Delay Control For UAV Based Emergency Communications

In recent years,frequent natural disasters such as earthquakes and tsunamis have caused serious damage to traditional wired and wireless terrestrial infrastructure facilities(such as ground base stations,etc.),which interrupts the communication in the disaster-stricken area.Unmanned Aerial Vehicles(UAVs)are widely deployed as mobile base stations in post-disaster emergency communications to achieve efficient recovery of communication connections in disaster areas.With the rapid growth of various types of emergency services,such as real-time disaster awareness and normalized monitoring,the delay-bounded quality-of-service(Qo S)requirements of emergency services are different.The traditional deterministic delay-bounded Qo S provisioning(DDQP)can only meet single delay requirement,how to guarantee the heterogeneous delay requirements of emergency services in UAV-enabled emergency communications networks is a major problem.In addition to long packet services such as video streaming,the next-generation UAV-enabled emergency communications networks will support short packet services such as massive machine-tomachine(MM2M)communication,which places extremely strict requirements on end-toend(E2E)delay.The existing long frame transmission strategy is only applicable to long packet services,how to achieve ultra-low E2 E delay data transmission is a problem that the next generation of UAV-enabled emergency communications networks must solve.In order to solve the above problems,this article mainly studies the delay control of the UAV-enabled emergency communications networks.The main work in this paper is listed as follows:1.This paper mainly studies delay control of UAV-enabled emergency communications networks based on statistical delay-bounded Qo S provisioning(SDQP).This paper establishes a heterogeneous delay demand mapping model based on SDQP to support the heterogeneous delay demand of various services.UAV is deployed as temporary base stations to provide users with data transmission instead of ground base stations that have been severely damaged due to disasters.In addition to UAV,underlying spectrum sharing device-to-device(D2D)communication transmission links are deployed to enhance spectrum efficiency and expand communication coverage.This paper formulates the effective capacity optimization problem for UAV-enabled emergency communications networks under different statistical delay Qo S requirements,which subjects to average power and buffer size constraints.By solving the optimization problem,the optimal power allocation scheme based on SDQP is designed.The numerical simulation results analyze the impact of D2 D distribution density,interference power level,and interference link quality on the effective capacity under the coexistence of UAV users(UUs)and D2 D users,and verify that the optimal power allocation strategy proposed in this paper can achieve the maximum effective capacity under different delay Qo S requirements than other existing schemes.2.This paper mainly analyzes E2 E delay control of UAV-enabled emergency communications networks based on finite block length(FBL)transmission.There is a trade-off relationship between queuing delay and transmission delay.Existing researches are focused on optimizing a single delay.The joint optimization of queuing delay and transmission delay is studied in this paper.This paper proposes a wireless channel transmission strategy based on the FBL transmission theory,constructs a statistical queuing model,and analyzes the uncertain queuing delay.On the basis of the above-mentioned transmission strategy and queuing model,the problem of minimizing the upper bound of the average E2 E delay under the constraint of the queuing delay violation probability is proposed.With the probability of queuing delay exceeding the queuing delay boundary is maintained at an extremely low level,the upper bound of the average E2 E delay can be minimized by designing the optimal FBL to ensure the strict E2 E delay of the UAV-enabled emergency communications networks.In particular,we propose the problem of minimizing the upper bound of the average E2 E delay when the arrival rate is stochastic and the service rate is constant,and the arrival and service rate are both stochastic.For the FBL transmission with a stochastic arrival and constant service rate,this paper proposes an optimal FBL design based on parameter approximation and bipartite iterative method,to minimize the average upper bound of E2 E delay.Extending the problem to FBL transmission where both the arrival and the service rate are stochastic,we propose an optimal FBL design based on function fitting and traversal searching,to minimize the average upper bound of E2 E delay.Numerical simulation results analyze the average E2 E delay upper bound performance under different packet transmission error probabilities,signal transmission powers,and queuing delay overrun probabilities,verifying that there is indeed a trade-off relationship between transmission delay and queuing delay.The relationship indicates that there is an optimal FBL to minimize the upper bound of the average E2 E delay.Compared with the long term evolution(LTE)frame,the proposed scheme in this paper can reduce the E2 E delay to the sub-millisecond level.