A resource allocation scheme for D2D communications with unknown channel state information

Device-to-device (D2D) communication is a dramatic departure from the conventional cellular architecture as it allows for user equipment (UE) in a cellular network to act as transmission relays without the involvement of network infrastructure, effectively realizing a co-existing massive ad-hoc network. While a hybrid D2D-cellular architecture can enhance the spectral efficiency, resource allocation in such a two-tier system is faced with unique challenges to ensure minimal impact on the performance of existing cellular users (CUs). In this paper, we address the D2D resource allocation problem under unknown channel state information (CSI). CSI-free schemes are substantial given that certain practical limitations (e.g., finite CSI feedback delay) make the knowledge of instantaneous CSI impossible in systems with fading channels. Also, statistical CSI needs pre-deployment training/model extraction and becomes invalid as the operating environment changes. Our proposed D2D resource optimization scheme for unknown CSI settings is applicable to a sub-class of scenarios which can be framed as the graph-theoretic weighted bipartite matching (WBM) problem. In the absence of CSI, WBM becomes a combinatorial problem with unknown random edge weights and generally unknown distributions. To compensate for this lack of knowledge, we resort to the formalism of combinatorial multi-armed bandit (CMAB) from machine learning theory. A CMAB-based network controller is able to reach optimal D2D configuration by following efficient rules which strike a balance between exploring alternative matchings and exploiting the gradually gained experience. We formulate and numerically experiment with two problems as proof of concept, namely (i) D2D relay selection, and (ii) joint mode selection and channel allocation. We also compare our results with existing/baseline schemes with various flavors of CSI availability assumptions, including perfect instantaneous CSI, perfect statistical CSI, erroneous CSI, and static CSI.

Automated summary

In this paper, the problem of D2D resource allocation under unknown channel state information (CSI) is addressed. A network controller based on combinatorial multi-armed bandit (CMAB) is proposed to achieve optimal D2D configuration without CSI. Numerical experiments are conducted to verify the effectiveness of the proposed scheme in D2D relay selection and joint mode selection and channel allocation.