CN106341216B - A kind of wireless communication link adaptive approach and uplink downlink adaptive approach - Google Patents

Abstract

The invention discloses a wireless communication link self-adaptation method and an uplink and downlink self-adaptation method. During downlink transmission, at each user, according to the statistical channel information of each user, a deterministic equivalent method is used to calculate the corresponding transmission of the user. Then, the adaptive transmission parameters are sent to the base station side as the parameters of downlink adaptive transmission through the feedback channel; during uplink transmission, the base station side uses the deterministic equivalent method to calculate the corresponding parameters of each user according to the statistical channel information Then, the optimal coding and modulation mode of each user is transmitted to each user through the feedback channel as the uplink adaptive transmission parameter. The invention can greatly reduce the complexity of large-scale antenna downlink adaptive parameter calculation, and has lower performance loss and better robustness.

Description

A wireless communication link adaptation method and uplink and downlink adaptation method

technical field

The present invention relates to the field of large-scale array antenna communication, in particular to a wireless communication link adaptation method and an uplink and downlink adaptation method.

Background technique

The technology of large-scale array antenna communication system requires the configuration of large-scale antennas on the base station side to serve a large number of terminals at the same time. In large-scale array antenna communication systems, link adaptation techniques are also required to improve transmission efficiency—making the transmission suitable for different channel conditions.

The link adaptation scheme in LTE includes a typical closed-loop space division multiplexing model. Three parameters are calculated at the receiving end: RI (Rank Indicator), PMI (Precoding Matrix Indicator) and CQI (Channel Quality Indicator). After the precoding method is known, the selection of CQI becomes the core parameter in link adaptation. The traditional CQI calculation method mainly includes equivalent signal-to-noise ratio mapping (ESM, Effective SINR Mapping), the most important of which is Exponential Effective SINR Mapping (EESM, Exponential Effective SINR Mapping) --- that is, in each resource particle. Calculate the signal-to-noise ratio for the corresponding receiver, and then use the exponential mapping method for the signal-to-noise ratio obtained on all resource blocks to obtain an equivalent signal-to-noise ratio representing the average level of the signal-to-noise ratio on the entire resource block. The table method determines the CQI as the final adaptive feedback parameter.

In a large-scale array antenna communication system, the above adaptive scheme is still available as a whole, but due to the sharp increase in the number of antennas on the base station side and the number of served terminals, it is complex to use the original method of calculating the equivalent signal-to-noise ratio on resource blocks. Therefore, it is necessary to study and adopt a new adaptive parameter calculation method suitable for large-scale antenna scenarios.

Therefore, the present invention proposes a technique for calculating adaptive parameters based on a deterministic equivalent algorithm in a random matrix, which can greatly reduce the complexity of large-scale antenna downlink adaptive parameter calculation, and at the same time, is more complex than The Monte Carlo traversal algorithm of , with a small loss in performance.

SUMMARY OF THE INVENTION

Purpose of the invention: The purpose of the present invention is to provide a wireless communication link adaptation method and an uplink and downlink adaptation method that can solve the defects existing in the prior art.

Technical scheme: in order to achieve this purpose, the present invention adopts the following technical scheme:

The wireless communication link adaptation method of the present invention includes an uplink adaptation process and a downlink adaptation process, wherein:

The uplink adaptation process includes the following steps:

S1.1: The base station obtains the corresponding channel energy coupling array according to the uplink channel estimation of each user;

S1.2: The base station obtains the function expression of the correlation matrix of each user's receiving and transmitting ends according to the energy coupling array of each user's channel;

S1.3: The base station substitutes the function expression of the correlation matrix of each user's receiving and transmitting ends into the iterative equation for iterative solution, and uses the solution of the iterative equation to obtain a deterministic equivalent approximation of the lower bound of each user's achievable rate;

S1.4: According to the bit error rate-rate table, the base station uses the deterministic equivalent approximation of the lower bound of the achievable rate of each user to obtain the corresponding channel quality indication, and then feeds back the channel quality indication to the user to indicate the current period Transmission mode, after the end of this cycle, return to step S1.1 to calculate the transmission parameters of the next cycle;

The downlink adaptation process includes the following steps:

S1.5: The base station performs user scheduling and beam selection in the beam domain to determine the user set and the corresponding beam set;

S1.6: On the basis of the scheduled user set and beam set, the certainty of calculating the average achievable rate of each user on all resource blocks under the minimum mean square error receiver is equivalent;

S1.7: According to the average achievable rate, by looking up the bit error rate-rate table, the corresponding channel quality indication of the user is obtained, and the coding and modulation mode of the downlink transmission of the user in this period is determined according to the channel quality indication, and the current period ends Then, return to step S1.5 to calculate the transmission parameters of the next transmission period.

The wireless communication uplink adaptive method of the present invention includes the following steps:

S2.1: The base station obtains the corresponding channel energy coupling array according to the uplink channel estimation of each user;

S2.2: The base station obtains the function expression of the correlation matrix of each user's receiving and transmitting ends according to the energy coupling array of each user's channel;

S2.3: The base station substitutes the function expression of the correlation matrix of each user's receiving and transmitting ends into the iterative equation for iterative solution, and uses the solution of the iterative equation to obtain a deterministic equivalent approximation of the lower bound of each user's achievable rate;

S2.4: According to the bit error rate-rate table, the base station uses the deterministic equivalent approximation of the lower bound of the achievable rate of each user to obtain the corresponding channel quality indication, and then feeds back the channel quality indication to the user for indicating the next cycle. the transmission mode, after the end of the next cycle, return to step S2.1.

Further, the channel energy coupling matrix Ω _k in the step S2.1 is calculated according to formula (1):

In formula (1), N is the total number of subcarriers, L is the total number of sampling time points, U _r,k is the eigenmatrix of the correlation matrix at the receiving end, h _k,l,t is the frequency domain channel estimation of user k, U _t,k is the characteristic matrix of the correlation matrix at the sending end, and ⊙ represents the matrix hardmad product.

Further, the receiver correlation matrix function in the step S2.2 As shown in formula (2):

In formula (2), U _r,k is the characteristic matrix of the correlation matrix at the receiving end, is a diagonal matrix of order M×M, and the diagonal elements are M is the number of antennas at the transmitting end, Ω _k is the channel energy coupling array of user k, if the user is a single antenna, then Ω _k degenerates into a vector, [Ω _k ] _i represents the element at the ith position in the vector Ω _k , is an M×M order matrix variable;

The correlation matrix function η _k (D) of the transmitter is shown in formula (3):

In Equation (3), U _t,k is the characteristic matrix of the correlation matrix at the transmitting end,

Further, the iterative equations in the step S2.3 are shown in equations (4)-(7):

where z is a scalar independent variable, The operator-valued Cauchy transform function, Intermediate function variables, Φ(z), is the intermediate matrix function variable, I _M is the identity matrix, K is the number of users, is the correlation matrix function of the user k receiving end, and n _k is the correlation matrix function of the transmitting end of the user k;

The process of one iteration is: set the initial value of equation (7), then substitute equation (7) into equation (5), then substitute equation (5) into equation (4), and finally substitute equation (4) into equation (6).

The wireless communication downlink adaptive method of the present invention includes the following steps:

S6.1: The base station performs user scheduling and beam selection in the beam domain to determine the user set and the corresponding beam set;

S6.2: On the basis of the scheduled user set and beam set, the certainty of calculating the average achievable rate of each user on all resource blocks under the minimum mean square error receiver is equivalent;

S6.3: According to the average achievable rate, by looking up the bit error rate-rate table, the channel quality indication corresponding to the user is obtained, and the coding and modulation mode of the downlink transmission of the user is determined according to the channel quality indication.

Further, the user scheduling and beam selection in the step S6.1 adopts the maximizing system and rate criterion, and adopts the greedy algorithm to search for the user set and the beam set.

Beneficial effects: The present invention discloses a wireless communication link adaptive method and an uplink and downlink adaptive method, which adopts a deterministic equivalent algorithm based on a random matrix to calculate adaptive parameters, which can greatly reduce large-scale antenna downlinks. The complexity of the road adaptive parameter calculation is low, and the performance loss is low, and it has better robustness.

Description of drawings

1 is a schematic diagram of an uplink adaptation method in a specific embodiment of the present invention;

FIG. 2 is a schematic diagram of a downlink adaptation method in a specific embodiment of the present invention.

Detailed ways

The technical solutions of the present invention will be further introduced below with reference to the accompanying drawings and specific embodiments.

This specific embodiment discloses a wireless communication uplink adaptive method. As shown in FIG. 1 , it is assumed that a Massive MIMO system includes a base station configured with M antennas and K single-antenna terminals. Assuming that M and K are both large, with typical values such as M=64, K=32, etc., K single-antenna terminals transmit data to the base station independently, and simultaneously receive data from K terminals on the base station side:

In formula (1), y is the received signal at the base station side, h _k is the uplink channel of user k, x _k is the data signal of user k, and n is complex white Gaussian noise.

The LMMSE receiver is used for joint signal detection and decoding. According to the ergodic reachable rate expression of the LMMSE receiver, the deterministic equivalence of its lower bound is obtained. At the same time of calculating certainty, it is necessary to calculate the correlation matrix of the transceiver Through the operation and deformation of the matrix, it is converted into a function of the channel energy coupling matrix Ω _k . Then, according to the deterministically equivalent core iterative fixed point equation, the deterministic equivalence with the lower bound of the rate is obtained. Use the certainty of the lower bound of the sum rate as an approximation of the sum rate, and look up the CWER-RATE table according to this value to obtain the CQI value. CQI is also a channel quality indicator. The method is described in detail below:

S1.1: The base station obtains the corresponding channel energy coupling matrix Ω _k according to the uplink channel estimation of each user, as shown in formula (1):

In formula (1), N is the total number of subcarriers, L is the total number of sampling time points, U _r,k is the eigenmatrix of the correlation matrix at the receiving end, h _k,l,t is the frequency domain channel estimation of user k, U _t,k is the characteristic matrix of the correlation matrix at the sending end, and ⊙ represents the matrix hardmad product.

S1.2: The base station obtains the receiver correlation matrix function of each user according to the channel energy coupling array of each user And the sender correlation matrix function η _k :

Receiver related functions can be written as U _r,k is the characteristic matrix of the correlation matrix at the receiver, is a diagonal matrix of order M×M, and the diagonal elements are M is the number of antennas at the transmitting end, Ω _k is the channel energy coupling array of user k, if the user is a single antenna, then Ω _k degenerates into a vector, [Ω _k ] _i represents the element at the ith position in the vector Ω _k , is an M×M order matrix variable.

In addition, when M is large enough, U _r,k can be approximated by an _M -dimensional DFT matrix, that is, U _r,k =FM , and _FM is a DFT matrix.

Sender related functions can be written as in Because the terminal is a single antenna, U _t,k degenerates to a scalar 1, and Π _k (D) is also a scalar.

S1.3: The base station substitutes the function expression of the correlation matrix of each user's receiving and transmitting ends into the iterative equation for iterative solution, and uses the solution of the iterative equation to obtain a deterministic equivalent approximation of the lower bound of each user's achievable rate. The iterative equations are shown in equations (2)-(5):

where z is a scalar independent variable, The operator-valued Cauchy transform function, Intermediate function variables, Φ(z), is the intermediate matrix function variable, I _M is the identity matrix, K is the number of users, is the correlation matrix function of the receiver of user k, and η _k is the correlation matrix function of the transmitter of user k.

The process of one iteration is: set the initial value of equation (5), then substitute equation (5) into equation (3), then substitute equation (3) into equation (2), and finally substitute equation (2) into equation (4). After one round of iteration is over, the next round starts to update equation (5) with the value calculated by equation (4) in the previous round, and the cycle repeats. The iteration termination condition is a K-dimensional vector The relative error of the two iterations before and after is less than the given value ε, indicating that the iterative process has converged.

S1.4: According to the bit error rate-rate table, the base station uses the deterministic equivalent approximation of the lower bound of the achievable rate of each user to obtain the corresponding channel quality indication, and then feeds back the channel quality indication to the user to indicate the current period In the transmission mode, after the end of the current cycle, return to step S1.1 to calculate the transmission parameters of the next cycle. The specific process is as follows:

The solution to the convergence of equation (5) in the iterative equation is the lower bound of the ergodic achievable rate of user k signal during uplink transmission The deterministic equivalent approximation of , where,

In practice, z=-σ ² is substituted. is the ergodic reachable rate under the MMSE receiver, I _K is the identity matrix, σ ² is the noise variance, and H is the channel matrix.

In order to obtain the CQI, it is necessary to obtain the CWER-RATE look-up table through channel simulation, that is, the code word rate-rate table. The specific method is, for each MCS mode:

1. Set a suitable SNR (signal-to-noise ratio) interval;

2. Generate a large number of channel realization samples N _h , and the number of samples is large enough so that the simulation can cover all channel conditions;

3. For each channel sample, traverse enough white noise realization samples;

4. For each SNR value, count the average CWER _i value and the corresponding reachable rate R _i , so that a CWER _i ～R _i curve can be obtained;

5. By synthesizing 15 MCSs, a graph composed of 15 sets of CWER _i to R _i curves can be obtained, and a corresponding look-up table can be obtained according to the CWER threshold value (typical value 0.1).

Using the deterministic equivalent approximation of the lower bound of each user's achievable rate, the CQI corresponding to the MCS mode that satisfies the threshold value and has the highest code rate is found in the look-up table as the final feedback value.

This specific embodiment also discloses a wireless communication downlink adaptive method. As shown in FIG. 2, the calculation of the adaptive parameters is also performed on the base station side. Compared with the uplink adaptation, the main difference is that the downlink transmission needs to be considered. The precoding adopted at the base station side uses the latest BDMA scheme in Massive MIMO, that is, beam domain precoding is considered at the base station side. In this precoding mode, the traversable reachable rate of the receiving end is still estimated at the base station side, and then the CWER-RATE table is searched according to the reachable rate to determine the CQI of downlink transmission. In addition, considering that the terminal contains multiple antennas, the specific steps are as follows:

S1.5: The base station performs user scheduling and beam selection in the beam domain, and determines the user set and the corresponding beam set. The user scheduling and beam selection here adopt the maximizing system and rate criteria, and the greedy algorithm is used to search for user sets and beam sets.

S1.6: On the basis of the scheduled user set and beam set, the certainty of calculating the average achievable rate of each user on all resource blocks under the minimum mean square error receiver is equivalent. The average achievable rate is calculated using the deterministic equivalent method and depends on the statistical channel information - the channel energy coupling matrix of the downlink channel. Since the channel energy coupling matrix has the reciprocity of uplink and downlink, the channel energy coupling matrix of the downlink can be calculated according to the channel energy coupling matrix Ω _k of the uplink obtained in step S1.1, and the channel energy coupling matrix of the downlink can be calculated according to the channel energy coupling matrix. , the originating correlation matrix function. Next, the average reachable rate can be calculated, the specific process is as follows:

The traversal reachable rate expression of user k is:

In formula (7), Q is the power allocation matrix (a diagonal matrix) composed of the energy of all users' transmitted signals, Q _\k =QQ _k , and Q _k is the power matrix of the transmitted signal of user k.

definition For a deterministic equivalence, define but

The deterministic equivalent calculation expression of is:

Φ _k (-σ ² ), The calculation is as described in step S1.3.

Because the statistical channel information is independent of subcarriers, the average rate of user k on all subcarriers can be finally calculated.

S1.7: According to the average achievable rate, by looking up the bit error rate-rate table, the channel quality indication corresponding to the user is obtained, and the coding and modulation mode of the downlink transmission of the user is determined according to the channel quality indication. This part is as described in step S1.4.

Claims (7)

1. a kind of wireless communication link adaptive approach, it is characterised in that: including adaptive uplink process and downlink Adaptive process, in which:

Adaptive uplink process the following steps are included:

S1.1: base station obtains corresponding channel energy according to the uplink channel estimation of each user and couples battle array；

S1.2: base station obtains the function representation of the Correlation Matrix of each user's Transmitting and Receiving End according to each subscriber channel energy coupling battle array；

S1.3: the function representation of each user's Transmitting and Receiving End Correlation Matrix is substituted into iterative equation and is iterated solution by base station, utilizes iteration The certainty that non trivial solution finds out each user's achievable rate lower bound is equivalent；

S1.4: base station is equally found out corresponding according to error code word rate-Speedometer Drive using the certainty of each user's achievable rate lower bound Channel quality instruction, then by channel quality indication feedback to user, the transmission mode being used to indicate in this period, this period knot The calculating of Shu Hou, return step S1.1 progress next cycle configured transmission；

Down link self adaption process the following steps are included:

S1.5: base station carries out user's scheduling and beam selection in Beam Domain, determines user's collection and corresponding beam collection；

S1.6: on the basis of the user of scheduling collection and beam collection, each user is calculated under least mean-square error receiver in institute There is the certainty of the average achievable rate on resource block equivalent；

S1.7: obtaining the corresponding channel quality instruction of user according to average achievable rate by searching for error code word rate-Speedometer Drive, The code modulation mode that the downlink transmission in determining this period of user is indicated according to channel quality, after this period, is returned It returns step S1.5 and carries out the calculating of the configured transmission in next transmission cycle.

2. a kind of wireless communications uplink adaptive approach, it is characterised in that: the following steps are included:

S2.1: base station obtains corresponding channel energy according to the uplink channel estimation of each user and couples battle array；

S2.2: base station obtains the function representation of each user's Transmitting and Receiving End Correlation Matrix according to each subscriber channel energy coupling battle array；

S2.3: each user's Transmitting and Receiving End Correlation Matrix function representation substitution iterative equation is iterated solution by base station, utilizes iteration side The certainty that the solution of journey finds out each user's achievable rate lower bound is equivalent；

S2.4: base station is equally found out corresponding according to error code word rate-Speedometer Drive using the certainty of each user's achievable rate lower bound Channel quality instruction, then by channel quality indication feedback to user, the transmission mode being used to indicate in next cycle is next After a end cycle, return step S2.1.

3. wireless communications uplink adaptive approach according to claim 2, it is characterised in that: in the step S2.1 Channel energy couple battle array Ω_kIt is calculated according to formula (1):

In formula (1), N is the sum of subcarrier, and L is the sum of sampling time point, U_r,kFor the eigenmatrix of receiving end Correlation Matrix, h_k,l,nFor the channel estimation in frequency domain of user k, U_t,kFor the eigenmatrix of transmitting terminal Correlation Matrix, ⊙ representing matrix hardmad product.

4. wireless communications uplink adaptive approach according to claim 2, it is characterised in that: in the step S2.2 Receiving end Correlation Matrix functionAs shown in formula (2):

In formula (2), U_r,kFor the eigenmatrix of receiving end Correlation Matrix,For M × M rank diagonal matrix, diagonal element isM is transmitting terminal number of antennas, Ω_kBattle array is coupled for the channel energy of user k, if user is Dan Tian Line, then Ω_kDegenerate is vector, [Ω_k]_iRepresentation vector Ω_kIn i-th of position element,For M × M rank matrix variables；

Transmitting terminal Correlation Matrix function η_k(D) as shown in formula (3):

In formula (3), U_t,kFor the eigenmatrix of transmitting terminal Correlation Matrix,

5. wireless communications uplink adaptive approach according to claim 2, it is characterised in that: in the step S2.3 Iterative equation such as formula (4)-(7) shown in:

Wherein z is scalar arguments,Operator value Cauchy's transforming function transformation function,Intermediate function variable, Φ (z),For Intermediary matrix function variable, I_MFor unit battle array, K is number of users,For the receiving end user k Correlation Matrix function, η_kFor user k transmission Hold Correlation Matrix function；

The process of one wheel iteration are as follows: the initial value of setting formula (7) then substitutes into formula (7) in formula (5), then formula (5) are substituted into formula (4) in, finally formula (4) are substituted into formula (6).

6. a kind of wireless communication descending chain circuit self-adaptive method, it is characterised in that: the following steps are included:

S6.1: base station carries out user's scheduling and beam selection in Beam Domain, determines user's collection and corresponding beam collection；

S6.2: on the basis of the user of scheduling collection and beam collection, each user is calculated under least mean-square error receiver in institute There is the certainty of the average achievable rate on resource block equivalent；

S6.3: obtaining the corresponding channel quality instruction of user according to average achievable rate by searching for error code word rate-Speedometer Drive, The code modulation mode of determining user's downlink transmission is indicated according to channel quality.

7. wireless communication descending chain circuit self-adaptive method according to claim 6, it is characterised in that: in the step S6.1 User's scheduling and beam selection use maximization system and rate criterion, user's collection and beam collection are carried out using greedy algorithm Search.