Scheduling Of Distributed Collaborative Tasks And Adaptive Forwarding On MANET

In a challenging scenario of executing distributed collaborative tasks on a Mobile Ad hoc Network(MANET),such as disaster relief,every mobile node can provide certain services.Meanwhile,each task can be decomposed into a sequential composi-tion of these services.Usually,multiple tasks are issued to the MANET,so there is a requirement to improve the efficiency of the task execution.The existing solutions for scheduling distributed collaborative tasks on MANET are inefficient,because they rarely consider the situation of the execution sequence of services in a specific task,where the output of previous service can be used as the input of the next one.In addition,they ignore the redundancy of services execution that have the same requests,which results a waste of network resources.In this paper,we propose a solution for scheduling distributed collaborative tasks on MANET.The solution adopts NDN(Named Data Networking)as the network layer and considers the dynamics of task scheduling and service composition.Moreover,the solution can improve the execution efficiency by leveraging the characteristics of request aggregation and data reuse of NDN.The main work of this dissertation is summarized as follows:1.A solution for scheduling of distributed collaborative tasks on MANET(SD-CTM)is provided.We build a service logical network whose nodes are diverse services that physical nodes can provide.At the task layer,the granularity of task execution is refined,i.e.,a task is regarded as a sequential composition of several services.The ser-vice logical network is responsible for how to schedule a task to different nodes so as to minimize the total execution time.After scheduling,the selected service logical nodes will execute the corresponding services sequentially.2.An NDN-based adaptive forwarding scheme on MANET is presented.NDN can be more beneficial to MANET because of many inbuilt features.In this disser-tation,we offer a solution to adopt NDN for communication in the specific context of disaster relief.Firstly,we give a proactive routing protocol which is also reactive-routing-enabled.Every node broadcasts its existence to neighbors and synchronizes the Network Information Base in order to update its own FIB(Forwarding Information Base).It also remains multiple next-hops for every prefix in FIB in order to support multipath forwarding.By duplicating or splitting the queue of Interests at a node,it's able to utilize multipath so as to improve transmission reliability and efficiency.3.We designed and implemented a prototype of a UGV team that contains some smart cars equipped with Raspberry Pi.We implemented our scheduling scheme of distributed collaborative tasks and adaptive forwarding scheme.The performance of our solution was demonstrated through extensive experiments.