BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:Canada/Eastern BEGIN:DAYLIGHT DTSTART:20210314T030000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZNAME:EDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20201101T010000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZNAME:EST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20201215T045740Z UID:AB94B017-EEC6-4C21-955A-F75EED8CFAAC DTSTART;TZID=Canada/Eastern:20201214T114500 DTEND;TZID=Canada/Eastern:20201214T151500 DESCRIPTION:Walid Saad\, ECE Department\, Virginia Tech\, Blacksburg\, USA\ n\nProfessor\, Fellow of IEEE\n\nTitle: Can Terahertz Communications Provi de High-Rate Highly Reliable Low Latency Communications in 6G Networks?\n\ nAbstract: Communication at high-frequency terahertz (THz) bands is seen a s a staple of the sixth generation (6G) of wireless cellular networks\, du e to the large amount of available bandwidth. However\, 6G systems will ha ve to support\, not only high data rates\, but also highly reliable commun ication links for emerging applications such as advanced wireless virtual reality (VR) systems. In particular\, advanced wireless VR applications wi ll impose new visual and haptic requirements that are directly linked to t he quality-of-experience (QoE) of VR users. These QoE requirements can onl y be met by wireless 6G connectivity that offers high-rate and high-reliab ility low latency communications (HRLLC)\, unlike the low rates usually co nsidered in vanilla 5G ultra-reliable low latency communication scenarios. Guaranteeing HRLLC in THz-enabled 6G systems requires dealing with the un certainty that is specific to the THz channel. Therefore\, in this talk\, after a brief overview on our vision of 6G systems\, we will explore the p otential of THz for meeting HRLLC requirements. In this regard\, we first quantify the risk for an unreliable VR performance through a novel and rig orous characterization of the tail of the end-to-end (E2E) delay. Then\, w e perform a thorough analysis of the tail-value-at-risk (TVaR) to concrete ly characterize the behavior of extreme wireless events crucial to the rea l-time VR experience. We use this analysis to derive system reliability fo r scenarios with guaranteed line-of-sight (LoS) as a function of THz netwo rk parameters. We then present simulation results that show how abundant b andwidth and low molecular absorption are necessary to improve the reliabi lity\, although their effect remains secondary compared to the availabilit y of LoS\, which significantly affects the THz HRLLC performance. We concl ude our talk with an overview on other key open problems in the realms of THz communications and 6G systems.\n\nHalim Yanikomeroglu\, ECE Department \, Carleton University\, Ottawa\, ON\, Canada\n\nProfessor\, Fellow of IEE E\, Fellow of Canadian Academy of Engineering\, Fellow of Engineering Inst itute of Canada\n\nTitle: Wireless Access Architecture: The Next 20+ Years \n\nAbstract: The very fundamental principles of digital and wireless comm unications reveal that the provision of ubiquitous super-connectivity in t he global scale – i.e.\, beyond indoors\, dense downtown or campus-type areas – is infeasible with the legacy terrestrial network architecture a s this would require prohibitively expensive gross over-provisioning. The problem will only exacerbate with even more demanding use-cases of 2030s s uch as UAVs requiring connectivity (ex: delivery drones)\, thus the 3D sup er-connectivity.\n\nThe roots of today’s wireless access architecture (t he terrestrial 4G & 5G cellular network) go back to 1940s. The access arch itecture has evolved substantially over the decades. However\, rapid devel opments in a number of domains outside telecommunications\, including thos e in aerospace and satellite industries as well as in artificial intellige nce\, will likely result in a disruptive transformation in the wireless ac cess architecture in the next 20+ years.\n\nIn this talk\, an ultra-agile\ , dynamic\, distributed\, and partly-autonomous vertical heterogeneous net work (VHetNet) architecture with very low earth orbit satellites (VLEOs)\, high-altitude platform station (HAPS) systems\, and UAV-BSs (UxNB in 3GPP terminology) for almost-ubiquitous super-connectivity will be presented.\ n\nIn the envisioned VHetNet architecture\, the HAPS systems constitute ar guably the most promising tier. For more information on HAPS systems\, ple ase refer to the below papers under review:\n\nhttps://arxiv.org/abs/2007. 15088\n\nhttps://arxiv.org/abs/2007.08747\n\nhttps://arxiv.org/abs/2009.02 771\n\nhttps://arxiv.org/abs/2006.09328\n\nhttps://arxiv.org/abs/2009.0947 7\n\nVincent Wong\, ECE Department\, University of British Columbia\, Vanc ouver\, BC\, Canada\n\nProfessor\, Fellow of IEEE\n\nTitle: Throughput Opt imization for Grant-Free Multiple Access with Multiagent Deep Reinforcemen t Learning\n\nAbstract: Grant-free multiple access (GFMA) is a promising p aradigm to efficiently support uplink access of Internet of Things (IoT) d evices. In this talk\, we present a deep reinforcement learning (DRL)-base d pilot sequence selection scheme for GFMA systems to mitigate potential p ilot sequence collisions. We formulate a pilot sequence selection problem for aggregate throughput maximization in GFMA systems with specific throug hput constraints as a Markov decision process (MDP). By exploiting multiag ent DRL\, we train deep neural networks (DNNs) to learn near-optimal pilot sequence selection policies from the transition history of the underlying MDP without requiring information exchange between the users. While the t raining process takes advantage of global information\, we leverage the te chnique of factorization to ensure that the policies learned by the DNNs c an be executed in a distributed manner. Simulation results show that the p roposed scheme can achieve an average aggregate throughput that is close t o the optimum\, and has a better performance than several heuristic algori thms.\n\nMehdi Bennis\, ECE Department\, University of Oulu\, Finland\n\nP rofessor\, IEEE Fellow\n\nAbstract: This talk will break down the vision o f wireless network edge intelligence at scale in terms of theoretical and algorithmic principles in addition to a number of applications in beyond 5 G/6G.\n\nWei Yu\, ECE Department\, University of Toronto\, Toronto\, ON\, Canada\n\nFellow of IEEE and a Fellow of Canadian Academy of Engineering\n \nTitle: Data-Driven Approaches to Massive MIMO System Design\n\nAbstract: In this talk\, I will illustrate how machine learning can be used to impr ove the design of wireless communication systems. In particular\, we propo se deep learning approaches to design the channel sensing and downlink pre coding strategies for both the time-domain duplex (TDD) and the frequency- domain duplex (FDD) massive MIMO systems at the mmWave frequency. The firs t part of the talk considers the design of TDD system employing hybrid ana log and digital beamforming. In the second part of the talk\, we further s how that deep learning can be used for efficient and distributed channel e stimation\, quantization\, feedback\, and downlink multiuser precoding for an FDD cellular system with rate-limited feedback.\n\nSpeaker(s): Walid S aad\, Mehdi Bennis\n\nAgenda: \n11:45 PM-- 12:00 PM\; Opening and Welcome\ , Mansour Naslcheraghi\, Chair of IEEE YP Montreal\n\n12:00 PM-- 12:30 PM\ ; Talk: Walid Saad\n\n12:30 PM -- 12:35 PM\; Q&A: Walid Saad\n\n12:35 PM-- 01:05 PM\; Talk: Halim Yanikomeroglu\n\n01:05 PM -- 01:10 PM\; Q&A: Halim Yanikomeroglu\n\n01:10 PM-- 01:40 PM\; Talk: Vincent Wong\n\n01:40 PM -- 01:45 PM\; Q&A: Vincent Wong\n\n01:45 PM-- 02:15 PM\; Talk: Mehdi Bennis\n \n02:15 PM -- 02:20 PM\; Q&A: Mehdi Bennis\n\n02:20 PM-- 02:50 PM\; Talk: Wei Yu\n\n02:50 PM -- 02:55 PM\; Q&A: Wei Yu\n\n02:55 PM -- 03:10 PM\; Q&A with all speakers\n\nMontreal\, Quebec\, Canada\, Virtual: https://events .vtools.ieee.org/m/241334 LOCATION:Montreal\, Quebec\, Canada\, Virtual: https://events.vtools.ieee.o rg/m/241334 ORGANIZER:mansour.nch@ieee.org SEQUENCE:28 SUMMARY:Current and Future Trends in 5G/B5G/6G URL;VALUE=URI:https://events.vtools.ieee.org/m/241334 X-ALT-DESC:Description: <br /><p>&nbsp\;&nbsp\;</p>\n<p><strong>Walid Saad\ , ECE Department\, Virginia Tech\, Blacksburg\, USA<br /></strong></p>\n<p ><strong>Professor\, Fellow of IEEE</strong></p>\n<p><strong>Title:</stron g>&nbsp\;Can&nbsp\;Terahertz&nbsp\;Communications Provide High-Rate&nbsp\; Highly&nbsp\;Reliable&nbsp\;Low&nbsp\;Latency&nbsp\;Communications in 6G N etworks?</p>\n<p><strong>Abstract</strong><strong>:&nbsp\;</strong>Communi cation at high-frequency&nbsp\;terahertz&nbsp\;(THz) bands is seen as a st aple of the sixth generation (6G) of wireless cellular networks\, due to t he large amount of available bandwidth. However\, 6G systems will have to support\, not only high data rates\, but also highly&nbsp\;reliable&nbsp\; communication links for emerging applications such as advanced wireless vi rtual reality (VR) systems. In particular\, advanced wireless VR applicati ons will&nbsp\;impose&nbsp\;new visual and haptic requirements that are di rectly linked to the quality-of-experience (QoE) of VR users. These QoE re quirements can only be met by wireless 6G connectivity that offers high-ra te and high-reliability&nbsp\;low&nbsp\;latency&nbsp\;communications (HRLL C)\, unlike the&nbsp\;low&nbsp\;rates usually considered in vanilla 5G ult ra-reliable&nbsp\;low&nbsp\;latency&nbsp\;communication scenarios.&nbsp\; Guaranteeing HRLLC&nbsp\;in THz-enabled 6G systems requires dealing with t he uncertainty that is specific to the THz channel. Therefore\, in this ta lk\, after a brief overview on our vision of 6G systems\, we will explore the potential of THz for meeting HRLLC requirements. In this regard\, we f irst quantify the risk for an unreliable VR performance through a novel an d rigorous characterization of the tail of the end-to-end (E2E) delay. The n\, we perform a thorough analysis of the tail-value-at-risk (TVaR)&nbsp\; to concretely characterize the behavior of extreme wireless events crucia l to the real-time VR experience. We use this analysis to derive system re liability for scenarios with guaranteed line-of-sight (LoS) &nbsp\;as a fu nction of THz network parameters. We then present simulation results that show how abundant bandwidth and&nbsp\;low&nbsp\;molecular absorption are n ecessary to improve the reliability\, although their effect remains second ary compared to the availability of LoS\, which significantly affects the THz HRLLC performance. We conclude our talk with an overview on other key open problems in the realms of THz communications and 6G systems.</p>\n<p> &nbsp\;</p>\n<p><strong>Halim Yanikomeroglu\, ECE Department\, Carleton Un iversity\, Ottawa\, ON\, Canada<br /></strong></p>\n<p><strong>Professor\, Fellow of IEEE\, Fellow of Canadian Academy of Engineering\, Fellow of En gineering Institute of Canada<br /> <br /> </strong></p>\n<p><strong>Title : </strong>Wireless Access Architecture: The Next 20+ Years</p>\n<p><stron g>Abstract</strong><strong>:&nbsp\;</strong>The very fundamental principle s of digital and wireless communications reveal that the provision of ubiq uitous super-connectivity in the global scale &ndash\; i.e.\, beyond indoo rs\, dense downtown or campus-type areas &ndash\; is infeasible with the l egacy terrestrial network architecture as this would require prohibitively expensive gross over-provisioning. The problem will only exacerbate with even more demanding use-cases of 2030s such as UAVs requiring connectivity (ex: delivery drones)\, thus the 3D super-connectivity.</p>\n<p>The roots of today&rsquo\;s wireless access architecture (the terrestrial 4G &amp\; 5G cellular network) go back to 1940s. The access architecture has evolve d substantially over the decades. However\, rapid developments in a number of domains outside telecommunications\, including those in aerospace and satellite industries as well as in artificial intelligence\, will likely r esult in a disruptive transformation in the wireless access architecture i n the next 20+ years.&nbsp\;</p>\n<p>In this talk\, an ultra-agile\, dynam ic\, distributed\, and partly-autonomous vertical heterogeneous network (V HetNet) architecture with very low earth orbit satellites (VLEOs)\, high-a ltitude platform station (HAPS) systems\, and UAV-BSs (UxNB in 3GPP termin ology) for almost-ubiquitous super-connectivity will be presented.</p>\n<p >In the envisioned VHetNet architecture\, the HAPS systems constitute argu ably the most promising tier. For more information on HAPS systems\, pleas e refer to the below papers under review:</p>\n<p><a href="https://arxiv.o rg/abs/2007.15088">https://arxiv.org/abs/2007.15088</a></p>\n<p><a href="h ttps://arxiv.org/abs/2007.08747">https://arxiv.org/abs/2007.08747</a></p>\ n<p><a href="https://arxiv.org/abs/2009.02771">https://arxiv.org/abs/2009. 02771</a></p>\n<p><a href="https://arxiv.org/abs/2006.09328">https://arxiv .org/abs/2006.09328</a></p>\n<p><a href="https://arxiv.org/abs/2009.09477" >https://arxiv.org/abs/2009.09477</a></p>\n<p>&nbsp\;</p>\n<p><strong>Vinc ent Wong\, ECE Department\, University of British Columbia\, Vancouver\, B C\, Canada<br /></strong></p>\n<p><strong>Professor\, Fellow of IEEE </str ong></p>\n<p><strong>Title:&nbsp\;</strong>Throughput&nbsp\;Optimization f or Grant-Free Multiple Access with Multiagent Deep Reinforcement Learning< /p>\n<p><strong>Abstract</strong><strong>:&nbsp\;</strong>Grant-free multi ple access (GFMA) is a promising&nbsp\;paradigm to efficiently support upl ink access of Internet&nbsp\;of Things (IoT) devices. In this talk\, we pr esent a deep&nbsp\;reinforcement learning (DRL)-based pilot sequence selec tion&nbsp\;scheme for GFMA systems to mitigate potential pilot sequence&nb sp\;collisions. We formulate a pilot sequence selection problem for aggreg ate throughput maximization in GFMA systems with&nbsp\;specific throughput constraints as a Markov decision process (MDP). By exploiting multiagent DRL\, we train deep neural&nbsp\;networks (DNNs) to learn near-optimal pil ot sequence selection&nbsp\;policies from the transition history of the un derlying MDP&nbsp\;without requiring information exchange between the user s. While&nbsp\;the training process takes advantage of global information\ , we&nbsp\;leverage the technique of factorization to ensure that the poli cies&nbsp\;learned by the DNNs can be executed in a distributed manner.&nb sp\;Simulation results show that the proposed scheme can achieve&nbsp\;an average aggregate throughput that is close to the&nbsp\;optimum\, and has a better performance than several heuristic algorithms.&nbsp\;</p>\n<p>&nb sp\;</p>\n<p><strong>Mehdi Bennis\, ECE Department\, University of Oulu\, Finland</strong></p>\n<p><strong>Professor\, IEEE Fellow</strong></p>\n<p> <strong>Abstract</strong><strong>:&nbsp\;</strong>This talk will break dow n the vision of wireless network edge intelligence at scale in terms of th eoretical and algorithmic principles in addition to a number of applicatio ns in beyond 5G/6G.</p>\n<p>&nbsp\;</p>\n<p><strong>Wei Yu\, ECE Departmen t\, University of Toronto\, Toronto\, ON\, Canada<br /></strong></p>\n<p>< strong>Fellow of IEEE and a Fellow of Canadian Academy of Engineering </st rong></p>\n<p><strong>Title:&nbsp\;</strong>Data-Driven Approaches to Mass ive MIMO System Design</p>\n<p><strong>Abstract</strong><strong>:&nbsp\;</ strong>In this talk\, I will illustrate how machine learning can be used t o improve the design of wireless communication systems. In particular\, we propose deep learning approaches to design the channel sensing and downli nk precoding strategies for both the time-domain duplex (TDD) and the freq uency-domain duplex (FDD) massive MIMO systems at the mmWave frequency. Th e first part of the talk considers the design of TDD system employing hybr id analog and digital beamforming. In the second part of the talk\, we fur ther show that deep learning can be used for efficient and distributed cha nnel estimation\, quantization\, feedback\, and downlink multiuser precodi ng for an FDD cellular system with rate-limited feedback.&nbsp\;</p><br /> <br />Agenda: <br /><p><strong>11:45 PM-- 12:00 PM\; </strong>Opening and Welcome\, Mansour Naslcheraghi\, Chair of IEEE YP Montreal<strong><br /></ strong></p>\n<p><strong>12:00 PM-- 12:30 PM\;</strong> Talk: Walid Saad</p >\n<p><strong>12:30 PM -- 12:35 PM\;</strong> Q&amp\;A: Walid Saad</p>\n<p ><strong>12:35 PM-- 01:05 PM\;</strong> Talk: Halim Yanikomeroglu</p>\n<p> <strong>01:05 PM -- 01:10 PM\;</strong> Q&amp\;A: Halim Yanikomeroglu</p>\ n<p><strong>01:10 PM-- 01:40 PM\;</strong> Talk: Vincent Wong</p>\n<p><str ong>01:40 PM -- 01:45 PM\;</strong> Q&amp\;A: Vincent Wong</p>\n<p><strong >01:45 PM-- 02:15 PM\;</strong> Talk: Mehdi Bennis</p>\n<p><strong>02:15 P M -- 02:20 PM\;</strong> Q&amp\;A: Mehdi Bennis</p>\n<p><strong>02:20 PM-- 02:50 PM\;</strong> Talk: Wei Yu</p>\n<p><strong>02:50 PM -- 02:55 PM\;</ strong> Q&amp\;A: Wei Yu</p>\n<p><strong>02:55 PM -- 03:10 PM\;</strong> Q &amp\;A with all speakers</p>\n<p>&nbsp\;</p> END:VEVENT END:VCALENDAR