arXiv subjects

Ramkumar Raghu

Publications and source records attributed to Ramkumar Raghu.

4 recordsLinked to original sources

arXiv2021

Queueing Theoretic Models for Multiuser MISO Content-Centric Networks with SDMA, NOMA, OMA and Rate-Splitting Downlink

Multiuser, Multiple Input, Single Output (MU-MISO) systems are proving to be indispensable in the next generation wireless networks such as 5G and 6G. The spatial diversity of MISO systems have been leveraged in physical layer designs in these wireless systems to improve the capacity. Several recent studies have utilised redundancies in the content request along with the spatial diversity of a MISO system to improve the capacity further. It is shown that Max-Min Fair (MMF) Beamforming schemes for MISO based on SDMA, NOMA, OMA and Rate-Splitting could be used to improve the content delivery rates. However, in most of these studies the key aspects such as the queueing delays in the downlink and the user dynamics have generally been ignored. In this work, we study how the interplay between queueing, beamforming and the user dynamics affects the Quality-of-Service (user experienced delay) of downlink in MU-MISO content centric networks (CCNs). We propose queueing theoretic models that are simple in nature and can be directly adapted to MU-MISO CCNs to perform optimal multi-group multicast downlink transmissions. We show that a recently developed Simple Multicast Queue (SMQ) for SISO systems can be directly used for MU-MISO systems and that it provides superior performance due to its always-stable nature. Further, we observe that MMF Beamforming schemes coupled with SMQ can be quite unfair to users with good channels. Thus, we propose an improvement to SMQ called Dual SMQ which addresses this issue. We also provide theoretical analysis of the mean delay experienced by the users in such MU-MISO CCNs.

cs.NI↗

arXiv2020

Scheduling and Power Control for Wireless Multicast Systems via Deep Reinforcement Learning

Multicasting in wireless systems is a natural way to exploit the redundancy in user requests in a Content Centric Network. Power control and optimal scheduling can significantly improve the wireless multicast network's performance under fading. However, the model based approaches for power control and scheduling studied earlier are not scalable to large state space or changing system dynamics. In this paper, we use deep reinforcement learning where we use function approximation of the Q-function via a deep neural network to obtain a power control policy that matches the optimal policy for a small network. We show that power control policy can be learnt for reasonably large systems via this approach. Further we use multi-timescale stochastic optimization to maintain the average power constraint. We demonstrate that a slight modification of the learning algorithm allows tracking of time varying system statistics. Finally, we extend the multi-timescale approach to simultaneously learn the optimal queueing strategy along with power control. We demonstrate scalability, tracking and cross layer optimization capabilities of our algorithms via simulations. The proposed multi-timescale approach can be used in general large state space dynamical systems with multiple objectives and constraints, and may be of independent interest.

cs.NI↗

arXiv2019

Deep Reinforcement Learning Based Power control for Wireless Multicast Systems

We consider a multicast scheme recently proposed for a wireless downlink in [1]. It was shown earlier that power control can significantly improve its performance. However for this system, obtaining optimal power control is intractable because of a very large state space. Therefore in this paper we use deep reinforcement learning where we use function approximation of the Q-function via a deep neural network. We show that optimal power control can be learnt for reasonably large systems via this approach. The average power constraint is ensured via a Lagrange multiplier, which is also learnt. Finally, we demonstrate that a slight modification of the learning algorithm allows the optimal control to track the time varying system statistics.

cs.NI↗

arXiv2018

Queuing Theoretic Models for Multicast and Coded-Caching in Downlink Wireless Systems

We consider a server connected to $L$ users over a shared finite capacity link. Each user is equipped with a cache. File requests at the users are generated as independent Poisson processes according to a popularity profile from a library of $M$ files. The server has access to all the files in the library. Users can store parts of the files or full files from the library in their local caches. The server should send missing parts of the files requested by the users. The server attempts to fulfill the pending requests with minimal transmissions exploiting multicasting and coding opportunities among the pending requests. We study the performance of this system in terms of queuing delays for the naive multicasting and several coded multicasting schemes proposed in the literature. We also provide approximate expressions for the mean queuing delay for these models and establish their effectiveness with simulations.

cs.PF↗