Adaptive channel sensing and tracking method based on channel knowledge base

Abstract

The invention discloses an adaptive channel sensing and tracking method based on a channel knowledge base. The method comprises the following steps: defining a wireless channel, a channel knowledge base and a channel observation model in an MIMO system; based on the channel knowledge base and the channel sensing sequence, defining joint probability distribution based on the channel measurement sequence, the channel sequence and the user position sequence; a switching Kalman filtering framework fused with a channel knowledge base is adopted, adaptive design of a channel sensing matrix is achieved, and user positioning and channel tracking are carried out. According to the method, a high-dimensional channel is predicted based on historical low-dimensional (relative to the high-dimensional channel) channel observation, and a channel sensing matrix is designed according to a channel prediction error for real-time channel measurement, so that the difference entropy minimization of channel estimation is realized; and on the basis of low-dimensional channel measurement observed by the self-adaptive channel sensing matrix, channel tracking is realized by using channel space-time statistical distribution (channel covariance matrix) provided in a channel knowledge base and a switching Kalman filtering technology.

Description

Technical Field

[0001] The present invention relates to the field of wireless channel estimation and tracking, and in particular to an adaptive channel sensing and tracking method based on a channel knowledge base. Background Art

[0002] Fast and accurate estimation and tracking of wireless channels is a key challenge to improving the quality of service in wireless communication networks. For example, in large-scale Multiple-Input Multiple-Output (MIMO) systems, multi-antenna technology enables beamforming and efficient signal transmission, but beam alignment requires a relatively accurate estimation of the channel. In some static scenarios, although beam dictionaries can be used to traverse all possible beams and select the best beam based on channel feedback, the process of such beam scanning is extremely complex and its effect is strongly related to the quality of the beam dictionary.

[0003] At present, the estimation and tracking technologies of wireless channels can be classified into three categories: methods based on known channel statistical distribution, methods based on compressed sensing, and machine learning methods based on historical channel data. Assuming that the statistical distribution of the channel is known, existing technologies can use the minimum root mean square error or Bayesian reasoning to obtain channel estimates. If the statistical distribution of the channel is unknown, but the channel is known to be sparse, the compressed sensing algorithm can be used to recover the channel from sparse observations. If the historical channel is known, Kalman filtering can be used to achieve channel tracking, or deep learning methods such as long short-term memory networks, attention networks, and generative networks can be used to achieve data-driven channel prediction.

[0004] When using Kalman filtering technology to achieve channel tracking, Kalman filtering is based on the stable and unchanging spatial statistical distribution of the channel. However, the spatial statistical distribution of the channel in actual scenarios will change due to the obstruction of obstacles. Therefore, Kalman filtering technology cannot capture the spatial statistical distribution of the channel in a timely manner, and the actual channel tracking accuracy will be low, and the channel-assisted beam alignment accuracy will be even lower.

[0005] Channel prediction based on deep learning is costly. The technology requires a large number of channel vectors. In large-scale multi-input and multi-output systems, the cost of collecting these channel vectors is extremely high. Channel prediction based on deep learning requires accurate channel vectors for model training and cannot be performed directly based on low-dimensional noisy channel observations.

[0006] In short, the limitations of existing technologies are mainly reflected in

[0007] 1) Most existing technologies assume that the spatial statistical characteristics of the channel are constant. This assumption is often not met when the user moves or the environment changes. When the communication link between the user and the base station changes from being unobstructed to being obstructed by obstacles such as buildings and vehicles, the spatial statistical characteristics of the channel will be significantly changed.

[0008] 2) Most existing technologies assume that the user's position is known. This assumption can be used to obtain the occlusion status of the link between the user and the base station, thereby inferring the spatial statistical characteristics of the channel. For example, in the presence of a three-dimensional spatial obstacle map, user location, and base station location, the user location and base station location can be connected to check whether there is an obstacle blocking the line, and different channel empirical models can be adopted according to whether there is an obstacle blocking the line.

[0009] 3) Most existing technologies require a large amount of accurate channel data for model training. However, in multi-antenna systems, due to the extremely high channel dimension, the cost of obtaining such channel data is extremely high. Summary of the invention

[0010] The purpose of the present invention is to overcome the deficiencies of the prior art and to provide an adaptive channel sensing and tracking method based on a channel knowledge base.

[0011] The objective of the present invention is achieved through the following technical solution: an adaptive channel sensing and tracking method based on a channel knowledge base, comprising the following steps:

[0012] Define wireless channels, channel knowledge base and channel observation model in MIMO systems;

[0013] Based on the channel knowledge base and the channel sensing sequence, a joint probability distribution based on the channel measurement sequence, the channel sequence, and the user position sequence is defined;

[0014] A switching Kalman filter architecture that integrates a channel knowledge base is used to achieve adaptive design of the channel sensing matrix, and perform user positioning and channel tracking.

[0015] The beneficial effects of the present invention are:

[0016] 1) The technology proposed in the present invention does not need to assume the stability of the channel. During the user's movement, the channel may be suddenly blocked by a building. The assumption of the stability of the channel in the traditional technology is not valid, and the present invention can cope with the scenario of sudden channel changes.

[0017] 2) The present invention does not require user location information for estimating and tracking wireless channels, and the present invention can output the estimated user location information.

[0018] 3) The implementation cost of the present invention is extremely low, and the amount of channel observation data required is extremely small. In actual use, only a small amount, such as 1, of pilot signals is needed at each discrete moment for channel sensing, and the technology proposed by the present invention can be used to perform channel estimation and tracking, as well as user positioning. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a schematic diagram of the principle of the present invention;

[0020] Figure 2 Schematic diagram of channel tracking performance based on wireless spectrum map and adaptive channel sensing matrix. DETAILED DESCRIPTION

[0021] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings, but the protection scope of the present invention is not limited to the following.

[0022] The present invention investigates adaptive channel sensing and tracking based on a channel knowledge base in a wireless network. A channel knowledge base is a knowledge base that describes the spatiotemporal statistical characteristics of a channel. It can map any spatial position coordinate to the spatiotemporal statistical characteristics of a channel. These spatiotemporal statistical characteristics include but are not limited to the channel covariance matrix, signal departure angle, signal arrival angle, received signal strength, signal-to-noise ratio, etc. Specifically, a high-dimensional channel is predicted based on low-dimensional (relative to high-dimensional channels) channel observations, and a channel sensing matrix is ​​designed based on the channel prediction error for channel measurement to minimize the differential entropy of channel estimation. Based on the channel measurement observed by the adaptive channel sensing matrix, channel tracking is achieved using the spatiotemporal statistical distribution of the channel (channel covariance matrix) and the switching Kalman filter technology provided in the channel knowledge base.

[0023] The present invention is applicable to the design of adaptive sensing matrix and wireless channel measurement and tracking in scenarios including but not limited to single antenna and multiple antennas, and the technical solution proposed by the present invention is not affected by the environmental topology, the number of base stations and the number of users. Specifically, for scenarios with multiple base stations and multiple users, it is only necessary to build a channel knowledge base for each base station, and use the channel knowledge base to perform adaptive channel sensing measurement, positioning and channel tracking for a specific user.

[0024] Below, the present invention takes a wireless channel in a multiple-input multiple-output (MIMO) network as an example to illustrate the technical solution of the present invention. The channel knowledge base in this example is visualized as a mapping from spatial position to channel covariance matrix, but the channel spatial characteristics carried by the channel knowledge base are not limited to the channel covariance matrix, but can also be variables such as signal departure angle, signal arrival angle, received signal strength, signal-to-noise ratio, etc. that are related to the spatial position of the signal transceiver device and can characterize a certain aspect of the channel.

[0025] like Figure 1 As shown, an adaptive channel sensing and tracking method based on a channel knowledge base includes the following steps:

[0026] Define wireless channels, channel knowledge base and channel observation model in MIMO systems;

[0027] Consider a MIMO system consisting of one base station and one mobile user. The base station is equipped with a massive MIMO antenna with N t antenna units, and mobile users use a single antenna for communication.

[0028] First, define the wireless channel as h. Since the base station has N t antenna elements, so the channel h is an N t Since the channel is related to both time and space, let h be the channel at time t. t Modeled as a first-order autoregressive model:

[0029] h t =γh t-1 +(1-γ 2 ) t (1)

[0030] Among them, γ is the first-order autoregressive coefficient, h t-1 is the channel at time t-1, u t It is a vector related to the channel spatial statistical characteristics. Define the user's position at time t as p t , here we assume that u t ~CN(0,C(p t )), that is u t Subject to mean 0 and variance p t The channel covariance matrix C(p t ) complex Gaussian distribution, where C(p t ) is provided by the channel knowledge base. According to the above channel model (1), it can be deduced that: h t ~CN(0,C(p t )).

[0031] Define a channel knowledge base as

[0032]

[0033] Where x = (x1, x2, x3) T is a three-dimensional position vector, x is the set of all spatial grid center positions after discretizing the target area into equidistant grids, and C(x) is the average channel covariance matrix at the position in the grid with the center position x, which is defined as here Represents the covariance hh of all channels h in the grid H Taking the expectation, according to the definition, C(x) is an N t ×N t Dimensional matrix. Since the user position p t falls in the grid of the target area. If the center of the grid is x, then u in the channel model t The covariance C(p t ) can be replaced by C(x).

[0034] Given M pilot signals, we can observe y at time t t Create the following model:

[0035] y t =A t h t +n t (3)

[0036] Here, A t is the channel sensing matrix at the base station at time t, each row of which corresponds to the observation vector of a pilot signal on the MIMO antenna at the receiving end. Therefore, A t There are M rows and N t Column, h t is an N t The column vector of rows is the real channel at time t, n t is the measurement noise at time t, which is a column vector with M rows. Assume That is n t Each element of is subject to a mean of O and a variance of Gaussian distribution, where I is a column vector with M rows. According to the above channel measurement model (3), it can be deduced that in Yes A t The conjugate transpose of .

[0037] definition For the observation sequence from t = 1 to t = T, define is the user trajectory from t = 1 to t = T, and we define For the channel sequence from t = 1 to t = T, define is the channel sensing matrix sequence from t=1 to t=T.

[0038] Based on the channel knowledge base and the channel sensing sequence, a joint probability distribution based on the channel measurement sequence, the channel sequence, and the user position sequence is defined;

[0039] Based on the above model, we define Channel measurement sequence Channel Sequence User location sequence The joint probability distribution of and channel sensing sequence In the process of constructing the channel knowledge base, this technique only assumes that the channel measurement sequence and channel knowledge base It is known that the real channel sequence and user location sequence Unknown, and the channel sensing sequence Need to be designed, these unknown quantities and known quantities need to use the channel knowledge base The channel space features in the channel are connected in series, that is, the user position indicates the channel space features (through the channel knowledge base The channel covariance matrix in ), the channel spatial characteristics reflect the spatial distribution characteristics of the real channel, and some characteristics of the real channel are captured by the channel sensing matrix and saved in the channel observation.

[0040] Channel Knowledge Base The channel spatial characteristics of each position can be provided, such as the channel covariance matrix C(x). According to the channel model (1), the channel h t ~CN(0,C(p t ), therefore, the channel h t The complex Gaussian distribution of is known, and combined with the relationship between the current channel and the historical channel, channel tracking can be performed.

[0041] A switching Kalman filter architecture that integrates a channel knowledge base is used to achieve adaptive design of the channel sensing matrix, and perform user positioning and channel tracking.

[0042] 1) Input: Channel knowledge base Output: Adaptive channel sensing matrix A t , user location estimation Channel Estimation

[0043] 2) Initialization (including but not limited to the following methods): transmit M pilots to obtain sparse observation y1 at time t = 1. Obtain or estimate an initial position through prior knowledge Initialize the channel estimate for the current time in, Obtained from the channel knowledge base. Initialize a channel estimation error matrix in, A1 is the adaptive channel sensing matrix at time t=1 randomly generated in the initialization phase;

[0044] 3) Channel estimation and tracking (t≥2):

[0045] a) Predict the current channel based on historical information (tracking):

[0046] Channel Prediction: in, represents the channel predicted based on the information at time t-1, represents the channel estimate at time t-1.

[0047] Channel prediction error matrix update: in From the location to p t The transition probability can be estimated from the user's historical trajectory data, assuming that this information is known.

[0048] b) Adaptive sensing matrix design:

[0049] The channel prediction error matrix Q t|t-1 Perform eigenvalue decomposition, that is in, is a diagonal element Q t|t-1 The diagonal matrix of the eigenvalues ​​of is the value corresponding to each eigenvalue λ n The eigenvector w n The matrix composed of.

[0050] In order to minimize the differential entropy of channel estimation (detailed derivation is described later), the adaptive sensing matrix is ​​designed as: t =[w1, w2, ..., w M ] H

[0051] c) Transmitting pilot signals for channel observation based on the adaptive sensing matrix A t Get channel observation y t .

[0052] d) Channel prediction error matrix correction:

[0053] Among them, π t (p t ) is a function that combines the spatial statistical characteristics of the channel and is defined as

[0054] e) Channel estimation based on current observations:

[0055] Kalman filter optimal coefficient calculation:

[0056] Channel estimation:

[0057] Channel estimation error matrix update: Q t =(IK t A t )Q t|t-1 .

[0058] f) User positioning:

[0059] Estimate the user location by maximizing the user location conditional probability distribution , where the conditional probability can be obtained from the channel model (1), because h t |h t-1 ~CN(γh t-1 , (1-γ 2 )C(p t )).

[0060] g) Output channel estimation and user location estimation

[0061] The following is a simulation performance of a specific embodiment of the present invention. This embodiment is deployed in a 740m×710m urban environment. The target area contains 7 base stations. Each base station is configured with an N t =MIMO antenna composed of 64 antenna units, and base stations are randomly deployed on the top of some buildings.

[0062] Consider a user moving on the road at a speed of 10 m / s and an interval of 10 ms for transmitting pilot signals. Figure 2 The performance of the proposed technical solution under different signal-to-noise ratios (SNRs) is shown. The horizontal axis of the image represents the value of the signal-to-noise ratio in dB, and the vertical axis represents the channel capacity efficiency ratio, which is expressed as the ratio of the maximum channel capacity that can be achieved by the current channel estimation scheme to the maximum channel capacity that can be achieved by a perfect channel. It can be seen from the image that when the signal-to-noise ratio (SNR) is greater than or equal to 20 dB, this technology can achieve more than 97% of the perfect channel effect, while the performance of other compared technical solutions is less than 83%.

[0063] The above is a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used in other combinations, modifications and environments, and can be modified within the scope of the concept described herein through the above teachings or the technology or knowledge of the relevant field. The changes and modifications made by those skilled in the art do not depart from the spirit and scope of the present invention, and should be within the scope of protection of the claims attached to the present invention.

Claims

1. An adaptive channel sensing and tracking method based on a channel knowledge base, characterized in that: The following steps are involved: Define wireless channels, channel knowledge base and channel observation models in MIMO systems; Based on the channel knowledge base and the channel sensing sequence, a joint probability distribution based on the channel measurement sequence, the channel sequence, and the user position sequence is defined; A switching Kalman filter architecture that integrates a channel knowledge base is used to achieve adaptive design of the channel sensing matrix, and perform user positioning and channel tracking.

2. The adaptive channel sensing and tracking method based on a channel knowledge base according to claim 1, characterized in that: Defining the wireless channel, channel knowledge base and channel observation model in the MIMO system includes: Assume that the MIMO system includes one base station and one mobile user. The base station is equipped with a massive MIMO antenna, which has N t antenna units, mobile users use a single antenna for communication; Define the wireless channel as h: Since the base station has N t antenna units, channel h is an N t Dimensional column vector, since the channel is related to both time and space, the channel h at time t t Modeled as a first-order autoregressive model: h t =γh t-1 +-γ 2 )u t (1) Among them, γ is the first-order autoregressive coefficient, h t-1 is the channel at time t-1, u t It is a vector related to the channel spatial statistical characteristics; the user's position at time t is defined as p t , assuming u t ~CN(0,C(p t )), that is u t Subject to mean 0 and variance p t The channel covariance matrix C(p t ) complex Gaussian distribution, where C(p t ) is provided by the channel knowledge base; According to the above channel model (1), it is obtained that: h t ~CN(0,C(p t )); Define a channel knowledge base as: Where x = (x1, x2, x3) T is a three-dimensional position vector, χ is the set of all spatial grid center positions after discretizing the target area into equidistant grids, and C(x) is the average channel covariance matrix of the positions in the grid with the center position x, which is defined as here Represents the covariance hh of all channels h in the grid H Taking the expectation, C(x) is an N t ×N t Dimensional matrix; since the user position p t falls in the grid of the target area. If the center of the grid is x, then u in the channel model t The covariance C(p t ) is replaced by C(x); Given M pilot signals, the channel observation y at time t is t Create the following model: y t =A t h t +n t (3) Among them, A t is the channel sensing matrix at the base station at time t, each row of which corresponds to the observation vector of a pilot signal on the MIMO antenna at the receiving end. t There are M rows and N t Column, h t is an N t The column vector of rows is the real channel at time t, n t is the measurement noise at time t, which is a column vector with M rows; assuming That is n t Each element of has a mean of 0 and a variance of Gaussian distribution, I is a column vector of M rows, according to the above channel measurement model (3) we can deduce in Yes A t The conjugate transpose of ; definition For the observation sequence from t = 1 to t = T, define the user position sequence is the user trajectory from t = 1 to t = T, and we define For the channel sequence from t = 1 to t = T, define is the channel sensing matrix sequence from t=1 to t=T.

3. The adaptive channel sensing and tracking method based on a channel knowledge base according to claim 1, characterized in that: The defining of the joint probability distribution based on the channel measurement sequence, the channel sequence, and the user position sequence based on the channel knowledge base and the channel sensing sequence includes: definition Channel measurement sequence Channel Sequence User location sequence The joint probability distribution of and channel sensing sequence In the process of building the channel knowledge base, it is assumed that the channel measurement sequence and channel knowledge base It is known that the real channel sequence and user location sequence Unknown, and the channel sensing sequence Need to be designed, these unknown quantities and known quantities need to use the channel knowledge base The channel space features in the channel knowledge base are connected in series. The channel covariance matrix in , the channel spatial characteristics reflect the spatial distribution characteristics of the real channel, and some characteristics of the real channel are captured by the channel sensing matrix and saved in the channel observation.

4. The adaptive channel sensing and tracking method based on a channel knowledge base according to claim 1, characterized in that: The channel knowledge base Provide the channel spatial characteristics of each position, including the channel covariance matrix C(x), and derive the channel h according to the channel model (1) t ~CN(0,C(p t ), therefore, the channel h t The complex Gaussian distribution of is known, and combined with the relationship between the current channel and the historical channel, user positioning and channel tracking can be achieved.

5. The adaptive channel sensing and tracking method based on a channel knowledge base according to claim 1, characterized in that: The switching Kalman filter architecture using the fused channel knowledge base realizes the adaptive design of the channel sensing matrix, user positioning and channel tracking, including: A1. Input: Channel knowledge base Output: Adaptive channel sensing matrix A t , user location estimation Channel Estimation A2. Initialization: Transmit M pilots, obtain sparse observation y1 at time t = 1, and obtain or estimate an initial position through prior knowledge Initialize the channel estimate for the current time in, Obtained from the channel knowledge base; initialize a channel estimation error matrix Q1 = (1-γ 2 )∑ x∈x p(y1|x)C(x); in, A1 is the adaptive channel sensing matrix at time t=1 randomly generated in the initialization phase; A3. When t≥2, perform user positioning and channel tracking.

6. The adaptive channel sensing and tracking method based on a channel knowledge base according to claim 5, characterized in that: The step A3 comprises: A301. Predict the current channel based on historical information to achieve current channel prediction: Channel Prediction: in, represents the channel predicted based on the information at time t-1, represents the channel estimate at time t-1; Channel prediction error matrix update: in From the location to p t The transition probability is estimated from the user's historical trajectory data, which is known; A302, Adaptive Perception Matrix Design: The channel prediction error matrix Q t|t-1 Perform eigenvalue decomposition, that is in, is a diagonal element Q t|t-1 The eigenvalue of The diagonal matrix of is the value corresponding to each eigenvalue λ n The eigenvector w n The matrix composed of In order to minimize the differential entropy of channel estimation, the adaptive sensing matrix is ​​designed as: t =[w1, w2, ..., w M ] H ; A303, transmit pilot signal to observe the channel, based on the adaptive sensing matrix A t Get channel observation y t ; A304, channel prediction error matrix correction: Among them, π t (p t ) is a function that combines the spatial statistical characteristics of the channel and is defined as: A305. Channel estimation based on current observation: Kalman filter optimal coefficient calculation: Channel estimation: Channel estimation error matrix update: Q t =(IK t A t )Q t|t-1 ; A306. User location: Estimate the user location by maximizing the conditional probability distribution of the user location , where the conditional probability From the channel model (1), this is because h t |h t-1 ~CN(γh t-1 , (1-γ 2 )C(p t )); A307, output channel estimation and user location estimation

Images