Mobility-aware federated self-supervised learning in vehicular network

Abstract

The development of the Internet of Things has led to a significant increase in the number of devices, consequently generating a vast amount of data and resulting in an influx of unlabeled data. Collecting these data enables the training of robust models to support a broader range of applications. However, labeling these data can be costly, and the models dependent on labeled data are often unsuitable for rapidly evolving fields like vehicular networks and mobile Internet of Things, where new data continuously emerge. To address this challenge, Self-Supervised Learning (SSL) offers a way to train models without the need for labels. Nevertheless, the data stored locally in vehicles are considered private, and vehicles are reluctant to share data with others. Federated Learning (FL) is an advanced distributed machine learning approach that protects each vehicle’s privacy by allowing models to be trained locally and the model parameters to be exchanged across multiple devices simultaneously. Additionally, vehicles capture images while driving through cameras mounted on their rooftops. If a vehicle’s velocity is too high, the captured images, donated as local data, may be blurred. Simple aggregation of such data can negatively impact the accuracy of the aggregated model and slow down the convergence speed of FL. This paper proposes a FL algorithm for aggregation based on image blur levels, which is called FLSimCo. This algorithm does not require labels and serves as a pre-training stage for SSL in vehicular networks. Simulation results demonstrate that the proposed algorithm achieves fast and stable convergence.

Keywords

Federated learning / Self-supervised learning / Vehicular network / Mobility

References

[1.] Wu Q, Wang S, Ge H, Fan P, Fan Q, Letaief KB. Delay-sensitive task offloading in vehicular fog computing-assisted platoons. IEEE Trans Netw Serv Manag, 2023, 21: 2012-2026.

[2.] Wu Q, Shi S, Wan Z, Fan Q, Fan P, Zhang C. Delay-sensitive task offloading in vehicular fog computing-assisted platoons. Chin J Electron, 2023, 32: 1230-1244.

[3.] Zhu H, Wu Q, Wu X, Fan Q, Fan P, Wang J. Decentralized power allocation for mimo-noma vehicular edge computing based on deep reinforcement learning. IEEE Internet Things J, 2021, 9: 12770-12782.

[4.] Anagnostopoulos C, Gkillas A, Piperigkos N, Lalos AS (2023) Federated deep feature extraction-based slam for autonomous vehicles. In: Paper presented at the 2023 24th International Conference on Digital Signal Processing (DSP), Rhodes (Rodos), Greece. https://doi.org/10.1109/DSP58604.2023.10167897

[5.] Wu Q, Wang W, Fan P, Fan Q, Wang J, Letaief KB. Urllc-awared resource allocation for heterogeneous vehicular edge computing. IEEE Trans Veh Technol, 2024, 73: 11789-11805.

[6.] Wu Q, Wang W, Fan P, Fan Q, Zhu H, Letaief KB. Cooperative edge caching based on elastic federated and multi-agent deep reinforcement learning in next-generation networks. IEEE Trans Netw Serv Manag, 2024, 21: 4179-4196.

[7.] Wu Q, Wang X, Fan Q, Fan P, Zhang C, Li Z. High stable and accurate vehicle selection scheme based on federated edge learning in vehicular networks. China Commun, 2023, 20: 1-7.

[8.] Wu Q, Xia S, Fan Q, Li Z (2019) Performance analysis of ieee 802.11p for continuous backoff freezing in iov. Special Issue Intelligent and Cooperation Communication and Networking Technologies for IoT 8. https://doi.org/10.3390/electronics8121404

[9.] Yan R, Qu L, Wei Q, Huang SC, Shen L, Rubin DL, Xing L, Zhou Y. Label-efficient self-supervised federated learning for tackling data heterogeneity in medical imaging. IEEE Trans Med Imaging, 2023, 42: 1932-1943.

[10.] Shao Z, Wu Q, Fan P, Cheng N, Fan Q, Wang J (2024) Sematic-aware resource allocation based on deep reinforcement learning for 5g–v2x hetnets. IEEE Commun Lett. https://doi.org/10.1109/LCOMM.2024.3443603

[11.] Shao Z, Wu Q, Fan P, Cheng N, Chen W, Wang J, B LK, (2024) Semantic-aware spectrum sharing in internet of vehicles based on deep reinforcement learning. IEEE Internet Things J. https://doi.org/10.1109/JIOT.2024.3448538

[12.] Zhang C, Zhang K, Pham TX, Niu A, Qiao Z, Yoo CD, Kweon I (2022) Dual temperature helps contrastive learning without many negative samples: towards understanding and simplifying moco, 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), New Orleans. https://doi.org/10.1109/CVPR52688.2022.01404

[13.] Doersch C, Gupta A, Efros AA (2015) Unsupervised visual representation learning by context prediction. 2015 IEEE International Conference on Computer Vision (ICCV), Santiago. https://doi.org/10.1109/ICCV.2015.167

[14.] Wu Z, Xiong Y, Yu SX, Lin D (2018) Unsupervised feature learning via non-parametric instance discrimination (IEEE). 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City. https://doi.org/10.1109/CVPR.2018.00393

[15.] Chen T, Kornblith S, Norouzi M, Hinton G (2002) A simple framework for contrastive learning of visual representations. https://doi.org/10.48550/arXiv.2002.05709

[16.] Kong S (2023) Self-supervised image classification using convolutional neural network. In: Paper presented at 2023 IEEE 3rd International Conference on Power, Electronics and Computer Applications(ICPECA), Shenyang. https://doi.org/10.1109/ICPECA56706.2023.10075949

[17.] Hjelm RD, Fedorov A, Lavoie-Marchildon S, Grewal K, Bachman P, Trischler A, Bengio Y (2019) Learning deep representations by mutual information estimation and maximization. https://doi.org/10.48550/arXiv.1808.06670

[18.] He K, Fan H, Wu Y, Xie S, Girshick R (2020) Momentum contrast for unsupervised visual representation learning. In: Paper presented at 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle. https://doi.org/10.1109/CVPR42600.2020.00975

[19.] Chen X, Fan H, Girshick R, He K (2020) Improved baselines with momentum contrastive learning. https://doi.org/10.48550/arXiv.2003.04297

[20.] Zhao J, Xiong X, Zhang Q, Wang D. Extended multi-component gated recurrent graph convolutional network for traffic flow prediction. IEEE Trans Intell Transp Syst, 2023, 25: 4634-4644.

[21.] Cai T, Gan H, Peng B, Huang Q, Zou Z (2022) Real-time classification of disaster images from social media with a self-supervised learning framework. In: Paper presented at IGARSS 2022 - 2022 IEEE International Geoscience and Remote Sensing Symposium(IEEE), Kuala Lumpur. https://doi.org/10.1109/IGARSS46834.2022.9883129

[22.] Zhao J, Li R, Wang H, Xu Z (2021) Hotfed: Hot start through self-supervised learning in federated learning. In: Paper presented at 2021 IEEE 23rd Int Conf on High Performance Computing & Communications; 7th Int Conf on Data Science & Systems; 19th Int Conf on Smart City; 7th Int Conf on Dependability in Sensor, Cloud & Big Data Systems & Application (HPCC/DSS/SmartCity/DependSys), Haikou, Hainan. https://doi.org/10.1109/HPCC-DSS-SmartCity-DependSys53884.2021.00046

[23.] Zhao J, Li Q, Ma X, Yu FR. Computation offloading for edge intelligence in two-tier heterogeneous networks. IEEE Trans Netw Sci Eng, 2023, 11: 1872-1884.

[24.] Zhao J, Quan H, Xia M, Wang D. Adaptive resource allocation for mobile edge computing in internet of vehicles: A deep reinforcement learning approach. IEEE Trans Veh Technol, 2023, 73: 5834-5848.

[25.] Zhang C, Zhang W, Wu Q, Fan P, Fan Q, Wang J, Letaief KB (2024) Distributed deep reinforcement learning based gradient quantization for federated learning enabled vehicle edge computing. IEEE Internet Things J. https://doi.org/10.1109/JIOT.2024.3447036

[26.] Feng M, Kao CC, Tang Q, Sun M, Rozgic V, Matsoukas S, Wang C (2022) Federated self-supervised learning for acoustic event classification. In: Paper presented at ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing, Singapore, Singapore, 23–27 May 2022

[27.] Wu Q, Zhao Y, Fan Q, Fan P, Wang J, Zhang C. Mobility-aware cooperative caching in vehicular edge computing based on asynchronous federated and deep reinforcement learning. IEEE J Sel Top Signal Process, 2022, 17: 66-81.

[28.] Yu Z, Hu J, Min G, Zhao Z, Miao W, Hossain MS. Mobility-aware proactive edge caching for connected values using federated learning. IEEE Trans Intell Transp Syst, 2020, 22: 5341-5351.

[29.] Shirmohammadi S, Ferrero A. Camera as the instrument: the rising trend of vision based measurement. IEEE Instrum Meas Mag, 2014, 17: 41-47.

[30.] Cortes-Osorio JA, Gomez-Mendoza JB, Riano-Rojas JC. Velocity estimation from a single linear motion blurred image using discrete cosine transform. IEEE Trans Instrum Meas, 2018, 68: 4038-4050.

[31.] Hou K, Lv X, Zhang W (2020) An adaptive fusion panoramic image mosaic algorithm based on circular lbp feature and hsv color system. In: Paper presented at 2020 IEEE International Conference on Information Technology, Big Data and Artificial Intelligence (ICIBA), Chongqing. https://doi.org/10.1109/ICIBA50161.2020.9277348

[32.] Hsieh K, Phanishayee A, Mutlu O, Gibbons PB (2020) The non-iid data quagmire of decentralized machine learning. https://doi.org/10.48550/arXiv.1910.00189

[33.] Lu Z, Pan H, Dai Y, Si X, Zhang Y (2024) Federated Learning With Non-IID Data: A Survey, IEEE Internet of Things Journal 11:19188-19209. https://doi.org/10.1109/JIOT.2024.3376548

[34.] Wu H, Wang M, Zhou W, Li H (2021) 2021 IEEE/CVF International Conference on Computer Vision (ICCV). Montreal. https://doi.org/10.1109/ICCV48922.2021.01122

[35.] Zhang Z, Ma S, Yang Z, Xiong Z, Kang J, Wu Y, Zhang K, Niyato D. Robust semi-supervised federated learning for images automatic recognition in internet of drones. IEEE Internet Things J, 2022, 10: 5733-5746.

[36.] Wei S, Cao G, Dai C, Dai S, Guo B (2022) Fedco: self-supervised learning in federated learning with momentum contrast. In: Paper presented at 2022 IEEE 24th Int Conf on High Performance Computing & Communications; 8th Int Conf on Data Science & Systems; 20th Int Conf on Smart City; 8th Int Conf on Dependability in Sensor, Cloud & Big Data Systems & Application (HPCC/DSS/SmartCity/DependSys, Hainan. https://doi.org/10.1109/HPCC-DSS-SmartCity-DependSys57074.2022.00192

[37.] Tang Z, Shi S, Chu X (2020) Communication-Efficient Decentralized Learning with Sparsification and Adaptive Peer Selection. 2020 IEEE 40th International Conference on Distributed Computing Systems (ICDCS), Singapore. https://doi.org/10.1109/ICDCS47774.2020.00153

Funding

National Natural Science Foundation of China http://dx.doi.org/10.13039/501100001809(61701197)