Opportunistic Offloading And Band Accessing-based Cellular Networks Congestion Optimization

The popularity of smart mobile devices and smart applications has caused explosive growth in mobile data,which causes cellular networks to gradually face network congestion problems.In order to alleviate cellular networks con-gestion problem caused by the surge of data volume,"opportunity offloading"and "unlicensed spectrum accessing' as methods that do not require additional deployment costs are proposed."Opportunistic offloading' is a method using short-distance wireless communication opportunities between mobile nodes to transmit data.It can reduce the amount of data transferred by cellular networks.However,since the uncertainty of short-range wireless communication oppor-tunities between mobile nodes,the data will not be guaranteed to be delivered to the requesting node within its valid time."Unlicensed spectrum accessing"extends cellular networks to the unlicensed spectrum increasing the capacity of cellular networks.To ensure fair coexistence among the cellular network and other networks occupied the unlicensed spectrum,3GPP has proposed two types of accessing mechanisms for cellular networks:Duty-Cycle and Listen-Before-Talk.The accessing mechanisms are controlled by cellular networks.Cellular networks cannot consider the performance loss of other networks oc-cupied the unlicensed spectrum for its own throughput optimization.There-fore,research on the opportunistic offloading strategies that optimize the data offloading ratio,as well as unlicensed spectrum accessing mechanisms that op-timize the throughput of cellular networks while satisfying the fair coexistence restriction,are of great significance for alleviating cellular networks congestion.To the opportunistic offloading strategy,the limitations of most studies focused on improving data offloading ratio are illustrated as follows:(1)there is no research to distributed and real-time quantify the ability of a node of of-floading the data to a specified destination;(2)no study has comprehensively considered the impact of the opportunistic contact duration,frequency,direct and indirect contact Wi-Fi access points,on the data offloading ratio.To the un-licensed spectrum accessing mechanism,we focus on extending LTE network to the unlicensed spectrum.Most research optimizes the throughput of the LTE network,and at the same time ensures fair coexistence by adjusting the param-eter of the access mechanism.Their limitations are illustrated as follows:(3)in the process of optimizing the throughput of the LTE network on the unlicensed spectrum,the Duty-Cycle accessing mechanism ignores whether the fair coex-istence criterion based on 3GPP fairness is satisfied;(4)the Listen-Before-Talk accessing mechanism ignores the influence of its channel occupation time pa-rameter on the throughput of the network.The article proposed methods to solve the above-mentioned problems,the methods are illustrated as follows:(1)To solve the problem that the performance of opportunistic offloading including offloading ratio and data redundancy is influenced by inaccurately evaluating the offloading ability of the node in the process of offloading down-link data,we focused on the scenario that only one user requests to download data and proposed a downlink data opportunistic offloading strategy based on real-time forwarding capability evaluation.The strategy defines the node car-rying data only transfers data to the node with higher forwarding capability.To accurately evaluate the capability of a node of transmitting data to a specific destination within the valid time of the data,we proposed a distributed node forwarding capability evaluation method based on transient clusters.Simula-tion results prove our proposed method find a balance between offloading ratio and data redundancy.The value of offloading ratio/data redundancy is larger than other opportunistic offloading strategies.(2)To solve the problem in the process of offloading uplink data that data offloading ratio is influenced by incompletely offloading of data within one op-portunistic contact because of the large data size or short opportunistic contact duration,we proposed a uplink data opportunistic offloading strategy based on the heuristic algorithm.First,the data is divided into segments with the same size,and each segment can be completely transferred within one contact.Sec-ond,considering the factors such as the frequency and duration of opportunistic contact we construct a non-linear integer programming model aiming to maxi-mize offloading ratio based on the proposed offloading probability calculation method of opportunistic paths.In the constructed model,the number of of-floaded data segments of each path is the optimized variable.At last,we pro-posed a segment assignment method based on the heuristic algorithm to solve the constructed model.Simulation results show that the offloading ratio of our proposed strategy is superior to other opportunistic offloading strategies in var-ious simulation scenarios.(3)To solve the problem that the criterion of fair coexistence based on 3GPP fairness can not be satisfied in the process of optimizing the through-put of the LTE network on the unlicensed spectrum,we proposed an optimized Duty-Cycle mechanism based on adaptive adjustment of duty cycle.The mech-anism can adaptively adjust the value of duty cycle to optimize the throughput of the LTE network on the unlicensed spectrum while ensuring this fair co-existence criterion.To adjust the duty cycle value,we comprehensively con-sidering factors such as the bandwidth of the authorized spectrum of the LTE network,its users' downlink data rate requirements and the channel fair coexis-tence criteria,construct a non-linear programming model aiming to maximize the throughput of LTE network on the unlicensed spectrum.In the constructed model,duty cycle of unlicensed channels are optimized variables.To solve the optimization problem,the channel duty cycle adjustment algorithm based on interior point method is proposed.Simulation results prove that the proposed Duty-Cycle mechanism based on adaptive adjustment of duty cycle optimizes the throughput of the LTE network on the unlicensed spectrum while ensuring fair coexistence with Wi-Fi network.(4)To solve the problem that the throughput of the LTE network on unli-censed spectrum can not be optimized or the throughput of the Wi-Fi network is excessively affected because of fixing the channel occupancy time of the Listen-Before-Talk mechanism,we proposed an optimized Listen-Before-Talk mech-anism based on adaptive adjustment of the channel occupancy time.The mech-anism can adaptively adjust the value of channel occupancy time to optimize the throughput of the LTE network on the unlicensed spectrum while ensuring fair coexistence.For the channel occupancy time adjustment,we comprehen-sively considering factors such as LTE users' downlink data rate requirements and channel fair sharing evaluation criteria,construct a non-linear programming model aiming to maximize the throughput of LTE network on the unlicensed spectrum based on the proposed method of calculating the throughput of net-work on the coexistence unlicensed spectrum.In the constructed model,the channel occupancy time of the channel is the optimized variable.To solve the optimization problem,a channel occupation time adjustment algorithm based on the interior point method is proposed.Simulation results prove that the pro-posed Listen-Before-Talk mechanism based on adaptive adjustment of channel occupancy time optimizes the throughput of the LTE network on the unlicensed spectrum while ensuring fair coexistence with Wi-Fi network.