Game-Based Multitype Task Offloading Among Mobile-Edge-Computing-Enabled Base Stations

A comprehensive review on internet of things task offloading in multi-access edge computing

“…Over 40 % of the literature surveyed in this survey focused on task allocation as the primary problem-solving goal [ 73 , 74 , 77 , 79 , 81 , [84] , [85] , [86] , 88 , [138] , [139] , [140] , 143 , 145 , [90] , [91] , [147] , 94 , [96] , [97] , [98] , 100 , 104 , 106 , [108] , [109] , [110] , 125 , 130 , 131 , 137 ]. The goal of resource allocation is addressed in approximately 10 % of the reviewed literature [ 99 , 103 , 105 , 111 , 116 , 121 ].…”

“…Device-to-edge-cloud (D2EC) task offloading in a MEC network involves offloading a task from a user's device to a cloud infrastructure that is located at the network's edge. This approach combines the benefits of both edge computing and cloud computing to optimize task execution [ 81 , 85 , [139] , [140] , [141] , [142] , 90 , 92 , 95 , 96 , 98 , 99 , 101 , [108] , [109] , [110] , [111] , [112] , 115 , 119 , 121 , 122 , 124 , 125 , 129 , 131 , 133 , 134 , 136 , 149 , 156 , 168 ].…”

“…This study adopted a transfer learning method, which improved the offloading efficiency. A study [ 110 ] modeled the problem as a Markov decision process (MDP) and solved it using the DRL algorithm. The state space takes into account the network environment, wireless communication resources, channel state changes, etc.…”

The offloading algorithm of mobile edge computing considering mobility in the intelligent inspection scenario

“…where (20) represents A(t), B(t) are the decision sets of all MIEs within the period t. Equations ( 21) and ( 22) indicate that the selection of service nodes cannot exceed its service scope. Equations ( 23) and (24) indicate that the offloading task can only be offloaded to one service node, and cannot be offloaded to multiple service nodes at the same time.…”

“…In this algorithm, users select access conditions based on the interference generated when connecting to each other, and finally optimize the energy consumption and time delay of each mobile device. Reference 24 developed a game theory‐based cooperative offloading algorithm among multiple MECs in response to the shortcomings of a single MEC to balance the delay of each MEC's execution of tasks and achieve the goal of minimizing the total delay. Reference 25 use reinforcement learning to solves the user's server selection problem in a multi‐user and multi‐server environment.…”