A framework for globally optimal energy-efficient resource allocation in wireless networks

Zappone, Alessio; Björnson, Emil; Sanguinetti, Luca; Jorswieck, Eduard A.

doi:10.1109/icassp.2016.7472351

Cited by 13 publications

(26 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First, we convert problem (54) into a monotonic optimization problem in canonical form, and then, we utilize the polyblock algorithm, to globally solve it. In this regard, the optimization problem (54) can be reformulated as

\begin{matrix} alignleft \\ align-1 & align-2 \max_{P} q^{+} (P) - q^{-} (P) \\ align-1 & align-2 s . t . P \in P, (54 d), (54 c), \end{matrix}

where

q^{+} false(boldP false) = \sum_{f \in scriptF} \sum_{m \in {scriptM}_{f}} \sum_{n \in scriptN} q_{m, f}^{n +} false(boldP false)

and

q^{-} false(boldP false) = \sum_{f \in scriptF} \sum_{m \in {scriptM}_{f}} \sum_{n \in scriptN} q_{m, f}^{n -} false(boldP false)

.…”

Section: Optimal Solutionmentioning

confidence: 99%

“…Moreover, q + ( P ) and q − ( P ) are two increasing functions with respect to P . The optimization problem is not monotonic yet, since the difference of two increasing functions, q + ( P ) − q − ( P ), is not monotonic; for more details, please refer to definition 2 in the works of Zappone et al Let

{boldP}^{\max} = false{p_{m, f}^{false(n false), \max} false} \forall f \in scriptF, n \in scriptN, m \in {scriptM}_{f}

and Λ = q − ( P max ) − q − ( P ) as an auxiliary variable. Then, the optimization problem can be reformulated as follows:

\begin{matrix} alignleft \\ align-1 & align-2 \max_{(Λ, P)} q^{+} (P) + Λ \\ align-1 & align-2 s . t . (Λ, P) \in Q, (54 d), (54 c), \end{matrix}

where

scriptQ = ({}, false(normalΛ, boldP false) false| boldP \in scriptP, 0 \leq normalΛ + q^{-} false(boldP false) \leq q^{-} {false(boldP false)}^{\max}, 0 \leq normalΛ \leq q^{-} {false(boldP false)}^{\max} - q^{-} false(bold0 false)) .

…”

Section: Optimal Solutionmentioning

confidence: 99%

“…The optimization problem is a monotonic problem in canonical form, because it fulfills definition 5 in the works of Zappone et al The feasible set of the optimization problem can be described as the intersection of the following two sets:

\begin{matrix} alignleft \\ align-1 Υ_{1} = & align-2 \{(Λ, Λ^{'}, Λ^{''}, P) : P ⪯̸ P^{\max}, (53), (54 b), Λ + q^{-} (P) \leq q^{-} {(P)}^{\max}, C_{1}^{+} (P) + Λ_{m, f}^{^{'} n} \leq C_{1}^{+} {(P)}^{\max}, \\ align-1 & align-2 \forall f \in F, n \in N, m \in M_{f}, C_{2}^{+} (P) + Λ_{m, i, f}^{'' n} \leq C_{2}^{+} {(P)}^{\max}, \forall f \in F, n \in N, m \in M_{f}, i \in M_{f}, i < m\} \end{matrix}

and

\begin{matrix} alignleft \\ align-1 Υ \end{matrix}

…”

Section: Optimal Solutionmentioning

confidence: 99%

“…Let  = {P ∈ R MN + ; (12c), (53), (54b)} and r n m, (P) = q n+ m, (P) − q n− m, (P), where q n− m, (P) = log 2 ( n m, + I n m, ) and q n+ m, (P) = log 2 ( n m, + I n m, + p n m, h n m, ). First, we convert problem (54) into a monotonic optimization problem in canonical form, and then, we utilize the polyblock algorithm 54,55 to globally solve it. In this regard, the optimization problem (54) can be reformulated as…”

Section: Optimal Solutionmentioning

confidence: 99%

“…where q + (P) = ∑ ∈F ∑ m∈ ∑ n∈ q n+ m, (P) and q − (P) = ∑ ∈F ∑ m∈ ∑ n∈ q n− m, (P). Moreover, q + (P) and q − (P) are two increasing functions with respect to P. The optimization problem (55) is not monotonic yet, since the difference of two increasing functions, q + (P) − q − (P), is not monotonic; for more details, please refer to definition 2 in the works of Zappone et al 54,55 Let P max = {p (n),max…”

Section: Optimal Solutionmentioning

confidence: 99%

See 4 more Smart Citations

Optimal radio resource allocation to achieve a low BER in PD‐NOMA–based heterogeneous cellular networks

Moltafet

Azmi

Javan

et al. 2019

Trans Emerging Tel Tech

View full text Add to dashboard Cite

In this paper, the problem of downlink radio resource allocation for a power‐domain nonorthogonal multiple‐access–based heterogeneous cellular network is investigated. We propose two scenarios: (1) resource allocation with exclusive subcarrier allocation and (2) resource allocation considering intercell interference. The aim of this paper is to obtain joint subcarrier and power allocation for the two considered system models by addressing the maximization of the users' sum rate subject to the bit error rate constraint, the successive interference cancellation constraint, subcarrier constraints, and the transmit power restriction of each base station. To solve the corresponding optimization problems, an alternating method is applied by decoupling the proposed problems into two subproblems, namely, subcarrier allocation problem and power allocation problem, where they are solved by using the mesh adaptive direct search algorithm and different successive convex approximation methods, respectively. Moreover, to evaluate the optimality gap of the alternating method, the proposed problems are solved using an optimal method, namely, monotonic optimization with high complexity as a benchmark. As the results show, with the bit error rate constraint, the link‐level performance, at the cost of a decreasing system sum rate, is improved.

show abstract

\begin{matrix} alignleft \\ align-1 & align-2 \max_{P} q^{+} (P) - q^{-} (P) \\ align-1 & align-2 s . t . P \in P, (54 d), (54 c), \end{matrix}

where

q^{+} false(boldP false) = \sum_{f \in scriptF} \sum_{m \in {scriptM}_{f}} \sum_{n \in scriptN} q_{m, f}^{n +} false(boldP false)

and

q^{-} false(boldP false) = \sum_{f \in scriptF} \sum_{m \in {scriptM}_{f}} \sum_{n \in scriptN} q_{m, f}^{n -} false(boldP false)

.…”

Section: Optimal Solutionmentioning

confidence: 99%

{boldP}^{\max} = false{p_{m, f}^{false(n false), \max} false} \forall f \in scriptF, n \in scriptN, m \in {scriptM}_{f}

and Λ = q − ( P max ) − q − ( P ) as an auxiliary variable. Then, the optimization problem can be reformulated as follows:

\begin{matrix} alignleft \\ align-1 & align-2 \max_{(Λ, P)} q^{+} (P) + Λ \\ align-1 & align-2 s . t . (Λ, P) \in Q, (54 d), (54 c), \end{matrix}

where

scriptQ = ({}, false(normalΛ, boldP false) false| boldP \in scriptP, 0 \leq normalΛ + q^{-} false(boldP false) \leq q^{-} {false(boldP false)}^{\max}, 0 \leq normalΛ \leq q^{-} {false(boldP false)}^{\max} - q^{-} false(bold0 false)) .

…”

Section: Optimal Solutionmentioning

confidence: 99%

\begin{matrix} alignleft \\ align-1 Υ_{1} = & align-2 \{(Λ, Λ^{'}, Λ^{''}, P) : P ⪯̸ P^{\max}, (53), (54 b), Λ + q^{-} (P) \leq q^{-} {(P)}^{\max}, C_{1}^{+} (P) + Λ_{m, f}^{^{'} n} \leq C_{1}^{+} {(P)}^{\max}, \\ align-1 & align-2 \forall f \in F, n \in N, m \in M_{f}, C_{2}^{+} (P) + Λ_{m, i, f}^{'' n} \leq C_{2}^{+} {(P)}^{\max}, \forall f \in F, n \in N, m \in M_{f}, i \in M_{f}, i < m\} \end{matrix}

and

\begin{matrix} alignleft \\ align-1 Υ \end{matrix}

…”

Section: Optimal Solutionmentioning

confidence: 99%

Section: Optimal Solutionmentioning

confidence: 99%

Section: Optimal Solutionmentioning

confidence: 99%

See 3 more Smart Citations

Optimal radio resource allocation to achieve a low BER in PD‐NOMA–based heterogeneous cellular networks

Moltafet

Azmi

Javan

et al. 2019

Trans Emerging Tel Tech

View full text Add to dashboard Cite

show abstract

Energy‐efficient radio resource allocation for dual‐hop relay‐assisted orthogonal frequency division multiple access systems with quality of service provisioning

Saraiva

Lima

Pessoa

et al. 2021

Trans Emerging Tel Tech

View full text Add to dashboard Cite

Besides the conventional spectral efficiency metric, energy efficiency (EE) has been adopted as one of the new mandatory performance metrics for green future mobile networks. In this sense, in this article we study energy-efficient radio resource allocation for dual-hop orthogonal frequency division multiple access relay networks considering multiple relays and users. Resource allocation in our scenario is comprised of relay selection, subcarrier pairing, subcarrier assignment, and transmit power allocation. Particularly, in the context of EE with quality of service constraints, three different problems, namely, power consumption minimization, global EE maximization, and minimum individual EE maximization are addressed. For tackling them, first, we demonstrate a property that exploits the use of the decode and forward protocol employed in the relays and we show how it can be applied to simplify the investigated problems. Although some of these problems are nonconvex due to the fractional objective functions and integer optimization domain, optimal solutions are provided using generalized fractional programming theory to transform the fractional objective functions to the subtractive forms. Then, by employing Dinkelbach or bisection methods, we propose iterative algorithms where a parametric problem is solved in each iteration until reaching convergence. By introducing a penalty function to handle integer variables and applying majorization-minimization approach, we make the parametric problem convex and we propose suboptimal iterative solutions with reduced computational cost. Regarding obtained simulation results, we discuss the performance of the involved solutions in terms of global EE and fairness with respect to different parameters of the system.

show abstract

Energy-Efficient Power Control and Clustering in Underlay Device to Multi-device Communications

Hmila

Fernández‐Veiga

2017

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Underlay device to device multicast communication brings great benefits to cellular networks in term of energy and spectral efficiency, since a group of users can communicate directly by reusing cellular resources. In this paper, we propose a scheme to maximize network energy efficiency (EE) through transmission power control. In addition, to get a compromise between fairness and the overall EE, we analyse a second scheme with fairness factors to maximize minimum individual EE. The problem is formulated as a fractional programming optimization and solved with iterative algorithm. Two different EE metrics are analysed via extensive numerical simulations with a spatial Poisson process for the users' locations, and applying different clustering algorithms as K-nearest neighbour, distance limit and DBSCAN.

show abstract

A framework for globally optimal energy-efficient resource allocation in wireless networks

Cited by 13 publications

References 29 publications

Optimal radio resource allocation to achieve a low BER in PD‐NOMA–based heterogeneous cellular networks

Optimal radio resource allocation to achieve a low BER in PD‐NOMA–based heterogeneous cellular networks

Energy‐efficient radio resource allocation for dual‐hop relay‐assisted orthogonal frequency division multiple access systems with quality of service provisioning

Energy-Efficient Power Control and Clustering in Underlay Device to Multi-device Communications

Contact Info

Product

Resources

About