# Improving Energy Efficiency Fairness of Wireless Networks: A Deep Learning Approach

^{1}

^{2}

^{3}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. System Model

## 3. Proposed DL-Based Power Control Scheme

#### 3.1. DL Basics

#### 3.2. DL Approach for EE Fairness Maximization

**Theorem**

**1.**

#### 3.3. Training and Implementation

Algorithm 1 Mini-batch SGD training algorithm. |

Construct the training set $\mathcal{H}$. Initialize $t=0$, the DNN parameter ${\mathsf{\Theta}}^{\left[0\right]}$, and the number of iterations T. for $t=1:T$ doSample a mini-batch channel set $\mathcal{S}\subset \mathcal{H}$. Update ${\mathsf{\Theta}}^{\left[t\right]}\leftarrow {\mathsf{\Theta}}^{[t-1]}+\alpha \frac{1}{S}{\sum}_{\mathbf{H}\in \mathcal{S}}{\nabla}_{\mathsf{\Theta}}{\eta}_{\widehat{i}}^{\mathrm{DNN}}(\mathbf{H};{\mathsf{\Theta}}^{[t-1]})$. end forObtain the trained DNN parameter ${\mathsf{\Theta}}^{\left[T\right]}$. |

## 4. Numerical Results

#### 4.1. DNN Validation

- SCA [3]: The SCA-based iterative optimization algorithm in [3] was implemented with the CVX and MOSEK solver.
- Maximum power: The transmit power was set to it maximum value as ${p}_{i}\left(\mathbf{H}\right)=P$, $\forall i\in \mathcal{K}$.
- Random power: A random transmission policy was employed where ${p}_{i}\left(\mathbf{H}\right)$ is uniformly distributed over $[0,P]$.

#### 4.2. Performance Comparison

## 5. Conclusions

## Author Contributions

## Funding

## Acknowledgments

## Conflicts of Interest

## References

- Isheden, C.; Chong, Z.; Jorswieck, E.; Fettweis, G. Framework for link-level energy efficiency optimization with informed transmitter. IEEE Trans. Wirel. Commun.
**2012**, 11, 2946–2957. [Google Scholar] [CrossRef] - Zappone, A.; Jorswieck, E. Energy efficiency in wireless networks via fractional programming theory. Found. Trends Commun. Inf. Theory
**2015**, 11, 185–396. [Google Scholar] [CrossRef] - Nguyen, K.-G.; Vu, Q.-D.; Juntti, M.; Tran, L.-N. Distributed solutions for energy efficiency fairness in multicell MISO downlink. IEEE Trans. Wirel. Commun.
**2017**, 16, 6232–6247. [Google Scholar] [CrossRef] - Xu, Z.; Yang, C.; Li, G.Y.; Liu, Y.; Xu, S. Energy-efficient CoMP precoding in heterogeneous networks. IEEE Trans. Signal Process.
**2014**, 64, 1005–1017. [Google Scholar] [CrossRef] - He, S.; Huang, Y.; Jin, S.; Yang, L. Coordinated beamforming for energy efficient transmission in multicell multiuser systems. IEEE Trans. Commun.
**2013**, 61, 4961–4971. [Google Scholar] [CrossRef] - He, S.; Huang, Y.; Yang, L.; Ottersten, B. Coordinated multicell multiuser precoding for maximizing weighted sum energy efficiency. IEEE Trans. Signal Process.
**2014**, 62, 741–751. [Google Scholar] [CrossRef] - Tervo, O.; Tran, L.-N.; Juntti, M. Optimal energy-efficient transmit beamforming for multi-user MISO downlink. IEEE Trans. Signal Process.
**2015**, 63, 5574–5588. [Google Scholar] [CrossRef] - He, S.; Huang, Y.; Jin, S.; Yu, F.; Yang, L. Max-min energy efficient beamforming for multicell multiuser joint transmission systems. IEEE Commun. Lett.
**2013**, 17, 1956–1959. [Google Scholar] [CrossRef] - Du, B.; Pan, C.; Zhang, W.; Chen, M. Distributed energy-efficient power optimization for CoMP systems with max-min fairness. IEEE Commun. Lett.
**2014**, 18, 999–1002. [Google Scholar] [CrossRef] - Nguyen, K.-G.; Tran, L.-N.; Tervo, O.; Vu, Q.-D.; Juntti, M. Achieving energy efficiency fairness in multicell MISO downlink. IEEE Commun. Lett.
**2015**, 19, 1426–1429. [Google Scholar] [CrossRef] - Goodfellow, I.; Bengio, Y.; Courville, A. Deep Learning; MIT Press: Cambridge, MA, USA, 2016. [Google Scholar]
- LeCun, Y.; Yoshua, B.; Hinton, G. Deep learning. Nature
**2015**, 521, 436–444. [Google Scholar] [CrossRef] - O’Shea, T.; Hoydis, J. An introduction to deep learning for the physical layer. IEEE Trans. Cog. Commun. Netw.
**2017**, 3, 563–575. [Google Scholar] [CrossRef] - Dörner, S.; Cammerer, S.; Hoydis, J.; Brink, S. Deep learning-based communication over the air. IEEE J. Sel. Top. Signal Process.
**2018**, 12, 132–143. [Google Scholar] [CrossRef] - Lee, H.; Lee, I.; Lee, S.H. Deep learning based transceiver design for multi-colored VLC systems. Opt. Express
**2018**, 26, 6222–6238. [Google Scholar] [CrossRef] [PubMed] - Lee, H.; Lee, I.; Quek, T.Q.S.; Lee, S.H. Binary signaling design for visible light communication: A deep learning framework. Opt. Express
**2018**, 26, 18131–18142. [Google Scholar] [CrossRef] - Lee, H.; Lee, S.H.; Quek, T.Q.S.; Lee, I. Deep learning framework for wireless systems: Applications to optical wireless communications. IEEE Commun. Mag.
**2019**, 57, 35–41. [Google Scholar] [CrossRef] - Sun, H.; Chen, X.; Shi, Q.; Hong, M.; Fu, X.; Sidiropoulos, N.D. Learning to optimize: Training deep neural networks for interference management. IEEE Trans. Signal Process.
**2018**, 66, 5438–5453. [Google Scholar] [CrossRef] - Lee, W.; Cho, D.-H.; Kim, M. Deep power control: Transmit power control scheme based on convolutional neural network. IEEE Commun. Lett.
**2018**, 22, 1276–1279. [Google Scholar] [CrossRef] - Lee, H.; Lee, S.H.; Quek, T.Q.S. Constrained deep learning for wireless resource management. In Proceedings of the ICC 2019—2019 IEEE International Conference on Communications (ICC), Shanghai, China, 20–24 May 2019; pp. 1–6. [Google Scholar]
- Lee, W. Resource allocation for multi-channel underlay cognitive radio network based on deep neural network. IEEE Commun. Lett.
**2018**, 22, 1942–1945. [Google Scholar] [CrossRef] - Lee, W.; Kim, M.; Cho, D.-H. Transmit power control using deep neural network for underlay device-to-device communication. IEEE Wirel. Commun. Lett.
**2019**, 8, 141–144. [Google Scholar] [CrossRef] - Lee, W.; Kim, M.; Cho, D.-H. Deep learning based transmit power control in underlaid device-to-device communication. IEEE Sys. J.
**2019**, 13, 2551–2554. [Google Scholar] [CrossRef] - Boyd, S.; Vandenberghe, L. Convex Optimization; Cambridge University Press: Cambridge, UK, 2004. [Google Scholar]
- Hong, M.; Razaviyayn, M.; Luo, Z.-Q.; Pang, J.-S. A unifined algorithmic framework for block-structured optimization involving big data: With applications in machine learning and signal processing. IEEE Signal Process. Mag.
**2016**, 33, 57–77. [Google Scholar] [CrossRef] - Kingma, D.; Ba, J. Adam: A method for stochastic optimization. In Proceedings of the the 3rd International Conference for Learning Representations. Represent. (ICLR), San Diego, CA, USA, 7–9 May 2015. [Google Scholar]
- Hornik, K.; Stinchcombe, M.; White, H. Multilayer feedforward networks are universal approximators. Neural Netw.
**1989**, 2, 359–366. [Google Scholar] [CrossRef] - Lu, Z.; Pu, H.; Wang, F.; Hu, Z.; Wang, L. The expressive power of neural networks: A view from the width. In Proceedings of the Advances in Neural Information Processing Systems 30 (NIPS), Long Beach, CA, USA, 4–9 December 2017; pp. 6231–6239. [Google Scholar]
- Boyd, S.; Xiao, L.; Mutapcic, A. Subgradient Methods; Technical Report EE364b; Stanford Univ: Stanford, CA, USA, 2014. [Google Scholar]
- Glorot, X.; Bengio, Y. Understanding the difficulty of training deep feedforward neural networks. In Proceedings of the International Conference on Artificial Intelligence and Statistics, Sardinia, Italy, 13–15 May 2010; pp. 249–256. [Google Scholar]
- Ioffe, S.; Szegedy, C. Batch normalization: Accelerating deep network training by reducing internal covariance shift. In Proceedings of the International Conference on Machine Learning, Lille, France, 6–11 July 2015; pp. 448–456. [Google Scholar]

**Figure 4.**Convergence of training and validation performance with $K=2$ and ${P}_{C}=33$ dBm for different P.

**Figure 5.**Convergence of validation performance with $K=4$, ${P}_{C}=33$ dBm, and $P=30$ dBm for different L.

**Figure 6.**Convergence of validation performance with $K=4$, ${P}_{C}=33$ dBm, and $P=30$ dBm for different S.

SCA [3] | DNN | ||
---|---|---|---|

$\mathit{P}=10$ dBm | $\mathit{P}=20$ dBm | ||

$K=2$ | 0.79 | 1.67 | 6.00 × 10${}^{-4}$ |

$K=3$ | 0.97 | 2.64 | 9.85 × 10${}^{-4}$ |

$K=4$ | 1.10 | 3.13 | 1.24 × 10${}^{-3}$ |

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Lee, H.; Jang, H.S.; Jung, B.C.
Improving Energy Efficiency Fairness of Wireless Networks: A Deep Learning Approach. *Energies* **2019**, *12*, 4300.
https://doi.org/10.3390/en12224300

**AMA Style**

Lee H, Jang HS, Jung BC.
Improving Energy Efficiency Fairness of Wireless Networks: A Deep Learning Approach. *Energies*. 2019; 12(22):4300.
https://doi.org/10.3390/en12224300

**Chicago/Turabian Style**

Lee, Hoon, Han Seung Jang, and Bang Chul Jung.
2019. "Improving Energy Efficiency Fairness of Wireless Networks: A Deep Learning Approach" *Energies* 12, no. 22: 4300.
https://doi.org/10.3390/en12224300