THz communications: general issues THz devices for coms (Tx and Rx) Some Reported com links Some conclusions

Transcription

1 THz communications for next generation HD rate wireless links TENXSYS Talk, 2015, June 17th G. Ducournau, M. Zaknoune, P. Szriftgiser, Jean-François Lampin (Tx and Rx) (Tx and Rx) 2 3 THz coms: general issues (1/3): why? Context: how to place T-ray coms on the telecom map? [P.J. Winzer; IEEE Proceedings] - Line of Sight - Point to point (actual) - Single/multi-carrier? THz coms: general issues (2/3): why? Some years ago Changing activities How handle this «big» wireless data??? Today THz coms: general issues (3/3): THz windows «400 GHz» window «600 GHz» window «200/300 GHz» window 1 THz 60 GHz LAN THz 1 GHz Radio comms Satellite comms LMDS WPAN Future fixed pencil links Future mobile accesses 1 MHz Marconi We will use THz carrier frequencies by 2020, a f t e r T. S. B i r d, K e y n o t e t a l k at Asia-Pacific Microwave Conference, Melbourne, A u s t r a l i a, D e c e m b e r Increase backhaul network capability [xxx] THz coms may index that range Applications indoor? (LiFi like?) Concept for optical-to-thz radio links outdoor? Has to be direct line Wall/objects reflexions? Advantage of Photonics: multi-user service (Tx and Rx) Courtesy of SEEM service, Lille University

2 Very usefull for first links Commercially available Major issue Integration with free space From micro-waves or optics Electronic side: Bench-top, waveguide based T X Electronic side: Integrated Sources for T X to THz Multiplication chains: Phase noise (driven by synthesizers) Cost!! Oscillators (fundamental): Can be powerfull (mw) but narrow band Bulky - HEMT InP 35 nm GHz Usefull for first tests (trials, channel investigation) Can be modulated using SHM at output or vectorial THz signaling: - SiGe Bipolar GHz nm CMOS GHz [600 GHz source, IEMN] Oscillators (harmonic): GaAs Schottky Technology nm CMOS GHz THz : 100 GHz (l = 3 mm)? 300 GHz (l = 1 mm)? GaN based future? - 65 nm CMOS GHz (Depend on electrical/optical point of view!) [Electronics Letters 50 (23), , 2015] - 65 nm CMOS GHz THz photomixers Key advantage: «fiber-optic» compatible THz photomixers: some devices Rx: maybe electronic (more efficient) Several structures: photo-conductors, photo-diode are most common in THz now: Photoconductor pionners (Auston, Grischkowski, Brown) Unitraveling carrier photodiodes pionners (Ishibashi, Ito, Nagatsuma) Traveling wave structures, resonant structures UTC-PD devices: up to mw level (0 dbm) demonstrated around 300 GHz band [Song et al., IEEE MWCL (2012)] Silicon photonics (Ge Photodiodes) is pushing the limits: may become a competitor? Less powerfull than THz UTC-PD devices BUT level of integration due to silicon photonics processes! SiGe photodiodes GHz for ex. [Hajimiri, IMS 2015] [Rouvalis et al., [Song et al., IEEE IEEE MTT (2012)] MWCL (2012)] 2 UTC-PD integrated (module): GHz TW-UTC: GHz ( V) NBUTC-PD: GHz Flip-chip on AlN [Wun et al., IEEE PTL (2014)] Advantage of PM devices: the relative bandwidth. 1 device = compatible with multi-carrier THz emission Receivers (Rx): Photonic devices? - photodiodes? - photo-conductors? - (No really efficient material at 1.55 µm) - At 0.8 µm OK but no telecom industry Photonics devices for THz detection? 1.55 µm TDS systems were integrated video bandwidth is narrow. Seems to be not appropriate for coms Good performances (Conv losses 8 db around 300 GHz) Electronic devices: sub-harmonic mixers Usefull for first tests (field trials, channel investigation) GaAs Schottky Technology Can also be modulated using SHM at Tx output for vectorial THz signaling Very usefull for first links Commercially available Inside the Rx the LNAs! Future: Mixed technologies? Depend on application! User (<< cost) or specific (Space, P2P) Other Rx/circuits or specific devices Receivers (Rx): Bench top Electronic devices Ways for THz detection: direct and heterodyne schemes using diodes HETERODYNE DIRECT + LO THz-modulated THz-modulated signal Data signal signal Data signal - 50 ohms loading - 50 ohms loading - Lost of phase info./quadratic detection - Phase information conserved/linear - Phase noise not a problem - Phase noise: high quality required - No L.O. required - L.O. required - Very sensitive 16 Receivers (Rx): Integrated Electronic devices / First chipsets (Tx/Rx modules) IAF (Germany) : mhemt Single devices (Lg -> 20 nm), f T/f max 660/1000 GHz LNA MMICs: GHz, (NF 5 db), 4 stages [M. Schlechtweg, IMS 2015] Sch. Electrical Engineering, Korea University/Teledyne: 0.25 µm DHBT InP Tx/Rx 300 GHz Tx: -2.3 dbm, 18 GHz (3dB BW) [S. Kim et al. TTST 2015] Rx: 20 GHz BW, NF > 12 db. IHCT/Wuppertal SiGe 240 GHz I/Q Tx/Rx, 65 GHz Bandwidth [U. Pfeiffer, IMS 2015] Validated in Tx/Rx transmission over 30 cm Infineon/IHP J-Band amplifiers ( 220 GHz): > 20 db (LNA, 0.13 µm SiGe) Northrop Grumman 480 GHz amplifier (2010), HEMT InP sub 50 nm. [W. Deal, MWCL 2010] 660 GHz amplifier (2012) THz amplifier (2014) 17 Receivers (Rx): # 300 GHz Fully integrated Rx (ASK) InP amp. Up to 12.5 Gbps (BER = -16 dbm) [Song, TMTT July 2013] 0.3 m (25 dbi Tx, UTC-PD) # 300 GHz QPSK Tx and Rx (250 nm InP HBT) Tested on-wafer up to 50 Gbps Tested on-wafer [Song, TMTT March 2014] Resonant Tunelling Diode (RTD) Tx for short distance THz coms (Rx = Schottky) 542 GHz, 2 Gbps / 1 cm distance / BER [Ishigaki Elec. Lett., 2012] Could not be exhaustive!!! 18 2

3 Reported working data links used mainly III/V electronic and/or photonic devices First Links: 120 GHz at NTT: the T-ray coms pionners (Tx and Rx) Integrated dual-freq laser W-band (multiplexed channels) Highest data GHz Including LNA at Rx Highest data 300 GHz: 48 Gbit/s Dual polarization UTC-PD Data formats Propagation of complex THz signaling With environnemental effects??? THz window for first applications (data com): seems to be GHz (Huge BW) Simple amplitude coding: easy but not efficient (0.5 bit/s/hz of spectral data density) Bandwith in THz is huge but some other techniques are possible: High level signaling Multi-carrier frequencies 400 GHz Up to 625 GHz + Comparative tests IR vs THz (Same data rates) Interests: - Sharing bandwidth Basic Medium More Complex - co-existence of THz links with observation services F F

4 Fiber-to-THz bridge using Photomixing Easy Integration with coherent optical networks! Other THz link budget THz links: will be first for Point to Point (P2P), up to km range? Key advantage of THz photonics devices: fiber-optic compatibility Link budget: Create a fiber-to-thz bridge? Only Use photonics with a optical signal!!! QPSK at GHz Order of magnitude of L db 1 km: easy to have 150 db! Large Optics (P2P): 2*50 dbi [Ducournau, Elec. Lett, June 2015] IEEE THz group to study the release of a 100 Gbps standard for THz coms -> 2020? Leader: Thomas Kuerner Still 50 db + Margin required! Links: up to km range? Why not free space optics?? Free space optics has been developped also «Current state of the art?» Devices with affordable cost / Commercially available for Gbps back-haul It works! Technology is very close to be fully operational in the lab. Quite simple to operate First Out-of-the-lab tests ok with packages devices! Weakness with envrionnemental effects compared to T-rays (wavelenght effects) GHz band (Tx and Rx) [Adapted from K. Miyauchi, Millimeter-Wave Communications, ISBN , 1983] TRL achieved between 4 and 5? depending on technologies used «What your vision of the field is in ~20 years time?» Perceived challenges in Terahertz Band New-Generation Mobile Communication Systems Demonstrate THz High Data rates links in intented environment GHz band in use for operator needs (fixed links) or backhauls: very probable Find a way (normalization for ex.) to make THz radio compatible with existing systems Every smartphone connected with THz:??? Depend on back-haul capability! Physical layer but also MAC (Medium Access Control) BUT required: robust components, channel effects known and taken into account in an operational system Show a real gain beyond E-band coms (71-76 and GHz) recently opened Still many things to develop in that field! Technology? (III-V, Photomixers, SiGe, CMOS, ): Will they all be used? Need outdoor tests, environnemental effects, weather effects, Field trials, mixed/hybrid technologies Next 10 years: complete efficient packaged systems to be used in outdoor! Interface with fiber optic world : could be III/V for specific purposes. Perceived opportunities in Terahertz Band New-Generation Mobile Communication Systems It is new => so new activities, new markets, applications, usages, opportunities New frequencies: mandatory to release the saturated electromagnetic spectra! Live THz com demo! Need channel effects modeling for indoor (multipath, signal processing )!!! It should be used it is only a question of time-scale! Silicon photonics technologies applied to THz

5 THz coms at IEMN, Lille THz coms are still in their infancy Photonics-based devices are very attractive for fiber optic compatibility and bandwidth. High distance P2P links Channel effects, Next-gen. UTC-PD devices Solid state InP THz coherent coms THz QPSK 400 GHz Brillouin lasers for THz Narrow THz gen. 1 Gbps@200 GHz UTC-PD for THz Thank you! Year