Joint Path Selection and Rate Allocation Framework for 5G Self-Backhauled mmWave Networks

In this project, a new system design, which exploits multiple antenna diversity, mmWave bandwidth, and traffic splitting techniques, is proposed to improve the downlink transmission. The studied problem is cast as a network utility maximization, subject to an upper delay bound constraint, network stability and network dynamics. 

By leveraging stochastic optimization, the problem is decoupled into: (i) path selection and (ii) rate allocation sub-problems, whereby a framework which selects the best paths is proposed using reinforcement learning techniques. Moreover, the rate allocation is a non-convex program, which is converted into a convex one by using the successive convex approximation method. In this method, furthermore, the results showcase the key trade-off between latency and network arrival rate.

Keywords: Ultra-low latency and reliable communication (URLLC), self-backhaul, mmWave communications, multi-hop scheduling, ultra-dense small cells, stochastic optimization, reinforcement learning.