Computation offloading from a mobile device to edge server is an emerging paradigm to reduce completion latency of intensive computations in mobile edge computing (MEC). In order to satisfy the delay-sensitive computing tasks, offloading time including task uploading time, task execution time and results downloading time is adopted as the computational performance metrics for offloading nodes that perform offloaded computing tasks for mobile devices. Therefore, how to minimize the offloading time by selecting an optimal offloading node in MEC is of research importance. This work first investigates an MEC system consisting of mobile devices and heterogeneous edge severs that support various radio access technologies. Then, based on the available bandwidth of heterogeneous edge severs and the location of mobile devices, an optimal offloading node selection strategy is formulated as a Markov decision process (MDP), and solved by employing value iteration algorithm (VIA). Finally, extensive numerical results demonstrate the effectiveness of the proposed strategy over classic strategies in terms of offloading time.

In Wireless Sensor Networks, the node reliability can affect the reliability of data transmission. To analyze the node reliability, this paper defines both the betweenness centrality and the dependency degree for a node to reflect its global reliability and local reliability, respectively. Based on the above definition, a Global and Local Reliability based Routing (GLRR) protocol is proposed to guarantee the reliability of data transmission between a source and destination node in network. In GLRR, at first, some nodes, usually with greater betweenness centrality among their neighbors in a limited range, will be selected as the backtracking node. Moreover, all backtracking nodes then construct their backtracking paths from themselves to the source node separately, and each node on a backtracking path should calculate the dependency degree on its previous hop node. All the betweenness centrality and dependency degree will be forwarded to the source node along the backtracking paths, and be combined to design the routing metric, based on which, the source node can calculate an optimal backtracking path to forward packets to the corresponding backtracking node, meanwhile, this backtracking node will act as a new source node to launch another routing process until the packets be forwarded to the destination node. At last, the simulation results demonstrate that the proposed protocol is superior to the classical algorithms in terms of the network reliability and the network efficiency.

Mobile Crowd Sensing is an emerging sensing paradigm that employs massive number of workers' mobile devices to realize data collection. Unlike most task allocation mechanisms that aim at optimizing the global system performance, stable matching considers workers are selfish and rational individuals, which has become a hotspot in MCS. However, existing stable matching mechanisms lack deep consideration regarding the effects of workers' competition phenomena and complex behaviors. To address the above issues, this paper investigates the competition-congestion-aware stable matching problem as a multi-objective optimization task allocation problem considering the competition of workers for tasks. First, a worker decision game based on congestion game theory is designed to assist workers in making decisions, which avoids fierce competition and improves worker satisfaction. On this basis, a stable matching algorithm based on extended deferred acceptance algorithm is designed to make workers and tasks mapping stable, and to construct a shortest task execution route for each worker. Simulation results show that the designed model and algorithm are effective in terms of worker satisfaction and platform benefit.

Obstacle avoidance is the crux of formation control for a UAV swarm, no matter in urban or wild application environments. But there is a dilemma to promote both of obstacle avoidance flexibility and consistency in UAV formation control, which has not been solved effectively by existing researches. In view of this, this paper proposes a consensus based reinforcement learning method for differentiated formation control (DFC) of UAVs, to balance the obstacle avoidance flexibility and consistency. In this method, to promote obstacle avoidance flexibility of a UAV swarm, a consensus mechanism with differentiated formation control strategies is designed, it allows each UAV in a swarm changes its formation control strategy among aggregation, formation keeping and obstacle avoidance according to its local environment, and calculates its own current subgoal based on the selected strategy. Further, to improve the flight efficiency of the UAV swarm, a reinforcement learning model is provided to generate the optimal offset vector for each UAV according to its current subgoal. Moreover, to enhance the obstacle avoidance consistency of the UAV swarm, a collaborative obstacle avoidance algorithm is designed in the obstacle avoidance strategy, it requires UAVs to share their obstacle information and obstacle avoidance actions, and provides obstacle avoidance consensus rules to help UAVs to choose consistent obstacle avoidance directions. The experiment results show that the proposed method can combine obstacle avoidance flexibility and consistency of UAVs, thereby achieving higher flight efficiency and maintaining stable network connectivity.

Task allocation, which can be divided into offline task allocation and online task allocation, is a significant issue in mobile crowd sensing (MCS). Unlike offline task allocation, in the online task allocation scenario, since participants arrive at the service dynamically, the quantity of participants in a specific time and space is uncertain, hence it could affect the quality and efficiency of task completion. However, due to the difficulty of predicting the quantity of real-time participants in a specific time and space accurately, the existing studies of online task allocation lack deep consideration of the quantity of participants. Therefore, this paper investigates a participant-quantity-aware online task allocation problem. First of all, in view of the difficulty of predetermining the participant quantity in MCS, a fuzzy time series analysis (FTSA) method is developed to predict the participant quantity available for each task in a specific time and space. Then, according to the predicted quantity, two reasonable attributes for each task, including the task’s threshold on participant’s sensing ability and the reward provided for participants to execute the task, can be calculated separately. On this basis, considering the participant’s willingness, the participant’s sensing ability, the sensor types of the participant’s device, and the participant’s time coverage jointly, we design an online task allocation algorithm based on an improved genetic algorithm (OTAGA) to allocate appropriate set of tasks to each participant who arrives in real-time, so as to maximize the platform utility and minimize the movement cost of the participant. Simulation results show that the proposed method is effective in terms of the accuracy of prediction, the platform utility and the movement cost of the participant.

As a key research issue in mobile crowdsensing (MCS), recent studies on task recommendation have begun to focus on recommending tasks to participants according to the learned participant preferences. The common drawbacks of these studies are that, on the one hand, the factors affecting participant preferences are predefined, which is not practical as the influential factors are quite complex and a full map of participant profiles needs to be preexisted. On the other hand, they do not consider how to update the recommendation dynamically. To overcome these drawbacks, a profile-free and real-time task recommendation method is proposed in this work. First, we apply the recommendation systems to MCS to realize profile-free task recommendations. Second, a participant-task-location tensor is constructed, based on which an improved tensor factorization method is presented to provide task recommendations for participants at a given location. Finally, we design a real-time update algorithm based on the idea of one update at a time to update task recommendation lists for participants in real time. Based on real-world trace data sets, extensive evaluations show that the proposed method has obvious advantages over other baselines in terms of accuracy and time cost.