Minimizing The Age Of Information In Industrial Internet Of Things

With the development of Internet of Things(Io T)technology,industrial network has undergone great changes,and formed the Industrial Internet of Things(IIo T).The IIo T enhances the data transmission and data processing capabilities between machines,improves the efficiency of production and manufacturing,and provides strong convenience for tasks such as network deployment and resource scheduling.As a result,IIo T is empowering the process of intelligent manufacturing and Industry 4.0,bringing a new technological revolution.In this technological revolution,there are many opportunities and challenges,including the challenge of data timeliness in the IIo T.In the IIo T,there are strict timeliness requirements on tasks such as remote monitoring.Thus,there is an urgent need that data should be transmitted from the perspective of freshness.This thesis studies the freshness of information with the aid of age of information(Ao I)in IIo T.To reduce the Ao I,we leverage mobile edge computing(MEC)to partially offload information to the mobile edge server.Aiming to cope with the packet error in the setting of short packet communication(SPC)in IIo T,we consider the standard automatic repeat request(ARQ)protocol with two policies,i.e.,either retransmitting an out-of-date packet(RO)or transmitting a freshest packet(TF),when a packet error occurs.We derive the closed-form of average Ao I under these two policies respectively,and then formulate the average Ao I minimization problem by jointly optimizing the short packet blocklength and MEC offloading ratio.Due to the nonconvexity nature of the problem,we tackle it by employing block coordinate descent(BCD)and successive convex approximation(SCA)methods and then prove their convergence.Our extensive numerical results show that the optimal average Ao I yielded by our proposed approach is almost identical to that from the high-complexity exhaustive search method,and has significant improvement over the benchmark methods.Furthermore,it is found that the RO policy is suitable for the relatively small bandwidth and large local computing capability scenario,whilst the TF policy is better for the large bandwidth and small local computing capability case.Different from the Ao I which just considers the freshness of data,age of incorrect information(Ao II)considers also the content of information,which will help the monitoring system to make more accurate decisions.Therefore,we further study the Ao II minimization problem in IIo T.Considering the multiple sensors case,we model the problem as constrained Markov decision process(CMDP).Due to the large state space and large action space in CMDP,we resort to deep reinforcement learning to solve it.We propose two algorithms,i.e.,asynchronous advantage actor-critic(A3C)and deep deterministic policy gradient(DDPG),to make decisions which can meet bandwidth constraint.The simulation results show that the two algorithms we proposed can achieve the best result in theory in state-homogeneous system without bandwidth constraint,whilst the algorithm based on DDPG is better in stateheterogeneous system or system with bandwidth constraint.Furthermore,we find that the performance by Ao I-oriented approach is worse than the one by Ao II,since the Ao I-oriented method only considers the age of information and ignores the content of information,which incurs many unnecessary updates.