Deep Reinforcement Learning for Performance-Aware Adaptive Resource Allocation in Mobile Edge Computing

Table 1

Major notations.

Symbols	Semantics
	The number of mobile devices in MEC
	The time slot duration
	The task workload
	The result data size per unit workload
	The arrival rate of mobile device
	The data buffer for mobile device
	The size of data buffer
	The number of arrival tasks for in time slot
	The number of computed tasks for in time slot
	The number of transmitted computation results for
	The front-end queue of
	The back-end queue of
	The transmission rate of a wireless channel from the edge server to
	The process capacity with a single CPU core
	The number of CPU cores
	The number of timeout tasks in time slot
	The number of dropped tasks in time slot
	The immediate reward in time slot