IEEE Computational Intelligence Magazine - August 2022 - 33

TABLE I Detection rate with different PN sequence powers
for Fashion-MNIST and Cifar-10 datasets.
SNR
Fashion-MNIST
Cifar-10
9 dB
0.931
0.925
6 dB
0.989
0.975
3 dB
0.999
0.996
have been investigated for BLADE-FL including privacy,
resource allocation, and lazy clients, and possible solutions have
been provided to address those issues and explored with experiments.
These results provide guidelines for the design of the
BLADE-FL framework.
Some directions for further study in this area include asyn0.4
0.5
0.6
0.7
0.8
0.9
1
Fashion-MNIST
Cifar-10
Without
Lazy Clients
30% Lazy Clients Without Detection Scheme
The Proposed Lazy Client Detection Scheme
FIGURE 4 Learning performance with/without lazy client detection.
that of the injected PN sequence, and Table I represents the
detection rate of lazy clients under different SNRs. If a high
peaks in terms of the cross-correlation coefficient surpasses a predefined
threshold, this client is identified as a lazy one. A PN
sequence of length 215 is generated and the first 25400 values are
used to add onto the parameters. From the results with different
SNRs, the detection performance is remarkable, and a nearly
100% rate of lazy client recognition when SNR=3 dB can be
obtained. Fig. 4 shows the PN sequence-protection performance
(SNR = 6 dB) when there are 30% (6) lazy clients in each communication
round. As can be seen in this figure, the system performance
with a certain percentage of lazy clients degrades
sharply, i.e., 22.1% and 19.6% reduction for the Fashion-MNIST
and Cifar-10 datasets, respectively. In addition, the proposed PN
sequence-protection method achieves 18% and 13.8% performance
gain for each dataset, respectively.
V. Future Directions and Conclusions
In this paper, the weaknesses of FL have been reviewed and a
blockchain-assisted decentralized FL architecture, called
BLADE-FL, has been proposed to address some of these weaknesses.
The effectiveness of BLADE-FL has been shown in
addressing these issues, notably the problem of a single point of
failure that exists in conventional FL. In addition, further issues
chronous and heterogenetic investigations for different client
capabilities, such as computing capability, training-data size, and
transmitting diversity, and SC design, which provides reasonable
reward allocation for training and mining. In addition, lightweight
model transmission using quantization and sketch may
be an alterative way of reducing the transmission cost.
Acknowledgments
This work was supported in part by the National Natural
Science Foundation of China under Grant No. 62002170, and
61872184, and in part by the Fundamental Research Funds for
the Central Universities under Grant No. 30919011274, in part
by the Natural Science Foundation of Jiangsu Province under
Grant BK20210331, in part by the Jiangsu Specially-Appointed
Professor Program in 2021, in part by the Natural Science
Fund of Guangdong Province under Grant 2020A1515010708
and the Natural Science Fund of Shenzhen under Grant
JCYJ20210324094609027, and in part by the U.S. National
Science Foundation under Grants ECCS-2039716,
CNS-2107216 and CNS-2128368.
References
[1] C. Ma et al., " On safeguarding privacy and security in the framework of federated learning, "
IEEE Netw., vol. 34, no. 4, pp. 242-248, Jul./Aug. 2020, doi: 10.1109/MNET.001.1900506.
[2] Z. Liu, P. Longa, G. C. C. F. Pereira, O. Reparaz, and H. Seo, " FourQ on embedded
devices with strong countermeasures against side-channel attacks, " IEEE Trans. Dependable Secure
Comput., vol. 17, no. 3, pp. 536-549, May/Jun. 2020, doi: 10.1007/978-3-319-66787-4_32.
[3] Y. Chen, X. Qin, J. Wang, C. Yu, and W. Gao, " Fedhealth: A federated transfer
learning framework for wearable healthcare, " IEEE Intell. Syst., vol. 35, no. 4, pp. 83-93,
Jul./Aug. 2020, doi: 10.1109/MIS.2020.2988604.
[4] T. Li, A. K. Sahu, A. Talwalkar, and V. Smith, " Federated learning: Challenges, methods,
and future directions, " IEEE Signal Process. Mag., vol. 37, no. 3, pp. 50-60, May 2020,
doi: 10.1109/MSP.2020.2975749.
[5] H. Kim, J. Park, M. Bennis, and S. Kim, " Blockchained on-device federated learning, "
IEEE Commun. Lett., vol. 24, no. 6, pp. 1279-1283, Jun. 2020, doi: 10.1109/
LCOMM.2019.2921755.
[6] Y. Lu, X. Huang, Y. Dai, S. Maharjan, and Y. Zhang, " Blockchain and federated
learning for privacy-preserved data sharing in industrial IoT, " IEEE Trans. Ind. Informat.,
vol. 16, no. 6, pp. 4177-4186, Jun. 2020, doi: 10.1109/TII.2019.2942190.
[7] X. Bao, C. Su, Y. Xiong, W. Huang, and Y. Hu, " Flchain: A blockchain for auditable
federated learning with trust and incentive, " in Proc. 2019 5th Int. Conf. Big Data Comput.
Commun. (BIGCOM), pp. 151-159, doi: 10.1109/BIGCOM.2019.00030.
[8] P. K. Sharma, J. H. Park, and K. Cho, " Blockchain and federated learning-based distributed
computing defence framework for sustainable society, " Sustain. Cities Soc., vol. 59, p. 102,220, 2020.
[9] S. Wang, " Blockfedml: Blockchained federated machine learning systems, " in Proc.
2019 Int. Conf. Intell. Comput., Autom. Syst. (ICICAS), pp. 751-756, doi: 10.1109/ICICAS48597.2019.00162.
[10]
Y. Qu, S. R. Pokhrel, S. Garg, L. Gao, and Y. Xiang, " A blockchained federated
learning framework for cognitive computing in industry 4.0 networks, " IEEE Trans. Ind.
Informat., vol. 17, no. 4, pp. 2964-2973, Apr. 2021, doi: 10.1109/TII.2020.3007817.
[11] S. Otoum, I. Al Ridhawi, and H. Mouftah, " Blockchain-supported federated learning
for trustworthy vehicular networks, " Dec. 2020, pp. 1-6.
[12] M. Jelasity, " Gossip, " in Self-Organising Software. New York, NY, USA: Springer-Verlag,
2011, pp. 139-162.
[13] L. Ismail, H. Materwala, and S. Zeadally, " Lightweight blockchain for healthcare, "
IEEE Access, vol. 7, pp. 149,935-149,951, 2019, doi: 10.1109/ACCESS.2019.2947613.
[14] P. Fairley, " Blockchain world - feeding the blockchain beast if bitcoin ever does go
mainstream, the electricity needed to sustain it will be enormous, " IEEE Spectr., vol. 54,
no. 10, pp. 36-59, Oct. 2017, doi: 10.1109/MSPEC.2017.8048837.
[15] K. Wei et al., " User-level privacy-preserving federated learning: Analysis and performance
optimization, " IEEE Trans. Mobile Comput., early access, Feb. 4, 2021, doi: 10.1109/
TMC.2021.3056991.
AUGUST 2022 | IEEE COMPUTATIONAL INTELLIGENCE MAGAZINE 33
Testing Accuracy
IEEE Computational Intelligence Magazine - August 2022

Table of Contents for the Digital Edition of IEEE Computational Intelligence Magazine - August 2022
