Smartphone Enabled Data Aggregation And Computation In The Future Mobile Networks

In the future of mobile networks,incoming data traffic will not only generated from the mobile devices but also from low powered devices such as wearable,sensor and Machine-Type Communication.Most of these data will demand the computation resources allocated toward the centralized mobile clouds,which deployed far away from the data sources.With the massive data generated and distant allocation of the centralized mobile clouds will cause the burden on backhaul communication link and long latency.Different literatures present several techniques in industry and academia to solve these problems.However,the effort is succeeded to some extent but not sat:isfactorily,especially at the massive data generated region.In this thesis,we have leveraged the advantages of mobile devices improvement in terms of functionalities and capabilities from low feature phones to advanced smart phones.Firstly,we examined the practicability of enabling the data aggregation to save communication overhead on the mobile network using advanced smartphone devices as the device to collect and pre-process the data instead of processing on the centralized mobile cloud stations.We presented the batch latency update(BLU)scheme whereby the smartphone users aggregate and apply pre-data processing functions such as data cleaning,data redundancy,and integration,and then send to the mobile operators.We have applied the calling data records to verify the performance of the presented scheme.The experimental results has showed that,the presented scheme outperform each time update in terms of communication consumption.Secondly,the data collected by low powered devices specifically Machine-Type communication(MTC)devices demand the computation resources that located far away from the data sources.By considering the smartphone users having the computation capability that can satisfy the computation resources required from the MTC devices.We presented the smartphone-user selection technique whereby the MTC devices select a user that can satisfy the required computation resources.However,the smartphone user has limited computation resources;sometimes the selected user cannot fulfill the computation resources required from MTC devices,we proposed the utility function offloading strategy to choose the nearest smartphone users that can satisfy the required computational resources at the completion time,The evaluated results showed that the proposed technique minimize the execution time and energy consumption on the MTC devices compared with other techniques.Thirdly,with massive computation requests from MTC devices,smartphone devices experience higher energy consumption.Besides,the MTC devices are deployed to perform specific tasks collectively;the data collected from each device are not completely independent rather correlated.By examining the existence of correlated between the MTC device we presented three grouping techniques namely k-means,k-medoids and hierarchical algorithm to combine the proximity MTC devices.To minimize the energy consumption on the smartphone we presented an offloading technique,in which the data that requires higher computation resources are offloaded to the powerful server that allocated at the base station,called mobile edge computing(MEC).As a detecting capability adopted on MTC devices,a power exponential function model is used to compute a correlation coefficient existing between the MTC devices.Based on this framework,.the energy consumption when all data processed at the smartphone or offloaded at the mobile edge computing server is compared to the optimal solution obtained using the exhaustive search method.The simulation results reveal that the presented grouping techniques reduce the energy consumption at a smartphone while satisfying the required completion time.Our research work proved the benefits of enabling the smartphone devices on data aggregation and computation that can be applicable for the future mobile networks.In fact,the results validation proved that our research work is practical useful in the real life scenarios.