TY - JOUR
T1 - Towards Intelligent Zone-Based Content Pre-Caching Approach in VANET for Congestion Control
AU - Nazar, Khola
AU - Saeed, Yousaf
AU - Ali, Abid
AU - Algarni, Abeer D.
AU - Soliman, Naglaa F.
AU - Ateya, Abdelhamied A.
AU - Muthanna, Mohammed Saleh Ali
AU - Jamil, Faisal
N1 - Funding Information:
The authors would like to acknowledge the support of the Norwegian University of Science and Technology for paying the Article Processing Charges (APC) of this publication. Special acknowledgment to Princess Nourah bint Abdulrahman University Researchers Supporting Project number (PNURSP2022R66), Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia. Faisal Jamil was in part supported through an ERCIM Fellowship.
Funding Information:
This research was funded by the Norwegian University of Science and Technology and Princess Nourah bint Abdulrahman University Researchers Supporting Project number (PNURSP2022R66), Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia.
Publisher Copyright:
© 2022 by the authors.
PY - 2022/12/1
Y1 - 2022/12/1
N2 - In vehicular ad hoc networks (VANETs), content pre-caching is a significant technology that improves network performance and lowers network response delay. VANET faces network congestion when multiple requests for the same content are generated. Location-based dependency requirements make the system more congested. Content pre-caching is an existing challenge in VANET; pre-caching involves the content’s early delivery to the requested vehicles to avoid network delays and control network congestion. Early content prediction saves vehicles from accidents and road disasters in urban environments. Periodic data dissemination without considering the state of the road and surrounding vehicles are considered in this research. The content available at a specified time poses considerable challenges in VANET for content delivery. To address these challenges, we propose a machine learning-based, zonal/context-aware-equipped content pre-caching strategy in this research. The proposed model improves content placement and content management in the pre-caching mode for VANET. Content caching is achieved through machine learning, which significantly improves content prediction by pre-caching the content early to the desired vehicles that are part of the zone. In this paper, three algorithms are presented, the first is zone selection using the customized algorithm, the second is the content dissemination algorithm, and the third is the content pre-caching decision algorithm using supervised machine learning that improves the early content prediction accuracy by 99.6%. The cache hit ratio for the proposed technique improves by 13% from the previous techniques. The prediction accuracy of the proposed technique is compared with CCMP, MLCP, and PCZS+PCNS on the number of vehicles from 10 to 150, with an improved average of 16%. Finally, the average delay reduces over time compared with the state-of-the-art techniques of RPSS, MLCP, CCMP, and PCZS+PCNS. Finally, the average delay shows that the proposed method effectively reduces the delay when the number of nodes increases. The proposed solution improves the content delivery request while comparing it with existing techniques. The results show improved pre-caching in VANET to avoid network congestion.
AB - In vehicular ad hoc networks (VANETs), content pre-caching is a significant technology that improves network performance and lowers network response delay. VANET faces network congestion when multiple requests for the same content are generated. Location-based dependency requirements make the system more congested. Content pre-caching is an existing challenge in VANET; pre-caching involves the content’s early delivery to the requested vehicles to avoid network delays and control network congestion. Early content prediction saves vehicles from accidents and road disasters in urban environments. Periodic data dissemination without considering the state of the road and surrounding vehicles are considered in this research. The content available at a specified time poses considerable challenges in VANET for content delivery. To address these challenges, we propose a machine learning-based, zonal/context-aware-equipped content pre-caching strategy in this research. The proposed model improves content placement and content management in the pre-caching mode for VANET. Content caching is achieved through machine learning, which significantly improves content prediction by pre-caching the content early to the desired vehicles that are part of the zone. In this paper, three algorithms are presented, the first is zone selection using the customized algorithm, the second is the content dissemination algorithm, and the third is the content pre-caching decision algorithm using supervised machine learning that improves the early content prediction accuracy by 99.6%. The cache hit ratio for the proposed technique improves by 13% from the previous techniques. The prediction accuracy of the proposed technique is compared with CCMP, MLCP, and PCZS+PCNS on the number of vehicles from 10 to 150, with an improved average of 16%. Finally, the average delay reduces over time compared with the state-of-the-art techniques of RPSS, MLCP, CCMP, and PCZS+PCNS. Finally, the average delay shows that the proposed method effectively reduces the delay when the number of nodes increases. The proposed solution improves the content delivery request while comparing it with existing techniques. The results show improved pre-caching in VANET to avoid network congestion.
KW - congestion control
KW - content pre-caching
KW - machine learning
KW - VANET
KW - zone
UR - http://www.scopus.com/inward/record.url?scp=85143788866&partnerID=8YFLogxK
U2 - 10.3390/s22239157
DO - 10.3390/s22239157
M3 - Article
C2 - 36501861
AN - SCOPUS:85143788866
VL - 22
JO - Sensors
JF - Sensors
SN - 1424-3210
IS - 23
M1 - 9157
ER -