| Cloud computing(CC)is a centralized computing paradigm that accumulates resources centrally and provides these resources to users through Internet.Although CC holds a large number of resources,it may not be acceptable by real-time mobile applications,as it is usually far away from users geographically.On the other hand,edge computing(EC),which distributes resources to the network edge,enjoys increasing popularity in the applications with low-latency and high-reliability requirements.EC provides resources in a decentralized manner,which can respond to users’ requirements faster than the normal CC,but with limited computing capacities.As both CC and EC are resource-sensitive,several big issues of resource management arise,such as how to conduct job scheduling,resource allocation,and task offloading,which significantly influence the performance of the whole system.To tackle these issues,many optimization problems have been formulated.These optimization problems usually have complex properties,such as non-convexity and NP-hardness,which may not be addressed by the traditional convex optimization-based solutions.Computational intelligence(CI),consisting of a set of nature-inspired computational approaches,recently exhibits great potential in addressing these optimization problems in CC and EC.In this dissertation,we focus on CI techniques to solve complex optimization problems in mobile EC(MEC).The core works reported in this dissertation are listed as follows:1.This dissertation proposes a multi-unmanned aerial vehicle(UAV)-enabled autonomous MEC system,in which several UAVs are deployed to provide services to users’ equipment(UEs).The aim is to reduce/minimize the overall energy consumption of the autonomous system via designing the optimal trajectories of multiple UAVs.The problem is very complicated to be solved by traditional methods,as one has to take into account the deployment updation of stop points(SPs),the association of SPs with UAVs,and the optimal trajectories designing of UAVs.To tackle this problem,we propose a variable-length trajectory planning algorithm(VLTPA)consisting of two phases.In the first phase,the deployment of SPs is updated via presenting a genetic algorithm(GA)having variable-length individuals.Accordingly,a multi-chrome GA is proposed to jointly handle the association of SPs with UAVs and their order for UAVs.The proposed VLTPA is tested via performing extensive experiments on eight instances ranging from 60 to 200 UEs,which reveal that the proposed VLTPA outperforms other compared state-ofthe-art algorithms.2.This dissertation presents an energy and task completion time minimization scheme for the UAVs-empowered MEC system,where several UAVs are deployed to serve large-scale UEs.The aim is to minimize the weighted sum of energy consumption and task completion time of the system by planning the trajectories of UAVs.The problem is NPhard,non-convex,non-linear,and mixed-decision variables.Therefore,it is very challenging to be solved by conventional optimization techniques.To handle this problem,this dissertation proposes an energy and task completion time minimization algorithm(ETCTMA)that solves the above problem in three steps.In the first step,the deployment updation of SPs is handled by adopting a differential evolution algorithm with a variable population size.Then,in the second step,the association between SPs and UAVs is determined.Specifically,a clustering algorithm is proposed to associate SPs with UAVs.Finally,in the third step,a low-complexity tabu search algorithm is adopted to construct the trajectories of all UAVs.The performance of the proposed ETCTMA is tested on seven instances with up to 700 UEs.The results reveal that our proposed algorithm ETCTMA outperforms other variants in terms of the weighted sum of energy consumption and task completion time of the system.3.This dissertation presents a multi-intelligent reflecting surface(IRS)-and multi-UAV-assisted MEC system,where several UAVs are deployed to serve large-scale UEs with the help of multiple IRSs.This dissertation aims to minimize the overall cost including energy consumption,completion time,and maintenance cost of UAVs by jointly optimizing the trajectories of UAVs and phase shifts of IRSs.When solving this problem,one has to consider the deployment of SPs of UAVs,the association among UEs,SPs,and UAVs(i.e.,which UE will send data to which UAV at which SP),the order of SPs,and the phase shifts of IRSs.Therefore,it is very challenging to solve this problem by using traditional optimization techniques.To tackle this problem,this dissertation proposes an algorithm called TPa PBA that consists of four phases.In the first phase,the deployment of SPs is optimized by adopting a differential evolution algorithm with a variable population size.As a result,all the SPs of UAVs can be obtained.Then,in the second phase,the association among UEs,SPs,and UAVs is optimized.Specifically,a clustering algorithm is first adopted to associate SPs with UAVs,and then a close criterion is introduced to associate UEs with SPs.Subsequently,in the third phase,a low-complexity greedy algorithm is adopted to optimize the order of SPs for all UAVs.Finally,the phase shifts of IRSs are optimized to enhance the data rate between UEs and UAVs.The performance of the proposed TPa PBA is tested on ten instances with up to 1000 UEs.The results reveal that the proposed TPa PBA performs better than other compared algorithms in terms of the overall cost of UAVs. |
