1. Introduction
Cloud radio access network (C-RAN) [
1] is a promising and flexible architecture to accommodate the exponential growth of mobile data traffic in the next-generation cellular network. In a C-RAN, all the base-band signal processing is shifted to a single base-band unit (BBU) pool [
2]. The conventional base-stations (BSs), however, are replaced by geographically distributed remote antenna ports (RAPs) with only antenna elements and power amplifiers, which are connected to the BBU pool via high-speed low-latency fronthaul links by fiber. Thanks to its simple structure, it is promising to deploy ultra-dense RAPs in a C-RAN with low cost.
With highly dense geographically distributed RAPs, significant rate gains can be expected over that with the same amount of co-located antennas in both the single-user and multi-user cases [
3,
4,
5]. Due to the huge differences among the distance between the user and the geographically distributed RAPs, it has been shown in [
3] that the capacity in the single-user case is crucially determined by the access distance from the user to its closest RAP. This motivates us to investigate whether it is possible to achieve a significant proportion of the sum rate by using a subset of the RAPs. When the number of active RAPs is small, the other RAPs will operate in the sleeping mode with very low power consumption, which reduces the network power consumption, and significantly improves the energy efficiency [
6,
7]. Moreover, it is also able to improve the rate performance under limited fronthaul link capacity, as each RAP will only serve users with small access distances [
8,
9,
10,
11]. In the single-user case, it is straightforward to avoid distant RAPs transmitting, as they have little contribution to improving the capacity. In the multi-user case, the problem becomes challenging, as the beamformers of all users should be jointly designed.
To tackle this problem, a branch of sparse beamforming technologies are therefore proposed, where the beamforming vectors are designed to be sparse with respect to the total number of transmit antennas [
8,
9,
12]. Intuitively, a sparse beamforming vector implies that the number of active BS antennas is much smaller than the total number of BS antennas, leading to a significant reduction on both active fronthaul links and the circuit power consumption. Typically, the sparse constraint is introduced by imposing the
norm of the beamforming vector as a regularization of the original objective function. As the
norm is neither convex nor continuous, the sparse beamforming problem is in general a mixed-integer optimization problem, which is difficult to be globally optimized. Motivated by the recent theoretical breakthroughs in compressive sensing [
13], the sparse beamforming problem is formulated by including the
norm of the beamforming vectors as a regularization such that the problem becomes convex. By iteratively updating the weights of the
norm, the sparse beamformer that minimizes the total transmit power can be obtained by iteratively solving a second-order-conic-programming (SOCP) [
8] or a semi-definite-programming (SDP) [
12]. The problem can be further simplified to an uplink beamformer design problem via uplink-downlink duality [
9].
Nevertheless, when each RAP includes multiple antennas, one RAP will be switched off only when all the coefficients in its beamformer are set to be zero. In other words, all antennas at a RAP should be selected or ignored simultaneously, otherwise the number of active links from the BBU pool to the RAPs and the circuit power consumption cannot be reduced. Recently, the group sparse beamforming problem has been proposed where the antennas at the same RAP are restricted to be switched on or off simultaneously [
6,
7,
10,
14,
15,
16,
17], which further complicates the problem. Luong et al. [
17] formulated the sum rate and power consumption tradeoff problem as a mixed-integer-second-order-conic-programming (MI-SOCP) problem to obtain the global optimum by using branch-and-reduce-and-bound (BRB) algorithm, which imposes prohibitively computational complexity when the numbers of the RAPs and the users are large. To reduce the complexity, convex approximations are usually used to make it convex, continuous, and differentiable. For instance, Dai and Yu [
14] introduces a reweighted
norm of the vector that identifies the transmission power at each RAP to approximate the number of active RAPs, and the non-convex weighted sum rate maximization problem is approximated by weighted minimum mean square error (WMMSE) minimization and can be solved via a quadratical-constrained-quadratic-programming (QCQP). A more commonly used method is to replace the rate constraint by its equivalent signal-to-interference-plus-noise ratio (SINR) constraint, so as to solve the problem via SOCP [
6,
15,
16,
17]. Compared to the individual sparse beamforming, the group sparse beamforming further reduces the network power consumption, and the energy efficiency can be improved as well.
So far, most algorithms focus on the situation where each user has a single antenna [
6,
8,
9,
12,
14,
15,
16,
17]. As suggested by the multiple-input-multiple-output (MIMO) theory, the capacity increases linearly with the minimum number of transmit and receive antennas [
18]. In a C-RAN, it is desirable to employ multiple antennas at each user to exploit the potential multiplexing gains, which, however, further complicates the sparse precoder design. In fact, most techniques [
6,
8,
9,
12,
15,
16,
17] developed for single-antenna users cannot be directly applied to the multiple-antenna user case. The difficulty originates from the fact that the rate is determined by not only the SINR but also by the power allocated to the multiple sub-channels, and thus the problem cannot be transformed into an SOCP problem as in the group sparse beamformer design. Therefore, Pan et al. [
19] proposed to adopt the reweighted
norm to make the sparse constraint smooth and use the WMMSE method to make the rate expression convex, and a low-complexity algorithm is proposed to solve the network power consumption minimization problem by exploiting the special structure of the WMMSE approximation. This motivates us to extend some well-structured precoding scheme to C-RAN, and design a group sparse precoding approach with low computational complexity. In particular, we focus on designing a group sparse precoder based on a orthogonal precoding scheme, block diagonalization (BD) [
20], which has gained widespread popularity thanks to its low complexity and near-capacity performance when the number of transmit antennas is large [
21,
22,
23,
24]. With BD, the receiving beamformer can be directly calculated from the channel gain matrix and the transmit precoder, and the design of the receive beamformer, which is mutually coupled with the transmit beamformer and difficult to be optimized in the multiple antenna user case [
25], can be further simplified.
In this paper, we address the joint problem of RAP selection and joint precoder design in a C-RAN with multiple antennas at each user and each RAP. Whereas the problem is typically non-deterministic polynomial-time hard (NP-hard), we show that the problem becomes convex by inducing the reweighted norm of a vector that indicates the transmit power at each RAP as a regularization. Based on its Lagrangian dual problem, we propose an algorithm by iteratively updating the weights of the norm to generate a sparse solution. Simulation results verify that the proposed algorithm can achieve almost the same sum rate as that from exhaustive search.
The rest of this paper is organized as follows:
Section 2 introduces the system model and formulates the problem.
Section 3 proposes an iterative algorithm to solve the group sparse precoding problem. The complexity analysis of the proposed algorithm is presented in
Section 4. Simulation results are presented and discussed in
Section 5.
Section 6 concludes this paper.
Notation: Italic letters denote scalars, and boldface upper-case and lower-case letters denote matrices and vectors, respectively. denotes the norm of vector . , , and denote the transpose, conjugate transpose, trace and determinant of matrix , respectively. denotes an diagonal matrix with diagonal entries . denotes an identity matrix. and denote matrices with all entries zero and one, respectively. denotes the cardinality of set . and denote the ceiling and expectation operators, respectively.
2. System Model and Problem Formulation
Consider a C-RAN with a set of remote antenna ports (RAPs), denoted as
, and a set of users, denoted as
, with
and
, as shown in
Figure 1. Suppose that each RAP is equipped with
antennas, and each user is equipped with
N antennas. Then we have a total number of
BS antennas. The baseband units (BBUs) are moved to a single BBU pool which are connected to the RAPs via high-speed fronthaul links, such that the BBU pool has access to the perfect channel state information (CSI) between the RAPs and the users, and the signals of all RAPs can be jointly processed. With a high density of geographically distributed RAPs, the access distances from each user to the RAPs varies significantly, and the distant RAPs have little contribution to improve the capacity. This motivates us to find a subset of RAPs that can provide near optimal sum rate performance.
In particular, let
denote the set of active RAPs, with
. To utilize the multiplexing gains from the use of multiple user antennas, we assume that
. The received signal at user
k can be then modeled as
where
and
denote the transmit and receive signal vectors, respectively.
is the channel gain matrix between the active RAPs and user
k.
denotes the additive noise, which is modeled as a Gaussian random vector with zero mean and covariance
. With linear precoding, the transmit signal vector
can be expressed as
where
is the precoding matrix.
is the information bearing symbols. It is assumed that Gaussian codebook is used for each user at the transmitter, and therefore
. The transmit covariance matrix for user
k can be then written as
. It is easy to verify that
. The sum rate can be written from (
1) as
Note that it is difficult to find the optimal linear precoder that maximizes the sum rate
R due to the non-convexity of (
3), and the mutual coupling of the transmit and receive beamformers makes it difficult to jointly optimize the beamformers [
25]. In this paper, we assume that block diagonalization (BD) is adopted, where the desired signal is projected to the null space of the channel gain matrices of all the other users, such that
, or equivalently
for all
. With BD, the sum rate can be obtained as
As the signals come from more than one RAP, they need to satisfy a set of per-RAP power constraints, i.e.,
where
denotes the per-RAP power constraint.
is a diagonal matrix, whose diagonal entries is defined as
This paper focuses on the tradeoff between the sum rate and the group sparsity. In particular, the problem can be formulated as the following optimization problem:
where (8) is the zero-forcing (ZF) constraint, which ensures that the inter-user interference can be completely eliminated at the optimum.
is the tradeoff constant, which controls the sparsity of the solution, and thus the number of active RAPs. With
, the problem reduces to a BD precoder optimization problem with per-RAP power constraint. The group sparsity can be improved by assigning a larger
.
Note that the optimization problem defined in (
7)–(10) needs to jointly determine the subset
and design the transmit covariance matrices
for
K users, which is a combinatorial optimization problem and is NP-hard. A brute-force solution to a combinatorial optimization problem like (
7)–(10) is exhaustive search. Specifically, we must check all possible combinations of the active RAPs. For each combination, we must search for the optimal
that satisfies the constraints (8)–(10). In the end, we pick out the combination that maximizes the sum rate. However, the complexity grows exponentially with
L, which cannot be applied to real-world application. Instead, we use the concept of
norm to reformulate problem (
7)–(10). In particular, define
as
where
, with
denoting the transmit signal vector from all RAPs in
to user
k. The
-th to the
-th entries of
are zero if RAP
l is inactive. It is clear that the
l-th entry
is the transmit power of RAP
l, which is non-zero if and only if RAP
. It is easy to verify that
. We then have the following lemma:
Lemma 1. The problem defined in (7)–(10) is equivalent to the following optimization problem:where is the channel gain matrix from all RAPs in to user k, . Lemma 1 indicates that instead of searching over the possible combinations of
and then optimizing according to the corresponding channel gain matrices
, (
7)–(10) can be solved based on the channel between the users and all RAP antennas, i.e.,
. However, the problem (
12) is non-convex due to the existence of the
norm, making it difficult to find the global optimal solution.
In compressive sensing theory, the
norm is usually replaced by a
norm, and sparse solution can be achieved. However, simply substituting
by
in (
12) will not necessarily produce sparse solution in general, as
equals the sum power consumption instead of the number of non-zero entries. By replacing
by
, the transmit power at all RAPs still tend to satisfy the power constraints with equality at the optimum, leading to a non-sparse solution. In this paper, we propose to solve (
12) heuristically by iteratively relaxing the
norm as a weighted
norm. In particular, at the
t-th iteration, the
norm
is approximated by
where
, with
denoting a small positive constant. Equation (
12) can be then reformulated as
where
By noting that the first item of the objective function, i.e.,
, is concave with respect to
, and the second item
is affine with respect to
, we can then conclude that (
14) is a convex optimization problem. The problem can be solved by standard convex optimization techniques, e.g., interior point method [
26], which, however, is typically slow. In fact, by utilizing the structure of BD precoding, the problem can be efficiently solved by its Lagrangian dual.
3. Reweighted Based Algorithm
In this section, the algorithm to solve the group sparse linear precoding problem will be presented. To solve (
14), it is desirable to remove the set of ZF constraints in the first place. It has been proved in [
27] that the optimal solution for BD precoding with per-RAP constraint is given by
where
.
is given from the singular value decomposition (SVD) of
as
where
is the last
columns of the right singular matrix of
. It is easy to verify that
for all
.
Therefore, by substituting (
16) into (
14), the problem reduces to
Note that (
18) is a convex problem, its Lagrangian dual can be written as
where
denoting the Lagrangian dual variables. The Lagrangian dual function can be given as
where
. We can then obtain the Lagrangian dual problem of (
18) as
Since the problem (
18) is convex and satisfies the Slater’s condition, strong duality holds. The respective primal and dual objective values in (
18) and (
21) must be equal at the global optimum, and the complementary slackness must hold at the optimum, i.e.,
where
and
are the optimal primal and dual variables, respectively.
For fixed
, the Lagrangian dual function
can be obtained by solving
where
.
Appendix B shows that the optimal
can be obtained as
where
is obtained from the following reduced SVD:
, with
where
.
The optimal
for given
can be then obtained as
With the optimal
achieved for given
, we can then find the Lagrangian dual variables
by the projected subgradient method. Projected subgradient methods following, e.g., the square summable but not summable step size rules, have been proven to converge to the optimal values [
28]. In particular, a subgradient with respect to
is
. With a step size
, the dual variables can be updated as
From the complementary slackness in (
22), a stopping criterion for updating (
28) can be
Once the optimal
is obtained, the optimal precoding matrices
can be achieved by using the fact that
as
The optimal set and the corresponding precoding matrices can be obtained from . The algorithm is summarized as Algorithm 1.
Algorithm 1 Reweighted Norm Based Sparse Precoding Design. |
Initilization: Set iteration counter , Lagrangian dual variable , . Repeat:Calculate and , , according to ( 15) and ( 27), respectively. Update according to ( 28). Update the iteration counter and
Stop if , where is a pre-defined tolerance threshold. |
4. Complexity Analysis
In this section, we provide our complexity analysis of Algorithm 1. Note that can be calculated before the iterations. The main complexity of the proposed algorithm lies in step 1 and step 2.
Let us first focus on step 1. According to [
29], the calculations of matrix multiplication
and matrix inversion
have complexities on the orders of
and
, respectively, where
,
and
. Therefore, the complexity to calculate
is on the order of
. On the other hand, as the complexity of the SVD in (
25) is on the order of
[
29], the complexity to calculate
is on the order of
. By considering that the matrix calculation in (
27) has a complexity on the order of
, we can then conclude that the complexity of step 1 is on the order of
.
In step 2, as and can be calculated before updating , step 2 has a complexity on the order of . By noting that the number of RAPs is typically larger than the number of users, i.e., , and that the total number of BS antennas , the overall complexity of the proposed algorithm is on the order of , where is the average number of iterations. As we will show in the following section, the algorithm is able to converge within 15 iterations for a properly selected step size and tolerance threshold.
5. Simulation Results
In this section, simulation results are presented to illustrate the results in this paper. We assume that the channel gain matrices
from RAP
l to user
k are independent over
k and
l for all
and
, and all entries of
are independent and identically distributed (iid) complex Gaussian random variables with zero mean and variance
. The path-loss model from RAP
l to the user
k is
and
, where
is the distance between user
k and RAP
l in the unit of kilometer. The large-scale fading coefficient from RAP
l to user
k can be then obtained as
The transmit power constraint at each RAP are assumed to be identical, which is set to be dBm/Hz, , and the noise variance is set to be dBm/Hz.
We consider the case that
RAPs with
antennas each and
users with
antennas each. The positions of users and RAPs are generated in a circular area with radius 1000 km following uniform distribution. An example of the randomly generated antenna and user layout is plotted in
Figure 2.
Figure 3 shows the convergence behavior of the proposed algorithm under the layout given in
Figure 2 under different step size
,
and
. As we can see from
Figure 3, the value of (
29) rapidly decreases with the number of iterations. When the step size is large, i.e.,
, despite that the algorithm converges in general, the value increases after several iterations. Such observation comes from the fact that we use the reweighted
norm to approximate the non-convex
norm, and a large step in
will generally leading to a significant change of
in (
18). When the step size is smaller,
changes moderately, and a monotonic decrease can be observed in
Figure 3 when
and
.
Figure 4 shows how the number of active RAPs varies with iterations in the layout shown in
Figure 3 with step size
and the tolerance threshold
, where a RAP is said to be active if its transmit power
. The tradeoff constant
is set to be 0,
and
.
Figure 4 shows that the number of active RAPs
decreases with
. Specifically, with
, the problem reduces to a BD precoder design with per-RAP constraint, and all RAPs are active. With
, the sparsest solution can be achieved, i.e.,
. We should mention that the value of
that corresponds to the sparsest solution varies with the system configuration.
As
Figure 4 shows, the first several iterations lead to the biggest improvement. As iterations go on, there is no further improvement after the 15th iteration. Compared to the full cooperation case, i.e.,
, as
RAPs are switched off, 70% of the circuit power consumption can be saved with, however, limited rate performance loss, which will be illustrated later in this section.
Figure 5 plots the transmit power distribution over all 10 RAPs. Intuitively, the RAPs that have the smaller access distances contribute most to the sum rate. As shown in
Figure 2 and
Figure 5, RAP 1 and RAP 9, which are close to User 1, and RAP 7, which is close to User 2, are selected to be active, while all the other RAP’s transmit power eventrally goes to zero after 15 iterations.
Figure 6 plots how the average sum rate varies with the number of active RAPs. The average sum rate is obtained by averaging over 20 realizations of small-scale fading and 30 realizations of the positions of RAPs and users. The results of the exhaustive search is also presented for comparison, which is obtained by searching over all possible combinations of active RAPs, and computing its achievable sum rate by the algorithm given in [
27]. For the proposed algorithm, we simulate a series of different
’s to get different points along the curve. As we can see from
Figure 6, our proposed algorithm achieves almost the same average sum rate as the exhaustive search, which verifies the optimality of our proposed algorithm.
Figure 6 further plots the average sum rate with a fixed number of
instead of
L uniformly distributed antennas are deployed for comparison, which is denoted as “Fixed
L” in
Figure 6. We can clearly see that the proposed algorithm can achieve much better rate performance over that with the same amount of transmit RAP. With six selected antennas, for instance, the group sparse precoder reduces the average sum rate for only 3 bit/s/Hz, whereas if only 6 antennas were installed instead of
, an additional 9 bit/s/Hz rate loss can be observed compared to the proposed algorithm. Moreover, the rate gap between the proposed algorithm and that with a fixed number of
full cooperative RAPs further increases as the number of selected antennas
decreases. This highlights the importance of group sparse precoder design in C-RAN with a large number of distributed RAPs.