CN112738883B - A method and device for determining the position of an air base station - Google Patents

Abstract

The invention provides a method and a device for determining the position of an air base station, wherein the method comprises the following steps: acquiring historical ground terminal quantity data of a target area, and predicting the ground terminal quantity of the target area in a target time period according to the historical ground terminal quantity data; if the number of the ground terminals of the target area in the target time period is larger than a preset threshold value, acquiring current ground terminal position information in the target area; and determining the position of the air base station according to the current ground terminal position information in the target area. According to the method for determining the position of the aerial base station, the number of the ground terminals of the target area in the target time period is predicted in advance before the service capacity of the current base station cannot meet the requirements of users, and when the predicted number of the ground terminals is large and a preset threshold value is reached, the position of the aerial base station is determined according to the position information of the current ground terminals in the target area, so that the problem of communication interruption caused by untimely deployment of the aerial base station is solved.

Description

A method and device for determining the position of an air base station

technical field

The present invention relates to the field of communication technologies, in particular to a method and device for determining the position of an air base station.

Background technique

The construction of smart grid is an important direction for the development and construction of my country's power system, but in actual construction, there is often a problem of incoordination between planning and construction, resulting in the situation that local hotspot areas cannot get good communication services. On the one hand, due to the lag in the construction of power grid communication nodes, the planned power grid communication nodes have not been completed yet, while the power projects in hotspot areas have been completed early and are in urgent need of communication; on the other hand, some hotspot areas are in the original grid planning It is not a hot spot, but it has gradually become a hot spot with the strengthening of the government's development efforts, such as airports, business districts, etc. There are gaps in the planning of these areas, and the construction of ground base stations has not kept up.

Due to the advantages of UAV-based aerial base stations such as hovering capability, easy deployment, flexible action, and low deployment cost, the use of UAV-based aerial base stations for ad hoc communication is regarded as an important supplement to ground communication networks. , which can effectively enhance the wireless capacity and coverage on the ground to meet the requirements of 5G and B5G cellular mobile communications. When a hotspot area appears and the ground base station cannot meet the communication needs of the user, the UAV can be equipped with a temporary base station to be placed over the hotspot area to enhance the capacity of the hotspot area.

However, at present, the deployment location of the air base station is determined after the hot spot area appears. It takes a certain period of time from the discovery of the hot spot area to the determination of the deployment location of the air base station. During this period, the service capacity of the ground deployed base station will be unable to meet the needs of users. The problems of communication interruption caused by the existing technology are caused by the defect that the deployment position of the air base station cannot be determined in time.

SUMMARY OF THE INVENTION

Therefore, the technical problem to be solved by the present invention is to overcome the defect in the prior art that the deployment position of the air base station cannot be determined in time, so as to provide a method and apparatus for determining the position of the air base station.

A first aspect of the present invention provides a method for determining the location of an air base station, comprising: acquiring historical ground terminal quantity data in a target area, and predicting the ground terminal quantity in the target area within a target time period according to the historical ground terminal quantity data; When the number of ground terminals in the target time period is greater than a preset threshold, the current ground terminal position information in the target area is obtained; the position of the air base station is determined according to the current ground terminal position information in the target area.

Optionally, in the method for determining the location of an air base station provided by the present invention, acquiring the historical ground terminal quantity data of the target area includes: acquiring initial historical ground terminal quantity data in multiple areas; according to the average value of the initial historical ground terminal quantity data; and standard deviation, filter the initial historical ground terminal quantity data; according to the geographic location information of multiple areas, group the filtered initial historical ground terminal quantity data to obtain the filtered historical ground terminal quantity of the target area data; according to the average value of the filtered historical ground terminal quantity data of the target area, fill in the filtered historical ground terminal quantity data of the target area to obtain the historical ground terminal quantity data of the target area.

Optionally, in the method for determining the position of the air base station provided by the present invention, the step of determining the position of the air base station according to the current ground terminal position information in the target area includes: determining the initial position of the air base station according to the current ground terminal position information; The initial position information of the base station and the current ground terminal position information determine the capacity of the air base station when it is at the initial position; the capacity when the air base station is at the initial position is input into the reinforcement learning model to determine the position of the air base station.

Optionally, in the method for determining the position of the air base station provided by the present invention, the step of determining the initial position of the air base station according to the current ground terminal position information includes: classifying the ground terminal according to the current ground terminal position information and the first preset clustering algorithm. Perform clustering to obtain the initial cluster center; perform clustering on the ground terminals according to the current ground terminal location information, the initial cluster center and the second preset clustering algorithm to obtain the target cluster center; determine the target cluster center as the air The initial location of the base station.

Optionally, in the method for determining the position of an air base station provided by the present invention, the air base station is a millimeter-wave air base station, and according to the initial position information of the air base station and the current ground terminal position information, the steps of determining the capacity of the air base station when the air base station is at the initial position, Including: determining the line-of-sight link loss and non-line-of-sight link loss from the ground terminal to the millimeter-wave air base station and the antenna gain during the millimeter-wave transmission process according to the initial position information of the millimeter-wave air base station and the current position information of the ground terminal; The environmental parameters of the current position of the air base station in the air, determine the line-of-sight link probability and non-line-of-sight link probability from the ground terminal to the millimeter-wave air base station according to the environmental parameters and the initial position of the millimeter-wave air base station; The distance link probability and the corresponding line-of-sight link loss and non-line-of-sight link loss are used to obtain the overall path loss from the ground terminal to the millimeter-wave air base station; according to the antenna gain and overall path loss of each ground terminal, determine the millimeter-wave air The total signal-to-noise ratio of the receiver when the base station is in the initial position; according to the total signal-to-noise ratio of the receiver, determine the capacity of the base station in the millimeter-wave air when the base station is in the initial position.

其中，

表示地面终端i到毫米波空中基站的视距链路损耗，

表示地面终端i到毫米波空中基站的非视距链路损耗，

ρ是由ρ＝32.4+20log(f)给出的固定路径损耗，f为毫米波空中基站装载的毫米波的频率，x_L对数正态随机变量，表示视距链路场景中的阴影效应，x_N是对数正态随机变量，表示非视距链路场景中的阴影效应，α_LoS表示是视距链路场景中的路径损耗指数，α_NLos表示非视距链路场景中的路径损耗指数，d_i为地面终端i距离空中基站的距离，PLoS_i为地面终端i与毫米波空中基站的视距链路概率，PNLoS_i为地面终端i与毫米波空中基站的非视距链路概率，

a和b是取决于环境的参数，

表示地面终端i到毫米波空中基站的仰角,h表示初始位置的高度，x_i表示地面终端i距离毫米波空中基站在地面上垂直投影的水平距离，PNLoS_i＝1-PLoS_i。Optionally, in the method for determining the position of an air base station provided by the present invention, the distance from the ground terminal to mm is obtained according to the line-of-sight link probability and the non-line-of-sight link probability and the corresponding line-of-sight link loss and non-line-of-sight link loss. The steps for the overall path loss of a base station in the air include:

in,

represents the line-of-sight link loss from the ground terminal i to the millimeter-wave air base station,

represents the non-line-of-sight link loss from the ground terminal i to the millimeter-wave air base station,

ρ is the fixed path loss given by ρ=32.4+20log(f), f is the frequency of the millimeter-wave carried by the millimeter-wave air base station, and x _L is a log-normal random variable representing the shadowing effect in the line-of-sight link scenario , x _N is a log-normal random variable, representing the shadowing effect in the non-line-of-sight link scenario, α _LoS is the path loss index in the line-of-sight link scenario, α _NLos is the path in the non-line-of-sight link scenario Loss index, d _i is the distance between the ground terminal i and the air base station, PLoS _i is the line-of-sight link probability between the ground terminal i and the millimeter-wave air base station, PNLoS _i is the non-line-of-sight link between the ground terminal i and the millimeter-wave air base station probability,

a and b are parameters that depend on the environment,

represents the elevation angle from the ground terminal i to the millimeter-wave air base station, h represents the height of the initial position, _xi represents the horizontal distance between the ground terminal i and the millimeter-wave air base station in the vertical projection on the ground, PNLoS _i =1-PLoS _i .

其中，SNR表示接收机总信噪比，G_a表示天线增益，G_a＝G_{i_main}G_{r_main}，G_{i_main}表示毫米波空中基站的主瓣增益，G_{r_main}表示毫米波空中基站的旁瓣增益,P_t为毫米波空中基站的发射功率，σ²为噪声，PL_i表示地面终端i到毫米波空中基站的总体路径损耗。Optionally, in the method for determining the position of the air base station provided by the present invention, according to the antenna gain and the overall path loss of each ground terminal, determine the total signal-to-noise ratio of the receiver when the millimeter-wave air base station is at the initial position, including:

Among them, SNR represents the total signal-to-noise ratio of the receiver, _Ga represents the antenna gain, _{Ga =G i_main G r_main} , G _{i_main} represents the main lobe gain of the millimeter-wave air base station, G _{r_main} represents the side lobe gain of the millimeter-wave air base station, P _t is the transmit power of the millimeter-wave air base station, σ ² is the noise, and PL _i represents the overall path loss from the ground terminal i to the millimeter-wave air base station.

A second aspect of the present invention provides an apparatus for determining the position of an aerial base station, comprising: a ground terminal quantity prediction module, configured to obtain historical ground terminal quantity data of a target area, and predict the target area's quantity within a target time period according to the historical ground terminal quantity data. The number of ground terminals; the current ground terminal location information acquisition module, if the number of ground terminals in the target area within the target time period is greater than the preset threshold, the current ground terminal location information acquisition module is used to acquire the current ground terminal location information in the target area; the base station The position determination module is used for determining the position of the air base station according to the current position information of the ground terminal in the target area.

A third aspect of the present invention provides a computer device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are processed by the at least one processor The controller executes, thereby executing the method for determining the position of an air base station provided by the first aspect of the present invention.

A fourth aspect of the present invention provides a computer-readable storage medium, characterized in that the computer-readable storage medium stores computer instructions, and the computer instructions are used to cause the computer to execute the method for determining the location of an air base station provided in the first aspect of the present invention.

The technical scheme of the present invention has the following advantages:

1. The method for determining the location of an air base station provided by the present invention predicts the number of ground terminals in the target area within the target time period in advance before the service capability of the current base station cannot meet the needs of users. When the threshold is preset, the position of the air base station is determined according to the current ground terminal position information in the target area, which avoids the problem of communication interruption caused by untimely deployment of the air base station.

2. In the method for determining the position of an air base station provided by the present invention, after obtaining the initial ground terminal quantity data, first filter the initial ground terminal quantity data, and then fill in the filtered data, thereby obtaining the history of the target area. The data on the number of ground terminals, through filtering processing, filtered out abnormal data in the initial ground terminal quantity data, and ensured the integrity of the data through filling processing. Therefore, the method for determining the position of the air base station provided by the present invention can be used for the target area in the target time period. The number of ground terminals in the system can be accurately predicted, so as to more timely and accurately determine the location where the air base station needs to be deployed.

3. In the method for determining the position of the air base station provided by the present invention, after determining the initial position of the air base station according to the current ground terminal position information, the capacity of the air base station is combined with reinforcement learning, so that the reinforcement learning model can determine the air base station according to the capacity of the initial position. The location of the base station improves the flexibility of the air base station deployment and the effectiveness of the communication supplementation of the air base station.

4. The device for determining the location of an air base station provided by the present invention pre-predicts the number of ground terminals in the target area within the target time period before the service capability of the current base station cannot meet the needs of users. When the threshold is preset, the position of the air base station is determined according to the current ground terminal position information in the target area, which avoids the problem of communication interruption caused by untimely deployment of the air base station.

Description of drawings

In order to illustrate the specific embodiments of the present invention or the technical solutions in the prior art more clearly, the following briefly introduces the accompanying drawings that need to be used in the description of the specific embodiments or the prior art. Obviously, the accompanying drawings in the following description The drawings are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without creative efforts.

1-5 are flowcharts of specific examples of a method for determining the position of an over-the-air base station in an embodiment of the present invention;

6 is a schematic block diagram of a specific example of an apparatus for determining the position of an over-the-air base station in an embodiment of the present invention;

FIG. 7 is a principle block diagram of a specific example of a computer device in an embodiment of the present invention.

Detailed ways

The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are a part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

In the description of the present invention, it should be noted that the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as there is no conflict with each other.

An embodiment of the present invention provides a method for determining the position of an over-the-air base station, as shown in FIG. 1 , including:

Step S10: Acquire historical ground terminal quantity data of the target area.

In a specific embodiment, the target area can be any area where ground terminals exist, such as airports, business districts, and residential areas. The historical ground terminal quantity data of the target area refers to the number of ground terminals in the target area in different time periods.

Step S20: Predict the number of ground terminals in the target area within the target time period according to the historical ground terminal quantity data. The target time period can be the next six months, one year, etc.

In a specific embodiment, the number of ground terminals in the target area in the target time period can be predicted by a differential autoregressive moving average model (Autoregressive Integrated Moving Average, ARIMA), and ARIMA consists of three parts, namely autoregressive (AR), Difference (I) and Moving Average (MA). AR is used to predict future values by a linear combination of past values, MA is used to predict future values by extracting the influence of past values, and I means differential time series to smooth it out.

In this embodiment of the present invention, the formula for predicting the number of ground terminals in the target area within the target time period by using ARIMA is as follows:

where y _t represents the prediction result of the target time period t, a _i is the AR parameter, θ _j is the MA parameter, ε _t is the random error term, and p, q are the orders of AR and MA, respectively. In a specific embodiment, for p and q, after obtaining the historical ground terminal quantity data, the data can be plotted to observe whether it is a stationary time series. The autocorrelation coefficient ACF and the partial autocorrelation coefficient PACF of the stationary time series are obtained respectively. Through the analysis of the autocorrelation graph and the partial autocorrelation graph, the optimal level p and order q are obtained, and finally the optimal ARIMA model is obtained. For AR parameters and MA parameters, the auto.arima function in R language can be used for automatic order determination.

Step S30: Determine whether the number of ground terminals in the target area in the target time period is greater than the preset threshold, if the number of ground terminals in the target area in the target time period is greater than the preset threshold, perform steps S40 and S50, if the target area is in the target area. The number of ground terminals in the time period is less than or equal to the preset threshold, and no operation is performed.

In a specific embodiment, the preset threshold may be determined according to the number of terminals that the base station in the current target area can support. For example, the preset threshold may be set to 85% of the number of terminals that the base station in the current target area can support. When the number of ground terminals in the target area in the target time period is greater than 85% of the number of terminals that can be supported by the base station in the current target area, it is determined that the number of ground terminals in the target area in the target time period is greater than the preset threshold, which means that the target area In the target time period, there may be a problem that the service capability of the current base station cannot meet the user's needs, causing communication interruption. Therefore, it is necessary to perform steps S40 and S50 to determine the location of the air base station and implement the deployment of the air base station.

Step S40: Acquire current ground terminal position information in the target area. The current ground terminal position information in the target area may be two-dimensional coordinate information of all terminals in the current target area.

Step S50: Determine the location of the air base station according to the current location information of the ground terminal in the target area.

Since the relative position of the terminal in the target area and the air base station will have a certain influence on the capacity of the air base station, in order to increase the capacity of the air base station as much as possible, the method for determining the position of the air base station provided in the embodiment of the present invention is calculated in the air base station. The location of the base station will be combined with the current ground terminal location information in the target area.

In the method for determining the location of an air base station provided by the embodiment of the present invention, the number of ground terminals in the target area in the target time period is predicted in advance before the service capability of the current base station cannot meet the needs of users. When the threshold is preset, the position of the air base station is determined according to the current ground terminal position information in the target area, which avoids the problem of communication interruption caused by untimely deployment of the air base station.

In an optional embodiment, in the method for determining the position of an over-the-air base station provided by the embodiment of the present invention, as shown in FIG. 2 , the foregoing step S10 specifically includes:

Step S11: Acquire initial historical ground terminal quantity data in multiple areas.

Step S12: Perform filtering processing on the initial historical ground terminal quantity data according to the average value and standard deviation of the initial historical ground terminal quantity data.

The purpose of filtering the initial historical ground terminal quantity data is to filter outliers in the initial historical ground terminal quantity data. When the data obeys the normal distribution, according to the definition of the normal distribution, the probability of being 3σ away from the mean value is P(|x-μ|>3σ)<=0.003, which belongs to a very small probability event. When the number of ground terminals in a certain period of time is greater than 3σ from the average value of the initial historical ground terminal number data, the ground terminal is determined Amounts are outliers. where μ and σ are the mean and standard deviation, respectively, based on the historical data set {x ₁ ,x ₂ ,…,x _m }, which can be calculated by the following formula:

Step S13: Group the filtered initial historical ground terminal quantity data according to the geographic location information of the multiple areas to obtain the filtered historical ground terminal quantity data of the target area.

In a specific embodiment, the initial historical ground terminal quantity data obtained in the early stage is data of multiple areas. In order to accurately analyze each area, in this embodiment of the present invention, the geographic location information and the acquisition of the initial historical ground terminal quantity data are obtained. Time groups the filtered initial historical ground terminal quantity data: U={U ₁ ,U ₂ ,...,U _i ,...,U _k }, where U _i represents the i-th group of data with the same time and location. Then, the filtered historical ground terminal quantity data of the target area is obtained according to the geographic location information of the target area to be studied.

Step S14: Fill in the filtered historical ground terminal quantity data of the target area according to the average value of the filtered historical ground terminal quantity data of the target area to obtain the historical ground terminal quantity data of the target area.

After the above step S13 is performed to delete the outliers, there will be a problem of missing data, and it is difficult to accurately predict the number of ground terminals using incomplete data. Therefore, the filtered data needs to be filled. In the embodiment of the present invention, conditional mean interpolation (CMI) is used to process the missing values in the data, and the average value of each group is calculated by the following steps as follows:

where X represents the number of filtered historical ground terminals in one of the groups, and |U _i | represents the size of U _i . After filling the corresponding missing values with the mean of each group, you can remove noise and get more accurate predictions.

In an optional embodiment, in the method for determining the position of an over-the-air base station provided by the embodiment of the present invention, as shown in FIG. 3 , the foregoing step S50 specifically includes:

Step S51: Determine the initial position of the air base station according to the current ground terminal position information.

In a specific embodiment, the establishment of the air base station is to meet the communication requirements of more ground terminals, so the terminals in the target area can be clustered according to the current ground terminal location information, the cluster center is determined, and the cluster center is determined as the air terminal. The initial location of the base station.

Step S52: Determine the capacity of the air base station when the air base station is at the initial position according to the initial position information of the air base station and the current ground terminal position information.

Step S53: Input the capacity of the air base station at the initial position into the reinforcement learning model to determine the position of the air base station. In the embodiment of the present invention, the capacity of the air base station at the initial position is input into the reinforcement learning model, so as to determine the position of the air base station, in order to maximize the capacity of the air base station.

Exemplarily, the reinforcement learning model uses the capacity of the air base station at the current position and the capacity of the position at the previous moment as the reward, and is obtained by training with the highest capacity as the optimization goal.

Reinforcement learning is the continuous exploration of the agent in a given scene to obtain information about the state of the environment, and the environment will feed back a reward value to the agent according to the action taken by the agent, and the agent learns the best decision through continuous exploration to obtain maximum long-term rewards. The reinforcement learning model (Deep Q Network, DNQ) in this embodiment will train the millimeter-wave air base station as an agent, and perform training to complete the task of maximizing the system capacity by adjusting the position of the millimeter-wave air base station.

The specific reinforcement learning model is modeled as follows:

Agent: The millimeter-wave air base station can be regarded as a single agent. At the beginning of the task, the millimeter-wave air base station selects an action according to the ∈-greedy strategy, and then the environment sends the next state and returns the reward to the agent body. The agent updates its knowledge with the rewards returned by the environment, evaluating the last action. This cycle continues until the end of the air base station mission.

State (State): The state set is the current position of the air base station, that is, S={(x,y)}.

Action: The action set is the movable direction of the air base station, that is, A={10 meters forward, 10 meters backward, 10 meters left, 10 meters right, hover} five options.

Reward: The immediate reward is set as the system capacity difference between the current moment and the previous moment, expressed as: R=C _capacity (t+t _δ )-C _capacity (t), where t _δ is the current moment and the previous moment time difference.

The main process of the DQN algorithm is:

Step 1, first initialize the experience playback pool D, its capacity is N; let the agent explore the environment, accumulate the experience pool to a certain extent, and randomly select a batch of samples for training.

Step2, initialize the Q network and its neural network parameters ω; initialize the target Q network and its neural network parameters ω ^- .

DQN consists of two networks with the same structure but different parameters, the Q network and the target Q network. The Q-network uses the latest parameters, while the parameters of the target Q-network are used a few iterations ago. Q(S, A; ω) represents the output of the current Q network, which is used to evaluate the value function of the current state-action pair; Q(S, A; ω ^- ) represents the output of the target Q network, which can be updated according to the loss function. The parameters of: After a certain number of iterations, the parameters of the Q network are copied to the target Q network. The purpose of the target Q network is to improve the stability of the algorithm, because the target Q value remains unchanged for a period of time, which reduces the correlation between the current Q value and the target Q value to a certain extent.

Step3 loop through the round episode=1,2,...,M:

Step3.1 Initialize the state set S;

Step3.2 Loop through step=1,2,...,T:

Step3.2.1 Take action policy A with ∈-greedy policy;

Step3.2.2 Execute action A, calculate the cost reward R for the millimeter-wave air base station to take action A in state S, and the millimeter-wave air base station obtains the planning state S' at the next moment;

Step3.2.3 Store the samples (S, A, R, S′) in the experience playback pool D;

Step3.2.4 Calculate the target Q value using the uniformly randomly sampled sample Minibatch in the experience playback pool, y _i =R+γ·max _A Q(S',A;ω ^- ), where y _i is the target Q value, R is the reward function, γ is the discount factor (γ(DiscountRate) is between 0 and 1, indicating the importance of future returns relative to current returns. When γ is 0, it is equivalent to only consider immediate returns and not long-term returns. , when γ is 1, long-term and immediate returns are equally important). The Q network parameters ω are updated to reduce the loss function [y _i -Q(S,A;ω)] ² .

Step3.2.5 Update the parameters of the target Q network planned by the base station every C steps, that is, ω ^- =ω, that is, copy the neural network parameter ω of the Q network to the neural network parameter ω ^- of the target Q network.

The above pre-established reinforcement learning model can adjust the position of the drone carrying the millimeter-wave air base station according to the capacity of the location at all times, and the change in the capacity difference between the previous moment and the current moment can be used as a reward for the reinforcement learning model. (reward), the larger the capacity difference, the greater the reward. The UAV carrying the millimeter-wave air base station will select the action according to the ∈-greedy strategy at the next moment (the action is randomly selected with the probability of ∈, and the action that can obtain the maximum reward is selected with the probability of 1-∈). With high capacity, the UAV carrying the millimeter-wave air base station has a higher probability of choosing it. Through multiple iterations, the UAV carrying the millimeter-wave air base station can move to the position with the largest capacity and realize dynamic deployment, thereby obtaining Optimum communication effect.

The method for determining the position of the air base station provided by the embodiment of the present invention combines the capacity of the air base station with the reinforcement learning, so that the reinforcement learning model can determine the position of the air base station according to the capacity of the current location, which improves the flexibility of the air base station deployment and increases the The effectiveness of the communication supplement of the air base station is improved.

In the embodiment of the present invention, the simulation parameters of the reinforcement learning model are shown in Table 1.

Table 1

In the method for determining the position of the air base station provided by the embodiment of the present invention, after the initial position of the air base station is determined according to the current ground terminal position information, the capacity of the air base station is combined with reinforcement learning, so that the reinforcement learning model can determine the air base station according to the capacity of the initial position. The location of the base station improves the flexibility of the air base station deployment and the effectiveness of the communication supplementation of the air base station.

In an optional embodiment, as shown in FIG. 4 , in the method for determining the position of an over-the-air base station provided by the embodiment of the present invention, the foregoing step S51 specifically includes:

Step S511: Cluster the ground terminals according to the current ground terminal position information and the first preset clustering algorithm to obtain an initial cluster center.

In the embodiment of the present invention, the first preset clustering algorithm is a self-organizing mapping algorithm (Self-organizing Maps, SOM). The SOM algorithm is an artificial neural network algorithm with visualization and unsupervised characteristics, which can simulate the human brain to process signals It has high adaptive learning ability and robustness, and is suitable for initial clustering analysis of complex samples. Specifically, the ground terminals are clustered according to the current ground terminal location information and the SOM algorithm to obtain the initial The steps for clustering centers include:

1) Weight initialization. A random number is assigned to the weight vector W _j (j=1,2,...,p) connecting the input node to the jth output node, and the initial number of cycles is set.

2) SOM initial clustering. For the position information X _N (N=1,2,...,N) of each ground terminal, firstly find the distance between the winner weight vector W _g and X _N in W _j according to the following formula:

Then, define Ng(t) as the winner's neighborhood, and unit _g as the winner, and move the connection weight vector corresponding to each unit in the neighborhood to X _i . Repeat under different training times until the network is stable, and complete the initial clustering of samples according to the responses of the output nodes. Its learning equation is:

为第N个样本第i个输入节点的输入；w_ij为第i个输入节点与第j个输出节点之间的连接权值，其中j∈N_N(t)。In the formula: η(t) is the t-th learning rate, which decreases with the increase of training times;

is the input of the ith input node of the Nth sample; w _ij is the connection weight between the ith input node and the jth output node, where j∈N _N (t).

3) The cluster center Z=(Z ₁ , Z ₂ , . . . , Z _K ) obtained by applying the SOM algorithm is used as the initial center.

Step S512: Cluster the ground terminals according to the current ground terminal position information, the initial cluster center and the second preset clustering algorithm to obtain the target cluster center.

In the embodiment of the present invention, the second preset clustering algorithm is the K-means algorithm. The K-means algorithm is an objective function division method that uses the sample Euclidean distance as a similarity evaluation index. Hierarchical clustering algorithm. The principle of K-means is simple and the running speed is fast, but the location of the initialized centroid has a great impact on the final clustering result and the running time. If it is just a completely random selection, it may lead to unsatisfactory clustering results. And the algorithm convergence is very slow, so in the embodiment of the present invention, before using the K-means algorithm for clustering, the SOM algorithm is used to obtain the initial cluster center. Specifically, the steps of clustering through the K-means algorithm include:

First, calculate the value of the sum of squared distances J(C) of all samples to the cluster center of their class, and divide each ground terminal into the class center closest to it, and the same class center is a class.

In the formula: _umn is a binary variable, _umn =1 indicates that the nth ground terminal belongs to the mth class, and _umn = 0 indicates that it does not belong to this class; _d ( _cm ,xn) is the ground terminal to its class The distance of the cluster center; _{cm is the cluster center; x n is} the other ground terminal data in the class.

Then, the cluster centers are updated, and the centers cm of K clusters are updated according to the division result calculated in the previous step, the least squares method and the _Lagrangian principle, until the convergence conditions are met.

The final clustering result is obtained, with a total of k clusters. Each cluster is assigned a millimeter-wave air base station, and the millimeter-wave air base station provides communication services for the power grid terminal equipment in the cluster. The initial position of each millimeter-wave air base station is the center cm of the corresponding _cluster .

Step S513: Determine the target cluster center as the initial position of the air base station.

In an optional embodiment, in the method for determining the position of an air base station provided by the embodiment of the present invention, the air base station is a millimeter-wave air base station. As shown in FIG. 5 , the foregoing step S52 specifically includes:

S521: Determine the line-of-sight link loss, non-line-of-sight link loss, and antenna gain during millimeter-wave transmission from the ground terminal to the millimeter-wave air base station according to the initial location information of the millimeter-wave air base station and the current ground terminal location information;

x_i为地面终端i距离毫米波空中基站在地面上垂直投影的水平距离，h为毫米波空中基站的高度。Exemplarily, according to the initial position information of the millimeter-wave air base station and the position information of the ground terminal, the distance between the ground terminal i and the millimeter-wave air base station can be obtained, that is,

x _i is the horizontal distance from the ground terminal i to the vertical projection of the millimeter-wave air base station on the ground, and h is the height of the millimeter-wave air base station.

According to the distance between the ground terminal i and the millimeter-wave air base station, the line-of-sight link loss and the non-line-of-sight link loss can be obtained by the following formulas:

表示地面终端i到毫米波空中基站的视距链路损耗，

表示地面终端i的到毫米波空中基站的非视距链路损耗，ρ是由ρ＝32.4+20log(f)给出的固定路径损耗，f为毫米波空中基站装载的毫米波的频率，χ_L对数正态随机变量，表示视距链路场景中的阴影效应，x_N是对数正态随机变量，表示非视距链路场景中的阴影效应，α_LoS表示是视距链路场景中的路径损耗指数，α_NLoS表示非视距链路场景中的路径损耗指数，d_i为地面终端i距离空中基站的距离，即

在本发明实施例中，α_LoS的值为2，α_NLoS的值为3.3，std(χ_L)的值为5.2，std(χ_N)的值为7.2。in,

represents the line-of-sight link loss from the ground terminal i to the millimeter-wave air base station,

represents the non-line-of-sight link loss from the ground terminal i to the millimeter-wave air base station, ρ is the fixed path loss given by ρ=32.4+20log(f), f is the frequency of the millimeter wave carried by the millimeter-wave air base station, χ _L is a log-normal random variable, representing the shadowing effect in the line-of-sight link scene, x _N is a log-normal random variable, representing the shadowing effect in the non-line-of-sight link scene, and α _LoS indicates the line-of-sight link scene The path loss index in α _NLoS is the path loss index in the non-line-of-sight link scenario, d _i is the distance between the ground terminal i and the air base station, that is

In the embodiment of the present invention, the value of α _LoS is 2, the value of α _NLoS is 3.3, the value of std(χ _L ) is 5.2, and the value of std(χ _N ) is 7.2.

Antenna arrays are deployed on both the millimeter-wave air base station and the ground receiver to form a directional beam, and the antenna gain will also be affected during the signal transmission process. Impact Assuming that when establishing communication, the antennas on the millimeter-wave air base station and the ground power terminal can be adjusted in angle and aligned with each other, and the antenna gain is the product of the main lobe gain and the side lobe gain of the transmitter:

G _a =G _{i_main} G _{r_main} ,

G _{i_main} represents the main lobe gain of the millimeter-wave air base station, and G _{r_main} represents the side lobe gain of the millimeter-wave air base station. The main lobe gain and side lobe gain are inherent parameters of the base station antenna, which are related to their specific models. Exemplarily, the main lobe gain and side lobe gain can be selected. The lobe gain is 10dB, and the side lobe gain is -10dB.

S522: Obtain the environmental parameters of the current position of the millimeter-wave air base station, and determine the line-of-sight link probability and the non-line-of-sight link probability from the ground terminal to the millimeter-wave air base station according to the environmental parameters and the initial position of the millimeter-wave air base station;

According to the environmental parameters and the current location information of the millimeter-wave air base station, the line-of-sight link probability between the ground terminal and the millimeter-wave air base station can be determined by the following formula:

表示地面终端i到毫米波空中基站的仰角,h表示初始位置的高度，具体实施例中，初始位置的高度可以确定为100米，a和b是取决于环境的参数，如高层城市环境(a＝27.23；b＝0.08)，密集城市环境(a＝12.08；b＝0.11)，郊区环境(a＝4.88；b＝0.43)。Among them, PLoS _i is the line-of-sight link probability between the ground terminal i and the millimeter-wave air base station,

Represents the elevation angle from the ground terminal i to the millimeter-wave air base station, h represents the height of the initial position, in a specific embodiment, the height of the initial position can be determined to be 100 meters, a and b are parameters depending on the environment, such as high-rise urban environment (a = 27.23; b = 0.08), dense urban environment (a = 12.08; b = 0.11), suburban environment (a = 4.88; b = 0.43).

The corresponding non-line-of-sight link probability between the ground terminal i and the millimeter-wave air base station is:

PNLoS _i =1-PLoS _i .

S523, according to the line-of-sight link probability and the non-line-of-sight link probability and the corresponding line-of-sight link loss and non-line-of-sight link loss, obtain the overall path loss from the ground terminal to the millimeter-wave air base station;

Exemplarily, according to the line-of-sight link probability and the non-line-of-sight link probability and the corresponding line-of-sight link loss and non-line-of-sight link loss, the way to obtain the overall path loss from the ground terminal to the millimeter-wave air base station may be through The following formula determines:

表示地面终端i到毫米波空中基站的视距链路损耗，PNLoS_i表示地面终端i与毫米波空中基站的非视距链路概率，

表示地面终端i的到毫米波空中基站的非视距链路损耗。Among them, PL _i represents the overall path loss from ground terminal i to the millimeter-wave air base station, and PLoS _i represents the line-of-sight link probability between the ground terminal i and the millimeter-wave air base station,

represents the line-of-sight link loss from ground terminal i to the millimeter-wave air base station, PNLoS _i represents the non-line-of-sight link probability between ground terminal i and the millimeter-wave air base station,

represents the non-line-of-sight link loss of ground terminal i to the mmWave air base station.

S524: Determine the total signal-to-noise ratio of the receiver at the initial position of the millimeter-wave air base station according to the antenna gain and overall path loss of each ground terminal;

Exemplarily, according to the antenna gain and the overall path loss of each ground terminal, the total signal-to-noise ratio of the receiver at the initial position of the millimeter-wave air base station can be determined by the following formula:

S525: Determine the capacity of the millimeter-wave air base station at the initial position according to the total signal-to-noise ratio of the receiver.

Exemplarily, capacity is an important indicator of the communication capability of the communication system and terminal equipment. According to the signal-to-noise ratio formula, the capacity of the millimeter-wave air base station at the current location can be deduced as:

C _capacity =B log ₂ (1+SNR);

Among them, C _capacity represents the capacity of the millimeter-wave air base station at the initial position, B is the channel bandwidth, and SNR represents the total signal-to-noise ratio of the receiver.

In any of the above embodiments, the simulation parameters of the millimeter wave transmission process are shown in Table 2 below.

Table 2

Parameter Value Frequency 28GHz Bandwidth 2GHz Transmit Power 1W α_LoS,α_NLoS 2,3.3 std(χ_L), std(χ_N) 5.2, 7.2 Region of Interest 100×100m² Height of UAV 100m Overload thresholdτ 85%

An embodiment of the present invention further provides an apparatus for determining the position of an over-the-air base station, as shown in FIG. 6 , including:

The ground terminal quantity prediction module 10 is used to obtain the historical ground terminal quantity data of the target area, and predict the ground terminal quantity of the target area within the target time period according to the historical ground terminal quantity data. For details, see the above description of step S10 and step S20 , and will not be repeated here.

The current ground terminal position information acquisition module 20, if the number of ground terminals in the target area within the target time period is greater than the preset threshold, the current ground terminal position information acquisition module is used to acquire the current ground terminal position information in the target area. For details, see the above The description of step S30 and step S40 will not be repeated here.

The base station position determination module 30 is configured to determine the position of the air base station according to the current ground terminal position information in the target area. For details, refer to the above description of step S50, which will not be repeated here.

The apparatus for determining the location of an aerial base station provided by the embodiment of the present invention pre-predicts the number of ground terminals in the target area within the target time period before the service capability of the current base station cannot meet the needs of users. When the threshold is preset, the position of the air base station is determined according to the current ground terminal position information in the target area, which avoids the problem of communication interruption caused by untimely deployment of the air base station.

Another embodiment of the present invention provides a computer device. As shown in FIG. 7 , the computer device mainly includes one or more processors 61 and a memory 62 . In FIG. 7 , one processor 61 is used as an example.

The computer equipment may also include: an input device 63 and an output device 64 .

The processor 61 , the memory 62 , the input device 63 and the output device 64 may be connected by a bus or in other ways, and the connection by a bus is taken as an example in FIG. 7 .

The processor 61 may be a central processing unit (Central Processing Unit, CPU). The processor 61 may also be other general-purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuit (Application Specific Integrated Circuit, ASIC), Field-Programmable Gate Array (Field-Programmable Gate Array, FPGA) or Other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components and other chips, or a combination of the above types of chips. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The memory 62 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the air base station location determination device, and the like. Additionally, memory 62 may include high-speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, memory 62 may optionally include memory located remotely from processor 61, which remote memory may be connected via a network to the airborne base station location determination device. The input device 63 may receive user input of calculation requests (or other numerical or character information) and generate key signal input associated with the air base location determination device. The output device 64 may include a display device such as a display screen, and is used to output the calculation result.

An embodiment of the present invention further provides a computer-readable storage medium, where the computer-readable storage medium stores computer instructions, where the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions can execute any of the foregoing method embodiments. A method for determining the location of an air base station. Wherein, the storage medium may be a magnetic disk, an optical disk, a read-only memory (Read-Only Memory, ROM), a random access memory (Random Access Memory, RAM), a flash memory (Flash Memory), a hard disk (Hard) Disk Drive, abbreviation: HDD) or solid-state drive (Solid-State Drive, SSD), etc.; the storage medium may also include a combination of the above-mentioned types of memories.

Obviously, the above-mentioned embodiments are only examples for clear description, and are not intended to limit the implementation manner. For those of ordinary skill in the art, changes or modifications in other different forms can also be made on the basis of the above description. There is no need and cannot be exhaustive of all implementations here. And the obvious changes or changes derived from this are still within the protection scope of the present invention.

Claims (5)

1. A method for determining a location of an airborne base station, comprising:

acquiring historical ground terminal quantity data of a target area, and predicting the ground terminal quantity of the target area in a target time period according to the historical ground terminal quantity data;

if the number of the ground terminals of the target area in the target time period is larger than a preset threshold value, acquiring current ground terminal position information in the target area;

determining the position of the air base station according to the current ground terminal position information in the target area;

the step of determining the position of the air base station according to the current ground terminal position information in the target area comprises the following steps:

determining the initial position of the air base station according to the current ground terminal position information;

determining the capacity of the aerial base station when the aerial base station is positioned at the initial position according to the initial position information of the aerial base station and the current ground terminal position information;

inputting the capacity of the aerial base station at the initial position into a reinforcement learning model, and determining the position of the aerial base station;

the step of determining the initial position of the air base station according to the current ground terminal position information comprises the following steps:

clustering the ground terminals according to the current ground terminal position information and a first preset clustering algorithm to obtain an initial clustering center;

clustering the ground terminals according to the current ground terminal position information, an initial clustering center and a second preset clustering algorithm to obtain a target clustering center;

determining the target cluster center as an initial position of the airborne base station;

the air base station is a millimeter wave air base station, and the step of determining the capacity of the air base station at the initial position according to the initial position information of the air base station and the current ground terminal position information comprises the following steps:

according to the initial position information of the millimeter wave air base station and the current position information of the ground terminal, determining the line-of-sight link loss and the non-line-of-sight link loss from the ground terminal to the millimeter wave air base station and the antenna gain in the millimeter wave transmission process;

acquiring environmental parameters of the current position of the millimeter wave air base station, and determining the line-of-sight link probability and the non-line-of-sight link probability from the ground terminal to the millimeter wave air base station according to the environmental parameters and the initial position of the millimeter wave air base station;

obtaining the total path loss from the ground terminal to the millimeter wave air base station according to the line-of-sight link probability and the non-line-of-sight link probability and the corresponding line-of-sight link loss and the non-line-of-sight link loss;

determining the total signal-to-noise ratio of a receiver when the millimeter wave air base station is at the initial position according to the antenna gain and the total path loss of each ground terminal;

determining the capacity of the millimeter wave air base station at the initial position according to the total signal-to-noise ratio of the receiver;

the step of obtaining the total path loss from the ground terminal to the millimeter wave air base station according to the line-of-sight link probability and the non-line-of-sight link probability and the corresponding line-of-sight link loss and the non-line-of-sight link loss comprises the following steps:

，

wherein,

ground terminal

Line-of-sight link loss to millimeter wave air base stations,

；

ground terminal

Non-line-of-sight link loss to millimeter wave air base stations,

，

is composed of

Given the fixed path loss that is present,

the frequency of the millimeter wave carried by the millimeter wave air base station,

a lognormal random variable, representing shadowing effects in line-of-sight link scenarios,

is a lognormal random variable, represents the shadowing effect in a non-line-of-sight link scene,

the representation is the path loss exponent in the line-of-sight link scenario,

representing the path loss exponent in a non-line-of-sight link scenario,

for ground terminals

The distance from the base station in the air,

is the line-of-sight link probability of the ground terminal i and the millimeter wave air base station,

the probability of the non-line-of-sight link between the ground terminal i and the millimeter wave air base station,

，

and

is a parameter that depends on the circumstances and,

，

representing the elevation angle of the ground terminal i to the millimeter wave air base station,

which represents the height of the initial position of the device,

represents the horizontal distance of the ground terminal i from the millimeter wave air base station projected vertically on the ground,

；

determining the total signal-to-noise ratio of the receiver when the millimeter wave air base station is at the initial position according to the antenna gain and the total path loss of each ground terminal, wherein the determining comprises the following steps:

，

wherein,

which represents the overall signal-to-noise ratio of the receiver,

the gain of the antenna is represented by,

，

representing the main lobe gain of the millimeter wave airborne base station,

representing the side lobe gain of the millimeter wave air base station,

is the transmission power of the millimeter wave air base station,

in order to be a noise, the noise is,

representing the overall path loss from the ground terminal i to the millimeter wave air base station.

2. The method of claim 1, wherein obtaining historical ground terminal quantity data for a target area comprises:

acquiring initial historical ground terminal quantity data of a plurality of areas;

filtering the initial historical ground terminal quantity data according to the average value and the standard deviation of the initial historical ground terminal quantity data;

grouping the initial historical ground terminal quantity data subjected to filtering processing according to the geographic position information of the plurality of areas to obtain the historical ground terminal quantity data subjected to filtering processing of the target area;

and filling the filtered historical ground terminal quantity data of the target area according to the average value of the filtered historical ground terminal quantity data of the target area to obtain the historical ground terminal quantity data of the target area.

3. An over-the-air base station position determining apparatus, comprising:

the ground terminal quantity prediction module is used for acquiring historical ground terminal quantity data of a target area and predicting the ground terminal quantity of the target area in a target time period according to the historical ground terminal quantity data;

the current ground terminal position information acquisition module is used for acquiring current ground terminal position information in the target area if the number of ground terminals in the target area in the target time period is greater than a preset threshold value;

the base station position determining module is used for determining the position of the aerial base station according to the current ground terminal position information in the target area;

determining the position of the air base station according to the current ground terminal position information in the target area, wherein the determining comprises the following steps:

determining the initial position of the air base station according to the current ground terminal position information;

determining the capacity of the aerial base station when the aerial base station is positioned at the initial position according to the initial position information of the aerial base station and the current ground terminal position information;

inputting the capacity of the aerial base station at the initial position into a reinforcement learning model, and determining the position of the aerial base station;

the step of determining the initial position of the air base station according to the current ground terminal position information comprises the following steps:

clustering the ground terminals according to the current ground terminal position information and a first preset clustering algorithm to obtain an initial clustering center;

clustering the ground terminals according to the current ground terminal position information, an initial clustering center and a second preset clustering algorithm to obtain a target clustering center;

determining the target cluster center as an initial position of the airborne base station;

the air base station is a millimeter wave air base station, and the step of determining the capacity of the air base station at the initial position according to the initial position information of the air base station and the current ground terminal position information comprises the following steps:

according to the initial position information of the millimeter wave air base station and the current position information of the ground terminal, determining the line-of-sight link loss and the non-line-of-sight link loss from the ground terminal to the millimeter wave air base station and the antenna gain in the millimeter wave transmission process;

acquiring environmental parameters of the current position of the millimeter wave air base station, and determining the line-of-sight link probability and the non-line-of-sight link probability from the ground terminal to the millimeter wave air base station according to the environmental parameters and the initial position of the millimeter wave air base station;

obtaining the total path loss from the ground terminal to the millimeter wave air base station according to the line-of-sight link probability and the non-line-of-sight link probability as well as the corresponding line-of-sight link loss and the non-line-of-sight link loss;

determining the total signal-to-noise ratio of a receiver when the millimeter wave air base station is at the initial position according to the antenna gain and the total path loss of each ground terminal;

determining the capacity of the millimeter wave air base station at the initial position according to the total signal-to-noise ratio of the receiver;

the step of obtaining the total path loss from the ground terminal to the millimeter wave air base station according to the line-of-sight link probability and the non-line-of-sight link probability and the corresponding line-of-sight link loss and the non-line-of-sight link loss comprises the following steps:

，

wherein,

ground terminal

Line-of-sight link loss to millimeter wave air base stations,

；

ground terminal

Non line-of-sight link loss to millimeter wave air base stations,

，

is formed by

Given the fixed path loss that is present,

the frequency of the millimeter wave carried by the millimeter wave air base station,

a lognormal random variable, representing shadowing effects in line-of-sight link scenarios,

is lognormalA random variable, representing the shadow effect in a non-line-of-sight link scene,

the representation is the path loss exponent in the line-of-sight link scenario,

representing the path loss exponent in a non-line-of-sight link scenario,

for ground terminals

The distance from the base station in the air,

is the line-of-sight link probability of the ground terminal i and the millimeter wave air base station,

the probability of the non-line-of-sight link between the ground terminal i and the millimeter wave air base station,

，

and

is a parameter that depends on the circumstances and,

，

representing the elevation angle of the ground terminal i to the millimeter wave air base station,

the height of the initial position is indicated,

represents the horizontal distance of the ground terminal i from the millimeter wave air base station projected vertically on the ground,

；

determining the total signal-to-noise ratio of the receiver when the millimeter wave air base station is at the initial position according to the antenna gain and the total path loss of each ground terminal, wherein the determining comprises the following steps:

，

wherein,

which represents the overall signal-to-noise ratio of the receiver,

the gain of the antenna is represented by,

，

representing the main lobe gain of the millimeter wave airborne base station,

representing the side lobe gain of the millimeter wave air base station,

for transmission power of millimeter wave air base station，

In order to be a noise, the noise is,

representing the overall path loss from the ground terminal i to the millimeter wave air base station.

4. A computer device, comprising:

at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to perform the method of aerial base station position determination of claim 1 or 2.

5. A computer readable storage medium having stored thereon computer instructions for causing a computer to perform the method of aerial base station position determination of claim 1 or 2.

Images