A Multilevel Optimization Framework for Computation Offloading in Mobile Edge Computing

Algorithm 1

Multiuser multichannel task scheduling algorithm.

	Step 1: initialization
	Step 2: all computing tasks are done locally, i.e.
	Step 3: end initialization
	Step 4: repeat for each user n and server node in each decision slot
	Step 5: transmit the pilot signal on the chosen channel m to the mobile cloud server base-stations
	Step 6: receive the information of the received powers on all channels from each mobile edge user
	Step 7: compute the best response set in the base-stations
	Step 8: ifthen
	Step 9: send RTU message to the cloud for contending for the decision update opportunity
	Step 10: if receive the UP message from the cloud then
	Step 11: choose the decision for next slot
	Step 12: else choose the original decision for next slot
	Step 13: end if
	Step 14: else choose the original decision for next slot
	Step 15: end if
	Step 16: until END message is received from the mobile cloud server base-stations