Ecole d'ingénieur et centre de recherche en Sciences du numérique

Seminar: Comsys Talk Eurecom - Recent results in coded schemes

Babak H. Khalaj the B.Sc. degree from Sharif University of Technology

Systèmes de Communication

Date: 11 juin 2018

Lieu: Eurecom - Room FOURIER

COMSYS Talk EURECOM ************** Time : 03:00 pm , Monday, 11th of June, Place : Room FOURIER, level - 4 , EURECOM Speaker: Babak H. Khalaj the B.Sc. degree from Sharif University of Technology, Title: Recent results in coded schemes: from wireless coded cashing to coded computation of virtual network functions Abstract : The talk focuses on two topics relying on coded transmission: one in wireless caching and the second on Virtual Network Function computation. On the caching side, based on recent results on Multiserver Coded Caching, we extend the model from DOA computation to finite SNR scenarios. In this work, we consider a single-cell downlink scenario where a multiple-antenna base station delivers contents tomultiple cache-enabled user terminals. Using the ideas from multi-server coded caching (CC) scheme developed for wired networks, a joint designof CC and general multicast beamforming is considered to benefit from spatial multiplexing gain, improved interferencemanagement and the global CC gain, simultaneously. The proposed multicast beamforming strategies utilize the multiantenna multicasting opportunities provided by the CC technique and optimally balance the detrimental impact of both noise andinter-stream interference from coded messages transmitted in parallel. The proposed scheme is shown to provide the samedegrees-of-freedom at high SNR as the state-of-art methods and, in general, to perform significantly better than several baseline schemes including, the joint zero forcing and CC, max-min fair multicasting with CC, and basic unicasting with multiuser beamforming On the VNF computation side, we assume the model where the goal is to compute a sequence of functions in the network, where each function can be computed at only a specific subset of network nodes. In contrast to the conventional approaches that each function have to be computed in only one node, in this work, we allow functions to be computed in more than one node, redundantly in parallel, to respond to a given request. We argue that such redundancy in computation not only improves the reliability of the network, but would also, reduce the overall transmission delay compared to the conventional approaches. In some scenarios, such reduction can even surprisingly scale with the size of the network. Moreover, we show that while finding the subset of nodes for each computation, in general, is a complex integer linear program, approximation algorithms can be proposed to reduce the computational complexity. In fact, for the case where the number of computing nodes for a given function is upper-bounded by a constant, a dynamic programming scheme can be proposed to find the optimum subsets in polynomial times. Biography: Babak H. Khalaj received the B.Sc. degree from Sharif University of Technology, Tehran, Iran,and the M.Sc. and Ph.D. degrees from Stanford University, Stanford, CA, USA, all in Electrical Engineering, in 1989, 1993, and 1996, respectively. At Stanford, he has been among the pioneering team working on adoption of Multi-Antenna arrays in mobile networks. He joined KLA-Tencor in 1995, as a Senior Algorithm Designer working on advanced processing techniquesfor signal estimation. From 1996 to 1999, he has been working at startup companies in Silicon Valley. Since then, he joined Sharif University of Technology as a faculty and has also been a Visiting Professor at CEIT, San Sebastian, Spain, from 2006 to 2007 and the recipient of Alexander von Humboldt Fellowship in 2007–2008. His main research areas are design and analysis of communication networks.

Comsys Talk Eurecom - Recent results in coded schemes

Rechercher