Apparatus, method, and system for performing multiple-user multiple-input multiple-output communication, and non-transitory computer readable storage medium

1 . An apparatus performing multiple user multiple input multiple output (MU-MIMO) communication, the apparatus comprising:

a hybrid precoder for channel precoding signals to be transmitted to a plurality of users through user channels, the hybrid precoder comprising:

a digital precoder configured to precode the signals digitally by using a digital beamformer matrix;

a plurality of radiofrequency (RF) chains, and

an analog precoder configured to precode the signals analogically, the analog precoder includes a plurality of antennas to transmit the precoded signals processed by the hybrid precoder, the plurality of antennas are grouped, thereby allowing for building of antennas' array as an array of several subarrays of antennas;

wherein in the antennas' subarray, each antenna is connected to a two-way switch, having terminals connected to a phase inverter and a multiple-way switch and another terminal of the phase inverter connected to the multiple-way switch; and

the number of terminals of the multiple-way switch is equal to the number of the plurality of RF chains such that each terminal is connected to an RF chain and each antenna subarray is switchable among the plurality of RF chains.

2 . The apparatus according to claim 1 , wherein the apparatus is configured to execute a switching algorithm that sets a position of the multi-way switches and the two-way switches for increasing the sum rate, wherein the switching algorithm is based on:

the optimal setting of the multi-way switches and the two-way switches is addressed to maximize the sum-rate of a system and is formulated by the following optimization problem:

F

RF

*

=

arg

⁢

max

⁢

R

f

∈

ℱ

N

t

×

1

,

g

∈

𝒢

N

×

1

where F RF * is the optimum analog beamformer, f=[f 1 , f 2 , . . . , f Nt ] T represents the values of the N t two-way switches, g=[g 1 , g 2 , . . . , g N ] T represents the values of the N N RF -way switches, and R represents the sum-rate of the system; f i denotes the switch value related to the i-th antenna, i=1, . . . , N t , such that f i ∈ , where ={−1,1} represents the invert and not-invert options of the switches, g j represents the switch value of the j-th antenna subarray, j=1, . . . , N, where g j ∈ , and ={1,2, . . . , N RF } represents the N RF options of the switches; in addition, N t represents the number of antennas at the transmitter, N RF is the number of the plurality of RF chains, and N the number of antenna subarrays in the transmitter.

3 . The apparatus according to claim 2 , wherein the decimal value of g j is mapped to its corresponding binary representation such that G j,: ∈ 1×N b where ={0,1} and N b =log 2 N RF ; the vector g∈ N×1 is written as a binary matrix G∈ N×N b ; then, the binary matrix G is vectorized by stacking its rows as follows:

g =[G 1,: G 2,: . . . G N,: ] T

where G j,: denotes the j-th row of the binary matrix G, j=1, . . . , N.

4 . The apparatus according to claim 3 , wherein the apparatus is further configured to compute the probability of the N phase switches values, p=[p 1 , p 2 , . . . , p NN b ] T . For the binary vector g , where g j is the j-th entry of the vector g , and j=1,2, . . . , N N b , the probability of g j =1, is p j , and the probability of g j =0 is given by 1−p j ; and, due to there is no priori information, the probability is initialized to p (0) =(½)1 NN b ×1 , where 1 a×b is a ones matrix with size a×b.

5 . The apparatus according to claim 4 , wherein the apparatus is further configured to compute the probability of the N t non-zero elements in F RF , which is represented by the vector u=[u 1 , u 2 , . . . , u N t ] T , u i is the probability of f i =1, and 1−u i is the probability of f i =−1. The probability is initialized to u (0) =(½)1 N t ×1 because there is no priori information.

6 . The apparatus according to claim 5 , wherein the apparatus is further configured to generate S random vectors g s and f s , s=1,2, . . . , S, are generated according to p (m) and u (m) , respectively, where m represents the m-th iteration; and the value assignation in the analog beamformer is determined by (F RF ) i,g j =f i , when the i-th antenna is in the j-th antenna subarray.

7 . The apparatus according to claim 5 , wherein the digital precoder is a zero forcing (ZF) linear precoder, the matrix is determined by:

F BB =c n (HF RF ) † , where H=[h 1 T . . . h K T ] T , h k is the channel matrix between the base station and the user k, and c n is computed to satisfy the power constraint such that ∥F∥ F 2 =∥F RF F BB ∥ F 2 =E T , where E T represents the total available power for transmission.

8 . The apparatus according to claim 5 , wherein S hybrid precoders are calculated as F s =F RF s F BB s , s=1, . . . , S; and then, the achievable sum-rate R(F s ) is computed by making use of:

R

=

∑

k

=

1

K

log

2

(

1

+



h

k

⁢

F

:

,

k



2

∑

j

≠

k

K

⁢



h

k

⁢

F

:

,

j



2

+

σ

n

2

)

then, S elite candidates are taken that correspond to the largest sum-rate values such that R(F 1 )>R(F 2 )> . . . >R(F S elite ).

9 . The apparatus according to claim 6 , wherein weights of the S elite candidates are computed as follows;

w

s

=

S

elite

⁢

❘

"\[LeftBracketingBar]"

R

⁡

(

F

s

)

-

R

⁡

(

F

S

elite

)

❘

"\[RightBracketingBar]"

∑

s

=

1

S

elite

⁢

❘

"\[LeftBracketingBar]"

R

⁡

(

F

s

)

-

R

⁡

(

F

S

elite

)

❘

"\[RightBracketingBar]"

.

10 . The apparatus according to claim 5 , wherein the probabilities related to the N t N switches are updated by:

p

j

(

m

+

1

)

=

∑

s

=

1

S

slite

⁢

w

s

⁢

g

j

_

∑

s

=

1

S

slite

⁢

w

s

⁢

and

⁢

u

i

(

m

+

1

)

=

∑

s

=

1

S

slite

⁢

w

s

(

f

i

+

1

)

2

⁢

∑

s

=

1

S

slite

⁢

w

s

up to reach an established number of iterations; then, the optimum analog beamformer, F RF *, is taken from the best sum-rate value among all iterations.

11 . The apparatus according to claim 5 , wherein the adequate values of the switches in the architecture of the proposed hybrid precoder can be also obtained by machine learning, deep learning, reinforcement learning, or any algorithm from Intelligence Artificial area.

12 . The method for performing multiple user multiple input multiple output (MU-MIMO) communication with the proposed apparatus comprises:

processing signals digitally with a digital precoder;

processing the digitally processed signals with a plurality of radiofrequency (RF) chains;

processing the signals analogically with an analog precoder, wherein the analog precoder comprises:

a plurality of antennas for transmitting the signals processed by the proposed hybrid precoder, the plurality of antennas is grouped, allowing for the building of the antenna array can be made by several subarrays of antennas;

wherein in the antennas' subarray, each antenna is connected to a two-way switch, whose terminals are connected to a phase inverter and a multiple-way switch. The other terminal of the phase inverter is connected to the multiple-way switch; and

the number of terminals of the multiple-way switch is equal to the number of the plurality of RF chains such that each terminal is connected to an RF chain and each antenna subarray is switchable among the plurality of RF chains.

13 . A system performing multiple user multiple input multiple output (MU-MIMO) communication comprises a processor and a memory, comprising computer readable instructions that, when executed by the processor, causes the processor to perform the method according to claim 11 .

14 . A non-transitory computer readable storage medium, storing computer readable instructions that, when executed by a processor, causes the processor to perform the method according to claim 11 .