Approximating the Controllability Robustness of Directed Random-graph Networks Against Random Edge-removal Attacks

Yang LOU; Lin  WANG; Shengli XIE; Guanrong CHEN

doi:10.1007/s12555-021-0831-4

Approximating the Controllability Robustness of Directed Random-graph Networks Against Random Edge-removal Attacks

Yang LOU, Lin WANG, Shengli XIE, Guanrong CHEN^*

^*Corresponding author for this work

Department of Computing and Decision Sciences

Research output: Journal Publications › Journal Article (refereed) › peer-review

5 Citations (Scopus)

Abstract

Network controllability robustness reflects how well a networked system can maintain its controllability against destructive attacks. Its performance can be measured by a sequence of values that record the remaining controllability of the network after a sequential node-removal or edge-removal attacks. In this paper, a hybrid approximation (HyA) method is proposed to estimate the controllability robustness performance of large-scale directed random-graph (RG) networks under random edge-removal attacks. HyA sufficiently utilizes the similarity between the generation mechanism of the RG networks and the destructing process of random edge-removal attacks. Two threshold values are set to classify general RG networks as ‘dense’, ‘sparse’, or ‘median’, according to the average degree of each network. A two-phase approximation is applied to ‘sparse’ RG networks, while different three-phase approximations are applied to ‘dense’ and ‘median’ RG networks, respectively. Simulation results verify that 1) HyA is able to precisely approximate the controllability curves of RG networks under random edge-removal attacks; 2) HyA is time-efficient as compared to the conventional time-consuming attack simulations.

Original language	English
Pages (from-to)	376-388
Number of pages	13
Journal	International Journal of Control, Automation and Systems
Volume	21
Issue number	2
Early online date	30 Jan 2023
DOIs	https://doi.org/10.1007/s12555-021-0831-4
Publication status	Published - 1 Feb 2023

Bibliographical note

Funding Information:
This research was supported in part by the National Natural Science Foundation of China (No. 62002249, 61873167), in part by the Foundation of Key Laboratory of System Control and Information Processing, Ministry of Education, P. R. China (No. Scip202103), in part by the Lam Woo Research Fund at Lingnan University (No. LWP20012), and in part by the Hong Kong Research Grants Council under the GRF Grant CityU11206320.

Publisher Copyright:
© 2023, ICROS, KIEE and Springer.

Keywords

Complex network
controllability
directed random graph
random edge attack
robustness

Access to Document

10.1007/s12555-021-0831-4Licence: Other

https://link.springer.com/10.1007/s12555-021-0831-4

Cite this

@article{12a5a86ee42f4bf18e039371f866e3ea,

title = "Approximating the Controllability Robustness of Directed Random-graph Networks Against Random Edge-removal Attacks",

abstract = "Network controllability robustness reflects how well a networked system can maintain its controllability against destructive attacks. Its performance can be measured by a sequence of values that record the remaining controllability of the network after a sequential node-removal or edge-removal attacks. In this paper, a hybrid approximation (HyA) method is proposed to estimate the controllability robustness performance of large-scale directed random-graph (RG) networks under random edge-removal attacks. HyA sufficiently utilizes the similarity between the generation mechanism of the RG networks and the destructing process of random edge-removal attacks. Two threshold values are set to classify general RG networks as {\textquoteleft}dense{\textquoteright}, {\textquoteleft}sparse{\textquoteright}, or {\textquoteleft}median{\textquoteright}, according to the average degree of each network. A two-phase approximation is applied to {\textquoteleft}sparse{\textquoteright} RG networks, while different three-phase approximations are applied to {\textquoteleft}dense{\textquoteright} and {\textquoteleft}median{\textquoteright} RG networks, respectively. Simulation results verify that 1) HyA is able to precisely approximate the controllability curves of RG networks under random edge-removal attacks; 2) HyA is time-efficient as compared to the conventional time-consuming attack simulations.",

keywords = "Complex network, controllability, directed random graph, random edge attack, robustness",

author = "Yang LOU and Lin WANG and Shengli XIE and Guanrong CHEN",

note = "Funding Information: This research was supported in part by the National Natural Science Foundation of China (No. 62002249, 61873167), in part by the Foundation of Key Laboratory of System Control and Information Processing, Ministry of Education, P. R. China (No. Scip202103), in part by the Lam Woo Research Fund at Lingnan University (No. LWP20012), and in part by the Hong Kong Research Grants Council under the GRF Grant CityU11206320. Publisher Copyright: {\textcopyright} 2023, ICROS, KIEE and Springer.",

year = "2023",

month = feb,

day = "1",

doi = "10.1007/s12555-021-0831-4",

language = "English",

volume = "21",

pages = "376--388",

journal = "International Journal of Control, Automation and Systems",

issn = "1598-6446",

publisher = "Institute of Control, Robotics and Systems",

number = "2",

}

TY - JOUR

T1 - Approximating the Controllability Robustness of Directed Random-graph Networks Against Random Edge-removal Attacks

AU - LOU, Yang

AU - WANG, Lin

AU - XIE, Shengli

AU - CHEN, Guanrong

N1 - Funding Information: This research was supported in part by the National Natural Science Foundation of China (No. 62002249, 61873167), in part by the Foundation of Key Laboratory of System Control and Information Processing, Ministry of Education, P. R. China (No. Scip202103), in part by the Lam Woo Research Fund at Lingnan University (No. LWP20012), and in part by the Hong Kong Research Grants Council under the GRF Grant CityU11206320. Publisher Copyright: © 2023, ICROS, KIEE and Springer.

PY - 2023/2/1

Y1 - 2023/2/1

N2 - Network controllability robustness reflects how well a networked system can maintain its controllability against destructive attacks. Its performance can be measured by a sequence of values that record the remaining controllability of the network after a sequential node-removal or edge-removal attacks. In this paper, a hybrid approximation (HyA) method is proposed to estimate the controllability robustness performance of large-scale directed random-graph (RG) networks under random edge-removal attacks. HyA sufficiently utilizes the similarity between the generation mechanism of the RG networks and the destructing process of random edge-removal attacks. Two threshold values are set to classify general RG networks as ‘dense’, ‘sparse’, or ‘median’, according to the average degree of each network. A two-phase approximation is applied to ‘sparse’ RG networks, while different three-phase approximations are applied to ‘dense’ and ‘median’ RG networks, respectively. Simulation results verify that 1) HyA is able to precisely approximate the controllability curves of RG networks under random edge-removal attacks; 2) HyA is time-efficient as compared to the conventional time-consuming attack simulations.

AB - Network controllability robustness reflects how well a networked system can maintain its controllability against destructive attacks. Its performance can be measured by a sequence of values that record the remaining controllability of the network after a sequential node-removal or edge-removal attacks. In this paper, a hybrid approximation (HyA) method is proposed to estimate the controllability robustness performance of large-scale directed random-graph (RG) networks under random edge-removal attacks. HyA sufficiently utilizes the similarity between the generation mechanism of the RG networks and the destructing process of random edge-removal attacks. Two threshold values are set to classify general RG networks as ‘dense’, ‘sparse’, or ‘median’, according to the average degree of each network. A two-phase approximation is applied to ‘sparse’ RG networks, while different three-phase approximations are applied to ‘dense’ and ‘median’ RG networks, respectively. Simulation results verify that 1) HyA is able to precisely approximate the controllability curves of RG networks under random edge-removal attacks; 2) HyA is time-efficient as compared to the conventional time-consuming attack simulations.

KW - Complex network

KW - controllability

KW - directed random graph

KW - random edge attack

KW - robustness

UR - http://www.scopus.com/inward/record.url?scp=85147025893&partnerID=8YFLogxK

U2 - 10.1007/s12555-021-0831-4

DO - 10.1007/s12555-021-0831-4

M3 - Journal Article (refereed)

SN - 1598-6446

VL - 21

SP - 376

EP - 388

JO - International Journal of Control, Automation and Systems

JF - International Journal of Control, Automation and Systems

IS - 2

ER -

Approximating the Controllability Robustness of Directed Random-graph Networks Against Random Edge-removal Attacks

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Multi-Objective Optimization of Complex Network Robustness

Cite this

Approximating the Controllability Robustness of Directed Random-graph Networks Against Random Edge-removal Attacks

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Projects

Multi-Objective Optimization of Complex Network Robustness

Cite this