CN118509009A - Uplink and downlink rate distribution system and method for MIMO SWIPT network - Google Patents

Abstract

The invention discloses a system and a method for distributing uplink and downlink rates of a MIMO SWIPT network, which belong to the technical field of fifth-generation mobile communication and comprise the following steps: the terminal transmits an uplink pilot signal to the base station, and the base station performs channel estimation and precoding and transmitting power distribution for the downlink data signal; the terminal divides the received radio frequency signals according to the power division coefficients, one part carries out information decoding, and the other part carries out energy collection; the terminal distributes the collected energy according to the energy distribution coefficients of the uplink pilot frequency and the data signal, wherein one part supplements the energy consumption of the uplink pilot frequency signal, and the other part is used for transmitting the uplink data signal; finally, the uplink reachable rate and the downlink reachable rate of the terminal are calculated; and constructing and optimizing the problems of power distribution of the joint base station, power division coefficients of the terminal and energy distribution coefficients according to the obtained uplink reachable rate and downlink reachable rate, and realizing the maximum and minimum of the uplink reachable rate and the downlink reachable rate of the terminal. The invention can simultaneously realize the maximum and minimum of the terminal reachable rate in the uplink and downlink transmission directions, thereby not only increasing the technical comprehensiveness, but also providing fair bidirectional data transmission service for the terminal and solving the problems in the prior art.

Description

Uplink and downlink rate allocation system and method for MIMO SWIPT network

Technical Field

The present invention relates to an uplink and downlink rate allocation system and method of a MIMO SWIPT network, and in particular to a method for minimizing the uplink and downlink achievable rates of a large-scale MIMO SWIPT network, and belongs to the technical field of fifth-generation mobile communications.

Background Art

Massive MIMO technology, also known as massive antenna technology, is a communication technology that configures a large-scale antenna array at the base station end to simultaneously serve multiple terminals on the same frequency resources. The core idea of massive MIMO technology is to greatly increase the spatial multiplexing gain and resolution by greatly increasing the number of antennas at the base station end, thereby greatly improving the frequency efficiency and energy efficiency of the network without increasing the transmission power and bandwidth.

SWIPT technology is a technology that uses wireless radio frequency signals to transmit information to the terminal while also supplying energy to the terminal. Normally, it is impossible to decode information and collect energy on the same wireless radio frequency signal at the same time, so the hardware circuit of the terminal needs to be changed to a certain extent. In practice, the TS (Time Switching, TS) receiver and the PS (Power Splitting, PS) receiver are more widely used. The basic idea of the TS receiver is to divide each transmission frame period into two time slots, one time slot for information decoding and the other time slot for energy collection; the basic idea of the PS receiver is to divide the received wireless radio frequency signal into two parts according to power, one part for information decoding and the other part for energy collection.

However, in practice, due to path loss and shadow fading, the transmission efficiency of SWIPT is very low. To meet this challenge, a feasible method is energy beamforming. Massive MIMO technology is a good choice because it can use antenna arrays to converge wireless radio frequency signals into a very narrow beam and point it to the target terminal, greatly improving the transmission efficiency of the signal. Therefore, the integration of SWIPT technology with massive MIMO technology has become the focus of research by scientific researchers, and some research results have been achieved. However, the existing technologies only study the terminal achievable rate in a single data transmission direction, which limits their scope of application; in addition, although some technologies study the scenarios of uplink and downlink bidirectional data transmission, from the perspective of research objectives, the fairness of terminal services is ignored. Based on the problems existing in the existing technologies, how to apply them to the scenarios of uplink and downlink bidirectional data transmission, provide terminals with fair and high-quality services for bidirectional data transmission, and overcome the impact of the near-far effect has become a technical problem that urgently needs to be solved.

Summary of the invention

In view of the deficiencies of the prior art, the purpose of the present invention is to provide a system and method for allocating uplink and downlink rates of a MIMO SWIPT network, which solves the problems in the prior art.

The uplink and downlink rate allocation method of the MIMO SWIPT network of the present invention comprises the following steps:

S1: The terminal transmits an uplink pilot signal to the base station, and the base station performs channel estimation and precoding and transmit power allocation for the downlink data signal;

S2: The terminal divides the received RF signal according to the power division coefficient, one part is used for information decoding, and the other part is used for energy collection;

S3: The terminal distributes the collected energy according to the energy distribution coefficient of the uplink pilot signal and the data signal, with one part used to supplement the energy consumption of transmitting the uplink pilot signal and the other part used to transmit the uplink data signal;

S4: The base station decodes the received uplink data signal and finally calculates the uplink achievable rate and downlink achievable rate of the terminal;

S5: Based on the obtained uplink achievable rate and downlink achievable rate, an optimization problem is constructed by combining the power allocation of the base station, the power division coefficient of the terminal and the energy allocation coefficient to achieve the maximum minimization of the uplink achievable rate and the downlink achievable rate of the terminal.

Furthermore, in step S2, if the terminal sends all the power of the RF signal to the energy receiver, then the achievable rate of downlink data transmission is zero, and the network only has uplink data transmission. At this time, the power allocation of the base station and the energy allocation coefficient of the terminal are combined to construct an optimization problem to achieve the maximum minimization of the uplink achievable rate of the terminal; in step S3, if all the energy collected by the terminal is used to supplement the energy consumption of transmitting the pilot signal, the achievable rate of uplink data transmission is zero, and the network only has downlink data transmission. At this time, the power allocation of the base station and the power division coefficient of the terminal are combined to construct an optimization problem to achieve the maximum minimization of the downlink achievable rate of the terminal.

Furthermore, when the base station performs channel estimation in step S1, it is assumed that the channel coherence interval is frequency flat and obeys the Rayleigh distribution. Then, the channel between the kth terminal and the base station can be modeled as

Where M represents the number of antennas of the base station; K represents the number of all terminals; β _k represents the large-scale fading coefficient; represents the small-scale fading coefficient, each element of which obeys the CN(0,1) distribution. The length of the channel coherent interval structure is τ _c symbols, among which the first τ _p symbols are used for uplink pilot transmission, and of the remaining τ _c -τ _p symbols, half is used for downlink SWIPT transmission and the other half is used for uplink data transmission.

Furthermore, in the uplink pilot transmission phase, all terminals simultaneously send mutually orthogonal pilot sequences to the base station. is the pilot sequence sent by the kth terminal, then

After channel transmission, the pilot signal received by the base station is:

in, represents the channel matrix; represents the diagonal power matrix of the transmitted pilot, _qk represents the power of the pilot symbol transmitted by the kth terminal;

represents the pilot sequence matrix; (·) ^H represents the conjugate transpose operation; represents the additive white Gaussian noise matrix at the base station antenna, whose elements obey Based on formula (3), the channel estimation of the kth terminal is expressed as:

The base station uses the minimum mean square error method for estimation, and the estimation result is:

The channel estimation distribution of the kth terminal is

The channel estimation error distribution of the kth terminal is:

in, I _M represents the M×M identity matrix.

Furthermore, in the downlink SWIPT transmission phase, the transmission signal of the base station antenna is expressed as:

in, It indicates the downlink data symbol sent by the base station to the kth terminal, which follows the CN(0,1) distribution; represents the transmit power allocated by the base station to the kth terminal; _wk represents the precoding vector of the kth terminal, which is expressed as:

Where E{·} represents the mathematical expectation of the random variable. The RF signal transmitted by the base station reaches each terminal through the channel transmission. Assuming that the power division coefficient of the kth terminal is ρ _k ∈(0,1), then the signals received by its information receiver and energy receiver are respectively expressed as:

and

Where n _t,k represents the AWGN at the terminal antenna, which is distribution; n _c,k represents the AWGN introduced from RF to baseband conversion, obeying Distribution; According to formula (9), the downlink achievable rate of the kth terminal is calculated as:

in, represents the downlink signal to interference and noise ratio of the kth terminal, which is calculated as:

According to formula (10), the energy collected by the kth terminal is calculated as:

Where _Ts represents the symbol period, _ηk represents the energy conversion efficiency of the kth terminal; in a large-scale MIMO network, equation (13) is approximated as:

Assume that α _k ∈(0,1) is the energy allocation coefficient of the uplink pilot and data signals of the kth terminal, where Part of it is used to supplement the energy consumption of transmitting the pilot signal, and the rest Part is used to transmit uplink data signals. Based on this, γ _k is recalculated as:

The transmit power of the uplink data signal is calculated as:

Furthermore, in the uplink data transmission phase, each terminal uses the allocated energy to transmit an uplink data signal. After the signal is transmitted through the channel, the signal received by the base station is expressed as

in, represents the uplink data symbol sent by the kth terminal, which obeys the CN(0,1) distribution; n _b represents the AWGN at the base station antenna, and its elements obey Distribution; the decoding expression of the kth terminal is expressed as

By using formula (18), the uplink achievable rate of the kth terminal is calculated as:

in, represents the uplink signal to interference and noise ratio of the kth terminal, which is calculated as:

Further, in step S5, the maximum minimization of the uplink achievable rate and the downlink achievable rate of the terminal is specifically achieved by the following calculation formula:

Wherein, ω _k ^ul >0 and ω _k ^dl >0 represent weight factors of uplink and downlink signal to interference and noise ratio of the k-th terminal, respectively, which can be set according to the priority of the terminal; represents the total power allocated by the base station. In order to solve, equation (21) is equivalently transformed into:

Among them, SINR is an auxiliary variable introduced. The transformed formula (22) is a geometric programming optimization problem, which is solved by convex optimization tools to obtain the optimal base station power allocation. The power division coefficient {ρ _k } and energy allocation coefficient {α _k } of the terminal are used to minimize the maximum uplink achievable rate and the maximum downlink achievable rate.

Furthermore, if the terminal power divider sends all the power of the RF signal to the energy receiver, then {ρ _k } = 0, The network only transmits uplink data. The power allocation of the base station and the energy allocation coefficient of the terminal can be combined to construct an optimization problem to minimize the uplink achievable rate of the terminal.

In order to solve, equation (23) is equivalently transformed into:

The transformed equation (24) is a geometric programming optimization problem, which is solved by convex optimization tools to obtain the optimal base station power allocation. and the energy allocation coefficient {α _k } of the terminal to achieve the maximum minimization of the uplink achievable rate of the terminal.

Furthermore, if all the energy collected by the terminal is used to supplement the energy consumption of transmitting the pilot signal, then {α _k }＝1, The network only transmits downlink data. The optimization problem is constructed by combining the power allocation of the base station and the power division coefficient of the terminal to minimize the downlink rate of the terminal.

In order to solve, equation (25) is equivalently transformed into:

The transformed equation (26) is a geometric programming optimization problem, which is solved by convex optimization tools to obtain the optimal base station power allocation. and the power division coefficient {ρ _k } of the terminal to minimize the downlink achievable rate of the terminal.

The uplink and downlink rate allocation system of the MIMO SWIPT network described in the present invention includes a base station and several terminals. The base station is equipped with several antennas to provide uplink and downlink data transmission services for several terminals on the same time-frequency resources. Each terminal is equipped with a power splitter, and the power splitter is equipped with an information receiver and an energy receiver. The power splitter splits the received radio frequency signal according to the power splitting coefficient, and sends one part to the information receiver for information decoding; the other part is sent to the energy receiver for energy collection.

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

The uplink and downlink rate allocation system and method of the MIMO SWIPT network described in the present invention can simultaneously achieve the maximum minimization of the achievable rate in both the uplink and downlink transmission directions, which not only increases the comprehensiveness of the technology, but also provides fair two-way data services for the terminal. In particular, if the terminal power divider sends all the RF signal power to the energy receiver, the maximum minimization of the uplink achievable rate alone can be achieved; if all the energy collected by the terminal is used to supplement the energy consumption of the transmitted pilot, the maximum minimization of the downlink achievable rate alone can be achieved. These two special situations are compatible with existing technical methods, greatly expand the scope of application of the present invention, and solve the problems existing in the prior art.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic diagram of a massive MIMO SWIPT network based on PS protocol uplink and downlink data transmission according to the present invention;

FIG2 is a schematic diagram of a channel coherent spacing structure of the present invention;

FIG3 is a schematic diagram showing the influence of different power allocation methods on the maximum and minimum achievable rates of uplink and downlink according to the present invention;

FIG4 is a schematic diagram showing the influence of different weight factors on the maximum and minimum achievable rates of uplink and downlink according to the present invention;

FIG5 is a schematic diagram of the maximum and minimum achievable rates for different data transmission directions according to the present invention.

DETAILED DESCRIPTION

The present invention will be further described below in conjunction with the accompanying drawings and embodiments:

Embodiment 1:

The uplink and downlink rate allocation method of the MIMO SWIPT network of the present invention comprises the following steps:

S1: The terminal transmits an uplink pilot signal to the base station, and the base station performs channel estimation and precoding and transmit power allocation for the downlink data signal;

S2: The terminal divides the received RF signal according to the power division coefficient, one part is used for information decoding, and the other part is used for energy collection;

S3: The terminal distributes the collected energy according to the energy distribution coefficient of the uplink pilot signal and the data signal, with one part used to supplement the energy consumption of transmitting the uplink pilot signal and the other part used to transmit the uplink data signal;

S4: The base station decodes the received uplink data signal and finally calculates the uplink achievable rate and downlink achievable rate of the terminal;

S5: Based on the obtained uplink achievable rate and downlink achievable rate, an optimization problem is constructed by combining the power allocation of the base station, the power division coefficient of the terminal and the energy allocation coefficient to achieve the maximum minimization of the uplink achievable rate and the downlink achievable rate of the terminal.

As shown in Figure 1, assume that the base station is equipped with M antennas, providing uplink and downlink data transmission services for K single-antenna terminals on the same time-frequency resources, and satisfying M>>K. Each terminal is equipped with a power splitter, which can split the received RF signal according to the power split coefficient, sending one part to the information receiver for information decoding; the other part is sent to the energy receiver for energy collection. The network operates in TDD (Time-Division Duplexing, TDD) mode.

The channel coherence interval structure is shown in Figure 2, with a length of τ _c symbols. Specifically, the first τ _p symbols are used for uplink pilot transmission, and the remaining τ _c -τ _p symbols, half of which are used for downlink SWIPT transmission and the other half for uplink data transmission. Assuming that the channel coherence interval is frequency flat and obeys the Rayleigh distribution, the channel between the kth terminal and the base station can be modeled as

Where M represents the number of antennas of the base station; K represents the number of all terminals; β _k represents the large-scale fading coefficient; represents the small-scale fading coefficient, and each of its elements follows the CN(0,1) distribution. The following is a detailed analysis of the various components of the coherence interval.

(1) In the uplink pilot transmission phase, all terminals simultaneously send mutually orthogonal pilot sequences to the base station. is the pilot sequence sent by the kth terminal, then

After channel transmission, the pilot signal received by the base station is

in, represents the channel matrix; represents the diagonal power matrix of the transmitted pilot, _qk represents the power of the pilot symbol transmitted by the kth terminal;

represents the pilot sequence matrix; (·) ^H represents the conjugate transpose operation; represents the additive white Gaussian noise (AWGN) matrix at the base station antenna, and its elements obey Based on formula (3), the channel estimation of the kth terminal can be expressed as

The base station uses the minimum mean square error method for estimation, and the estimation result is:

The channel estimation distribution of the kth terminal is

The channel estimation error distribution of the kth terminal is:

in, I _M represents the M×M identity matrix.

(2) In the downlink SWIPT transmission phase, the transmission signal of the base station antenna can be expressed as

in, It indicates the downlink data symbol sent by the base station to the kth terminal, which follows the CN(0,1) distribution; represents the transmission power allocated by the base station to the kth terminal; _wk represents the precoding vector of the kth terminal. In this embodiment, MRT (Maximum Ratio Transmission, MRT) precoding is adopted, which is expressed as

Where E{·} represents the mathematical expectation of the random variable. The RF signal transmitted by the base station reaches each terminal through the channel transmission. Assuming that the power division coefficient of the kth terminal is ρ _k ∈(0,1), then the signals received by its information receiver and energy receiver can be expressed as

and

Where n _t,k represents the AWGN at the terminal antenna, which is distribution; n _c,k represents the AWGN introduced from RF to baseband conversion, obeying According to formula (9), the downlink achievable rate of the kth terminal can be calculated as

in, represents the downlink signal-to-interference-to-noise ratio of the kth terminal, which can be calculated as

According to formula (10), the collected energy of the kth terminal can be calculated as

Where _Ts represents the symbol period, and _ηk represents the energy conversion efficiency of the kth terminal. By observing equation (13), we can find that the first term on the right side of the equal sign is proportional to the number of base station antennas, while the second term is independent of the number of base station antennas. In a large-scale MIMO network, the number of base station antennas is very large. Therefore, the value of the first term far exceeds the second term. Then equation (13) can be approximated as

Assume that α _k ∈(0,1) is the energy allocation coefficient of the uplink pilot and data signals of the kth terminal, where Part of it is used to supplement the energy consumption of transmitting the pilot signal, and the rest Part is used to transmit uplink data signals. Based on this, γ _k is recalculated as

The transmit power of the uplink data signal is calculated as

(3) In the uplink data transmission phase, all terminals use the allocated energy to transmit uplink data signals. After the signals are transmitted through the channel, the signals received by the base station can be expressed as

in, represents the uplink data symbol sent by the kth terminal, which obeys the CN(0,1) distribution; n _b represents the AWGN at the base station antenna, and its elements obey The present invention assumes that the base station uses MRC (Maximum Ratio Combining, MRC) to decode the received uplink data signal. Then, the decoding expression of the kth terminal can be expressed as

By using formula (18), the uplink achievable rate of the kth terminal can be calculated as:

in, represents the uplink signal to interference and noise ratio of the kth terminal, which can be calculated as

In wireless communication networks, providing fair and high-quality services to each terminal is an important consideration for network administrators. In view of this consideration, the present invention studies the maximum minimization problem of uplink and downlink achievable rates.

(1) The power allocation of the base station, the power division coefficient of the terminal, and the energy allocation coefficient can be combined to construct an optimization problem to minimize the uplink and downlink achievable rates of the terminal.

Wherein, ω _k ^ul >0 and ω _k ^dl >0 represent weight factors of uplink and downlink signal to interference and noise ratio of the k-th terminal, respectively, which can be set according to the priority of the terminal; represents the total power allocated by the base station. Observing formula (21), it can be found that it is a non-convex optimization problem and it is very difficult to solve it directly. In order to obtain the solution of the optimization problem, formula (21) is equivalently transformed into

Among them, SINR is an auxiliary variable introduced. In addition, changing the original equal sign of C6 and C7 in equation (22) to a less than or equal sign does not affect the solution of the original optimization problem (21), because they are monotonically increasing functions of the objective function. Observing equation (22), it can be found that the objective function is a monomial function, and the constraint condition is a monomial function or a polynomial function. Therefore, this is a geometric programming optimization problem, which can be solved by convex optimization tools, so as to obtain the optimal base station power allocation. The power division coefficient {ρ _k } and energy allocation coefficient {α _k } of the terminal are used to minimize the uplink and downlink achievable rates of the terminal.

(2) If the terminal power divider sends all the power of the RF signal to the energy receiver, then {ρ _k } = 0. The network only transmits uplink data. The power allocation of the base station and the energy allocation coefficient of the terminal can be combined to construct an optimization problem to minimize the uplink achievable rate of the terminal.

By observing equation (23), we can find that it is a non-convex optimization problem and it is very difficult to solve it directly. Using the same treatment as equation (21), equation (23) can be equivalently transformed into

Similar to the analysis process of equation (22), equation (24) can be transformed into a geometric programming optimization problem, which can be solved by convex optimization tools to obtain the optimal base station power allocation. and the energy allocation coefficient {α _k } of the terminal to achieve the maximum minimization of the uplink achievable rate of the terminal.

(3) If all the energy collected by the terminal is used to supplement the energy consumption of transmitting the pilot signal, then {α _k } = 1, The network only transmits downlink data. The power allocation of the base station and the power division coefficient of the terminal can be combined to construct an optimization problem to minimize the downlink rate that can be achieved by the terminal.

By observing equation (25), we can find that it is a non-convex optimization problem and it is very difficult to solve it directly. Using the same treatment as equation (21), equation (25) can be equivalently transformed into

Similarly, similar to the analysis process of equation (22), equation (26) can be transformed into a geometric programming optimization problem, which can be solved by convex optimization tools to obtain the optimal base station power allocation. and the power division coefficient {ρ _k } of the terminal to minimize the downlink achievable rate of the terminal.

In this implementation, the above-mentioned method for minimizing the maximum uplink and downlink achievable rates was verified and the verification results were obtained. Assume that the maximum power transmitted by the base station is K = 4 terminals are located at 10m, 15m, 20m and 25m away from the base station respectively. The noise power at the base station antenna and each terminal antenna The noise power introduced by converting the RF signal of each terminal to the baseband signal The energy conversion efficiency of each terminal is {η _k } = 0.8, the length of each coherent interval is 200 symbols, each symbol period T _s = 0.005S, the length of the pilot sequence is equal to the number of terminals, and the large-scale fading coefficient is modeled as Where d _k represents the distance between the kth terminal and the base station, and the weight of the uplink and downlink signal-to-interference-noise ratio is set to Unless otherwise specified, the above settings are used in numerical verification.

As shown in Figure 3, in order to demonstrate the impact of different power allocation methods on the maximum and minimum achievable rates of uplink and downlink, this experiment compares the power allocation method provided by the present invention with the existing commonly used base station transmit power averaging method, terminal power division coefficient averaging method and energy allocation coefficient averaging method. Specifically, the base station transmit power averaging method is In Figure 3, it is marked with "base station transmission power averaging method"; the terminal power division coefficient averaging method is {ρ _k }＝1/2, which is marked with "power division coefficient averaging method" in Figure 3; the terminal energy allocation coefficient averaging method is {α _k }＝1/2, which is marked with "energy allocation coefficient averaging method" in Figure 3. It can be observed from the experimental results that the power allocation method provided by the present invention achieves higher uplink and downlink maximum and minimum achievable rates, which shows that compared with the existing commonly used power allocation methods, the present invention can provide better fair and high-quality services for terminals.

As shown in Figure 4, this experiment shows the impact of different weight factors on the maximum and minimum achievable rates of uplink and downlink. In order to facilitate the observation of the experimental results, the weight factor of the downlink signal-to-interference-noise ratio is fixed. Figure 4 shows the uplink signal-to-interference-noise ratio weight factors. and The maximum and minimum achievable rates of uplink in three cases. It can be observed from the experimental results that the maximum and minimum achievable rates of uplink increase with the increase of the weight factor. In addition, if different values are set for the uplink and downlink signal interference and noise ratio weight factors of each terminal, different achievable rates can be obtained. These show that by setting the uplink and downlink signal interference and noise ratio weight factors, the present invention can provide a flexible and variable achievable rate for the terminal, greatly improving the ability to serve the diversified needs of the terminal.

As shown in FIG5 , this experiment shows the maximum and minimum achievable rates in different data transmission directions. The maximum minimum achievable rates for uplink and downlink are obtained according to formula (21), which is indicated by “uplink and downlink data transmission” in FIG5 ; the maximum minimum achievable rate for uplink is obtained according to formula (23), which is indicated by “uplink data transmission only” in FIG5 ; the maximum minimum achievable rate for downlink is obtained according to formula (25), which is indicated by “downlink data transmission only” in FIG5 . This experiment shows that the present invention can provide fair and high-quality services in three data transmission directions, which greatly expands the scope of application of the present invention.

Embodiment 2:

As shown in FIG1 , the uplink and downlink rate allocation system of the MIMO SWIPT network of the present invention includes a base station and several terminals. The base station is equipped with several antennas to provide uplink and downlink data transmission services for several terminals on the same time-frequency resources. Each terminal is equipped with a power splitter, which is equipped with an information receiver and an energy receiver. The power splitter splits the received radio frequency signal according to the power splitting coefficient, and sends one part to the information receiver for information decoding; the other part is sent to the energy receiver for energy collection.

The system recorded in this embodiment provides a hardware architecture for the method recorded in Example 1, which can achieve the maximum minimization of the achievable rate in both the uplink and downlink transmission directions, which not only increases the comprehensiveness of the technology, but also provides fair two-way data services for the terminal, solving the problems existing in the prior art.

The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims (10)

1. A method for allocating uplink and downlink rates in a MIMO SWIPT network, characterized in that the method comprises the following steps: S1: The terminal transmits an uplink pilot signal to the base station, and the base station performs channel estimation and precoding and transmit power allocation for the downlink data signal; S2: The terminal divides the received RF signal according to the power division coefficient, one part is used for information decoding, and the other part is used for energy collection; S3: The terminal distributes the collected energy according to the energy distribution coefficient of the uplink pilot signal and the data signal, with one part used to supplement the energy consumption of transmitting the uplink pilot signal and the other part used to transmit the uplink data signal; S4: The base station decodes the received uplink data signal and finally calculates the uplink achievable rate and downlink achievable rate of the terminal; S5: Based on the obtained uplink achievable rate and downlink achievable rate, an optimization problem is constructed by combining the power allocation of the base station, the power division coefficient of the terminal and the energy allocation coefficient to achieve the maximum minimization of the uplink achievable rate and the downlink achievable rate of the terminal. 2. The uplink and downlink rate allocation method of the MIMO SWIPT network according to claim 1 is characterized in that: in the step S2, if the terminal sends all the power of the radio frequency signal to the energy receiver, then the achievable rate of the downlink data transmission is zero, and the network only has uplink data transmission. At this time, the power allocation of the base station and the energy allocation coefficient of the terminal are combined to construct an optimization problem to achieve the maximum minimization of the uplink achievable rate of the terminal; in the step S3, if the energy collected by the terminal is all used to supplement the energy consumption of transmitting the pilot signal, the achievable rate of the uplink data transmission is zero, and the network only has downlink data transmission. At this time, the power allocation of the base station and the power division coefficient of the terminal are combined to construct an optimization problem to achieve the maximum minimization of the downlink achievable rate of the terminal. 3. The uplink and downlink rate allocation method of the MIMO SWIPT network according to claim 2 is characterized in that: when the base station performs channel estimation in step S1, it is assumed that the channel coherence interval is frequency flat and obeys the Rayleigh distribution, then the channel between the kth terminal and the base station can be modeled as Where M represents the number of antennas of the base station; K represents the number of all terminals; β _k represents the large-scale fading coefficient; represents the small-scale fading coefficient, each element of which obeys the CN(0,1) distribution. The length of the channel coherent interval structure is τ _c symbols, among which the first τ _p symbols are used for uplink pilot transmission, and of the remaining τ _c -τ _p symbols, half is used for downlink SWIPT transmission and the other half is used for uplink data transmission. 4. The uplink and downlink rate allocation method of the MIMO SWIPT network according to claim 3 is characterized in that: in the uplink pilot transmission stage, all terminals simultaneously send mutually orthogonal pilot sequences to the base station, recording is the pilot sequence sent by the kth terminal, then After channel transmission, the pilot signal received by the base station is: in, represents the channel matrix; represents the diagonal power matrix of the transmitted pilot, _qk represents the power of the pilot symbol transmitted by the kth terminal; represents the pilot sequence matrix; (·) ^H represents the conjugate transpose operation; represents the additive white Gaussian noise matrix at the base station antenna, whose elements obey Based on formula (3), the channel estimation of the kth terminal is expressed as: The base station uses the minimum mean square error method for estimation, and the estimation result is: The channel estimation distribution of the kth terminal is The channel estimation error distribution of the kth terminal is: in, I _M represents the M×M identity matrix. 5. The uplink and downlink rate allocation method of the MIMO SWIPT network according to claim 4 is characterized in that: in the downlink SWIPT transmission stage, the transmission signal of the base station antenna is expressed as: in, It indicates the downlink data symbol sent by the base station to the kth terminal, which follows the CN(0,1) distribution; represents the transmit power allocated by the base station to the kth terminal; _wk represents the precoding vector of the kth terminal, which is expressed as: Where E{·} represents the mathematical expectation of the random variable. The RF signal transmitted by the base station reaches each terminal through the channel transmission. Assuming that the power division coefficient of the kth terminal is ρ _k ∈(0,1), then the signals received by its information receiver and energy receiver are respectively expressed as: and Where n _t,k represents the AWGN at the terminal antenna, which is distribution; n _c,k represents the AWGN introduced from RF to baseband conversion, obeying Distribution; According to formula (9), the downlink achievable rate of the kth terminal is calculated as: in, represents the downlink signal to interference and noise ratio of the kth terminal, which is calculated as: According to formula (10), the energy collected by the kth terminal is calculated as: Where _Ts represents the symbol period, _ηk represents the energy conversion efficiency of the kth terminal; in a large-scale MIMO network, equation (13) is approximated as: Assume that α _k ∈(0,1) is the energy allocation coefficient of the uplink pilot and data signals of the kth terminal, where α _k Part of it is used to supplement the energy consumption of transmitting the pilot signal, and the rest Part is used to transmit uplink data signals. Based on this, γ _k is recalculated as: The transmit power of the uplink data signal is calculated as: 6. The uplink and downlink rate allocation method of the MIMO SWIPT network according to claim 5 is characterized in that: in the uplink data transmission stage, each terminal uses the allocated energy to transmit an uplink data signal, and after the signal is transmitted through the channel, the signal received by the base station is expressed as in, represents the uplink data symbol sent by the kth terminal, which obeys the CN(0,1) distribution; n _b represents the AWGN at the base station antenna, and its elements obey Distribution; the decoding expression of the kth terminal is expressed as By using formula (18), the uplink achievable rate of the kth terminal is calculated as: in, represents the uplink signal to interference and noise ratio of the kth terminal, which is calculated as: 7. The uplink and downlink rate allocation method of the MIMO SWIPT network according to claim 6 is characterized in that: in the step S5, the maximum minimization of the uplink reachable rate and the downlink reachable rate of the terminal is specifically achieved by the following calculation formula: in, and They represent the weight factors of the uplink and downlink signal to interference and noise ratios of the kth terminal, respectively, which can be set according to the priority of the terminal; represents the total power allocated by the base station. In order to solve, equation (21) is equivalently transformed into: Among them, SINR is an auxiliary variable introduced. The transformed formula (22) is a geometric programming optimization problem, which is solved by convex optimization tools to obtain the optimal base station power allocation. The power division coefficient {ρ _k } and energy allocation coefficient {α _k } of the terminal are used to minimize the maximum uplink achievable rate and the maximum downlink achievable rate. 8. The method for allocating uplink and downlink rates of a MIMO SWIPT network according to claim 7, characterized in that: if the terminal power divider sends all the power of the radio frequency signal to the energy receiver, then {ρ _k } = 0, The network only transmits uplink data. The power allocation of the base station and the energy allocation coefficient of the terminal can be combined to construct an optimization problem to minimize the uplink achievable rate of the terminal. In order to solve, equation (23) is equivalently transformed into: The transformed equation (24) is a geometric programming optimization problem, which is solved by convex optimization tools to obtain the optimal base station power allocation. and the energy allocation coefficient {α _k } of the terminal to achieve the maximum minimization of the uplink achievable rate of the terminal. 9. The method for allocating uplink and downlink rates of a MIMO SWIPT network according to claim 8, characterized in that: if all the energy collected by the terminal is used to supplement the energy consumption of transmitting the pilot signal, then {α _k } = 1, The network only transmits downlink data. The optimization problem is constructed by combining the power allocation of the base station and the power division coefficient of the terminal to minimize the downlink rate of the terminal. In order to solve, equation (25) is equivalently transformed into: The transformed equation (26) is a geometric programming optimization problem, which is solved by convex optimization tools to obtain the optimal base station power allocation. and the power division coefficient {ρ _k } of the terminal to minimize the downlink achievable rate of the terminal. 10. A MIMO SWIPT network uplink and downlink rate allocation system, applied to the MIMO SWIPT network uplink and downlink rate allocation method according to any one of claims 1-9, characterized in that: the system comprises a base station and several terminals, the base station is equipped with several antennas, and provides uplink and downlink data transmission services for several terminals on the same time-frequency resources, each terminal is equipped with a power splitter, the power splitter is equipped with an information receiver and an energy receiver, the power splitter splits the received radio frequency signal according to the power splitting coefficient, and sends one part to the information receiver for information decoding; the other part is sent to the energy receiver for energy collection.