6G: Towards a Fully Digital and Connected World - 6G Wireless Summit
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
6G: Towards a Fully Digital and Connected World Marco Giordani◦, Michele Polese◦, Marco Mezzavilla†, Sundeep Rangan†, Michele Zorzi◦ ◦University of Padova, Department of Information Engineering, Italy † NYU WIRELESS, Tandon School of Engineering, New York University, Brooklyn, NY, USA 6G Wireless Summit, Levi, Finland March 26th, 2019
Fare clic per modificare stile Outline Ø Introduction and Motivations Ø 6G Key Performance Indices (KPIs) Ø 6G Potential Applications Ø 6G Enabling Technologies • Disruptive Communication Technologies • Novel 6G Communication Enabler • Innovative Network Architectures • Integrating Intelligence in the Network Ø Conclusions and Research Directions 6G: Towards a Fully Digital and Connected World 2 6G Wireless Summit, 26th March 2019
Fare clic per modificare stile Introduction and Motivations • From 1G to 5G, passing through UMTS and LTE innovations, each generation of mobile technology has tried to meet the needs of network operators and final consumers • The rapid development of data-centric and automated processes may exceed even the capabilities of emerging 5G systems, thereby calling for a new wireless generation innovation 6G 1-10 Gbps 5G 100-1000 Mbps 2 Mbps 4G 64 Kbps 2.4 Kbps 3G 1G 2G Internet of Massive broadband Voice calling SMS Internet Applications Internet of Things Towards a Fully Digital and Connected World 1980 1990 2000 2010 2020 2025-2030 time 6G: Towards a Fully Digital and Connected World 3 6G Wireless Summit, 26th March 2019
Fare clic per modificare stile Introduction and Motivations 5G is always associated with trade-offs: 6G will contribute to fill the gap between beyond-2020 societal and business demands and what 5G (and its predecessors) can support • Consider potential applications for future connected systems and estimate the key requirements in terms of throughput, latency, connectivity and other factors. • Identify use cases beyond the performance of 5G systems under development today. • Survey emerging technologies that are not available in networks today but have significant promise for future 6G systems (including developments at all layers of the protocol stack) New Spectrum Disaggregation/ virtualization Disruptive Technologies Energy Efficiency Cell-less Networks Artificial Intelligence 6G: Towards a Fully Digital and Connected World 4 6G Wireless Summit, 26th March 2019
6G: Towards a Fully Digital and Connected World 6G Applications and Use Cases 6G Wireless Summit, Levi, Finland March 26th, 2019 6G: Towards a Fully Digital and Connected World 5 6G Wireless Summit, 26th March 2019
Fare clic per modificare stile 6G Use Cases Massive Scale Communication Virtual Reality E-Health High-Speed Mobility Industry 4.0 Tactile Internet Smart Cities Holographic Telepresence Autonomous Driving 6G: Towards a Fully Digital and Connected World 6 6G Wireless Summit, 26th March 2019
Fare clic per modificare stile 6G Applications and Use Cases INDOOR COVERAGE • 80% of the mobile traffic is generated indoor • 5G current cellular networks never really targeted indoor coverage Ø High-frequencies cannot penetrate solid material Ø 5G densification through femtocells (proposed as a solution) presents scalability issues and high deployment and management costs for operators 6G 6G should target cost-aware indoor connectivity solutions autonomously deployed by end-users and managed by the network operators (e.g., wireless relays coupled with indoor communications) 6G: Towards a Fully Digital and Connected World 7 6G Wireless Summit, 26th March 2019
Fare clic per modificare stile 6G Applications and Use Cases MASSIVE SCALE COMMUNICATIONS • 5G networks are designed to support more than 1’000’000 connections per km2 • Mobile traffic will grow 3-fold from 2016 to 2021, pushing the number of connected devices to the extreme (> 500 billion connected things worldwide by 2030) • This will stress already congested networks, which will not guarantee the required QoS 6G 6G targets capacity expansion to offer high throughput and continuous connectivity, even when civil communication infrastructures may be compromised (public safety is main requirement) 6G: Towards a Fully Digital and Connected World 8 6G Wireless Summit, 26th March 2019
Fare clic per modificare stile 6G Applications and Use Cases eHEALTH • OBJECTIVE: revolutionize the health-care sector, e.g., eliminating time and space barriers through remote surgery and guaranteeing health-care workflow optimizations. • Current communication technologies cannot be applied in future health-care Ø high cost and lack of real-time tactile feedback Ø mmWaves can support low-latency, but do not guarantee connection continuity. 6G 6G enhancements will unleash the potential of eHealth applications through innovations like mobile edge computing, virtualization and artificial intelligence 6G: Towards a Fully Digital and Connected World 9 6G Wireless Summit, 26th March 2019
Fare clic per modificare stile 6G Applications and Use Cases INDUSTRY 4.0 and ROBOTICS • OBJECTIVE: digital transformation of manufacturing through Cyber Physical Systems (CPS) and Internet of Things (IoT) services. • Enabling, among other things, Internet-based diagnostics, maintenance, operation, and direct Machine to Machine (M2M) communications in a cost-effective, flexible and efficient way. • CPS will break the boundaries between the physical factory and the cyber space computation. 6G 6G will foster the Industry 4.0 revolution through new semiconductor and IC innovations (e.g., terahertz scale electronic packaging solutions) 6G: Towards a Fully Digital and Connected World 10 6G Wireless Summit, 26th March 2019
Fare clic per modificare stile 6G Applications and Use Cases SMART CITY • OBJECTIVE: life quality improvements, environmental monitoring, traffic control and city management automation • Current cellular systems have been mainly developed for broadband applications • Smart city applications build upon data generated by low-cost and low-energy consuming sensors, which efficiently interact with each other. 6G 6G will seamlessly include support for user- centric M2M communication, promoting ultra-long battery lifetime combined with energy harvesting approaches 6G: Towards a Fully Digital and Connected World 11 6G Wireless Summit, 26th March 2019
Fare clic per modificare stile 6G Applications and Use Cases HOLOGRAPHIC TELEPRESENCE • OBJECTIVE: remotely connect with an increasing amount of digital accuracy • ISSUE: raw hologram, without any optimization nor compression, with colors, full parallax, and 30 fps, would require a daunting 4.32 Tbps data rate. • ISSUE: latency requirement will hit the sub-millisecond, and thousands of synchronized view angles will be necessary (2 tiles for 4K/8K HD, and 12 tiles for VR/AR) 6G 6G will develop architectures and network designs able to digitalize and transfer all the 5 human senses, increasing the overall target data rate 6G: Towards a Fully Digital and Connected World 12 6G Wireless Summit, 26th March 2019
Fare clic per modificare stile 6G Applications and Use Cases UNMANNED MOBILITY (Autonomous Driving) • OBJECTIVE: fully autonomous connected and intelligent transportation systems, offering safer traveling, improved traffic management, and support for infotainment applications (>7000B$) • Unprecedented levels of communication reliability and low end-to-end latency, even in ultra-high mobility scenarios (up to an impressive 1000 km/h). • Sensors (more than 200 per vehicle by 2020) will demand increasing data rates (in the order of terabytes per driving hour), saturating the capacity of traditional technologies. 6G 6G will pave the way for the coming era of connected vehicles through hardware and software advancements and new technologies to eliminate typical 5G latency/reliability/throughput trade offs 6G: Towards a Fully Digital and Connected World 13 6G Wireless Summit, 26th March 2019
6G: Towards a Fully Digital and Connected World 6G Technologies and Innovations 6G Wireless Summit, Levi, Finland March 26th, 2019 6G: Towards a Fully Digital and Connected World 14 6G Wireless Summit, 26th March 2019
4 Fare clic per modificare stile creasing energy and Increasing energy and bandwidth 6G Technologies and Innovations bandwidth ncreasing wavelength Disruptive Communication Technologies – Terahertz Increasing wavelength TeraHertz Visible Light • Frequency bands between 100 GHz and 1 THz TeraHertz à bring Visible Light to the extreme the potentials and 430 THz 770 THz challenges of high-frequency communications. 300 GHz 10 THz 430 THz 770 THz GHz 300 GHz 10 THz • Huge data rates (possible to allocate contiguous chunks of up to 200 GHz of spectrum) 0 dB 0 dB Pathloss [dB] 150 dB Received 00 W0.002 0.004 0.006 Power 0.008 0.01 0.012 [W] 0.020.02W 0.014 0.016 0.018 150 dB 0 dB Pathloss [dB] 150 dB Received 00 W0.002 0.004 0.006 Power 0.008 0.01 0.012 [W] 0.020.02W 0.014 0.016 0.018 PROPAGATION LOSS (compensated using directional antenna arrays, also enabling spatial multiplexing LED1 without increasing the interference) Received Power [W] 102 LEDLED 1 2 Received Power [W] 102 LED2 Distance [m] Distance [m] 10-2 10-2 MOLECULAR ABSORPTION (it is important to choose deployments in frequency bands not severely affected 8 by molecular absorption) 8 86 6 10-5 4 8 6 4 00 0.1 5 10 42 2 6 10-5 4 300 0.1 5 Frequency [THz] 10 Y Room [m] 2 2X Room [m] Frequency [THz] Y Room [m] X Room [m] 6G: Towards a Fully Digital and Connected World 15 6G Wireless Summit, 26th March 2019
4 4 Fare clic per modificare stile 6G Technologies and Innovations Disruptive Communication Technologies – Visible Light Communications • Frequency bands between 430 GHz and 770 THz à complement RF communications by Visible piggybacking Visibleon Lightwide adoption of LED luminaries the Light 430430 THzTHz 770 770 THzTHz • VLC devices can switch between different light intensities to modulate a signal • More mature research than THz (standard for VLC – IEEE 802.15.7 – has been defined) B0 dB 00 W0.002 00 W0.002 0.004 0.004 Received 0.006 0.008 0.008 Received 0.006 Power 0.01 Power 0.01 0.012 0.014[W] 0.012 [W] 0.014 0.016 0.016 0.02 0.018 0.02 0.018 W 0.02W 0.02 INTERFERENCE (limited coverage range, require an illumination source and suffer from shot noise from LEDLED 1 1 other light sources) Received Power [W] Received Power [W] LEDLED 2 2 MULTI-CONNECTIVITY: need to be complemented by RF for the uplink 8 8 6 6 8 8 4 4 6 6 10 2 4 4 2 2 2 Y Room Y Room [m] [m] X Room X Room [m] [m] 6G: Towards a Fully Digital and Connected World 16 6G Wireless Summit, 26th March 2019
• • • • FOR SUBMISSION TO IEEE 30 [40] R. J. Weiler, M. Peter, T. Khne, M. Wisotzki, and W. Keusgen, “Simultaneous millimeter-wave multi-band channel sounding FOR SUBMISSION access in an urbanTO IEEEscenario,” in Proc. Eur. Conf. Antennas Propag. (EuCAP), May 2015, pp. 1–5. 30 [41] A. S. Poon and M. Ho, “Indoor multiple-antenna channel characterization from 2 to 8 GHz.” in Proc. IEEE Int. Conf. Commun. (ICC), 2003, pp. 3519–3523. [40] R. J. Weiler, M. Peter, T. Khne, M. Wisotzki, and W. Keusgen, “Simultaneous millimeter-wave multi-band channel sounding [42] S. Jaeckel, M. Peter, K. Sakaguchi, W. Keusgen, and J. Medbo, “5G Channel Models in mm-Wave Frequency Bands,” in in an urban access scenario,” in Proc. Eur. Conf. Antennas Propag. (EuCAP), May 2015, pp. 1–5. Proc. Eur. Wireless Conf., May 2016, pp. 1–6. [41] A. S. Poon and M. Ho, “Indoor multiple-antenna channel characterization from 2 to 8 GHz.” in Proc. IEEE Int. Conf. [43] A. Ö. Kaya, D. Calin, and H. Viswanathan. (2016) 28 GHz and 3.5 GHz Wireless Channels: Fading, Delay and Angular Commun. (ICC), 2003, pp. 3519–3523. Dispersion. [42] Jaeckel, [44]S.R. C. QiuM.andPeter, I.-T. K. Lu,Sakaguchi, “MultipathW. Keusgen, resolving with andfrequency “5G Channel J. Medbo,dependence for Models in mm-Wave wide-band Frequency wireless channel modeling,” in Bands,”IEEE FOR SUBMISSION TO IEEE 30 Proc. Trans. Wireless Eur.Veh. Conf., Technol., vol.May 2016, 48, no. 1, pp. pp. 1–6. 273–285, 1999. 6G Wireless Summit, 26th March 2019 [43] Kaya, D. [45]A.K.Ö.Haneda, A.Calin, Richter, andand Viswanathan. H.A. F. Molisch, (2016) 28 GHz “Modeling and 3.5 GHz the frequency dependence WirelessofChannels: ultra-wideband Delay and Angular Fading,spatio-temporal indoor FOR SUBMISSION TO IEEE 30 [40] Dispersion. J. Weiler, R.radio M. Peter, channels,” IEEET. Khne, Trans. M. Wisotzki, Antennas Propag., Keusgen, and W.vol. 60, no.“Simultaneous millimeter-wave 6, pp. 2940–2950, 2012. multi-band channel sounding [44] [46]in an R.D. Qiu and urban C.Dupleich, I.-T. access Lu, “Multipath R. S.scenario,” Thom, resolving Proc. G.inSteinb, Eur. J. Luo, with Conf. frequency Antennas E. Schulz, dependence X. Propag. Lu, G. Wang for (EuCAP),wide-band May et al., wireless 2015, channel modeling,” pp. 1–5. “Simultaneous multi-band IEEE channel sounding [41] A.at [40]Trans. Veh. Technol., R.S.mm-Wave J.Poon Weiler,and M. vol. Ho, frequencies,” M. Peter, 48,inno. T.“Indoor Khne, M. pp. 273–285, multiple-antenna Proc. Eur. Conf. 1, Wisotzki, 1999. channel andAntennas W. characterization Propag. Keusgen, (EuCAP),from “Simultaneous 2 to 2016, pp.81–5. millimeter-wave in Proc. IEEE GHz.”multi-band channel Conf. Int.sounding 6G: Towards a Fully Digital and Connected World [45] [47]Commun. P.Haneda, K.in Ky, an urban Richter, I.(ICC), 2003, Carton, A.access and pp. 3519–3523. Karstensen, A.scenario,” W. Fan, Proc. “Modeling Eur. A. F.inMolisch, G.Conf. the frequency F. Pedersen Antennas dependence al., “Frequency etPropag. of ultra-wideband (EuCAP),dependency May 2015, of spatio-temporal pp.channel 1–5. parameters indoor in urban [42] S.LOS [41]radio A. channels,” Jaeckel, S. scenario Poon M.and IEEE Peter, for Trans. mmwave M.K.Ho, Antennas Sakaguchi, W. Keusgen, communications,” “Indoor and Propag., invol. multiple-antenna 60, no. 6, J. Medbo, Proc. Eur. channel pp. “5G2940–2950, Conf. Antennas characterization 2012. Models Channel Propag. from in (EuCAP), 2 to 8mm-Wave GHz.”2016, Bands,” pp. 1–5. Proc. IEEE inFrequency in Int. Conf. [46] [48]Proc. D.Commun. (ICC),S. Thom, Eur. Wireless K.Dupleich, Haneda, R.J.-i. Takada, 2003, G. May Conf.,pp. Steinb, and T. Kobayashi, 2016,J. pp. 3519–3523. E. Schulz, Luo,1–6. X. Lu, G. “Experimental Investigation Wang et al.,of“Simultaneous Frequency Dependence channel multi-band in sounding Spatio-Temporal [43] A.Propagation [42]at S.mm-Wave Ö.Jaeckel, frequencies,” Kaya, D.Behaviour,” M. Calin, Peter, and in in Proc. H.Proc. K. Sakaguchi,Eur. Eur. Conf. Viswanathan. Antennas Antennas (2016) W.Conf. Keusgen, 28 GHz and Propag. (EuCAP), (EuCAP), and 3.5 J.Propag. Medbo, “5GGHz 2016, Wireless Channel pp. pp. 1–5. 1–6. Channels: 2007,Models in mm-Wave Fading, Frequency Delay and Bands,” Angular in [47] P. Ky, I. V. Nurmela Proc. [49]Dispersion. Carton, A. Eur. Wireless Karstensen, “METIS et al., Conf., May W. Fan, Channel G. F. Pedersen Models,” 2016, pp. et al., “Frequency dependency 1–6. Mobile and wireless communications ofEnablers parameters channel for in urban the Twenty-twenty [43]LOS [44] A. scenario Kaya, R.Information C.Ö. Qiu for D. mmwave Calin, and Society, I.-T. Lu,Tech. Rep., “Multipath 2015. andcommunications,” H. Viswanathan. in resolving with Proc. (2016) Eur. 28 GHz frequency Conf. andAntennas 3.5 GHz dependence Wireless for Propag. (EuCAP), wide-bandChannels: Fading, wireless2016, pp. Delay channel 1–5. and Angular modeling,” IEEE [48] K. Haneda, J.-i. Sayeed, Dispersion. [50]Trans. A. M.Veh. Takada, and T. “Deconstructing Technol., Kobayashi, vol. 48, no. multiantenna“Experimental fading 1, pp. 273–285, 1999. Investigation of Frequency Dependence in Spatio-Temporal channels,” IEEE Trans. Signal Process., vol. 50, no. 10, pp. 2563–2579, [44]Propagation [45] K.2002. R.Haneda, Behaviour,” and C. Qiu A. I.-T. Lu, Richter, in and Proc. “Multipath Eur.resolving A. F. Molisch, Antennas Conf.“Modeling Propag. with frequency the (EuCAP), dependence frequency 2007, pp. for wide-band dependence 1–6. wireless channel of ultra-wideband modeling,” spatio-temporal IEEE indoor [49] V. Nurmela Trans. [51]radio et A. Alkhateeb, al., G. “METIS Leus, Veh. Technol., channels,” IEEE and vol. Trans. Channel 48, R.no. Models,” W.1,Heath Antennas pp. 273–285, Propag., Mobile Jr., “Compressed vol.1999. and wireless 60, no. sensing communications based multi-user 6, pp. 2940–2950, for the Enablers wave 2012. millimeter Twenty-twenty systems: How many [45]Information [46] D.measurements K.Dupleich, R. are Haneda,Society, A.S. needed?” Richter, Tech.and Thom, in G.Rep., 2015. A. Proc. F. Steinb, Molisch, J.IEEE Luo, Int. E. Conf. Acoust., “Modeling Schulz, X.theLu, Speech frequency G. dependence et al.,Process. WangSignal (ICASSP), of ultra-wideband “Simultaneous multi-band 2015, pp. 2909–2913. Aprilspatio-temporal channel indoor sounding information on a higher frequency channel. [50] M. M. A.radio [52]at Sayeed, L. “Deconstructing Bencheikh, channels,” mm-Wave Y. Wang, IEEE Trans. frequencies,” multiantenna and H. Antennas in Proc. fading Propag., Eur.He, Conf. channels,” “Polynomial vol. Antennas 60,root IEEE no.finding Propag. pp.Trans. Signal technique 2940–2950, 6, (EuCAP), Process., joint 2012. 2016,for pp. vol. 50, 1–5.DOA DOD 10, pp. 2563–2579, no. estimation in bistatic the spatial correlation matrix derived at one [46]2002. [47] D. P. MIMO radar,”R.Signal Ky,Dupleich, I. Carton, A. Process., S. Thom, G. Steinb, Karstensen, W. 90, J. Luo, vol.Fan, G. no. F.9, pp. 2723–2730, Schulz, E.Pedersen X.et Lu, 2010. al., G. Wang et al., “Frequency “Simultaneous dependency multi-band of channel parameters channelinsounding urban [51] M. A.at [53]LOS Bengtsson Alkhateeb, mm-Wave G. and B. Leus, Ottersten, and frequencies,” R. in W.“Low-complexity Heath Proc. Eur.Jr., Conf. estimators “Compressed Antennas for sensing Propag. distributed based (EuCAP), sources,” multi-user 2016, IEEE Trans. millimeter pp. 1–5. scenario for mmwave communications,” in Proc. Eur. Conf. Antennas Propag. (EuCAP), 2016, pp. 1–5. waveSignal Process., systems: vol. 48, How many [48] K.no. 8, pp. [47]measurements Ky, P.Haneda, 2185–2194, I. Carton, J.-i. areTakada, 2000. needed?” in Proc. A. Karstensen, and IEEE W. T. Kobayashi, Conf. Fan,Int.G.“Experimental Speech Acoust., et F. Pedersen Process. al., “Frequency Investigation Signal of (ICASSP), dependency Frequency of April 2015, channel Dependence pp. 2909–2913. inparameters in urban Spatio-Temporal [52] M.LOS R.L.Keys, “Cubic scenario Bencheikh, [54]Propagation for convolution mmwave Y. Wang, interpolation communications,” and H. He, for “Polynomialdigital in Proc. image Eur. root processing,” Conf. finding Antennas technique Behaviour,” in Proc. Eur. Conf. Antennas Propag. (EuCAP), 2007, pp. 1–6. IEEE for Trans. Propag. joint Acoust., (EuCAP), DOA DOD Speech, 2016, pp. Signal 1–5. estimation in Process., bistatic frequency to another at a much different frequency [48]MIMO 29, Haneda, radar,” J.-i. Signal 1153–1160, pp.Takada, and Process., T.1981. vol. Kobayashi, “Experimental no. 9, pp. 2723–2730, Investigation Dependence Spatio-Temporal Need to define a “transformation function” to relate [49] V.vol. K.Nurmelano.et6,al., “METIS Channel 90,Models,” Mobile and wireless 2010. communications of FrequencyEnablers for the in Twenty-twenty Disruptive Communication Technologies [53] M.Propagation Z.Bengtsson Dai, Z. Behaviour,” and Wang,in and Ottersten, S. Proc. Chen, Eur. “Spatially Conf. “Low-complexity common Antennas estimators sparsity Propag. for based (EuCAP), distributed adaptive 2007, pp. sources,” channel 1–6. IEEE estimation Trans. Signal and feedback Process., vol. 48,for increase multiplexing and system throughput without [55]Information Gao, L.Society, B. Tech. Rep., 2015. IDEA: leverage channel state information acquiredapertures. sources – [49]no. [50] massive 8,Nurmela A.FDD V.M. pp. etMIMO,” al., “METIS 2185–2194, Sayeed, 2000. “Deconstructing Trans. ChannelSignal IEEE multiantenna Process., Models,” fading Mobile vol. and channels,” no. 63,IEEE 23, wirelesspp. Trans. 6169–6183, communications Signal 2015. Process., Enablers vol. 50, no. for10,the pp.Twenty-twenty 2563–2579, [54] M. Mishali R.Information Keys, [56]2002. and Society, “Cubic Y. C.Tech. convolution “Reduce Eldar,Rep., interpolation boost: 2015.andfor Recovering digital arbitrary sets image processing,” of jointly IEEE Trans. sparse Acoust.,vectors,” Speech,IEEE Trans. Signal Signal Process., [50]vol. [51] M. Sayeed, 29, no. vol. A.Process., A.Alkhateeb, 56, pp. 6, G. no. 10, “Deconstructing 1153–1160, Leus, and pp. 4692–4702, multiantenna R. 1981. W. fading channels,” Heath Jr.,2008. “Compressed IEEE sensing Trans. based Signal Process., multi-user millimeter 50, no. vol.wave systems: 2563–2579, 10, pp.How many [55] F. Gorodnitsky Z.2002. I.Gao, [57]measurements and B. D. Wang, L. Dai,areZ.needed?” andRao, S. “Sparse Chen, signal “Spatiallyreconstruction common from sparsity limited based data using adaptive FOCUSS: channel A estimationre-weighted and minimum feedback in Proc. IEEE Int. Conf. Acoust., Speech Signal Process. (ICASSP), April 2015, pp. 2909–2913. for areOBB at a lower Both sub-6 algorithm,” IEEE Leus, Trans. R. Signal W. Heath Jr., 45, “Compressed 3, pp. sensing 600–616, based 1997. multi-user millimeter wave many used at the [51]FDD [52] massive M.norm MIMO,” A.L.Alkhateeb, Bencheikh,G.Y. IEEE Wang, Trans. andand H. Signal Process., He,Process., vol. vol. “Polynomial root63, no. no.finding pp. 6169–6183, 23,technique for joint2015. DOA DOD estimation systems:inHow bistatic M.measurements [56] MIMOMishali and are Y. needed?” C. Eldar, in Proc. “Reduce IEEE and Int. boost: Conf. Acoust., Recovering radar,” Signal Process., vol. 90, no. 9, pp. 2723–2730, 2010. Speech arbitrary Signal sets Process. of jointly (ICASSP), sparse April vectors,” 2015, IEEE pp. 2909–2913. Trans. Signal MmWave System Continuous downlink transmission with uplink acknowledgments [53] L. Bencheikh, [52]Process., vol. 56, M.M.Bengtsson and no. 10,Wang, B.Y. Ottersten, and pp. 4692–4702, “Polynomial H. He,2008. “Low-complexity estimators finding rootfor technique distributed for joint sources,” IEEEDOA Trans.DOD estimation Signal Process., in vol.bistatic 48, [57] Sub-6 GHz System radar,” Signal Process., pp. 2723–2730, 2010. I. MIMO [57] no. F.8,Gorodnitsky pp. 2185–2194, D. Rao, “Sparse and B.2000. vol. 90,signal no. 9,reconstruction from limited data using FOCUSS: A re-weighted minimum SUBMISSION assume that the sub-6 GHz and [53] M. Bengtsson and B. Ottersten, “Low-complexity for sources,” IEEE Trans. Signal Process., vol. 48, 2012. RFnorm [54] R.Chain algorithm,” Keys, IEEE “Cubic convolution Signal Process., Trans. interpolation vol. estimators for digital 45,image pp. distributed no. 3,processing,” 1997. 600–616,IEEE Trans. Acoust., Speech, Signal Process., MIMO no. 8, pp. 2185–2194, 2000. channel DAC vol. 29, no. 6, pp. 1153–1160, 1981. MmWave System FORfrequency [54] R. Keys, “Cubic convolution interpolation for digital image processing,” IEEE Trans. Acoust., Speech, Signal Process., [55] Z. Gao, L. Dai, Z. Wang, and S. Chen, “Spatially common sparsity based adaptive channel estimation and feedback for fading model (e.g., Rayleigh or Ricean). GHz and mmWave mmWave Sub-6 vol. GHz 29, no. 6, pp. 1153–1160, 1981. System FOR SUBMISSION TO IEEE FDD massive MIMO,” IEEE Trans. Signal Process., vol. 63, no. 23, pp. 6169–6183, 2015. TO IEEE as [55] Z. Gao, L. Dai, Z. Wang, and S. Chen, “Spatially common sparsity based adaptive channel estimation and feedback for Transmitter Receiver [56] M. Mishali and Y. C. Eldar, “Reduce and boost: Recovering arbitrary sets of jointly sparse vectors,” IEEE Trans. Signal RF Chain Disruptive Communication Technologies – Full-Duplex H [58] M. Grant, S. Boyd, and Y. Ye, radar,” Signal Process., FDD massive MIMO,” IEEE Trans. Signal Process., vol. 63, no. 23, pp. 6169–6183, 2015. “IEEE standard for information , [56] I. F. Gorodnitsky and B. D. Rao, [58] the proposed strategies can be extended to other[51] arrays Process., vol. 56, no. 10, pp. 4692–4702, 2008. no. 8, pp. 2185–2194, Aug. 2000. 6G Technologies and Innovations systems DAC metropolitan area networks–Specific [56] M. Mishali and Y. C. Eldar, “Reduce and boost: Recovering arbitrary sets of jointly sparse vectors,” IEEE Trans. Signal H layer (PHY) specifications amendment “CVX: [57] I. F. Gorodnitsky and B. D. Rao, “Sparse signal reconstruction from limited data using FOCUSS: A re-weighted minimum array Sub-6 GHznorm 2012. vol.are [56]“Sparse Process., vol. 56, no. 10, pp. 4692–4702, 2008. vol. 29, no. 6, pp. 1153–1160, Dec. 1981. Fare clic per modificare stile 90, no. Base station norm algorithm,” IEEE Trans. Signal Process., vol. 45, no. 3, pp. 600–616, 1997. M. Grant, norm algorithm,” IEEE Trans. Signal Process., layer3:(PHY) metropolitan User equipment [57] I. F. Gorodnitsky and B. D. Rao, “Sparse signal reconstruction from limited data using FOCUSS: A re-weighted minimum a form MmWave System System I. F. Gorodnitsky using additional bandwidth. algorithm,” MmWave Process., vol. 56, no. 10, pp. 4692–4702, Oct. 2008. norm algorithm,” SystemIEEE Trans. Signal Process., vol. 45, no. 3, pp. 600–616, 1997. TX and the RX.estimation requirements-Part The transceiver in base stations and UEs will be capable of TX a signal while also TX vol. 45, MatlabS.software operateof FOR SUBMISSION TO IEEE Enhancements The sub-6 GHz channel is H and the mmWave channel is H. IEEE Boyd, and Sub-6 GHz System M. L. geometries MmWave System specifications forY. co-located, signal reconstruction Transmitter Receiver Bencheikh, Y. Wang, 17 no.Trans. III. S YSTEM AND CHANNEL MODEL [57]technology–Telecommunications for very RF Chain and B. D. from 9, pp. 2723–2730, Sep. Sub-6 GHz System no. 8, pp. 2185–2194, Aug. 2000. http://www.stanford.edu/ boyd/cvx http://www.stanford.edu/ boyd/cvx Gaussian or truncated Laplacian), and the complex coefficients ↵r with “IEEE standard for informationand disciplined area networks–Specific side or control messages à 3, pp.Signal c,i amendment Ye, “CVX: DAC 2010. RF Chain 11: Wireless LAN Rao, “Sparse 600–616, vol. 29, no. 6, pp. 1153–1160, Dec. 1981. aligned, simultaneously. convex limited data Matlab DAC Process., high throughput signal and H.suitable using Mar. 1997. Process., vol. 56, no. 10, pp. 4692–4702, Oct. 2008. He, “Polynomial requirements-Part 3: Enhancements [55] M. Mishali and Y. C. Eldar, “Reduce and boost: Recovering arbitrary sets of jointly sparse FDD massive MIMO,” IEEE Trans. Signal Process., vol. according software f information exchange medium access1 [54] Z. Gao, L. Dai, Z. Wang, and S. Chen, “Spatially comm Fig. 1: The multi-band MIMO system with co-located sub-6 GHz and mmWave a [54] Z. Gao, L. Dai, Z. Wang, and S. Chen, “Spatially common sparsity based adaptive channel technology–Telecommun [55] M. Mishali and Y. C. Eldar, “Reduce and boost: Recover FDD massive MIMO,” IEEE Trans. Signal Process., vol. 63, no. 23, pp. 6169–6183, Dec. 2 The ULAs are considered for ease of exposi [52] M. Bengtsson and B. Ottersten, “Low-complexity estimators for distributed sources,” IEEE Tr [52] M. Bengtsson and B. Ottersten, “Low-complexity estimato MIMO radar,” Signal Process., vol. 90, no. 9, pp. 2723–2 reconstruc [51] M. L. Bencheikh, Y. Wang, and H. He, “Polynomial root finding technique for joint DOA programming in the 60 f FOCU [53] R. Keys, “Cubic convolution interpolation for digital image processing,” IEEE Trans. Acou mod roo [53] R. Keys, “Cubic convolution interpolation for digital ima and hav vol. 45, no We consider a multi-band MIMO system shown in Fig. 1, where ULAs of is
Fare clic per modificare stile 6G Technologies and Innovations Disruptive Communication Technologies – Channel Sparsity • Estimating the mmWave channel is equivalent to estimating the parameters of the channel paths, i.e., the AoA, the AoD, and the gain of each path. • IDEA: exploit the poor scattering nature of the mmWave channel to formulate the mmWave channel estimation problem as a sparse compressed sensing problem: the channel power is concentrated in a few entries of a virtual channel matrix • It is sufficient to estimate the AoAs and AoDs of the dominant paths to be resolved. Example of 6 dominant paths (the location of each square reflects the AoA and time delay of each path) (a) Grayscale of angular-delay domain B. Wang, et al., "Spatial- and Frequency-Wideband Effects in Millimeter-Wave Massive MIMO Systems," in IEEE Trans. Sig. Proc., vol. 66, no. 13, pp. 3393-3406, Jul. 2018. Fig. 4. An illustration of a 6-path SFW channel, where M = 128, N = 12 6G: Towards a Fully Digital and Connected World 18 6G Wireless Summit, 26th March 2019 D. Sparse Channel Representation and Angular-Delay Or-
Fare clic per modificare stile 6G Technologies and Innovations Innovative Network Architectures – Disaggregation/virtualization • IDEA: decouple network/control plane (CP) and forwarding/user plane (UP). Ø SDN offers network programmability and centralization of the control Ø SDN is agile and responsive (traffic flow meets fluctuating needs and demands). Ø SDN is standards-based (e.g., OpenFlow) and vendor-neutral. • IDEA: replace network services provided by dedicated hardware (e.g., network switches) with virtualized software. Ø NFV saves capital and operating expenses NFV and SDN are complementary technologies (SDN executes on an NFV infrastructure) • C. J. Bernardos et al., "An architecture for software defined wireless networking," in IEEE Wireless Communications, vol. 21, no. 3, pp. 52-61, June 2014. • R. Mijumbi, et al., "Network Function Virtualization: State-of-the-Art and Research Challenges," in IEEE COMST, vol. 18, no. 1, pp. 236-262, 2016 6G: Towards a Fully Digital and Connected World 19 6G Wireless Summit, 26th March 2019
Fare clic per modificare stile 6G Technologies and Innovations Innovative Network Architectures – Access/Backhaul integration • Massive 6G data rates technologies à IAB node stack Backhaul Access PDCP PDCP adequate growth of the backhaul capacity RLC MAC RLC MAC PHY PHY SCHED Donor gNB • THz and VLC deployments will call for a massive IAB node increase in the density of access points, which should IAB node be provided with backhaul connectivity to the core network à expensive • IDEA: deploy a fraction of BSs with traditional fiber-like backhaul capabilities and the rest of the BSs connecting to the fiber infrastructures wirelessly. • 6G deployments will introduce new challenges and opportunities • The networks will need higher autonomous configuration capabilities • Out-of-band IAB can be realized to increase the overall network throughput. • M. Polese, et al. , "End- to-End Simulation of Integrated Access and Backahul at mmWaves,” to appear on IEEE CAMAD, Sep. 2018. • M. Polese, et al., "Distributed Path Selection Strategies for Integrated Access and Backhaul at mmWaves", to appear on IEEE GLOBECOM, Dec. 2018 6G: Towards a Fully Digital and Connected World 20 6G Wireless Summit, 26th March 2019
Fare clic per modificare stile 6G Technologies and Innovations Integrating Intelligence in the Network – Learning • BACKGROUND: the signal received at multiple BSs renders a defining signature for the user location and its interaction with the surrounding environment. • BACKGROUND: UEs typically move through predefined paths, and some movements are impossible due to the presence of obstacles, e.g., buildings, walls. • IDEA: account for previous access statistics and use machine learning tools to predict the network behaviors (e.g., by remembering/observing consequences of previous decisions). SUPERVISED LEARNING The amount of data generated will be massive, thus labeling the data may be infeasible. UNSUPERVISED LEARNING Does not need labeling, used to autonomously build complex network representations 6G: Towards a Fully Digital and Connected World 21 6G Wireless Summit, 26th March 2019
Fare clic per modificare stile 6G Technologies and Innovations Mission Planning Integrating Intelligence in the Network – Knowledge sharing and learning Reference Planner Fig. 4: Reference planners result: the desired path cut corners • 5G: At at thebandwidth high frequency, the massive turn to minimize path length and spatial degreesand ofgenerate freedombetter are Behavioral Planner handling unlikely to be fully used by any one cellular operator. Spectrum can be shared, in time and in space, with several performance and energy benefits: Controllers § Reducing deployment costs, if operators share bands and infrastructures Candidate Strategy Generation § Inter-operators access and interference coordination ACC Lane Merge Controller Centering Assist Multiple candidate strategies • 6G: operators and users may also be interested in sharing learned representations ed behavioral of specific network Prediction planning framework deployments Engine and/or use cases § Speed up the network configuration in new markets Predicted surrounding traffic scenarios § Better adapt to new unexpected scenarios which may emerge during network operations RAL P LANNING F RAMEWORK Cost function-based Evaluation ction II, there are shortcomings of Fuel Progress Comfort Safety Consum archical and parallel robot decision -ption Due to real-time constraints, the mo- nsider the effects of imperfect vehi- peration between cars. In this paper, Best Strategy behavioral planning framework that 6G: Towards a Fully Digital and Connected World Fig. 5: The block diagram of Prediction- and Cost-function s of the hierarchical 6G Wireless and2019parallel ar- Summit, 26th March 22
Fare clic per modificare stile 6G Challenges • Circuit design, high propagation loss • Integrate energy characteristics in protocols • Limited coverage, need for RF uplink • Energy vs. high-deployment / MIMO 5 • Need for reliable frequency mapping Efficient and low-power network operations Energy will be at the core of 6G protocols design Extreme multi-connectivity s g e stin Exploit THz, VLC, mmWave and nod sub-6 GHz links ve wer har -po rgy Low Ene Disaggregated and virtualized RAN The networking equipment will not require dedicated hardware Cloud Edge Cell-less architecture The UE connects to the RAN and not to a single cell Virtual Virtual MAC PHY Generic hardware Fig. 4: Architectural innovations introduced in 6G networks. • Slower network operations/security concerns • Scheduling, need for new network design • CU/UP functions ping arestudied has been widely tightly coupled [15], but such capabilities Scalability, The• overcoming of the and interference cell concept management will also enable have never been deeply integrated with the operations and a tight integration of the different 6G communication protocols of cellular networks. 6G networks will exploit technologies. The users will be able to seamlessly tran- a unified interface for localization and communications sition among sub-6 GHz, mmWave, terahertz or VLC to (i)and 6G: Towards a Fully Digital improve control Connected World operations, which can rely on links without manual interventions or configurations in 23 6G Wireless Summit,context 26th March 2019 information (without the need for exchanging the device, which will automatically select the best avail-
6G: Towards a Fully Digital and Connected World Marco Giordani◦, Michele Polese◦, Marco Mezzavilla†, Sundeep Rangan†, Michele Zorzi◦ ◦University of Padova, Department of Information Engineering, Italy † NYU WIRELESS, Tandon School of Engineering, New York University, Brooklyn, NY, USA 6G Wireless Summit, Levi, Finland March 26th, 2019
You can also read