NOMA and D2D-based dense cellular network performance optimization method

Abstract

The invention relates to the technical field of wireless communication network optimization, and discloses a dense cellular network performance optimization method based on NOMA and D2D, and the method comprises the following steps: S1, obtaining the real-time signal intensity through a base station monitoring unit; s2, acquiring base station preset signal strength of the user terminal; s3, comparing the real-time signal intensity of the specified base station acquired by the user terminal with the preset signal intensity, and if the real-time signal intensity is greater than the preset signal intensity, selecting a cellular mode; otherwise, entering the step S4; and S4, the user terminal searches the surrounding HTCU which is successfully connected to the specified base station, takes the HTCU as a D2D relay, and performs data transmission in a two-hop D2D mode, otherwise, the user terminal is in an interrupted state, and by deeply fusing NOMA and D2D communication technologies in the dense cellular network, the network delay is reduced, the burden of the base station is relieved, and the local multiplexing efficiency of spectrum resources is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of wireless communication network optimization, and in particular to a method for optimizing the performance of a dense cellular network based on NOMA and D2D. Background Art

[0002] With the rapid development of mobile technology and the global deployment of 5G networks, the use of smart terminal devices has entered an explosive growth; traditional cellular networks cannot provide high-quality networks for massive communication devices, so dense cellular networks are used to meet the network performance requirements of the 5G era of the Internet of Everything.

[0003] Dense cellular networks deploy densely packed small base stations under the base stations of cellular networks to reduce the load on large base stations and provide users with high-quality spectrum resources, thereby ensuring network transmission speeds for massive communication devices.

[0004] Dense cellular networks are mainly used for human type communication by human type communication users (HTCU). As the types of smart terminal devices gradually increase, there are also a large number of machine type communication devices (MTCD) in the scenario of the Internet of Everything. If machine type communication devices are also directly connected to the base station through the cellular mode, the burden on the base station will be increased, which may lead to a decline in network quality. Summary of the invention

[0005] The purpose of the present invention is to provide a method for optimizing the performance of a dense cellular network based on NOMA and D2D to solve the following technical problems:

[0006] How to reduce the burden on base stations and ensure the network quality of each user terminal.

[0007] The purpose of the present invention can be achieved by the following technical solutions:

[0008] A method for optimizing the performance of a dense cellular network based on NOMA and D2D is applied to a dense cellular network system, wherein the dense cellular network system includes multiple base stations, multiple user terminals, and multiple communication modes; the optimization method includes the following steps:

[0009] S1: By setting a base station monitoring unit in the user terminal, the real-time signal strength of the specified base station is obtained;

[0010] S2: Obtain the preset signal strength of the base station of the user terminal by establishing a Dec-POMDP model;

[0011] S3: Compare the real-time signal strength of the designated base station obtained by the user terminal with the preset signal strength. If the real-time signal strength is greater than the preset signal strength, select the cellular mode, and the user terminal directly transmits the data to the designated base station through the cellular link; otherwise, proceed to step S4;

[0012] S4: The user terminal searches for HTCUs in the surrounding area that have been successfully connected to the designated base station as potential relay devices. From the searched HTCUs, the HTCU that is closest to the real-time device and has been successfully connected to the designated base station is selected as the D2D relay. Data is transmitted through the two-hop D2D mode. The user terminal first sends data to the selected HTCU relay through the D2D link, and the relay then forwards the data to the designated base station through the cellular link; otherwise, the user terminal is in an interrupted state.

[0013] As a further solution of the present invention: the user terminal includes an HTCU and an MTCD.

[0014] As a further solution of the present invention: the specific method for obtaining the preset signal strength of the base station of the user terminal comprises the following steps:

[0015] S10: The base station is taken as the intelligent agent, the increase or decrease of the preset signal strength by the base station is taken as the action, the real-time signal strength of the base station is taken as the state of the base station, and the environment is fed back to the base station as a reward to establish the Dec-POMDP model;

[0016] S20: Each base station initializes a Q table, stores the Q values ​​of all possible state-action pairs, and updates the Q values;

[0017] S30: The base station selects the best action in the Q table with a probability of ε, and randomly selects an action with a probability of 1-ε;

[0018] S40: By formula Calculate the reward r t ;

[0019] in, and are the rates of the j-th HTCU to the i-th base link at step t and step t-1, respectively. and are the rates of the k-th MTCD k to the i-th base link at step t and step t-1, respectively;

[0020] S50: Repeat steps S30-S50 until the training end condition is reached; each base station finds its optimal preset signal strength.

[0021] As a further solution of the present invention: the process of determining the designated base station is:

[0022] S100: Each base station obtains base station transmission data through a base station monitoring unit;

[0023] S200: Each base station obtains a status evaluation index of each base station by analyzing base station transmission data;

[0024] S300: Determine a target area of ​​the user terminal with the user terminal as the center;

[0025] S400: numbering all base stations in the target area, with the number being i;

[0026] S500: The user terminal obtains the status evaluation index of each base station in the target area;

[0027] S600: Acquire a connection index between the user terminal and each base station by evaluating the status evaluation index of each base station and the real-time distance between the user terminal and each base station; and determine a base station corresponding to the maximum value of the connection index as a designated base station.

[0028] As a further solution of the present invention: the base station transmission data includes real-time delay time, real-time packet loss rate, real-time number of connected user terminals and real-time average transmission rate.

[0029] As a further solution of the present invention: the status evaluation index of each base station is obtained by the formula:

[0030]

[0031] Calculate the state evaluation index S of the base station numbered i i ;

[0032] Among them, Q Link is the number of user terminals connected to the base station numbered i; Q0 is the preset number of user terminals connected to the base station numbered i; t s is the real-time delay time of the base station numbered i; t0 is the preset delay time of the base station numbered i; is the real-time packet loss rate of the base station numbered i; is the preset packet loss rate of the base station numbered i; c is the real-time average transmission rate of the connected user terminals of the base station numbered i; v0 is the preset average transmission rate of the connected user terminals of the base station numbered i; ε is the unit-removing coefficient; C1 is the first preset constant; C2 is the second preset constant; C3 is the third preset constant; C4 is the fourth preset constant; γ1 is the first weight coefficient; γ2 is the second weight coefficient; γ3 is the third weight coefficient; γ4 is the fourth weight coefficient; S0 is the basic status evaluation index.

[0033] As a further solution of the present invention: the connection index is obtained by the formula:

[0034]

[0035] Calculate the connection index L between the user terminal and the base station numbered i in the target area i ;

[0036] Where f(X) is the judgment function. When X≥0, f(X)=X; when X<0, f(X)=0; (X i , Y i , Z i ) are the coordinates of the base station numbered i; (X0, Y0, Z0) are the coordinates of the user terminal; R i is the preset coverage radius of the base station numbered i; σ i is the building density between user terminals and base station numbered i.

[0037] As a further solution of the present invention: the cellular mode includes an OMA mode and a NOMA mode; the two-hop D2D mode includes a D2D-assisted OMA mode and a D2D-assisted NOMA mode.

[0038] Beneficial effects of the present invention:

[0039] (1) The present invention deeply integrates non-orthogonal multiple access (NOMA) and device-to-device (D2D) communication technology in dense cellular networks. D2D communication technology allows neighboring devices to directly exchange data without passing through base stations. This principle significantly reduces network latency, reduces the burden on base stations, and improves the local multiplexing efficiency of spectrum resources, ensuring the network quality of each user terminal. NOMA technology breaks the limitations of traditional orthogonal multiple access. By allowing signals to be superimposed and transmitted on non-orthogonal resources such as power domain, code domain or spatial domain, multiple users can share resources at the same time, frequency or code channel, thereby greatly improving user connection density and spectrum utilization. By combining these two technologies, not only can the communication performance of the system be significantly improved, but also large-scale device connections can be effectively supported while meeting the different communication requirements of HTCU and MTCD, effectively solving various technical problems arising in the integration process, and optimizing from the perspectives of user mode selection, user grouping and power allocation in dense cellular networks to improve the overall spectrum efficiency of the system, promote the widespread application of 5G and future communication networks in the era of the Internet of Everything, and bring users a smoother and more intelligent communication experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The present invention will be further described below in conjunction with the accompanying drawings.

[0041] Figure 1 The present invention is a method flow chart of an embodiment of the present invention. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] See also Figure 1 As shown, in one embodiment, a method for optimizing the performance of a dense cellular network based on NOMA and D2D is provided, which is applied to a dense cellular network system, wherein the dense cellular network system includes multiple base stations, multiple user terminals, and multiple communication modes; the user terminals include HTCU and MTCD; the optimization method includes the following steps:

[0044] S1: By setting a base station monitoring unit in the user terminal, the real-time signal strength of the specified base station is obtained;

[0045] S2: Obtain the preset signal strength of the base station of the user terminal by establishing a Dec-POMDP model;

[0046] S3: Compare the real-time signal strength of the designated base station obtained by the user terminal with the preset signal strength. If the real-time signal strength is greater than the preset signal strength, select the cellular mode, and the user terminal directly transmits the data to the designated base station through the cellular link; otherwise, proceed to step S4;

[0047] S4: The user terminal searches for HTCUs in the surrounding area that have been successfully connected to the designated base station as potential relay devices. From the searched HTCUs, the HTCU that is closest to the real-time device and has been successfully connected to the designated base station is selected as the D2D relay. Data is transmitted through the two-hop D2D mode. The user terminal first sends data to the selected HTCU relay through the D2D link, and the relay then forwards the data to the designated base station through the cellular link; otherwise, the user terminal is in an interrupted state and cannot perform data transmission.

[0048] Through the above technical scheme, the cellular mode of this embodiment includes OMA mode and NOMA mode; the two-hop D2D mode includes D2D-assisted OMA mode and D2D-assisted NOMA mode; by deeply integrating non-orthogonal multiple access (NOMA) and device-to-device (D2D) communication technology in dense cellular networks, D2D communication technology allows neighboring devices to directly exchange data without passing through base stations. This principle significantly reduces network latency, reduces the burden on base stations, and improves the local multiplexing efficiency of spectrum resources; NOMA technology breaks the limitations of traditional orthogonal multiple access, and allows signals to be superimposed and transmitted on non-orthogonal resources such as power domain, code domain or spatial domain, thereby achieving multiple users at the same time and frequency. By combining these two technologies, not only can the communication performance of the system be significantly improved, but also large-scale device connections can be effectively supported to meet the different communication requirements of HTCU and MTCD, effectively solving various technical problems arising in the integration process, and optimizing the user mode selection, user grouping and power allocation in dense cellular networks to improve the overall spectrum efficiency of the system, promote the widespread application of 5G and future communication networks in the era of the Internet of Everything, and bring users a smoother and smarter communication experience; any HTCU or MTCD will select the cellular mode only when the real-time signal strength from the designated base station is greater than the preset signal strength. , otherwise, it will select the two-hop D2D mode and select a HTCU (intermediate user) that is closest to the base station and successfully connected to the base station as a relay to establish a D2D link. On the one hand, this communication mode can enable the HTCU or MTCD communicating in the two-hop D2D mode to maintain a stable D2D link at a higher average rate; on the other hand, from the perspective of energy efficiency, it can also reduce the energy consumption of the UE (i.e., HTCU or MTCD) in uplink transmission; if the cellular mode or the two-hop D2D mode is not selected, the HTCU or MTCD is in an interrupted state; the user can send its data to the corresponding base station via a one-hop cellular link or a two-hop D2D link, where the user is the transmitter, the base station and the D2D The relayed HTCU is the receiver. During the two-hop D2D uplink transmission process, the device first sends data to the nearby HTCU (D2D relay) through the D2D link, and then the HTCU forwards the data to the corresponding base station through the cellular link; the base and user terminals are distributed using HPPP, and the distribution of HTCU and MTCD is independent of the base, and is modeled according to two other independent HPPPs. Based on the fact that the number of MTCDs in reality is much larger than HTCU, and only a part of MTCDs are active in each time slot, the scenario of the present invention considers that MTCDs are randomly active in each time slot, and all bases, HTCUs and MTCDs are equipped with single antennas, and the available channels are fully shared and reused in each cell.

[0049] As an implementation manner of the present invention, the specific method for obtaining the preset signal strength of the base station of the user terminal includes the following steps:

[0050] S10: The base station is taken as the intelligent agent, the increase or decrease of the preset signal strength by the base station is taken as the action, the real-time signal strength of the base station is taken as the state of the base station, and the environment is fed back to the base station as a reward to establish the Dec-POMDP model;

[0051] S20: Each base station initializes a Q table, stores the Q values ​​of all possible state-action pairs, and updates the Q values;

[0052] S30: The base station selects the best action in the Q table with a probability of ε, and randomly selects an action with a probability of 1-ε;

[0053] S40: By formula Calculate the reward r t ;

[0054] in, and are the rates of the j-th HTCU to the i-th base link at step t and step t-1, respectively. and are the rates of the k-th MTCD k to the i-th base link at step t and step t-1, respectively;

[0055] S50: repeating steps S30-S50 until the training end condition is reached; each base station finds its optimal preset signal strength;

[0056] Through the above technical solution, this embodiment takes the base station as the intelligent agent, the increase or decrease operation of the base station on the preset signal strength is the action, the real-time signal strength of the base station is the state of the base station, and the environment feedbacks the base station with rewards to establish the Dec-POMDP model: then initialize a Q table for each base station, store the Q values ​​of all possible state-action pairs, and use the Bellman equation to update the Q value; the base station selects the optimal action in the Q table with a probability of ε, and randomly selects the action with a probability of 1-ε; through the formula Calculate the reward r t ; is the rate difference between the j-th HTCU and the i-th base link at step t and step t-1; is the speed difference between the k-th MTCD k and the i-th base link at step t and step t-1; is the sum of the speed differences of all links at step t and step t-1; when the sum of the speed differences of all links at step t and step t-1 is The reward is positive; otherwise, the reward is negative; repeat steps S30-S50 until the training end condition is reached; each base station finds its optimal preset signal strength; by introducing an intermittent learning strategy, multi-agent reinforcement learning is applied to dense cellular networks to cope with the burstiness of MTCD traffic. In each training step, the base station learns and updates the preset signal strength only based on local observation information at the beginning, and maintains the threshold in subsequent time slots until the step ends and obtains a reward based on the overall performance. This method can effectively reduce unnecessary adjustments and improve system stability and performance optimization efficiency.

[0057] It should be noted that the method for updating the Q value is a prior art and will not be described in detail here;

[0058] As an implementation manner of the present invention, the process of determining the designated base station is:

[0059] S100: Each base station obtains base station transmission data through a base station monitoring unit;

[0060] S200: Each base station obtains a status evaluation index of each base station by analyzing base station transmission data;

[0061] S300: Determine a target area of ​​the user terminal with the user terminal as the center;

[0062] S400: numbering all base stations in the target area, with the number being i;

[0063] S500: The user terminal obtains the status evaluation index of each base station in the target area;

[0064] S600: Obtain connection indexes between the user terminal and each base station by evaluating the status evaluation index of each base station and the real-time distance between the user terminal and each base station; determine a base station corresponding to a maximum value in the connection index as a designated base station;

[0065] Through the above technical scheme, each base station in this embodiment obtains base station transmission data through the base station monitoring unit; each base station obtains the status evaluation index of each base station by analyzing the base station transmission data; then, with the user terminal as the center, the target area of ​​the user terminal is determined; all base stations in the target area are numbered, and the number is recorded as i; then the user terminal obtains the status evaluation index of each base station in the target area; finally, the connection index between the user terminal and each base station is obtained by comparing the status evaluation index of each base station and the real-time distance between the user terminal and each base station; the base station corresponding to the maximum value in the connection index is determined as the designated base station; because the current base stations are established relatively closely, there will be multiple base stations in the target area of ​​the user terminal. By calculating the connection index, it is ensured that the base station can be connected to a more suitable base station, thereby improving the quality of network connection.

[0066] As an implementation mode of the present invention, the base station transmission data includes real-time delay time, real-time packet loss rate, real-time number of connected user terminals and real-time average transmission rate;

[0067] It should be noted that the real-time delay time, real-time packet loss rate, real-time number of connected user terminals and real-time average transmission rate are obtained through the base station network monitoring equipment. The specific acquisition process is existing technology and will not be described in detail here.

[0068] As an implementation mode of the present invention, the status evaluation index of each base station is calculated by the formula:

[0069]

[0070] Calculate the state evaluation index S of the base station numbered i i ;

[0071] Among them, Q Link is the number of user terminals connected to the base station numbered i; Q0 is the preset number of user terminals connected to the base station numbered i; t s is the real-time delay time of the base station numbered i; t0 is the preset delay time of the base station numbered i; is the real-time packet loss rate of the base station numbered i; is the preset packet loss rate of the base station numbered i; c is the real-time average transmission rate of the connected user terminals of the base station numbered i; v0 is the preset average transmission rate of the connected user terminals of the base station numbered i; ε is the unit-removing coefficient; C1 is the first preset constant; C2 is the second preset constant; C3 is the third preset constant; C4 is the fourth preset constant; γ1 is the first weight coefficient; γ2 is the second weight coefficient; γ3 is the third weight coefficient; γ4 is the fourth weight coefficient; S0 is the basic status evaluation index;

[0072] Through the above technical solution, this embodiment Q Link - Q0 is the first difference between the number of connected user terminals of the base station numbered i and the preset number of connected user terminals; when the first difference Q Link -Q0<0, it means that the number of user terminals connected to the base station numbered i does not exceed the preset number of connected user terminals, so the first difference Q Link -The larger the absolute value of Q0 is, the better the base station status is. The status evaluation index S i The smaller the first difference Q Link - When Q0>0, it means that the number of user terminals connected to the base station numbered i exceeds the preset number of connected user terminals, so the first difference Q Link -The larger the Q0, the worse the base station status. iThe larger the s -t0 is the second difference between the real-time delay time of the base station numbered i and the preset delay time; when the second difference t s When -t0<0, it means that the real-time delay time of the base station numbered i does not exceed the preset delay time, so the second difference t s -The larger the absolute value of t0 is, the better the base station status is. The status evaluation index S i The smaller the second difference t s -t0>0, it means that the real-time delay time of the base station numbered i exceeds the preset delay time, so the second difference t s -The larger the t0 is, the worse the base station status is. The status evaluation index S i The bigger; is the third difference between the real-time packet loss rate of the base station numbered i and the preset packet loss rate; when the third difference , it means that the real-time packet loss rate of the base station numbered i does not exceed the preset packet loss rate, so the third difference The larger the absolute value, the better the base station status. i The smaller; the third difference , it means that the real-time packet loss rate of the base station numbered i exceeds the preset packet loss rate, so the third difference The larger the value, the worse the base station status. i The bigger; v c -v0 is the fourth difference between the real-time average transmission rate of the base station numbered i and the preset average transmission rate; when the fourth difference v c -v0<0, it means that the real-time average transmission rate of the base station numbered i does not exceed the preset average transmission rate, so the fourth difference v c -The larger the absolute value of v0 is, the better the base station status is. The status evaluation index S i The smaller the fourth difference v c -v0>0, it means that the real-time average transmission rate of the base station numbered i exceeds the preset average transmission rate, so the fourth difference v c -The larger the v0 is, the worse the base station status is. The status evaluation index S i The bigger;

[0073] It should be noted that the preset number Q0 of connected user terminals of the base station numbered i, the preset delay time t0 of the base station numbered i, and the preset packet loss rate of the base station numbered i are The preset average transmission rate v0, the unit coefficient ε, the first preset constant C1, the second preset constant C2, the third preset constant C3, the fourth preset constant C4, the first weight coefficient γ1, the second weight coefficient γ2, the third weight coefficient γ3, the fourth weight coefficient γ4 and the basic status assessment index S0 of the connected user terminals of the base station numbered i are preset values, which are obtained based on experience and will not be described in detail here.

[0074] As an embodiment of the present invention, the connection index is obtained by the formula:

[0075]

[0076] Calculate the connection index L between the user terminal and the base station numbered i in the target area i ;

[0077] Where f(X) is the judgment function. When X≥0, f(X)=X; when X<0, f(X)=0; (X i , Y i , Z i ) are the coordinates of the base station numbered i; (X0, Y0, Z0) are the coordinates of the user terminal; R i is the preset coverage radius of the base station numbered i; σ i is the building density between the user terminal and the base station numbered i;

[0078] Through the above technical solution, this embodiment is the distance between the base station numbered i and the user terminal; is the ratio of the distance between the base station numbered i and the user terminal to the preset coverage radius of the base station numbered i; x in the judgment function f(X) is when When , it indicates the ratio of the distance between the base station numbered i and the user terminal to the preset coverage radius of the base station numbered i is greater than 1, the distance between the base station numbered i and the user terminal exceeds the preset coverage radius of the base station numbered i, so Connection index L i =0; when 1- When , it indicates the ratio of the distance between the base station numbered i and the user terminal to the preset coverage radius of the base station numbered i is less than or equal to 1, the distance between the base station numbered i and the user terminal is within the preset coverage radius of the base station numbered i, so The smaller the distance between the base station numbered i and the user terminal, the smaller the connection index L. iThe larger the building density of user terminals and base stations numbered i, σ i The satellite image is analyzed and obtained. The specific analysis method is the existing technology and will not be described in detail here. When the building density σ between the user terminal and the base station numbered i i The larger the value, the greater the attenuation, so the connection index L i The smaller the distance between the base station numbered i and the user terminal, the smaller the connection index L i The larger the building density of user terminals and base stations numbered i, σ i The satellite image is analyzed and obtained. The specific analysis method is the existing technology and will not be described in detail here. When the building density σ between the user terminal and the base station numbered i i The larger the value, the greater the attenuation, so the state evaluation index S i The larger it is, the worse the base station status is, so the connection index L i The smaller;

[0079] It should be noted that the coordinates (X i , Y i , Z i ), the coordinates of the user terminal (X0, Y0, Z0), the building density σ between the user terminal and the base station numbered i i The acquisition method is the existing technology; the preset coverage radius R of the base station numbered i i It is the default value and will not be described in detail here.

[0080] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A method for optimizing the performance of a dense cellular network based on NOMA and D2D, applied to a dense cellular network system, wherein the dense cellular network system includes multiple base stations, multiple user terminals and multiple communication modes; characterized in that: The optimization method comprises the following steps: S1: By setting a base station monitoring unit in the user terminal, the real-time signal strength of the specified base station is obtained; S2: Obtain the preset signal strength of the base station of the user terminal by establishing a Dec-POMDP model; S3: Compare the real-time signal strength of the designated base station obtained by the user terminal with the preset signal strength. If the real-time signal strength is greater than the preset signal strength, select the cellular mode, and the user terminal directly transmits the data to the designated base station through the cellular link; otherwise, proceed to step S4; S4: The user terminal searches for HTCUs in the surrounding area that have been successfully connected to the designated base station as potential relay devices. From the searched HTCUs, the HTCU that is closest to the real-time device and has been successfully connected to the designated base station is selected as the D2D relay. Data is transmitted through the two-hop D2D mode. The user terminal first sends data to the selected HTCU relay through the D2D link, and the relay then forwards the data to the designated base station through the cellular link; otherwise, the user terminal is in an interrupted state.

2. According to claim 1, a method for optimizing the performance of a dense cellular network based on NOMA and D2D, characterized in that: The user terminal includes an HTCU and an MTCD.

3. According to a method for optimizing the performance of a dense cellular network based on NOMA and D2D according to claim 2, it is characterized in that: The specific method for obtaining the preset signal strength of the base station of the user terminal comprises the following steps: S10: The base station is taken as the intelligent agent, the increase or decrease of the preset signal strength by the base station is taken as the action, the real-time signal strength of the base station is taken as the state of the base station, and the environment is fed back to the base station as a reward to establish the Dec-POMDP model; S20: Each base station initializes a Q table, stores the Q values ​​of all possible state-action pairs, and updates the Q values; S30: The base station selects the best action in the Q table with a probability of ε, and randomly selects an action with a probability of 1-ε; S40: By formula Calculate the reward r t ; in, and are the rates of the j-th HTCU to the i-th base link at step t and step t-1, respectively. and are the rates of the k-th MTCD k to the i-th base link at step t and step t-1, respectively; S50: Repeat steps S30-S50 until the training end condition is reached; each base station finds its optimal preset signal strength.

4. According to claim 3, a method for optimizing the performance of a dense cellular network based on NOMA and D2D, characterized in that: The process of determining the designated base station is as follows: S100: Each base station obtains base station transmission data through a base station monitoring unit; S200: Each base station obtains a status evaluation index of each base station by analyzing base station transmission data; S300: Determine a target area of ​​the user terminal with the user terminal as the center; S400: numbering all base stations in the target area, with the number being i; S500: The user terminal obtains the status evaluation index of each base station in the target area; S600: Acquire a connection index between the user terminal and each base station by evaluating the status evaluation index of each base station and the real-time distance between the user terminal and each base station; and determine a base station corresponding to the maximum value of the connection index as a designated base station.

5. According to claim 4, a method for optimizing the performance of a dense cellular network based on NOMA and D2D, characterized in that: The base station transmission data includes real-time delay time, real-time packet loss rate, real-time number of connected user terminals and real-time average transmission rate.

6. The method for optimizing the performance of a dense cellular network based on NOMA and D2D according to claim 5, characterized in that: The status evaluation index of each base station is calculated by the formula: Calculate the state evaluation index S of the base station numbered i i ; Among them, Q Link is the number of user terminals connected to the base station numbered i; Q0 is the preset number of user terminals connected to the base station numbered i; t s is the real-time delay time of the base station numbered i; t0 is the preset delay time of the base station numbered i; is the real-time packet loss rate of the base station numbered i; is the preset packet loss rate of the base station numbered i; c is the real-time average transmission rate of the connected user terminals of the base station numbered i; v0 is the preset average transmission rate of the connected user terminals of the base station numbered i; ε is the unit-removing coefficient; C1 is the first preset constant; C2 is the second preset constant; C3 is the third preset constant; C4 is the fourth preset constant; γ1 is the first weight coefficient; γ2 is the second weight coefficient; γ3 is the third weight coefficient; γ4 is the fourth weight coefficient; S0 is the basic status evaluation index.

7. The method for optimizing the performance of a dense cellular network based on NOMA and D2D according to claim 6, characterized in that: The connectivity index is given by the formula: Calculate the connection index L between the user terminal and the base station numbered i in the target area i ; Where f(X) is the judgment function. When X≥0, f(X)=X; when X<0, f(X)=0; (X i , Y i , Z i ) are the coordinates of the base station numbered i; (X0, Y0, Z0) are the coordinates of the user terminal; R i is the preset coverage radius of the base station numbered i; σ i is the building density between user terminals and base station numbered i.

8. The method for optimizing the performance of a dense cellular network based on NOMA and D2D according to claim 1, characterized in that: The cellular mode includes an OMA mode and a NOMA mode; the two-hop D2D mode includes a D2D-assisted OMA mode and a D2D-assisted NOMA mode.

Images