# Channel Estimation for UAV Communication Systems Using Deep Neural Networks

^{1}

^{2}

^{3}

^{4}

^{*}

^{†}

## Abstract

**:**

## 1. Introduction

- Apply GBRBM and real measurements to estimate the received signal power at UAV from the cellular network during the flight, where GBRB machines and deep neural network are employed to extract features from the UAV channels as a series set of blocks for channel modeling.
- Develop an adaptive learning rate approach and a new enhanced gradient to improve the training performance. Specifically, an autoencoder is used to fine-tune the parameter, while the parameters are trained by using an encoder neural network to model the RSS for the prediction of the ground at different heights.
- Verify the effectiveness of the proposed method throughout experimental measurements and comparisons with other benchmark schemes. The numeral results show that the proposed scheme outperforms the conventional autoencoders.

## 2. System Model and Problem Formulation

#### 2.1. Problem Formulation

#### 2.1.1. Adaptive Learning Rate

#### 2.1.2. Enhanced Gradient

Algorithm 1: Pre-training the GBRBM blocks (Unsupervised learning) |

Algorithm 2: Fine tuning of autoencoder (Unsupervised Learning) |

Algorithm 3: Fine tuning of the data labels (Supervised Learning) |

## 3. UAV Measurements

## 4. Performance Evaluation

#### 4.1. GBRBM Blocks

#### 4.2. Adaptive Learning Rate

## 5. Conclusions

## Author Contributions

## Funding

## Institutional Review Board Statement

## Informed Consent Statement

## Data Availability Statement

## Conflicts of Interest

## References

- Zhao, N.; Lu, W.; Sheng, M.; Chen, Y.; Tang, J.; Yu, F.R.; Wong, K.K. UAV-Assisted Emergency Networks in Disasters. IEEE Wirel. Commun.
**2019**, 26, 45–51. [Google Scholar] [CrossRef][Green Version] - Alladi, T.; Naren; Bansal, G.; Chamola, V.; Guizani, M. SecAuthUAV: A Novel Authentication Scheme for UAV-Ground Station and UAV-UAV Communication. IEEE Trans. Veh. Technol.
**2020**, 69, 15068–15077. [Google Scholar] [CrossRef] - Abdalla, A.S.; Marojevic, V. Communications Standards for Unmanned Aircraft Systems: The 3GPP Perspective and Research Drivers. IEEE Commun. Stand. Mag.
**2021**, 5, 70–77. [Google Scholar] [CrossRef] - Zeng, Y.; Wu, Q.; Zhang, R. Accessing From the Sky: A Tutorial on UAV Communications for 5G and Beyond. Proc. IEEE
**2019**, 107, 2327–2375. [Google Scholar] [CrossRef][Green Version] - Khawaja, W.; Guvenc, I.; Matolak, D.W.; Fiebig, U.C.; Schneckenburger, N. A Survey of Air-to-Ground Propagation Channel Modeling for Unmanned Aerial Vehicles. IEEE Commun. Surv. Tuts.
**2019**, 21, 2361–2391. [Google Scholar] [CrossRef][Green Version] - Abdullah, O.A.; Al-Hraishawi, H.; Chatzinotas, S. Deep Learning-Based Device-Free Localization in Wireless Sensor Networks. arXiv
**2022**, arXiv:2206.08191. [Google Scholar] - Sun, L.; Wang, Y.; Swindlehurst, A.L.; Tang, X. Generative-Adversarial-Network Enabled Signal Detection for Communication Systems With Unknown Channel Models. IEEE J. Sel. Areas Commun.
**2021**, 39, 47–60. [Google Scholar] [CrossRef] - Ma, X.; Gao, Z. Data-Driven Deep Learning to Design Pilot and Channel Estimator for Massive MIMO. IEEE Trans. Veh. Technol.
**2020**, 69, 5677–5682. [Google Scholar] [CrossRef][Green Version] - Shlezinger, N.; Farsad, N.; Eldar, Y.C.; Goldsmith, A.J. Model-Based Machine Learning for Communications. arXiv
**2021**, arXiv:2101.04726. [Google Scholar] - Ozpoyraz, B.; Dogukan, A.T.; Gevez, Y.; Altun, U.; Basar, E. Deep Learning-Aided 6G Wireless Networks: A Comprehensive Survey of Revolutionary PHY Architectures. arXiv
**2022**, arXiv:2201.03866. [Google Scholar] [CrossRef] - Karra, D.; Goudos, S.K.; Tsoulos, G.V.; Athanasiadou, G. Prediction of Received Signal Power in Mobile Communications Using Different Machine Learning Algorithms:A Comparative Study. In Proceedings of the Panhellenic Conf. on Electronics Telecommunications (PACET), Volos, Greece, 8–9 November 2019; pp. 1–4. [Google Scholar]
- Cao, Y.; Zhang, L.; Liang, Y.C. Deep Reinforcement Learning for Channel and Power Allocation in UAV-enabled IoT Systems. In Proceedings of the IEEE Global Commun. Conf. (GLOBECOM), Waikoloa, HI, USA, 9–13 December 2019; pp. 1–6. [Google Scholar]
- Luo, X.; Zhang, Y.; He, Z.; Yang, G.; Ji, Z. A Two-Step Environment-Learning-Based Method for Optimal UAV Deployment. IEEE Access
**2019**, 7, 149328–149340. [Google Scholar] [CrossRef] - Goudos, S.K.; Athanasiadou, G. Application of an Ensemble Method to UAV Power Modeling for Cellular Communications. IEEE Antennas Wirel. Propag. Lett.
**2019**, 18, 2340–2344. [Google Scholar] [CrossRef] - Wang, X.; Gao, L.; Mao, S.; Pandey, S. CSI-Based Fingerprinting for Indoor Localization: A Deep Learning Approach. IEEE Trans. Veh. Technol.
**2017**, 66, 763–776. [Google Scholar] [CrossRef][Green Version] - Choo, S.; Lee, H. Learning framework of multimodal Gaussian–Bernoulli RBM handling real-value input data. Neurocomputing
**2018**, 275, 1813–1822. [Google Scholar] [CrossRef] - Koller, D.; Friedman, N. Probabilistic Graphical Models: Principles and Techniques–Adaptive Computation and Machine Learning; The MIT Press: Cambridge, MA, USA, 2009. [Google Scholar]
- Carreira-Perpiñán, M.A.; Hinton, G. On Contrastive Divergence Learning. In Proceedings of the Tenth International Workshop on Artificial Intelligence and Statistics, Bridgetown, Barbados, 6–8 January 2005; Volume R5, pp. 33–40. [Google Scholar]
- Le Roux, N.; Bengio, Y. Representational Power of Restricted Boltzmann Machines and Deep Belief Networks. Neural Comput.
**2008**, 20, 1631–1649. [Google Scholar] [CrossRef] [PubMed] - Cho, K.; Raiko, T.; Ilin, A. Enhanced Gradient for Training Restricted Boltzmann Machines. Neural Comput.
**2013**, 25, 805–831. [Google Scholar] [CrossRef] [PubMed] - Cho, K.H.; Raiko, T.; Ilin, A. Gaussian-Bernoulli deep Boltzmann machine. In Proceedings of the International Joint Conference on Neural Networks (IJCNN), Dallas, TX, USA, 4–9 August 2013; pp. 1–7. [Google Scholar]
- Khawaja, W.; Ozdemir, O.; Guvenc, I. UAV Air-to-Ground Channel Characterization for mmWave Systems. In Proceedings of the 2017 IEEE 86th Vehicular Technology Conference (VTC-2017), Toronto, ON, Canada, 24–27 September 2017; pp. 1–5. [Google Scholar] [CrossRef]

**Figure 8.**The average difference between the estimated and real RSS values in the pre-training stage.

**Figure 10.**Statistical distribution of the difference in the RSS value of estimated (SVM) and real RSS values (dB).

**Figure 12.**Statistical distribution of the difference in the RSS value of estimated (ANN) and real RSS values (dB).

**Figure 14.**Statistical distribution of the difference in RSS value of estimated (GBRBM-AE) and real RSS values (dB).

Scenario | Building Heights (m) | Number of Buildings |
---|---|---|

Urban | 100–120 | 100 |

Suburban | 20–30 | 25 |

Rural | 5–10 | 10 |

Overseas | - | - |

Key Parameters | Settings |
---|---|

Number of layers | 5 |

Neuron number of bottleneck layer | 7 |

Number of GBRBMs | 5 |

Epoch number of pre-training stage | 250 |

Epoch number of training stage | 500 |

Learning rate | 0.001 |

Algorithm | Best | Worst | Standard Deviation |
---|---|---|---|

GBRBM-AE | 5.53 | 6.35 | 0.12 |

ANN | 7.53 | 7.8 | 0.37 |

SVM | 10.37 | 13.48 | 0.63 |

Algorithm | GBRBM-AE | ANN | SVM |
---|---|---|---|

MAE (dB) | 4.13 | 6.65 | 8.12 |

MAPE (%) | 3.27 | 5.18 | 10.37 |

RMSE (dB) | 7.17 | 8.32 | 12.63 |

Algorithm | Time Cost (ms) |
---|---|

GBRBM-AE | 2.1 |

ANN | 7 |

SVM | 68 |

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2022 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 (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Al-Gburi, A.; Abdullah, O.; Sarhan, A.Y.; Al-Hraishawi, H.
Channel Estimation for UAV Communication Systems Using Deep Neural Networks. *Drones* **2022**, *6*, 326.
https://doi.org/10.3390/drones6110326

**AMA Style**

Al-Gburi A, Abdullah O, Sarhan AY, Al-Hraishawi H.
Channel Estimation for UAV Communication Systems Using Deep Neural Networks. *Drones*. 2022; 6(11):326.
https://doi.org/10.3390/drones6110326

**Chicago/Turabian Style**

Al-Gburi, Ahmed, Osamah Abdullah, Akram Y. Sarhan, and Hayder Al-Hraishawi.
2022. "Channel Estimation for UAV Communication Systems Using Deep Neural Networks" *Drones* 6, no. 11: 326.
https://doi.org/10.3390/drones6110326