 Research
 Open Access
 Published:
NetworkMIMO for downlink inband relay transmissions
EURASIP Journal on Wireless Communications and Networking volume 2013, Article number: 13 (2013)
Abstract
With the objective of improving the spectral efficiency and coverage homogeneity of wireless cellular systems in the downlink, we investigate how to take advantage of two promising transmission technologies envisioned in current standards: networkMIMO and relay stations (RSs). It is assumed that halfduplex RSs are deployed in a cellular system, where the duration of the relayreceive and the relaytransmit phases is fixed beforehand for all cells. In order to reduce the spectral efficiency loses associated to halfduplex relaying, we propose the use of base station (BS) cooperation under a networkMIMO precoding strategy based on blockdiagonalization zeroforcing, and optimize radio resources under the convex performance criteria constrained by the perBS power, modulation and coding schemes and the transmission rate in the relaytransmit phase. By applying convex optimization techniques, the optimal precoding strategy is derived and a suboptimal low complexity solution is also proposed. The obtained solutions are evaluated at system level and compared to other cooperative and noncooperative BSbased schemes.
1. Introduction
Due to the availability of new sophisticated terminals and bandwidthdemanding services, future cellular network designs are expected to provide high system spectral efficiency and homogeneous coverage. The design of such advanced network architectures is a challenging task, since radio resources like power and bandwidth are often scarce and the wireless channel presents unique impairments such as fading, multipath, and interference. In this regard, next generation standards are already considering that conventional paradigms need to be rethought in a way able to exploit multiantenna capacity increase, shadowing diversity, and path loss break concept. For that reason, mature enabling technologies, like multipleinput multipleoutput (MIMO) or full frequency reuse planning, are an integral part of the system. However, more efficient wireless technologies and novel deployment concepts, like RSbased deployments or coordinated base station (BS) transmissions (often called networkMIMO or coordinated multipoint), are part of the ambitious study items [1] to deal with the demanding traffic growth expected in the near future. Leveraging on the advantages offered by the joint use of all these techniques, to improve performance and cost efficiency, is a challenge faced by IEEE 802.16 m [2] and LTEA [3].
One way to increase network capacity is by means of increasing the radio access points and reducing the cell size. However, since a regular BS is expensive, this solution is not attractive to network operators. Pursuing the same objective with lower cost, relaybased deployments is a promising approach that will have a large effect on system’s functions and architectures [4, 5]. While implementation details of fullduplex RS are still under investigation, relaybased enhancements in standards consider halfduplex relay operation [5], which incur a rate penalty as they require at least two timeslots to relay a message from source to destination [6, 7]. It is therefore crucial to enhance the capacity of the inband wireless backhaul between source and relay (in our case, the BS–RS link) to increase the information rate. One of the solutions usually assumed is that relay stations (RSs) are placed in specifically planned positions above rooftop or in lampposts, ensuring lineofsight (LOS) conditions in the BS–RS link, and hence reducing the path loss and shadowing effects. However, the price to pay is twofold: the likely LOS propagating conditions also to othercell BSs (which will inject harmful interference) and the rank deficiency of the spatial channel when both BS and RS are equipped with multiple antennas. Both effects are detrimental to MIMO channel gains [8, 9].
In this respect, networkMIMO [10] seems especially suited to address inband backhauling in relay transmissions for the downlink. While coordination may be seen as an efficient way to combat the interference from neighbor cells [11, 12], it also creates a virtual MIMO broadcast channel whose number of degrees of freedom is boosted (if compared to a conventional singleuser MIMO under TDMA) and is hardly affected by the rank deficiency of singleuser MIMO channels in LOS. It has been observed that networkMIMO based on blockdiagonalization zeroforcing (BDZF) performs closely to dirtypaper coding [10] but, although its simplicity, it requires accurate channel knowledge from all involved links. However, networkMIMO is again appropriate for our problem, thanks to the long channel coherence time of BS–RS links.
An additional way to improve the efficiency of relay transmissions is by optimizing the duration of the relayreceive and the relaytransmit phases [6, 7]. In [13], we observed that the joint optimization of coordinated BS–RS links (through networkMIMO) and transmit duration phases brings large benefits. However, this approach is not convenient when considering multiple coordinated cells: if each group of coordinated cells adapts the duration of the transmission independently, the interference power observed in each transmission slot may be timevarying, a harsh, and undesirable situation for the cellular system. Therefore, so as to guarantee the predictability of other cellclusters interference within transmission frames, the transmit duration phases should be fixed beforehand based on some criterion. For example, in [14], a statistical criterion was used and it was observed that still large gains both in terms of spectral efficiency and outage rate could be obtained.
Following the above motivations, we focus our study in the development of precoding strategies when networkMIMO based on BDZF is applied to the first hop of a relaybased transmission and the duration of the relayreceive and relaytransmit phases is fixed according to statistical approach. In this regard, we derive the optimal precoding strategy and compare it with other suboptimal and simpler solutions. The optimal precoding strategy is obtained in order to maximize the transmission rates subject to perBS power constraints, modulation and coding schemes (MCS) constraints and maximum bitrates on the RS–UEs link, following similar steps used in [12] to obtain the optimal precoding for networkMIMO. On the other side, the suboptimal precoding strategy comes out by maximizing the transmission rates subject to a sumpower constraint [15] and after adapting the power allocated to each symbol stream so as to satisfy the perBS power constraints, as it was done in [14]. As a result, the optimal precoder turns into a complex structure and the questions arising are

How much a relaybased system can benefit in terms of cellular spectral efficiency and celledge spectral efficiency by BS cooperation?

How much gain can be achieved with the optimal MIMO symbol precoding compared to suboptimal and simpler precoding solutions?
This article is structured as follows. In Section 2, we state the system model. The resource allocation problem is formulated in Section 3, in which we derive the optimal precoding strategy and also a suboptimal lowcomplex precoding strategy is proposed. In Section 4, simulation results are presented. Conclusions are drawn in Section 5.
Notation: For a given matrix A, its transpose matrix is denoted as A^{T} and its transpose conjugate matrix as A^{H}; A operator is referred to the determinant of A, tr(A) to the trace, rank(A) to the rank (which is the minimum between the column and the row rank) and kernel(A) is denoting the null space of the matrix; righteigenvec(A) denotes a matrix containing the right eigenvectors of A and eigenval(A) denotes a matrix with the eigenvalues of A in its diagonal; A ≥ 0 means that matrix A is positive semidefinite; A = diag(a_{1},…,a_{ N }) describes a diagonal matrix with diagonal entries a_{1},…,a_{ N }. For given scalars a and b, min (a,b) denotes the minimum between a and b, and (a)^{+} denotes the maximum between a and 0.
2. System model
This section presents the model for a relayedbased deployment with networkMIMO applied to the inband backhaul. In Section 2.1, some necessary system assumptions are detailed. Signal model in BS–RS link and signal model in RS–UE link are devoted to Sections 2.2 and 2.3, respectively.
2.1. System assumptions
Our system definition is based on the following practical assumptions:

1.
The number of antennas at BS, RS, and UE is n _{B}, n _{R}, and n _{M}, respectively.

2.
The quality of the backhaul link supports the exchange of perfect channel state information at the transmitter side (CSIT) of the first hop among the cooperating BSs. CSIT is assumed for transmissions from BSs to RSs since long time coherence is expected in those links as both terminals have static positions.

3.
A maximum possible MCS is assumed.

4.
RSs are time halfduplexed terminals operating under decodeandforward. That is a suitable coding approach for BS–RS links, where high signaltonoise ratio (SNR) is expected if LOS propagation is met [16].

5.
User equipments (UEs) do not process the signals transmitted by the BS, only those transmitted by the RS. In other words, simple forwarding relaying is assumed.

6.
No CSIT is available at RSs. RS transmissions are not coordinated in the way BS transmissions are. Their transmissions are interfered as multiple RSs transmit simultaneously (see Figure 1).

7.
Each RS transmits to a single associated UE and therefore it is considered as a single user MIMO link with interference. Full CSIT may be exploited at the RS if sufficient feedback rate from the UE is allowed. Otherwise, only average CSIT is assumed.

8.
It is assumed that the duration of slots for the BS–RS links and for the RS–UE links is constant to preserve interference stationarity between cell clusters.

9.
The scheduler implements a roundrobin TDMA algorithm to serve users in groups of R users.

10.
The radio resource allocation (transmitted power, beamformers) is performed by prioritizing each UE selected by the scheduler, so that a certain quality of service (QoS) is provided. A reasonable criterion is that user’s priority is taken as the inverse of the capacity in the RS–UE link so as to avoid unfair service to deprived users. This way, in comparison with the sumrate (SR) criterion (equal priorities to all users), a weighted SR (WSR) criterion would perform better in terms of outage rate while losing in terms of spectral efficiency.
2.2. Signal model in BS–RS link
We shall adopt a downlink transmission setup where B BSs are coordinated and assisted by R RSs to transmit messages to R UEs. Each UE is associated to a single RS. All BSs transmit on a fixed fraction of time α_{1} on the first hop to the RSs following a networkMIMO strategy based on BDZF [17] (see Figure 1), which is appropriate for BS–RS links in LOS conditions (MMSE precoding provides improved performance at medium to low SNR [18]).
The signal x transmitted by all n_{B} · B antennas is given by
where x_{ k } is the signal transmitted by the k th BS, b_{ i } is the symbol stream vector of m_{ i } independent components associated to the i th RS and Q_{ i } is its associated precoding matrix. We adopt a BDZF precoding [17] defined by two matrices
where S_{ i } is the (B · n_{B}  (R  1) · n_{R}) × m_{ i } MIMO symbol precoding matrix, and V_{ i } is the B · n_{B} × (B · n_{B}  (R  1) · n_{R}) BDZF precoding matrix. By virtue of the ZF precoding, the signal received by the i th RS is affected by the n_{R} × B · n_{B} channel matrix H_{ i } (containing the channel gains between the transmitting antennas at the B BSs and its receiving antennas):
where n_{ i } is the additive complex white Gaussian noise.
The BDZF precoder design implies {\mathbf{V}}_{i}\in \mathrm{kernel}\left({\tilde{\mathbf{H}}}_{i}\right), where
The existence of a kernel of appropriate rank requires B · n_{B} > (R  1) · n_{R} and hence
In addition, symbol decidability at the receivers implies the following conditions on the number of streams
It must be remarked that in the case the i th RS observes all coordinated BSs in LOS (hence BS–RS link channels are rank deficient) the rank of H_{ i } grows up to fullrows rank provided that channels to the B BSs be linearly independent.
Regarding matrix S_{ i }, it is designed to maximize users’ transmission rates:
where N_{ i } refers to the correlation matrix of the noise n_{ i } plus external interference at the i th RS. In addition, we consider practical MCS in the BSresource allocation, so we have to upper limit the rate for each transmission mode of each user on the BS–RS link
where S_{max} is the maximum spectral efficiency that can be achieved by the LTEA MCS [3] for each transmission mode.
The total power transmitted by the k th BS is given by
where x_{ k } is the signal transmitted by the k th BS (from Equation 1) and B_{ k } selects the n_{B} rows of V_{ i }S_{ i } used to transmit from the k th BS
2.3. Signal model in RS–UE link
On the second hop, each RS transmits to its associated UE on a fixed fraction of time α_{2} = 1 – α_{1} (see Figure 1). Since R relays can transmit simultaneously, interference is created. We could also split the second hop in different time slots, reducing interference at the expenses of some loss in spectral efficiency [13].
As we are assuming no coordination among RSs and simple receivers at the UE, only single user MIMO transmissions can be appointed. Then, the achievable rate for each RS–UE link, r_{2}, follows the conventional MIMO capacity expression affected by the presence of interference from other RS transmissions. If no CSIT is assumed at the RS
where r_{2i} denotes the rate in the i th RS–UE link, {P}_{i}^{R} is the power transmitted by the i th RS to its UE, and {P}_{j}^{R} defines the power transmitted by the j th RS.
To preserve information flow through RSs, the rate at the i th UE served by the i th RS is constrained by the minimum of rates in both hops
where r_{1i} is transmission rate on the BS–RS link for the i th RS (6). Since the transmission time allocation is fixed and the transmission rate on the second hop r_{2i} is given by the propagation channel conditions in the RS–UE links, we should allocate power in the BS–RS link accordingly so as not to waste it. Then, Equation (11) can also be written as a constraint on the resource allocation in the BS–RS link
3. Resource allocation
This section is devoted to find the optimal power allocation and beamforming vectors for BS–RS links (both included in the MIMO symbol precoder S_{ i }) in order to maximize the WSR of the system, when there are perBS power constraints and maximum MCS. Following similar steps as in [12] we obtain the optimal solution in WSRbased optimal resource allocation. Similarly, in WSRbased suboptimal resource allocation, a suboptimal and simpler solution for S_{ i } is presented, which assumes the conventional structure for the beamforming in networkMIMO, [15], and then the power is allocated so as to satisfy the power and rate constraints [14].
3.1. WSRbased optimal resource allocation
We want to maximize users’ transmission rates subject to perBS power constraint to obtain the optimal MIMO symbol precoding matrix structure for matrices S_{ i } in (2). If we consider T_{ i } = S_{ i }S_{ i }^{H}, the WSR maximization problem for the downlink transmission in relayed networkMIMO with BDZF precoding and perBS power constraints is formulated as follows
where r_{1i} is the transmission rate in the BS–RS link for the i th UE and μ_{ i } the priority given to it in order to add some QoS over the served users. The first constraint accounts for the perBS power constraint (taken from (8)), where {P}_{k}^{\mathit{\text{max}}} denotes the perBS available power at the k th BS. The second constraint refers to the peruser transmission rate in the BS–RS link. The third constraint contains the peruser transmission rate, which is limited by the peruser transmission rate in the RS–UE link (from Equation 12) and the peruser MCS constraint in the BS–RS link (from Equation 7). Finally, the fourth constraint imposes that the transmission rates have to be positive, and the fifth constraint imposes that the transmit covariance matrices have to be positive and semidefinite.
With the inclusion of the new variable r_{1i} in the problem, it can be verified that problem (P_{1}) turns out to be a convex optimization problem (as the objective function is convex over {r_{1i}} and all the constraints specify a convex set over T_{ i }’s and {r_{1i}} [19]). Then, problem (P_{1}) can be solved using standard convex optimization techniques, e.g., interiorpoint method [19]. However, the optimal structure for T_{ i } and r_{1i} can be found in a closedform using Lagrange duality methods [19, 20].
Proposition 1
Denoting by {γ_{ k }} and {φ_{ i }} a set of nonnegative dual variables associated to the k th perBS power constraint and the i th peruser transmission rate, respectively, the optimal solution to problem (P_{1}) for a set of fixed {γ_{ k }} and {φ_{ i }} is given by
Therefore, the power allocation and attained rate result
Proof. See Appendix 1.
Once we have the optimal solution for T_{ i }’s and {r_{1i}} for a set of fixed {γ_{ k }} and {φ_{ i }}, we have to optimize {γ_{ k }} and {φ_{ i }}. The set {φ_{ i }} can be optimized applying the KKT conditions [19].
Proposition 2
Denoting by {γ_{ k }} a set of nonnegative dual variables associated to the k th perBS power constraint, the optimal solution to {φ_{ i }} for a set of fixed {γ_{ k }} is given by
where the second result is obtained by substituting the expression of the transmission rate in (15) into the third constraint of problem (P_{1}) and equaling the restriction. Notice that Equation (16) depends on the number of transmission modes per user (m_{ i }) and, in addition, we need to evaluate r_{1i} with Equation (15) which involves a waterfilling algorithm. In this regard, an algorithm to find φ_{ i }^{*}, which is based on hypothetic testing, is presented in Algorithm 1 (see Appendix 1 for details).
Algorithm 1: Efficient search of φ_{ i }for a set of fixed {γ_{ k }}
Inputs: {γ_{ k }}
Output: {\phi}_{\mathit{i}}^{*}

1.
Set m _{ i } = rank (H _{ i }V _{ i })

2.
Compute p _{ ij } ({γ _{ k }}, 0) and then r _{1i} ({γ _{ k }}, 0) using (15)

3.
if (r _{1i}  min(α _{2} r _{2i}/α _{1}, S _{max} m _{ i }) ≤ 0)

4.
• {\phi}_{\mathit{i}}^{*}=0

5.
• go to 17

6.
else

7.
• compute φ _{ i } using (16)

8.
• updated p _{ ij } ({γ _{ k }}, φ _{ i }) using (15)

9.
• {m}_{i}=\mathit{\text{count}}\left({P}_{ij}>0,\forall \mathit{j}\right)

10.
\mathbf{\text{if}}\left({m}_{i},=,{m}_{i}\right)

11.
◦ {\phi}_{\mathit{i}}^{*}={\phi}_{\mathit{i}}

12.
◦ go to 17

13.
else

14.
◦ \left({m}_{i},=,{m}_{i}\right)

15.
◦ go to 2

16.
end

17.
end
Proof. See Appendix 1.
Now, problem (P_{1}) has to be optimized in terms of {γ_{ k }}. In this regard, an effective algorithm is proposed in Algorithm 2 to compute {γ_{ k }} following the ellipsoid method [19]. For each set of {γ_{ k }} the optimal solution for T_{ i }’s, {r_{1i}}, and {φ_{ i }} are computed following (14), (15), and (16) (see details in Appendix 1).
Algorithm 2: Efficient search of {γ_{ k }} and {φ_{ i }} in problem P_{1}

1.
Initialize: γ _{ k } k = 1,…, B

2.
do

3.
for i = 1,…,R

4.
◦ {\phi}_{\mathit{i}}^{*} = Algorithm 1({γ _{ k }})

5.
◦ Compute T _{ i } using (14)

6.
end

7.
• Compute the subgradient
{d}_{k}=\text{tr}\left({\sum}_{i=1}^{R}{\mathbf{\text{B}}}_{k}{\mathbf{\text{V}}}_{i}{\mathbf{\text{T}}}_{i}{\mathbf{\text{V}}}_{i}^{H}\right){\mathit{P}}_{\mathit{k}}^{\text{max}}

8.
• Update {\gamma}_{k}\mathrm{\text{'}}\text{s} accordingly based on the ellipsoid method [19]

9.
until all {\gamma}_{k}\mathrm{\text{'}}\text{s} converge

10.
Compute optimal {\mathbf{\text{T}}}_{i}^{*} and r _{1i} given by (14) and (15).
Finally, from (14) we can directly obtain the optimal MIMO symbol precoding matrix:
It is worth to remark that S_{ i }^{*} consists of nonorthogonal columns if B_{γ} is a nonidentity diagonal matrix (i.e., the optimal {γ_{ k }} are not all equal). Moreover, it can be verified that for the sumpower constraint case, S_{ i }^{*} in (17) contains orthogonal columns (i.e. orthogonal beamforming) since V_{ i }^{H}V_{ i } = I, and hence its structure turns out to be independent of γ. For the same reason, when the optimal solution for S_{ i }^{*} (17) is used, the computational complexity is significantly higher than for the sumpower constraint case as singular value decomposition (SVD) must be computed for each algorithm’s iteration (i.e., for each value of {γ_{ k }}).
3.2. WSRbased suboptimal resource allocation
Let us assume a conventional precoding [15] for the MIMO symbol precoder S_{ i }:
where {\widehat{\mathbf{P}}}_{i} is an m_{ i } × m_{ i } diagonal matrix describing the power allocated per symbol stream b_{ i }. It can be verified that the structure in (18) is optimal only when a single constraint (normally the sumpower) is imposed to the minimization problem (P_{1}) in WSRbased optimal resource allocation. Even though, as we are interested in satisfying the perBS power constraints, we can set the whole problem by including this structure for the MIMO symbol precoding matrices and then optimize the diagonal elements of {\widehat{\mathbf{P}}}_{i} to satisfy the perBS power constraints and the MCS constraints. Following the precoding structure (18), the power transmitted by the k th BS in (8) can be reformulated as
where {\tilde{\mathbf{W}}}_{i}^{k} contains the n_{B} rows of {\mathbf{V}}_{i}{\widehat{\mathbf{W}}}_{i} used by the k th BS in the transmission of message to the i th RS and {\tilde{w}}_{\mathit{ij}}^{k} is the j th column of {\tilde{\mathbf{W}}}_{i}^{k}.
This way, the WSR maximization problem for the downlink transmission in relayed networkMIMO with BDZF precoding and perBS power constraints can be formulated as follows [14]
where {\widehat{r}}_{1i} is the transmission rate in the BS–RS link for the i th UE and μ_{ i } the priority given to it. The first constraint accounts for the perBS power constraint, see (19), where P_{ k }^{max} denotes the perBS available power at the k th BS; the second constraint refers to the peruser transmission rate in the BS–RS link; the third constraint contains the peruser transmission rate in the RS–UE link (from Equation 12), and the fourth constraint is related to the peruser MCS constraint in the BS–RS link (which is here included directly to the power to be allocated). Finally, the fifth and sixth constraints account for the positiveness of the user’s rate and allocated power.
Problem (P_{2}) is convex and can be solved using standard convex optimization techniques, like interior point methods [19]. Nevertheless, in [14] we elaborate an efficient numerical algorithm based on the dual update methods [19, 20] which defines a polynomial complexity algorithm along with a reduction of the number of variables to be optimized.
Proposition 3
Denoting by {ϕ_{ k }} and {κ_{ i }} a set of nonnegative dual variables associated to the k th perBS power constraint and the i th peruser transmission rate, respectively, the optimal solution to problem (P_{2}) for a set of fixed {ϕ_{ k }} and {κ_{ i }} is given by
Proof. See [14], where also an algorithm is proposed to effectively update {ϕ_{ k }} and {κ_{ i }} in (21).
4. Evaluations and results
The evaluation of the proposed approach is done on a radio access network based on LTEadvanced specifications [3] at the 2 GHz band and 20 MHz bandwidth. We consider a hexagonal cellular deployment where B = 3 BS and a total of R = 6 RSs are deployed on the macrocell (see Figure 1). On each scenario, six UEs are dropped, each one attached to a different RS. Simulation parameters are detailed in Table 1.
Evaluations and results are taken in two different propagation scenarios: outdoor homogenous and outdoor/indoor hotspot. First, a deployment with simple propagation channel conditions is considered where users are located outdoor and homogeneously in the macrocell. Second, a hotspot scenario is deployed with indoor and outdoor users located forming areas of high traffic demand and propagation channel conditions are taken from the literature [21]. See Figure 2 for detailed deployment of users and RSs in the macrocell.
In relaybased systems, it is very important to optimize the duration of the relayreceive and the relaytransmit phases; but as we are assuming that they are fixed beforehand, it is crucial to make a good choice of the time devoted to each phase. In this regard, in [13] we observed that the optimum α_{1} in terms of spectral efficiency is a random variable that depends on the particular scenario and the target function to be maximized. For that reason, the duration of time slot α_{1} used in the simulations results is obtained from experimental results taking into account the LTEA frame structure and the maximum delay for voice transmission [3]. The considered values are depicted in Table 2 for the different scenarios.
Two fundamental measures are adopted: cellular spectral efficiency (S_{ e }), as the sum rate of R users divided by 3 (there are R users and 3 cells deployed per macrocell) averaged over many deployments, and 95% outage rate (r_{ out }) (or celledge spectral efficiency), as the peak achievable rate of the 5percentile worst users in the cell over many deployments. Both capture most of the benefits offered by coordination and cooperation of BS and relaybased transmission.
4.1. Outdoor homogeneous scenario
In a first deployment, all RSs are at the same distance d_{BS–RS} to their associated BS, equal to 60% of the cell radius (experimentally found as the best position) and users are uniformly placed beyond 60% of the cell radius (so as to include only those users benefiting from the presence of relays). See Figure 2left for detailed deployment of the RS and users in the macrocell.
The propagation channel between each BS and RS, and between each RS’s and UE’s is characterized by a path loss component, a shadow fading and a Rayleigh fading. The path loss component is modeled as β (d/d_{o})^{–ρ}, where d is the distance in kilometer between transmitter and receiver, d_{ o } is the reference distance (equal to 1 km), β is the propagation constant (equal to (4πf_{ c }/(3 × 10^{8}))^{2} for BS–RS links and equal to 2πf_{ c }/(3 × 10^{8}) for RS–UE links) and ρ denotes the path loss propagation exponent which is also different for each link; assuming a LOS situation of RS with respect to BS: ρ_{BS–RS} = 2 (which then defines the free space propagation model), and ρ_{RS–UE} is an input variable for our simulations. The shadowing effect is assumed Gaussian in logarithmic scale with a standard deviation of 4 dB for BS–RS link and 8 dB for RS–UE link. Therefore, the channel from the l th transmit antenna of the k th transmitter and the m th receive antenna of the i th receiver is given by
where χ_{i,k} is the shadowing between the k th transmitter and the i th receiver, and δ_{i,k}^{m,l} ∊ CN(0, 1) represents the Rayleigh fading.
In this section, the following transmission strategies are evaluated:

Relayed networkMIMO optimal: relaybased transmission with cooperative BSs using networkMIMO and the optimal precoding strategy derived in WSRbased optimal resource allocation.

Relayed networkMIMO suboptimal: relaybased transmission with cooperative BSs using networkMIMO and the suboptimal precoding strategy described in WSRbased suboptimal resource allocation.

Relayed BDZFTDMA: relaybased transmission with noncooperative BSs, where one time slot is allocated per BS transmission and each BS serves simultaneously its associated relays following a multiuser MIMO BDZF strategy [17].

Relayed BFTDMA: relaybased transmission with noncooperative BSs, where each BS serves its associated relays under roundrobin TDMA.
These transmission strategies are evaluated for the SR over 1,000 random user deployments when no MCS constraints are taken into account.
Figure 3 displays S_{ e } versus the path loss propagation exponent of the RS–UE link (ρ_{RS–UE}). On the one side, significant gains are observed in terms of S_{ e } when using networkMIMO strategies in the BS–RS link as compared to the utilization of noncooperative BSbased strategies (BDZFTDMA or BFTDMA). On the other side, the performance gain achieved by the suboptimal MIMO symbol precoding is near to the gain obtained with the optimal precoding (moreover, the difference is depreciable when the system is limited by transmission in the RS–UE link), while the computational complexity is significantly increased when using the optimal precoding strategy. Regarding the performance in terms of propagation exponent of the second hop, ρ_{RS–UE}, the performance is nearly constant for low values of ρ_{RS–UE} because of the high level of received interference from nonserving RS. As the attenuation in the RS–UE link increases (i.e., higher values of ρ_{RS–UE}), the system performance is severely degraded.
Figure 4 displays S_{ e } versus the time devoted for BS–RS links (α_{1}) when ρ_{RS–UE} = 3. Relaybased transmission strategies are compared to the case where both precoders and durations for the relaytransmit and relayreceive phases are optimized on each deployment following a networkMIMO suboptimal precoding strategy and considering that interference is stationary during the transmission intervals, as it was done in [13]. However, let us remark that optimizing the duration of the phases on each realization is not possible in practice, since the interference between macrocells would be timevarying. In Figure 4, it can be observed that, for each transmission strategy, there exists an optimum value for the transmit duration phases so as to maximize the cellular spectral efficiency in average. These values of α_{1} are the ones used for the system evaluations, as it is explicitly detailed in Table 2. Furthermore, when comparing the ‘relayed networkMIMO optimal and suboptimal’ with the upper bound given by the joint optimization of the precoders and phase durations on each realization, it can be observed that the difference is not significant when properly selecting α_{1}.
Finally, in Figure 5, the capacity of individual BS–RS link is evaluated when the path loss propagation exponent of the BS–RS link (ρ_{BS–RS}) increases, without taking into account the constraints imposed by the RS–UE link capacity and the maximum MCS. We can observe a gain in the capacity of 300 and 600% due to the cooperation of BSs, when comparing networkMIMO schemes to BDZF TDMA and BFTDMA, respectively. In addition, comparing the optimal and suboptimal precoding structures, we can observe that the gain obtained by the suboptimal scheme is very close to that achieved with the optimal scheme. Results show that there is a constant gain irrespective of channel propagation conditions. This confirms the fact that when a relaybased transmission is limited by the capacity in the RS–UE link, the optimal precoding does not obtain a significant enhanced result compared to the suboptimal case.
To conclude, in all studies it is observed that the suboptimal precoding strategy for a relayed networkMIMO system nearly achieves the optimum solution. Thanks to the evaluation of the BS–RS capacity in Figure 5, it is verified that the similar performance between both solutions is due to the first hop precoding strategy itself and not due to the inclusion of RS–UE link constraints or maximum MCS constraints. However, we do not have a mathematical intuition to explain the similar performance between both precoding strategies, since it is very difficult to relate the eigenvalues of N_{ i }^{1/2}H_{ i }V_{ i } in (18) with the eigenvalues of N_{ i }^{1/2}H_{ i }V_{ i }(V_{ i }^{H}B_{ γ }V_{ i })^{1/2} in (14), whose values depend on the Lagrange multipliers associated to the power constraints {γ_{ k }} for which there is not an analytical expression.
4.2. Outdoor/indoor hotspot scenario
In a second deployment, a hotspot scenario is considered to evaluate a realistic and more complex deployment, where users are located in indoor/outdoor areas of high traffic demand. Each hotspot is modeled as two buildings (with 10 apartments per floor and several floors) and the streets between them, following the specifications of the dual stripe model for corporate urban deployments [22] (see minor details in Table 3). One relay is placed on each hotspot. See Figure 2right for detailed deployment of the RS and users in the macrocell. In addition, two tiers of surrounding macrocells are considered as interference for each macrocell, and the worstcase scenario is considered: interfering BSs transmit at their maximum allocated power.
Channel models adopted are obtained from [21]. We assume LOS conditions for all BS–RS links, distancedependent LOS/NLOS conditions for transmissions from BS and RS to outdoor UE and NLOS conditions with wall penetration losses for transmissions from BS and RS to indoor UE. See details of the models used in Table 3.
In this section, the following transmission strategies are evaluated:

Relayed networkMIMO optimal: relaybased transmission with cooperative BSs using networkMIMO and the optimal precoding strategy derived in WSRbased optimal resource allocation.

Relayed networkMIMO suboptimal: relaybased transmission with cooperative BSs using networkMIMO and the suboptimal precoding strategy described in WSRbased suboptimal resource allocation.

Relayed BDZFTDMA: relaybased transmission with noncooperative BSs, where one time slot is allocated per BS transmission and each BS serves simultaneously its associated relays following a multiuser MIMO BDZF strategy [17].

Direct networkMIMO optimal: nonrelayed transmission with cooperative BSs using networkMIMO and the optimal precoding strategy [12].

Direct networkMIMO suboptimal: nonrelayed transmission with cooperative BSs using networkMIMO and the suboptimal precoding strategy.
These transmission strategies are evaluated for the SR and for the WSR over 1,000 random user deployments when taking into account the MCS constraints (from LTEA specifications [3]: S_{max} = 4.8 bps/Hz). When using WSR, the weights are inversely proportional to the rates in the second hop (i.e., μ_{ i } = 1/r_{2i}), in order to avoid unfair service to deprived users.
Figure 6 displays r_{ out } versus S_{ e } for ISD of 500 and 1500 m, while Figure depicts the cumulative density function (CDF) of the user’s transmission rate for ISD equal to 500 m.
Figures 6 and 7 elucidate that deploying RS significantly improves the system coverage in a corporate hotspot scenario (for both indoor and outdoor users), as the outage rate is higher for both relayassisted transmissions (based on networkMIMO and BDZFTDMA) compared to the results achieved in a transmission without relays (direct networkMIMO). It is due to the fact that RS are nearer the UE and they are able to efficiently combat shadowing and path loss effects. Moreover, in [14] it is shown that the enhanced celledge spectral efficiency and coverage thanks to RS’s deployment is achieved with a lower transmitted power as compared to direct transmissions. Notice that the 95% outage rate is 0 in the direct transmission evaluated with networkMIMO and SR criteria, because there are a high number of users with bad propagation channel conditions. We observed the same results even assuming the WSR criterion.
When comparing the relayassisted transmissions in Figure 6 (‘relayed networkMIMO’ and ‘relayed BDZFTDMA’), networkMIMO achieves a nearly doubled cellular spectral efficiency. BDZFTDMA suffers some loses in spectral efficiency due to the time division associated to coordination. In addition, it achieves an improved celledge rate because the duration of the relayreceive and relaytransmit phases is optimized beforehand (see Table 2) on each transmission scheme to enhance the system spectral efficiency (and they differ from networkMIMO to BDZFTDMA).
If we compare the relayassisted transmissions with cooperation of BSs for BS–RS link (‘relayed networkMIMO’ in legend) when using the optimal and the suboptimal precoding structure, it can be observed that when using the optimal structure results in terms of spectral efficiency and outage rate are slightly enhanced. It has been observed that these results are achieved with a significantly increased computational complexity.
When comparing results for relayed networkMIMO with different object functions (‘SR’ or ‘WSR’ in Figure 6), it can be observed that WSR criterion performs better in terms of outage rate while it suffers some losses in terms of spectral efficiency, as it is more concerned about users with bad propagation channel conditions in the RS–UE link.
In Figure 7, we can observe the effect of the MCS constraints over the halfduplex relayassisted transmissions, which are applied to each link of the RSbased transmission. Hence, the maximum peruser transmission rate that can be achieved by a direct transmission is nearly doubled than for a relaybased transmission (even in the scenario deployed, it is achieved by less than 95% of the users).
To conclude, in a hotspot or corporate scenario, deploying relays using networkMIMO in the BS–RS link represents a competitive solution for enhanced services, since it allows improving cellular spectral efficiency and outage rate compared to a nonrelayed transmission with cooperating BSs or a relaybased transmission with no cooperating BSs.
5. Conclusions
This article derives the optimal MIMO symbol precoding matrix and power allocation for QoSconstrained relayassisted cellular systems, where cooperation between BS is appointed and the duration of the relaying phases is fixed beforehand. Results in terms of spectral efficiency and outage rate with the optimal MIMO symbol precoding matrix have been obtained and shown not to be significantly better than those obtained with the suboptimal precoding, which requires a lowcomplexity algorithm and lower computational time. Moreover, the suitability of applying networkMIMO in the BS–RS transmission has been proved in a corporate hotspot scenario. Enhanced results in cellular spectral efficiency and outage rate are obtained compared to nonrelayed transmission or relayed transmission with noncooperative BSs. Further work is oriented to study modulationconstrained resource allocation for the uplink, user’s grouping and scheduling strategies, and coordination of transmissions between cell clusters.
Appendix 1
The Lagrangian function of problem (P_{1}) in (13) can be written as
where {\widehat{L}}_{i} contains the terms which depend on {r_{1i}} and T_{ i }’s
where B_{ γ } = ∑_{k=1}^{B}γ_{ k }B_{ k }; γ_{ k }, ψ_{ i }, and φ_{ i } denote the Lagrange multipliers or dual variables associated to the perBS power constraint, the peruser transmission rate in the BS–RS link, and the peruser transmission rate, respectively. Finally, η_{ i } is the Lagrange multiplier needed for having positive values of transmission rates.
If we formulate the Lagrange dual function of problem (P_{1})
Formally, the dual function should be defined as the infimum of the Lagrangian function, but in our case the infimum coincides with the minimum since the infimum is contained in the convex set [19]. This way, we can define the dual problem of (P_{1}) as
Since (P_{DP}) is convex and satisfies the Slater’s condition [19], the duality gap between the optimal objective value of (P_{DP}) and (P_{1}) is zero. Then, (P_{1}) can be solved equivalently by solving (P_{DP}). Now, we focus on solving {r_{1i}} and T_{ i }’s for a set of fixed {γ_{ k }}, {ψ_{ i }}, {φ_{ i }}, and {η_{ i }}, as it is expressed in Equation (25). Notice that Equation (25) can be separated into R independent minimization subproblems (on for each UE, see (23)) and hence we focus on
If we define
the minimization problem in (27) can be reformulated as
On the one hand, the condition to minimize (29) as a function of r_{1i} is
On the other hand, the solution to minimize (29) as a function of {\widehat{\mathbf{T}}}_{i} can be found applying Hadamard’s inequality [23]. Since the logarithm is an increasing function, it is the same maximizing the logarithm than the values inside it. Hadamard’s inequality demonstrates that the optimal solution for those problems is such that diagonalizes the values inside the determinant of (27). Therefore, if we consider the following (reduced) SVD
where {\mathbf{\Lambda}}_{i}=\mathrm{diag}\left({\lambda}_{i1},\dots ,{\lambda}_{i{m}_{i}}\right) is an m_{ i } × m_{ i } diagonal matrix containing the positive eigenvalues and U_{ i } and W_{ i } are unitary matrix containing the m_{ i } left singular vectors and the m_{ i } right singular vectors, respectively. Then, the optimal solution for (27) for a set of fixed {γ_{ k }}, {ψ_{ i }}, {φ_{ i }}, and {η_{ i }} is
where {P}_{i}=\mathrm{diag}\left({p}_{i1},\dots ,{p}_{i{m}_{i}}\right) is a diagonal matrix, which diagonal values can be obtained from the standard waterfilling algorithm [23], when also considering ψ_{ i } = μ_{ i } – φ_{ i } from (30)
It can be verified that (32) satisfies the channel diagonalization property. Therefore, denoted {γ_{ k }} and {φ_{ i }} a set of nonnegative dual variables associated to the k th perBS power constraint and the i th peruser transmission rate, respectively, the optimal solution to (27) for fixed {γ_{ k }} and {φ_{ i }} is given by
Once we have the optimal solution for T_{ i }’s and {r_{1i}} for a set of fixed {γ_{ k }} and {φ_{ i }}, we have to optimize {γ_{ k }} and {φ_{ i }} to obtain the optimal solution to (P_{DP}).
On the one hand, we can apply the Karush–Kuhn–Tucker conditions [19] to obtain the optimal {φ_{ i }} for each set of fixed {γ_{ k }}. Specifically, the Slater’s condition [19] over the peruser transmission rate specifies
So, if the resource allocation for the i th user is nonlimited by the peruser transmission rate (i.e., r_{1i} < min{α_{2}r_{2i}/α 1, S_{max}m_{ i }) then φ_{ i } = 0. Otherwise, if the resource allocation for the i th user is limited by the peruser transmission rate, we need to impose r_{1i} = min{α_{2}r_{2i}/α 1, S_{max}m_{ i }} and φ_{ i } can be obtained from the imposition when substituting (33) into r_{1i} following (34). Then, the optimal value for {φ_{ i }} for a set of fixed {γ_{ k }} can be expressed as
Notice that the φ_{ i }^{*} depends on the number of transmission modes per user (m_{ i }) and, in addition, we need to evaluate r_{1i} with Equation (34) which involves a waterfilling algorithm for the power allocation. For that reason, the optimal value for φ_{ i } cannot directly be found. In this regard, an algorithm to find φ_{ i }^{*}, which is based on hypothetic testing is presented in Algorithm 1. The underlying idea is to form a hypothesis of the active and inactive transmission modes and check whether a consistent solution can be found or not. In the Algorithm 1, initially it is assumed that all transmissions modes are active (and hence m_{ i } = rank (H_{ i }V_{ i }) i = 1,…,R); then, φ_{ i }^{*} is computed following (36) and the power allocation is calculated with (33); in the case that the resulting active transmission modes due to the power allocation are lower than those initially supposed, we have to reduce the number of active transmission modes and make a new hypothesis; until consistence.
On the other hand, problem (P_{DP}) has to be optimized in terms of {γ_{ k }}. The optimum values for {γ_{ k }} are calculated using an iterative algorithm in Algorithm 2, which is based on the ellipsoid method [19]. The subgradients required to update them are
Finally, notice that once we have the optimal solution for {γ_{ k }}, we can compute the optimal solution for {φ_{ i }} following (36), and hence we can directly obtain the optimal solution to the problems (P_{DP}) and (P_{1}). In this respect, an effective algorithm is proposed in Algorithm 2 to solve the resource allocation on the BS–RS link subject to perBS power constraints, taking into account the MCS constraints and the constraints imposed by the RS–UE link. In the algorithm, {γ_{ k }} are updated following the ellipsoid method [19], and for each set of {γ_{ k }}, the optimal solution for T_{ i }’s, {r_{1i}} and φ_{ i }^{*} are computed following (34) and (36).
References
Bhat P, Nagata S, Campoy L, Berberana I, Derham T, Liu G, Shen X, Zong P, Yang J: LTEadvanced: an operator perspective. IEEE Commun. Mag. 2012, 50(2):104114.
EEE 802.16 Broadband Wireless Access Group: IEEE 802, 16 m system requirements, IEEE 802.16 m07/002r4. 2007.
3GPP TR 36.912: LTE, Feasibility study for Further Advancements for EULTRA (LTEAdvanced). 2010.
Artist4G Consortium: Advanced Relay Technical Proposals, Deliverable D.3.2, ICT247223 ARTIST4G. 2011.
Hoymann C, Chen W, Montojo J, Golitschek A, Koutsimanis C, Shen X: Relaying operation in 3GPP LTE: challenges and solutions. IEEE Commun. Mag. 2012, 50(2):156162.
HostMadsen A, Zhang J: Capacity bounds and power allocation for wireless relay channels. IEEE Trans. Inf. Theory 2005, 51(6):20202040. 10.1109/TIT.2005.847703
Agustin A, Vidal J, Munoz O: Protocols and resource allocation for the twoway relay channel with halfduplex terminals. In IEEE International Conference on Communications (ICC). Dresden; 14–18 June 2009:15.
Blum RS: MIMO capacity with interference. IEEE J. Sel. Areas Commun 2003, 21(5):793801. 10.1109/JSAC.2003.810345
Takahashi H, Ofuji Y, Fukumoto S, Abeta S, Nakamura T: Field experimental results on EUTRA downlink throughput using polarization diversity antennas. In IEEE Vehicular Technology Conference (VTC). Barcelona; 26–29 April 2009:16.
Karakayali MK, Foschini GJ, Valenzuela RA: Network coordination for spectrally efficient communications in cellular systems. IEEE Wireless Commun. 2006, 13(4):5661. 10.1109/MWC.2006.1678166
Lee D, Seo H, Clerckx B, Hardouin E, Mazzarese D, Nagata S, Sayana K: Coordinated multipoint transmission and reception in LTEadvanced: deployment scenarios and operational challenges. IEEE Commun. Mag. 2012, 50(2):148155.
Zhang R: Cooperative multicell block diagonalization with perbasestation power constraints. IEEE J. Sel. Areas Commun 2010, 28(9):14351445.
Vidal J, Agustin A, Lagen S, Valera E, Muñoz O, Garcia A, Sanchez M: NetworkMIMO backhauling for QoSconstrained relay transmission. In IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Prague; 2227 –May 2011:33323335.
Agustin A, Vidal J, Lagen S, Valera E: Network MIMO for downlink inband relay transmission with relaying phases of fixed duration. In 19th European Conference on Signal Processing (EUSIPCO). Barcelona; 29 August – 2 September 2011:15.
Zhang J, Chen R, Andrews J, Ghosh RW A: Heath, networked MIMO with clustered linear precoding. IEEE Trans. Wirel. Commun 2009, 8(4):19101921.
Kramer G, Gastpar M, Gupta P: Cooperative strategies and capacity theorems for relay networks. IEEE Trans. Inf. Theory 2005, 51(9):30373063. 10.1109/TIT.2005.853304
Spencer QH, Swindlehurst AL, Haardt M: Zeroforcing methods for downlink spatial multiplexing in multiuser MIMO channels. IEEE Trans. Signal Process. 2004, 52(2):461471. 10.1109/TSP.2003.821107
Shi S, Schubert M, Vucic N, Boche H: MMSE optimization with perbasestation power constraints for network MIMO systems. In IEEE International Conference on Communications (ICC). Beijing; 1923 –May 2008:41064110.
Boyd S, Vandenberghe L: Convex Optimization. Cambridge, MA: Cambridge University Press; 2004.
Yu W, Lui R: Dual methods for nonconvex spectrum optimization of multicarrier systems. IEEE Trans. Commun. 2006, 54(7):13101322.
WINNER II consortium: Channel Models Part II, Radio Channel Measurements and Analysis Results. 2007. Deliverable 1.1.2, IST4027756 WINNER II
Femto Forum: Interference management in OFDMA femtocells. 2010.
Cover T, Thomas J: Elements of Information Theory . New York: Wiley; 1991.
Acknowledgments
This study was done in the framework of the projects TROPIC FP7 ICT20118318784 and COST Action IC0902, funded by the European Community. Also, by the Spanish Science and Technology Commissions and EC FEDER funds through projects: TEC201019171/TCM and CONSOLIDER INGENIO CSD200800010 COMONSENS, and by project 2009SGR1236 (AGAUR) of the Catalan Administration.
Author information
Authors and Affiliations
Corresponding author
Additional information
Competing interests
The authors declare that they have no competing interests.
Authors’ original submitted files for images
Below are the links to the authors’ original submitted files for images.
Rights and permissions
Open Access This article is distributed under the terms of the Creative Commons Attribution 2.0 International License (https://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
About this article
Cite this article
Lagen, S., Agustin, A. & Vidal, J. NetworkMIMO for downlink inband relay transmissions. J Wireless Com Network 2013, 13 (2013). https://doi.org/10.1186/16871499201313
Received:
Accepted:
Published:
DOI: https://doi.org/10.1186/16871499201313
Keywords
 NetworkMIMO
 Relay transmissions
 Blockdiagonalization zeroforcing
 Convex optimization