PROJECT TITLE :

Asynchronous Incremental Stochastic Dual Descent Algorithm for Network Resource Allocation - 2018

ABSTRACT:

Stochastic network optimization problems entail finding resource allocation policies that are optimum on a median but must be designed in an on-line fashion. Such problems are ubiquitous in communication networks, where resources such as energy and bandwidth are divided among nodes to satisfy sure long-term objectives. This Project proposes an asynchronous incremental dual good resource allocation algorithm that utilizes delayed stochastic gradients for polishing off its updates. The proposed algorithm is well-suited to heterogeneous networks as it permits the computationally-challenged or energy-starved nodes to, sometimes, postpone the updates. The asymptotic analysis of the proposed algorithm is administrated, establishing twin convergence under each, constant and diminishing step sizes. It is conjointly shown that with constant step size, the proposed resource allocation policy is asymptotically near-optimal. An application involving multicell coordinated beamforming is detailed, demonstrating the usefulness of the proposed algorithm.