阅读说明：本技术 大规模mimo中的主动用户选择 (Active user selection in massive MIMO ) 是由 杨弘 T·马泽塔 于 2018-05-02 设计创作，主要内容包括：针对接入终端群中的每个接入终端对,计算第一信道相关性。基于第一信道相关性,从接入终端群中识别特定接入终端对。接入终端群包括特定接入终端对中的第一接入终端、特定接入终端对中的第二接入终端、以及来自接入终端群的至少一个剩余接入终端。计算第一接入终端与至少一个剩余接入终端中的每一个之间的第二信道相关性。计算第二接入终端与至少一个剩余接入终端中的每一个之间的第三信道相关性。基于第二信道相关性和第三信道相关性,选择第一接入终端和第二接入终端中的一个以从服务中丢弃。(A first channel correlation is calculated for each access terminal pair in a group of access terminals. A particular access terminal pair is identified from a group of access terminals based on the first channel correlation. The group of access terminals includes a first access terminal in the particular pair of access terminals, a second access terminal in the particular pair of access terminals, and at least one remaining access terminal from the group of access terminals. A second channel correlation is calculated between the first access terminal and each of the at least one remaining access terminals. A third channel correlation is calculated between the second access terminal and each of the at least one remaining access terminals. Based on the second channel correlation and the third channel correlation, one of the first access terminal and the second access terminal is selected to be dropped from service.)

1. A method, comprising:

calculating a first channel correlation for each access terminal pair in the group of access terminals;

identifying a particular access terminal pair from the group of access terminals based on the first channel correlation, the group of access terminals including a first access terminal of the particular access terminal pair, a second access terminal of the particular access terminal pair, and at least one remaining access terminal from the group of access terminals;

calculating a second channel correlation between the first access terminal and each of the at least one remaining access terminal;

calculating a third channel correlation between the second access terminal and each of the at least one remaining access terminals; and

selecting one of the first access terminal and the second access terminal to drop from service based on the second channel correlation and the third channel correlation.

2. The method of claim 1, further comprising:

in response to the selection, removing respective correlations for the particular access terminal pair from the first channel correlations to provide updated first channel correlations; and

repeating the identifying, the calculating the second channel correlation, the calculating the third channel correlation, the selecting, and the removing for the updated first channel correlation until a maximum correlation among the updated first channel correlations satisfies a threshold.

3. The method of claim 1, wherein selecting one of the first access terminal and the second access terminal to drop from service comprises:

selecting the first access terminal if a maximum correlation of the second channel correlations is greater than a maximum correlation of the third channel correlations.

4. The method of claim 1, wherein selecting one of the first access terminal and the second access terminal to drop from service comprises:

selecting the second access terminal if a maximum correlation of the third channel correlations is greater than a maximum correlation of the second channel correlations.

5. The method of claim 1, wherein identifying a particular access terminal pair from the group of access terminals based on the first channel correlation comprises:

identifying the particular access terminal pair from the group of access terminals as having a highest correlation of the first channel correlations.

6. The method of claim 1, wherein calculating a first channel correlation for each access terminal pair in a group of access terminals comprises:

for each respective access terminal in the group of access terminals, calculating the first channel correlation between the respective access terminal and each other access terminal in the group of access terminals.

7. The method of claim 1, wherein the first channel correlation comprises a correlation between channel vectors associated with access terminals in each access terminal pair.

8. The method of claim 1, further comprising:

discarding the selected one of the first access terminal and the second access terminal from service from the group of access terminals.

9. An apparatus, comprising:

a processor; and

a memory for storing computer program instructions that, when executed on the processor, cause the processor to perform operations comprising:

calculating a first channel correlation for each access terminal pair in the group of access terminals;

identifying a particular access terminal pair from the group of access terminals based on the first channel correlation, the group of access terminals including a first access terminal of the particular access terminal pair, a second access terminal of the particular access terminal pair, and at least one remaining access terminal from the group of access terminals;

calculating a second channel correlation between the first access terminal and each of the at least one remaining access terminal;

calculating a third channel correlation between the second access terminal and each of the at least one remaining access terminals; and

selecting one of the first access terminal and the second access terminal to drop from service based on the second channel correlation and the third channel correlation.

10. A non-transitory computer-readable medium storing computer program instructions that, when executed on a processor, cause the processor to perform operations comprising:

calculating a first channel correlation for each access terminal pair in the group of access terminals;

identifying a particular access terminal pair from the group of access terminals based on the first channel correlation, the group of access terminals including a first access terminal of the particular access terminal pair, a second access terminal of the particular access terminal pair, and at least one remaining access terminal from the group of access terminals;

calculating a second channel correlation between the first access terminal and each of the at least one remaining access terminal;

calculating a third channel correlation between the second access terminal and each of the at least one remaining access terminals; and

selecting one of the first access terminal and the second access terminal to drop from service based on the second channel correlation and the third channel correlation.

Technical Field

The described invention relates to wireless communication systems, and more particularly to active user selection in large-scale multiple-input multiple-output wireless systems.

Background

In a massive MIMO (multiple input multiple output) system, a base station is equipped with a very large number (e.g., hundreds or thousands) of serving antennas to serve multiple (e.g., tens or hundreds) of access terminals simultaneously. A massive MIMO system may provide most of the advantages of a conventional MIMO system, but with a larger scale. In particular, massive MIMO systems may provide high throughput, communication reliability, and power efficiency and linear processing.

In conventional massive MIMO systems, some access terminals may have similar channel vectors (e.g., due to being geographically close to each other), which may cause the throughput of all access terminals to drop significantly. This is especially true when max-min power control is employed. Max-min power control aims to maximize the minimum throughput, effectively equalizing the throughput for all access terminals. The net effect of such max-min power control is to reduce the performance of the best performing access terminal to improve the performance of the worst performing access terminal. It has been shown that uniform good service can be provided to all access terminals by selectively discarding multiple access terminals.

To determine which access terminals to discard, a conventional exhaustive search can theoretically be performed based on the calculated signal-to-interference-plus-noise ratio (SINR). However, such a conventional exhaustive search is impractical because the decision of which access terminals to discard must be determined within a coherence interval of typically 1 to 2 milliseconds, and the required calculations can be very complex. For a large number of access terminals, conventional searching is not possible.

Disclosure of Invention

In accordance with one or more embodiments, a system and method for selecting an access terminal from a group of access terminals to drop from service is disclosed. A first channel correlation is calculated for each access terminal pair in a group of access terminals. A particular access terminal pair is identified from a group of access terminals based on the first channel correlation. The group of access terminals includes a first access terminal in the particular pair of access terminals, a second access terminal in the particular pair of access terminals, and at least one remaining access terminal from the group of access terminals. A second channel correlation is calculated between the first access terminal and each of the at least one remaining access terminals. A third channel correlation is calculated between the second access terminal and each of the at least one remaining access terminals. Based on the second channel correlation and the third channel correlation, one of the first access terminal and the second access terminal is selected to be dropped from service.

In one embodiment, in response to selecting which of the first access terminal and the second access terminal to discard from the group of access terminals, the respective correlation for the particular access terminal pair is removed from the first channel correlation to provide an updated first channel correlation. For example, the correlation between a particular pair of access terminals may be set to zero. The steps of identifying, calculating a second channel correlation, calculating a third channel correlation, selecting and removing are repeated for the updated first channel correlations until a maximum correlation among the updated first channel correlations satisfies a threshold.

In one embodiment, selecting from the group of access terminals which of the first access terminal and the second access terminal to discard includes selecting the first access terminal if a maximum correlation in the second channel correlations is greater than a maximum correlation in the third channel correlations, and selecting the second access terminal if the maximum correlation in the third channel correlations is greater than the maximum correlation in the second channel correlations.

In one embodiment, a particular access terminal pair is identified as the access terminal in the group of access terminals having the highest correlation in the first channel correlations.

In one embodiment, calculating the first channel correlation comprises, for each access terminal pair in the group of access terminals: for each respective access terminal in the group of access terminals, a first channel correlation is calculated between the respective access terminal and each other access terminal in the group of access terminals. The first channel correlation comprises a correlation between channel vectors associated with access terminals in each of the pairs of access terminals.

In one embodiment, a selected one of the first access terminal and the second access terminal is dropped from service from the group of access terminals by reassigning the selected one of the first access terminal and the second access terminal to a different coherence time slot.

These and other advantages of the present disclosure will be apparent to those of ordinary skill in the art by reference to the following detailed description and the accompanying drawings.

Drawings

FIG. 1 depicts a high-level block diagram of a communication system according to one embodiment;

FIG. 2 illustrates a gaze propagation model in accordance with one embodiment;

FIG. 3 shows a graph of a cumulative distribution function of a correlation between two channel vectors according to one embodiment;

fig. 4 illustrates a flow diagram of a method for selecting one or more access terminals from a group of access terminals to drop from service, in accordance with one embodiment;

fig. 5 illustrates a flow diagram of another method for selecting one or more access terminals from a group of access terminals to drop from service, in accordance with one embodiment;

fig. 6 depicts a high-level block diagram of a computer, which may be implemented for selecting one or more access terminals from a group of access terminals for dropping from service, in accordance with one embodiment.

Detailed Description

Fig. 1 illustrates a high-level overview of a wireless communication system 100 in accordance with one or more embodiments. In one embodiment, the communication system 100 is a massive MIMO system in a line of sight (LOS) propagation environment, however, any suitable communication system employing any suitable propagation environment (e.g., an iid rayleigh propagation environment) may be used. Communication system 100 includes a base station 102 that simultaneously serves a group of K users or access terminals 106-a, 106-B, …, 106-K (collectively referred to as access terminals 106). Access terminal 106 communicates with base station 102 via respective propagation channels 112-a, 112-B. The access terminals 106 may be randomly located over a large geographic area, such as a city. Although communication system 100 is illustratively shown as a single cell system with base stations 102, it should be understood that communication system 100 may include any number of base stations serving K access terminals 106 simultaneously, and the principles discussed herein may be applied to any or all of those base stations.

The base station 102 may be connected to other base stations (not shown) and to the hub node 108 via a backhaul network 110. Although backhaul network 110 may generally be a fixed, landline network, such as a fiber optic network, backhaul network 110 may be any suitable communication network. For example, the backhaul network 110 may be a wireless network.

The base station 102 is equipped with an array of serving antennas 104. The service antenna array 104 may include any number M (e.g., tens or hundreds) of service antennas. For example, the serving antenna array 104 may be a single antenna element or a multiple antenna element array. In some embodiments, the service antenna array 104 may be a one-dimensional linear service antenna array or a two-dimensional service antenna array.

In one embodiment, transmissions from base station 102 to access terminal 106 (i.e., downlink transmissions) and transmissions from access terminal 106 to base station 102 (i.e., uplink transmissions) follow a Time Division Duplex (TDD) communication protocol. Such communication according to the TDD communication protocol may be performed using the Method described in U.S. patent application No.13/238,329 entitled "System and Method of wireless communication using Large-Scale Antenna Networks," which is incorporated herein by reference in its entirety.

According to the TDD communication protocol, the downlink and uplink channels are separated in time but use the same frequency. Then, through reciprocity, the channel estimate for each downlink channel can be considered equal to the estimate for the corresponding uplink channel. Any channel estimate will only be valid for the coherence interval. The coherence interval depends on the fading characteristics of the particular network of interest. In a typical mobile network, the coherence interval is approximately the time that the access terminal 106 travels a distance of one quarter of an operating wavelength.

In accordance with the TDD communication protocol, all K access terminals 106 transmit their uplink signals to base station 102 synchronously and simultaneously over their respective propagation channels 112. All K access terminals 106 transmit pilot sequences to the base station 102 simultaneously and simultaneously over their respective propagation channels 112. Thus, a complex-valued channel coefficient vector g is defined between base station 102 and each access terminal 106_k. Base station 102 uses channel vector g_kTo decode uplink signals previously transmitted by the access terminal 106. Base station 102 also uses channel vector g_kTo form a signal that includes data intended for all K access terminals 106.

Channel vector g between kth access terminal 106 and M serving antenna arrays 104_kModeled by the following equation:

wherein, beta_kModeling slow fading, z_kIs an M-dimensional column vector. z is a radical of_kDefined in equation 1 as follows:

where the superscript T denotes the matrix transpose, λ is the carrier wavelength, φ_kPhase shifts associated with random ranges between the M serving antenna arrays 102 and the kth access terminal 106 are modeled and assumed to be uniformly distributed at-pi, pi]In the above-mentioned manner,the distance difference is represented. Difference in distanceThe definition is as follows:

where d (k, m) is the distance between the kth access terminal 106 and the mth serving antenna in the array 104. By definition,

ξ -U (S) denote uniform distribution in the set

Random variable ξ in (1). Thus, phi_k～U[-π,π]。

FIG. 2 illustratively shows a LOS propagation model 200 in accordance with one or more embodiments. The LOS propagation model 200 will be described with reference to fig. 1. The LOS propagation model 200 shows that K access terminals 106 (i.e., access terminals 106-A, 106-B, 106-C, 106-D, 106-E, 106-F, and 106-G) are randomly distributed over a sphere S with a radius R and a serving antenna array 104 (at the base station 102 in FIG. 1) at its center²(R) above. While LOS propagation model 200 is shown as a sphere, it should be understood that LOS propagation model may cover any geographic area (e.g., a circle) and is not limited to a sphere.

The coordinates of the kth access terminal 106 are

In accordance with one or more embodiments, the LOS propagation model 200 will first be described below, where the array 104 is a one-dimensional linear service antenna array, and then the array 104 is a two-dimensional service antenna array.

One-dimensional array: according to one embodiment, the LOS propagation model 200 may be modeled with the array 104 as a linear service antenna array. The linear array of M service antennas is placed on a sphere S²(R) center. The coordinates of each serving antenna are:

so that the antenna pitch of the linear array is equal to

In the far field (i.e., when R is much larger than the size of the array, R>>M lambda), and is located in (x, y, z) ∈ S²Difference Δ for typical access terminals of (R)^(m)Can be calculated as:

note that the area of the spherical cap is 2 pi Rh, which is a multiple of its height h, giving lemma 1.

Introduction 1: for any R>0, if (x, y, z,) U (S)²(R)), then x to U ([ -R, R) ]])。

By symmetry, the theorem 1 also denotes y to U ([ -R, R)]) And z to U ([ -R, R)]). Although x, y and z are each individually uniformly distributed, they must collectively satisfy the relationship x²+y²+z²＝R²。

Therefore, they are not independent.

According to equations 1 and 2, for the signal at S²(R) onK access terminals 106 are uniformly distributed, and the channel vector of the kth access terminal 106 is:

wherein the content of the first and second substances,

according to one embodiment, the access terminals 106 are assumed to be evenly distributed over a one-dimensional sphere S¹On (R), i.e. on a circle of radius R. Therefore, the temperature of the molten metal is controlled,

wherein, theta_k～U([-π/2,π/2]) Is the angle of incidence with respect to the normal direction of the linear antenna array.

Two-dimensional array: according to one embodiment, the LOS propagation model 200 may be modeled with the array 104 as a two-dimensional array of service antennas. When the location of the access terminal 106 is such as S²Can naturally introduce additional random variables in the channel vector when randomized in the two-dimensional manifold of (2). Suppose M₁×M₂The rectangular service antenna array is positioned on the spherical surface S²(R) center, the coordinates of each serving antenna in the array is (x)_m,y_n0), wherein,

in the far field (i.e., when R is much larger than the size of the array, R > M₁λ and R > M₂λ), from being located at (x, y, z) ∈ S²Difference Δ in distance of access terminal 106 of (R) to the (1,1) th and (m, n) th serving antennas^(m,n)The method comprises the following steps:

from Lesion 1, for (x, y, z) — U (S)²(R))，z～U([-R,R]). Note that at S²The set of points on (R) whose Z coordinate is equal to uR being a circle

If the points are uniformly distributed on the spherical surface S²On (R), they must also be uniformly distributed on the circle

The above. Thus, the x-coordinate and the y-coordinate of the point are represented by

And

given, wherein, α to U ([0,2 π)]). This is summarized in lemma 2.

2, leading: for any R>0, if (x, y, z,) U (S)²(R)), then

And is

Wherein, alpha-U ([0,2 π)])。

Note that alpha is chosen independently of z, lemmas 1 and 2 are combined to obtain theorem 1.

Theorem 1: for any R>0, if (x, y, z,) U (S)²(R)), then

z ═ vR, where v and α are independent，v～U([-1,1]) And alpha to U ([0,2 π)])。

To on the spherical surface S²To generate uniformly distributed random points, theorem 1 only needs two uniform variables.

According to theorem 1, equation 5 can be written as:

thus, the channel vector for the kth access terminal according to equations 1 and 6

By the following M₁xM₂The entries of the matrix give:

the three LOS propagation models described above are summarized as follows:

1) ¹s one-dimensional model: is randomly located on the circle S¹Wherein the array 104 is arranged as a one-dimensional linear array. The channel vector is given by equation 4.

2) ²S one-dimensional model: randomly located on the ball S²Wherein the array 104 is arranged as a one-dimensional linear array. The channel vector is given by equation 3.

3) ²S two-dimensional model: randomly located on the ball S²Wherein the array 104 is arranged as a two-dimensional array. The channel vector is given by the terms of the matrix in equation 7.

It should be understood that the present principles may be applied to other embodiments of LOS propagation model 200. For example, in one embodiment, when the access terminal location is confined to circle S¹When (R) is used, it may be randomly located on the circle S¹Wherein the array 104 is arranged as a two-dimensional array, Δ^(m,n)Randomization is performed by only one random variable a.

For all three LOS models described above, it can be seen that:

wherein the superscript denotes the conjugate transpose. And for l ≠ k:

of interest is the degree of correlation between the channel vectors of two access terminals. Gamma ray_1,2Is the inner product size between two normalized channel vectors, where:

for simplicity, γ_1,2Will be referred to herein as two channel vectors z₁And z₂The correlation between them.

Fig. 3 illustrates a channel vector z for 256 serving antennas in the array 104, according to one embodiment₁And z₂Correlation between gamma_1,2Graph 300 of the Cumulative Distribution Function (CDF). Graph 300 shows a comparison of CDFs for correlations between channel vectors for the following cases: (1) an iid rayleigh propagation environment; (2) s¹One-dimensional model, where the access terminal 106 is randomly located on the circle S¹Wherein the array 104 is arranged as a one-dimensional 1x256 linear array; (3) s²One-dimensional model, where the access terminal 106 is randomly located on the ball S²Wherein the array 104 is arranged as a one-dimensional 1x256 linear array; and (4) S²Two-dimensional model, where the access terminal 106 is randomly located on a sphere S²Wherein the array 104 is arranged as a two-dimensional 16x16 array.

Most of the time, the channel vectors of the three LOS models are closer to orthogonal than the iid rayleigh channel vector, as shown in graph 300. Furthermore, the CDFs of the three LOS models were similar in that they had the same median, but with a 16x16 array of S²The two-dimensional model showed a slightly larger standard deviation.

It may be desirable to see S²The channel correlation in the two-dimensional model is further reduced because each channel vector is modeled using additional random variables compared to the one-dimensional model. However, this is a false expectation, as many random variables, e.g., uniformly distributed random variables, may be decomposed into a sum of two or more independent random variables.

Referring back to fig. 1, the communication system 100 includes a single cell with a base station 102, the base station 102 having M serving antenna arrays 104 while serving K access terminals 106. Order to

Is an M x K channel matrix between the M serving antenna arrays 104 and the K access terminals 106 at the base station 102.

The downlink data channel is modeled as:

wherein the content of the first and second substances,

is the received signal vector, ρ, at the K access terminals 106_dIs the normalized (with respect to the noise power of the mobile receiver) downlink SNR (signal-to-noise ratio),is the precoding input vector to the ports of the M serving antenna arrays 104 in the base station 102 and w is the noise vector. The downlink power constraint is specified as E (x)^*x)≤1。

Similarly, with the superscript' representing the corresponding variable, the uplink data channel is modeled as:

wherein the content of the first and second substances,

is the received signal vector, p, at the ports of the M serving antenna arrays 104 at the base station 102_uIs the normalized (relative to the noise power of the base station receiver) uplink SNR,

is the power control message carrying signal from the K access terminals 106 and w' is the noise vector. The uplink power constraint is specified as | E (x |)^*T⊙x`)‖_∞≦ 1, where ⊙ represents the element level multiplication.

Now, an expression of downlink and uplink SINR (signal to interference plus noise ratio) will be derived for signal cell massive MIMO systems with MR (maximum ratio) processing and ZF (zero forcing) processing.

The concept of traversal capability is associated with coding over many independent implementations of all random sources, and it allows averaging of the channel matrix to obtain a capability expression. In the case of rayleigh fading, this condition is usually satisfied by the frequency dependence of small scale fading and the possibility of performing coding over more than one time coherence interval. However, under LOS conditions, traversal capability cannot be obtained because the random angle of arrival is independent of frequency and may remain substantially unchanged for long periods of time.

Effective SINR: an explicit expression for the effective SINR is derived as follows.

Table 1: maximum ratio and zero forcing expressions for Downlink (DL) and Uplink (UL).

In this context, it is intended that,

and [ (Z X Z)^-1]_l,lDenotes (Z X Z)^-1The first diagonal element ofThe content of the element is as follows,

is power control. For the downlink, η must satisfy the total power constraint:

wherein the content of the first and second substances,

is a set of K-dimensional non-negative real vectors. For the uplink, the power control η must satisfy individual power constraints:

max-min power control: a maximum obtainable common SINR and corresponding max-min power control for all K access terminals is obtained.

Max-min power control for MR: as can be seen from table 1, the MR downlink and uplink SINR expressions are linear for power control η for a given SINR. The maximum achievable downlink and uplink SINRs and corresponding maximum-minimum power control for the MR can be obtained by bisection. For a given SINR ζ, the downlink SINR expression may be written as:

wherein the content of the first and second substances,

and diag (·) denotes a diagonal matrix.

Similarly, the uplink SINR expression may be written as:

wherein the content of the first and second substances,

algorithm 1 may be used to calculate the maximum-minimum SINR and corresponding power control for the downlink as follows.

Algorithm 1:

1) set ζ ═ S, and solve equation 14 for power control η.

2) If the obtained power control η satisfies the constraint of equation 12, the optimal SINR > S, otherwise SINR < S.

3) The maximum SINR satisfying the constraint of equation 12 is searched using the bisection method.

The maximum-minimum SINR and corresponding uplink power control can be similarly calculated using equation 15 in step 1 of algorithm 1 and using the power control constraints in equation 13 in steps 2 and 3.

Order to

Representing max-min power control. For the MR downlink, it is necessary to have:

if it is not

Then all

All can pass through the factor

To scale so that

But for any K1, …, K,

the inequality of equation 17 holds because the function

For a>0 is increased, therefore, for α>1，f(α)>f(1)。

Similar arguments for uplink max-min power control indicate that it is necessary to:

therefore, the downlink max-min power control obtained from algorithm 1 must satisfy equation 16, and the similarly obtained uplink max-min power control must satisfy equation 18.

Max-min power control for ZF: from the SINR expression in table 1, downlink and uplink max-min power control for zero forcing can be obtained as follows.

For downlink max-min power control for ZF:

wherein, a_zfIs to satisfyThe corresponding maximum-minimum SINR is:

for uplink max-min power control for ZF:

the corresponding max-min SINR is:

in a massive MIMO system, such as the communication system 100 of fig. 1, two or more access terminals 106 with similar channel vectors may degrade the performance of the entire group of access terminals 106. By dropping one or more access terminals 106 from service, the overall performance of a population of access terminals 106 in the communication system 100 may be improved. According to one embodiment, one or more access terminals 106 are selected from a group of access terminals to drop from service based on a pair-wise correlation between each access terminal 106. As used herein, the term "dropped from service" or "dropped from service" refers to rescheduling a selected one or more access terminals 106 to separate transmission resources. For example, the dropped access terminals may be served, e.g., in separate coherent time slots or different frequency bands.

Fig. 4 illustrates a method 400 for selecting one or more access terminals to discard from a group of access terminals in accordance with one or more embodiments. The method 400 may be performed by a base station or any computing device communicatively coupled to a base station.

In step 402, a pairwise channel correlation γ is computed for each arbitrary two pairs of access terminals i and j in the K groups of access terminals 106_i，jSo that 1 is not more than i<j is less than or equal to K. For example, in an access terminal group having access terminals 1 through 10, pair-wise channel correlations γ are determined for each of access terminal 1 and access terminals 2-10, for each of access terminal 2 and access terminals 1 and 3-10, for each of access terminal 3 and access terminals 1-2 and 4-10, and so on_i，j. For a given access terminali and j pairs, pair-wise channel correlation gamma_i，jRepresenting a channel vector z_iAnd z_jThe correlation between them. As described above, the channel vector z_iAnd z_jA pair-wise channel correlation between_i，jThe following were used:

wherein z is_iAnd z_jWritten as a column vector, | · denotes the magnitude of the complex number, ＊ denotes the complex conjugate transpose, | · | ceiling₂Representing the 2 norm of the phasor.

At step 404, the maximum pairwise channel correlation γ is identified as being provided_maxNamely, it is

Access terminal i^*And j^*。

In step 406, the maximum pair-wise channel correlation γ is determined_maxWhether the correlation threshold γ is satisfied (e.g., less than, or less than or equal to)_hWherein γ is_hIs greater than 0. Correlation threshold gamma_hMay be any predetermined value, for example, 20%. In step 408, if the maximum pair-wise channel correlation γ is present_maxSatisfies a correlation threshold gamma_h(i.e.,. gamma.)_max≤γ_h) The method 400 ends.

In step 410, if the maximum pair-wise channel correlation γ is present_maxNot satisfying the correlation threshold gamma_h(i.e.,. gamma.)_max＞γ_h) Selecting an access terminal i from the group of access terminals to discard based on pair-wise channel correlations with remaining access terminals in the group of access terminals^*Or j^*One of them. In one embodiment, the dropped access terminal i^*Or j^*Is an access terminal that has a higher correlation with the remaining access terminals in the access terminal group. Therefore, if

This indicates the access terminal i^*And access terminalj^*Then the access terminal i is assigned to step 414, with a higher maximum correlation than the remaining access terminals^*Dropped from the access terminal population. Otherwise, in step 412, access terminal j is assigned^*Dropped from the access terminal population. Discarding access terminal i^*Or j^*The selected one of may include re-assigning the selected access terminal to a different resource (e.g., a different coherent time slot or frequency).

In step 416, the access terminal i^*And j^*Pair-wise channel correlation between

Set to zero and the method 400 returns to step 404. The method 400 repeats until the maximum pair-wise channel correlation γ_maxSatisfies a correlation threshold gamma_hUntil now.

In one embodiment, method 400 is repeated instead until the maximum pair-wise channel correlation γ_maxSatisfies a correlation threshold gamma_hBy this, the process 400 is repeated at a predetermined number of intervals or iterations (e.g., 3 intervals). Thus, in this embodiment, at step 406, the variable representing the current interval is compared to the variable representing the maximum predetermined number of intervals. If the variable representing the current interval satisfies (e.g., is greater than or equal to) the variable representing the maximum predetermined number of intervals, the method 400 ends at step 408. Otherwise, the method 400 proceeds to step 410. At step 416, the variable representing the current interval is incremented.

It has been found that the above-described method for selecting one or more access terminals from a group of access terminals to drop from service achieves substantially the same performance as conventional exhaustive search.

An example will now be described to demonstrate the performance of a communication system by selectively discarding one or more access terminals. In a macro-cellular massive MIMO base station in a suburban area with a radius of 2 km, the base station is equipped with 256 service antennas. The total available downlink radiated power is 10 watts. The transmit power for each access terminal is 200 milliwatts. The base station serves 18 access terminals. The carrier frequency is 1.0 GHz.

Tables 2 and 3 below summarize the performance of a communication system implemented by selectively dropping one or more access terminals. In all cases, a significant increase in SINR of 95% (i.e., five percent) is possible compared to baseline. The chance of 2 or more users being dropped is only 5% and the chance of 1 user being dropped at the most is only 50%. Once a small number of access terminals are dropped, the remaining LOS channel vectors are nearly mutually orthogonal. Therefore, in order for massive MIMO to perform well under LOS conditions, it is sufficient to selectively drop a small number of access terminals from service.

Table 2: maximum ratio processing

Table 3: zero breaking treatment

Fig. 5 illustrates a method 500 for selecting one or more access terminals from a group of access terminals to drop from service in accordance with one or more embodiments. The method 500 will be described with reference to fig. 1. The method 500 may be performed by any computing device communicatively coupled to the base station 102. For example, in one embodiment, method 500 may be performed by a computing device implemented at base station 102 (e.g., as shown in fig. 6). In another embodiment, the method 500 may be performed by a computing device implemented at a hub node 108 connected to the base station 102 via a backhaul network 110.

In step 502, a first channel correlation γ is calculated for each access terminal pair in a group of access terminals_i,j. For example, a first channel correlation is computed between each access terminal in a group of access terminals and each other access terminal in the group of access terminals. Channel correlation indicates the correlation of the channel for a given accessChannel vector z for terminal i and j pair_iAnd z_jA pair-wise correlation between them.

At step 504, a particular access terminal pair is identified from a group of access terminals based on the first channel correlation. The group of access terminals includes a first access terminal in the particular pair of access terminals, a second access terminal in the particular pair of access terminals, and at least one remaining access terminal from the group of access terminals. In one embodiment, a particular access terminal pair is identified as the access terminal i having the highest correlation among the first channel correlations^*And j^*。

At step 506, a second channel correlation is computed between the first access terminal and each of the at least one remaining access terminals. At step 508, a third channel correlation is computed between the second access terminal and each of the at least one remaining access terminals. At step 510, one of the first access terminal and the second access terminal is dropped from service from the group of access terminals based on the second channel correlation and the third channel correlation. In one embodiment, the first access terminal is discarded if the maximum correlation in the second channel correlations is greater than the maximum correlation in the third channel correlations (i.e., the first access terminal is more correlated with the remaining access terminals than the second access terminal). Otherwise, if the maximum correlation in the third channel correlations is greater than the maximum correlation in the second channel correlations (i.e., the second access terminal is more correlated with the remaining access terminals than the first access terminal), the second access terminal is discarded. Discarding the selected one of the first and second access terminals may include reallocating the selected access terminal to a different resource (e.g., a different coherent time slot or frequency).

In one embodiment, the steps of method 500 are repeated until the maximum correlation in the first channel correlations meets (e.g., is less than, or is less than or equal to) a correlation threshold. For example, after selecting one of the first access terminal and the second access terminal to drop from the group of access terminals, the channel correlation between the first access terminal and the second access terminal is removed (e.g., set to zero) from the first channel correlation, and the method repeats steps 504, 506, 508, and 510 until the maximum correlation in the first channel correlation satisfies the correlation threshold.

The systems, apparatus, and methods described herein may be implemented using digital circuitry or using one or more computers using well-known computer processors, memory units, storage devices, computer software, and other components. Generally, a computer includes a processor for executing instructions and one or more memories for storing instructions and data. The computer can also include or be coupled to one or more mass storage devices, such as one or more magnetic disks, internal hard disks and removable magnetic disks, magneto-optical disks, and the like.

The systems, apparatus and methods described herein may be implemented using a computer that operates in a client-server relationship. Typically, in such systems, client computers are located remotely from server computers and interact via a network. The client-server relationship may be defined and controlled by computer programs running on the respective client and server computers.

The systems, apparatuses, and methods described herein may be implemented within a network-based cloud computing system. In such a network-based cloud computing system, a server or another processor connected to a network communicates with one or more client computers via the network. The client computer may communicate with the server, for example, via a web browser application that resides on and runs on the client computer. Client computers may store data on servers and access data via a network. A client computer may send a request for data or a request for an online service to a server via a network. The server may perform the requested service and provide the data to the client computer. The server may also send data adapted to cause the client computer to perform specified functions, e.g., perform calculations, display specified data on a screen, etc. For example, the server may send a request adapted to cause the client computer to perform one or more of the method steps described herein, including one or more of the steps of fig. 4 and 5. Certain steps of the methods described herein, including one or more of the steps of fig. 4 and 5, may be performed by a server or by another processor in a network-based cloud computing system. Certain steps of the methods described herein, including one or more of the steps of fig. 4 and 5, may be performed by a client computer in a network-based cloud computing system. The steps of the methods described herein, including one or more of the steps of fig. 4 and 5, may be performed by a client computer in a server and/or network-based cloud computing system, in any combination.

The systems, apparatus, and methods described herein may be implemented using a computer program product tangibly embodied in an information carrier, e.g., in a non-transitory machine-readable storage device, for execution by a programmable processor; the method steps described herein, including one or more of the steps in fig. 4 and 5, may be implemented using one or more computer programs executable by such a processor. A computer program is a set of computer program instructions that can be used, directly or indirectly, in a computer to perform a certain activity or bring about a certain result. A computer program can be written in any form of programming language, including compiled or interpreted languages, and it can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment.

A high-level block diagram 600 of an exemplary computer that may be used to implement the systems, apparatus, and methods described herein is depicted in FIG. 6. The computer 602 includes a processor 604 operatively coupled to a data storage device 612 and a memory 610. The processor 604 controls the overall operation of the computer 602 by executing computer program instructions that define such operation. The computer program instructions may be stored in a data storage device 612 or other computer readable medium and loaded into memory 610 when execution of the computer program instructions is required. Thus, the method steps of fig. 4 and 5 may be defined by computer program instructions stored in the memory 610 and/or data storage device 612 and controlled by the processor 604 executing the computer program instructions. For example, the computer program instructions may be embodied as computer executable code programmed by one skilled in the art to perform the method steps of fig. 4 and 5. Thus, by executing the computer program instructions, the processor 604 performs the method steps of fig. 4 and 5. The computer 602 may also include one or more network interfaces 606 for communicating with other devices via a network. The computer 602 may also include one or more input/output devices 608 that enable a user to interact with the computer 602 (e.g., a display, keyboard, mouse, speakers, buttons, etc.).

The processor 604 may include a general purpose microprocessor, a special purpose microprocessor, and may be the sole processor or one of multiple processors of the computer 602. For example, processor 604 may include one or more Central Processing Units (CPUs). The processor 604, data storage device 612, and/or memory 610 may include, be supplemented by, or incorporated in, one or more Application Specific Integrated Circuits (ASICs) and/or one or more Field Programmable Gate Arrays (FPGAs).

The data storage device 612 and the memory 610 each include tangible, non-transitory computer-readable storage media. The data storage device 612 and the memory 610 may each include high speed random access memory, such as Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), double data rate synchronous dynamic random access memory (DDR RAM), or other random access solid state memory devices, and may include non-volatile memory, such as one or more magnetic disk storage devices, such as an internal hard disk and a removable disk, magneto-optical disk storage devices, flash memory devices, semiconductor memory devices, such as Erasable Programmable Read Only Memory (EPROM), Electrically Erasable Programmable Read Only Memory (EEPROM), compact disk read only memory (CD-ROM), a digital versatile disk read only memory (DVD-ROM) disk, or other non-volatile solid state memory devices.

Input/output devices 608 may include peripheral devices such as printers, scanners, display screens, and the like. For example, input/output devices 608 may include a display device such as a Cathode Ray Tube (CRT) or Liquid Crystal Display (LCD) monitor for displaying information to a user, a keyboard, and a pointing device such as a mouse or a trackball by which the user can provide input to computer 602.

Any or all of the systems and apparatus discussed herein, including the elements of communication system 100 of fig. 1, can be implemented using one or more computers (e.g., computer 602).

Those skilled in the art will recognize that an actual computer or computer system implementation may have other structures and may also contain other components, and that FIG. 6 is a high-level representation of some components of such a computer for illustrative purposes.

The foregoing detailed description is to be understood as being in every respect illustrative and exemplary, but not restrictive, and the scope of the invention disclosed herein is not to be determined from the detailed description, but rather from the claims as interpreted according to the full breadth permitted by the patent laws. It is to be understood that the embodiments shown and described herein are merely illustrative of the principles of the described invention and that various modifications may be made by those skilled in the art without departing from the scope and spirit of the invention. Various other combinations of features may be implemented by those skilled in the art without departing from the scope and spirit of the invention.