Rights statement: ©2021 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.
Accepted author manuscript, 2.81 MB, PDF document
Available under license: CC BY-NC: Creative Commons Attribution-NonCommercial 4.0 International License
Final published version
Research output: Contribution to Journal/Magazine › Journal article › peer-review
Research output: Contribution to Journal/Magazine › Journal article › peer-review
}
TY - JOUR
T1 - CoPace
T2 - Edge Computation Offloading and Caching for Self-Driving with Deep Reinforcement Learning
AU - Tian, H.
AU - Xu, X.
AU - Qi, L.
AU - Zhang, X.
AU - Dou, W.
AU - Yu, S.
AU - Ni, Q.
N1 - ©2021 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.
PY - 2021/12/31
Y1 - 2021/12/31
N2 - Currently, self-driving, emerging as a key automatic application, has brought a huge potential for the provision of in-vehicle services (e.g., automatic path planning) to mitigate urban traffic congestion and enhance travel safety. To provide high-quality vehicular services with stringent delay constraints, edge computing (EC) enables resource-hungry self-driving vehicles (SDVs) to offload computation-intensive tasks to the edge servers (ESs). In addition, caching highly reusable contents decreases the redundant transmission time and improves the quality of services (QoS) of SDVs, which is envisioned as a supplement to the computation offloading. However, the high mobility and time-varying requests of SDVs make it challenging to provide reliable offloading decisions while guaranteeing the resource utilization of content caching. To this end, in this paper we propose a \underline{co}llaborative com\underline{p}utation offlo\underline{a}ding and \underline{c}ont\underline{e}nt caching method, named CoPace, by leveraging deep reinforcement learning (DRL) in EC for self-driving system. Specifically, we resort to a deep learning model to predict the future time-varying content popularity, taking into account the temporal-spatial attributes of requests. Moreover, a DRL-based algorithm is developed to jointly optimize the offloading and caching decisions, as well as the resource allocation (i.e., computing and communication resources) strategies. Extensive experiments with real-world datasets in Shanghai, China, are conducted to evaluate the performance, which demonstrates that CoPace is both effective and well-performed.
AB - Currently, self-driving, emerging as a key automatic application, has brought a huge potential for the provision of in-vehicle services (e.g., automatic path planning) to mitigate urban traffic congestion and enhance travel safety. To provide high-quality vehicular services with stringent delay constraints, edge computing (EC) enables resource-hungry self-driving vehicles (SDVs) to offload computation-intensive tasks to the edge servers (ESs). In addition, caching highly reusable contents decreases the redundant transmission time and improves the quality of services (QoS) of SDVs, which is envisioned as a supplement to the computation offloading. However, the high mobility and time-varying requests of SDVs make it challenging to provide reliable offloading decisions while guaranteeing the resource utilization of content caching. To this end, in this paper we propose a \underline{co}llaborative com\underline{p}utation offlo\underline{a}ding and \underline{c}ont\underline{e}nt caching method, named CoPace, by leveraging deep reinforcement learning (DRL) in EC for self-driving system. Specifically, we resort to a deep learning model to predict the future time-varying content popularity, taking into account the temporal-spatial attributes of requests. Moreover, a DRL-based algorithm is developed to jointly optimize the offloading and caching decisions, as well as the resource allocation (i.e., computing and communication resources) strategies. Extensive experiments with real-world datasets in Shanghai, China, are conducted to evaluate the performance, which demonstrates that CoPace is both effective and well-performed.
KW - computation offloading
KW - content caching
KW - deep reinforcement learning
KW - Delays
KW - edge computing
KW - Optimization
KW - Prediction algorithms
KW - Quality of service
KW - Resource management
KW - Self-driving
KW - Software
KW - Task analysis
KW - Application programs
KW - Computer software reusability
KW - Edge computing
KW - Job analysis
KW - Motion planning
KW - Quality control
KW - Reinforcement learning
KW - Resource allocation
KW - Traffic congestion
KW - Computation offloading
KW - Content caching
KW - Delay
KW - Optimisations
KW - Quality-of-service
KW - Self drivings
KW - Deep learning
U2 - 10.1109/TVT.2021.3121096
DO - 10.1109/TVT.2021.3121096
M3 - Journal article
VL - 70
SP - 13281
EP - 13293
JO - IEEE Transactions on Vehicular Technology
JF - IEEE Transactions on Vehicular Technology
SN - 0018-9545
IS - 12
ER -