain, and polarization.

[0065] In this implementation, a Deep MIMO data set may be used to construct a virtual transmission environment, and the channel large-scale fading parameters and the channel small-scale fading parameters may be accurately obtained from the virtual transmission environment.

[0066] The RIS-assisted communication network CSI estimation prediction system and method provided by the present invention comprehensively considers the indoor communication transmission environment factors, and constructs the virtual transmission environment of the RIS-assisted indoor communication network based on the indoor communication transmission environment factors. In the virtual transmission environment, the indoor access point can easily obtain the channel large-scale fading parameters and the channel small-scale fading parameters, and directly fit the channel propagation characteristic curve. Therefore, the RIS-assisted communication network CSI estimation prediction system and method provided by the present invention can accurately predict the channel state information CSI of the RIS-assisted indoor communication network.

[0067] Furthermore, since the RIS-assisted communication network CSI estimation and prediction system and method provided by the present invention take into account the prior information in the field of channel modeling, the precision and accuracy of the RIS-assisted indoor communication network CSI estimation and prediction system and method can be significantly improved. At the same time, indoor access points can easily obtain channel large-scale fading parameters and channel small-scale fading parameters, which greatly reduces the difficulty of indoor access points in estimating and predicting the indoor communication transmission environment.

[0068] Based on this, the application of a RIS-assisted communication network CSI estimation and prediction system and method provided by the present invention can significantly improve the precision and accuracy of the RIS-assisted indoor communication network CSI estimation and prediction system and method, while improving the resource allocation accuracy and improving the performance of the RIS-assisted indoor communication network CSI estimation and prediction system and method.

[0069] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A RIS-assisted communication network CSI estimation and prediction system, characterized in that: include: The environmental information acquisition and preprocessing module is used to acquire indoor environmental information and preprocess the indoor environmental information using data preprocessing technology; A feature extraction module is used to extract features that have a significant impact on the indoor communication transmission environment from the indoor environmental information preprocessed by the environmental information acquisition and preprocessing module; The CSI prediction module is used to construct a virtual transmission environment, obtain the large-scale fading parameters and small-scale fading parameters of the channel, and predict the channel state information CSI of the RIS-assisted indoor communication transmission environment from the features extracted by the feature extraction module that have a significant impact on the indoor communication transmission environment.

2. A RIS-assisted communication network CSI estimation and prediction system according to claim 1, characterized in that: The feature extraction module takes into account prior information in the field of channel modeling when extracting features that have a significant impact on the indoor communication transmission environment.

3. A RIS-assisted communication network CSI estimation and prediction method, characterized in that: The method is implemented by using a RIS-assisted communication network CSI estimation and prediction system as described in claim 1 or 2, and comprises the following steps: Step S101: Acquire indoor environment information through an indoor access point or base station, use the acquired indoor environment information as input, and pre-process the indoor environment information using a data pre-processing technology; Step S102: using a feature extraction method, based on the actual communication scenario, extracting features that have a significant impact on the indoor communication transmission environment from the indoor environment information; Step S103: construct a virtual transmission environment using the features extracted in step S102, and obtain channel large-scale fading parameters and channel small-scale fading parameters from the virtual transmission environment; obtain channel large-scale fading parameters and channel small-scale fading parameters through indoor access points to fit the channel propagation characteristic curve, so as to obtain an accurate indoor communication transmission environment, and based on the obtained indoor communication transmission environment, predict and estimate the channel state information CSI of the indoor communication transmission environment in the virtual transmission environment based on the ray tracing method.

4. The RIS-assisted communication network CSI estimation and prediction method according to claim 3, characterized in that: In step S101, the indoor environment information is an indoor environment picture.

5. The RIS-assisted communication network CSI estimation and prediction method according to claim 3, characterized in that: In step S101, the indoor environment information is acquired using a depth camera or a camera.

6. The RIS-assisted communication network CSI estimation and prediction method according to claim 3, characterized in that: In step S101, before obtaining the indoor environment information, the update time for obtaining the indoor environment information is preset.

7. The RIS-assisted communication network CSI estimation and prediction method according to claim 3, characterized in that: In step S101, the indoor environment information is preprocessed using data preprocessing technology. The specific operation steps are as follows: (1) Outlier detection and deletion: Identify pixels in the indoor environment information that do not conform to the overall environmental characteristics and may be erroneous, and repair the erroneous pixels; (2) Data conversion: The indoor environment information is converted into binary vector information after outlier detection and removal.

8. The RIS-assisted communication network CSI estimation and prediction method according to claim 5, characterized in that: In step S102, the features that have a significant impact on the indoor communication transmission environment include the size of the indoor space, the placement of indoor objects, the distance between indoor objects, the shape of indoor objects, the material of indoor objects, and the speed and direction of indoor moving objects.

9. The RIS-assisted communication network CSI estimation and prediction method according to claim 5, characterized in that: In step S102, the feature extraction method is implemented based on LSTM neural network or CNN neural network.

10. The RIS-assisted communication network CSI estimation and prediction method according to claim 5, characterized in that: In step S103, a virtual transmission environment is constructed using the Deep MIMO data set.