A Stochastic Geometric Approach Towards the Temporal Interference Correlation Analysis of D2D Cache Enabled Networks /
Abdulmoneam Ali Hassan Ali
- 2021
- 64 p. ill. 21 cm.
Abstract: both academia and industry over the past few years. Exploiting the potential of direct device-to-device (D2D) communication has been of particular interest to relieve the burden on the cellular network and maximize the content delivery. In addition, deploying cache storage at the mobile devices plays a crucial role in enhancing the spectrum efficiency of the network. Nevertheless, massive and uncoordinated deployments of mobile devices induce significant interference that might substantially hinder the envisioned results. One aspect to overcome this sever interference is through retransmission upon decoding failures. In this thesis, we investigate the impact of temporal interference correlation in D2D cache-enabled cellular network. We focus on the effect of transmission activity on the performance of the network. To this end, tools from stochastic geometry theory is used to capture topological randomness together with the numerous stochastic constituents in such large scale network. We have characterized the temporally correlated interference introduced by the retransmissions and static nature of the cache-enabled devices under generalized Nakagami-m fading channels. In addition, the effect of the different network parameters are investigated. To take further more steps in order to improve network performance, we exploit the cached content information in scheduling D2D links instead of choosing the number of the D2D links blindly. We also characterize the average probability iv of successful content delivery under content dependent access probabilities. With this proposed scheduling scheme, the network performance outperforms in terms of the whole network performance. This achieved gain comes at the cost of performance degradation for the users with less files’ popularity. Hence, we consider enhancing the QoS for all the D2D users without sacrificing the gain attained by exploiting content information.Towards this objective, we formulate a constrained optimization problem for maximizing the overall network performance while guaranteeing the minimum average reliability at each user individually. Finally, for more understanding the network , we derive the distribution of signal to interference ratio under full interference correlation.