From: Proactive edge computing in fog networks with latency and reliability guarantees
CCDF | Complementary cumulative distribution function |
CSI | Channel state information |
DA | Deferred acceptance |
MCC | Mobile cloud computing |
MEC | Mobile edge computing |
QoS | Quality of service |
TDD | Time division duplex |
UN | User node |
URLLC | Ultra-reliable and low latency communication |
\(\mathcal {A}\) | Set of tasks |
\(\mathcal {C}_{i}\) | Set of UNs in cluster i |
\(\mathcal {E}\) | Set of edge cloudlets |
\(\mathcal {U}\) | Set of UNs |
B | System bandwidth |
c e | Computing power of cloudlet e |
D ea | Total delay of task a from cloudlet e |
\(D_{ea}^{\text {comp}}\) | Computing delay of task a from cloudlet e |
\(D_{ea}^{\text {comm}}\) | Transmission delay of task a from cloudlet e |
D th | Delay threshold |
L a | Size of task a |
n u | Task occurrence vector of UN u |
Q e | Task queue of cloudlet e |
r ue | Data rate of UN u from cloudlet e |
\(\bar {r}_{ue}\) | Estimated data rate of UN u from cloudlet e |
S d | Distance Gaussian similarity matrix |
S p | Task popularity similarity matrix |
s e | Storage size of cloudlet e |
v u | Geographical coordinates vector of UN u |
W ea | Waiting time of task a in the queue of cloudlet e |
x ea | Association indicator |
X max | Maximum number of associated cloudlets |
y ea | Caching indicator |
z | Zipf distribution parameter |
ε | Target maximum delay violation |
σ d | Similarity parameter |
κ | Task processing density |
λ u | Mean UN task arrival rate |
θ | Clustering parameter |
τ EP | Edge processing delay |
ν | Learning rate parameter |
ξ i | Popularity task vector of cluster i |