Caching data files directly on mobile user devices combined with device-to-device (D2D) communications has recently been suggested to improve the capacity of wireless networks. We investigate the performance of regenerating codes in terms of the total energy consumption of a cellular network. We show that regenerating codes can offer large performance gains. It turns out that using redundancy against storage node failures is only beneficial if the popularity of the data is between certain thresholds. As our major contribution, we investigate under which circumstances regenerating codes with multiple redundant data fragments outdo uncoded caching.

Abstract-As an alternative to downloading content from a cellular access network, mobile devices could be used to store data files and distribute them through device-to-device (D2D) communication. We consider a D2D-based storage community that is comprised of mobile users. Assuming that transmitting data from a base station to a mobile user consumes more energy than transmitting data between two mobile users, we show that it can be beneficial to use redundant storage to ensure that data files stay available to the community even if some of the storing users leave the network. We derive a tractable closed-form equation stating when redundancy should be used in order to minimize the expected energy consumption of data retrieval. We find that replication is the preferred method of adding redundancy as opposed to regenerating codes. Our findings are verified by computer simulations.

The Compute-and-Forward relaying strategy achieves high computation rates by decoding linear combinations of transmitted messages at intermediate relays. However, if the involved relays independently choose which combinations of the messages to decode, there is no guarantee that the overall system of linear equations is solvable at the destination. In this article it is shown that, for a Gaussian fading channel model with two transmitters and two relays, always choosing the combination that maximizes the computation rate often leads to a case where the original messages cannot be recovered. It is further shown that by limiting the relays to select from carefully designed sets of equations, a solvable system can be guaranteed while maintaining high computation rates. The proposed method has a constant computational complexity and requires no information exchange between the relays.

We consider a geographically constrained caching community where popular data files are cached on mobile terminals and distributed through Device-to-Device (D2D) communications. To ensure availability, data files are protected against user mobility, or churn, with select caching and erasure coding methods. Communication and storage costs are considered, with an objective of minimizing the consumption of radio resources, given an available storage size. We focus on finding the coding method that minimizes the overall cost. Closed-form expressions for the expected consumption of radio resources incurred by data delivery and redundancy maintenance are derived. Closed form transmission costs in a circular caching community with a specific node density and caching method are calculated, when cost obeys a power law of distance. Our results are illustrated by numerical examples and verified by extensive computer simulations.

No abstract