A Novel Biomimicry-based Analysis of D2D User Association Retention for Achieving Maximal Throughput | IEEE Conference Publication | IEEE Xplore

A Novel Biomimicry-based Analysis of D2D User Association Retention for Achieving Maximal Throughput


Abstract:

In 5G network designs, Device-to-Device (D2D) communication is a fast and reliable approach that enables two or more devices to communicate directly with each other wirel...Show More

Abstract:

In 5G network designs, Device-to-Device (D2D) communication is a fast and reliable approach that enables two or more devices to communicate directly with each other wirelessly, without the need to involve a base station. Considering that most commonly these devices are mobile, enabling seamless D2D support for frequent associations and disassociations between mobile devices in proximity of each other requires careful optimization. Association of two or more user devices should be guaranteed based on reliability and stability of the communication link. However, mobility tends to be disruptive for wireless communication, and thus D2D communication approaches are needed that are robust against the mobility impact and can leverage it to their advantage. In our work, we have observed mobility and corresponding device association as a biological phenomenon. In biological tissue we can observe that some cells can collectively be affine to similarly shaped cells, retain the affinity, and form their own `clique' periodically. Bringing the concept to device association - using retention of association between two devices - we present two new biologically inspired algorithms. The algorithms show how in different mobility models, i.e. Brownian, Random Walk, and Levy flight, user association can be retained 6-10% of the time while improving achievable throughput by up to 10% compared to random and social association.
Date of Conference: 24-28 June 2019
Date Added to IEEE Xplore: 22 July 2019
ISBN Information:

ISSN Information:

Conference Location: Tangier, Morocco

Contact IEEE to Subscribe

References

References is not available for this document.