Adaptive Antennas for Wireless Communications

Transcription

1 Adaptive Antennas or Wireless Communications Jan Hesselbarth University o Stuttgart Institute or Radio Frequency Technology < 1 >

2 Adaptive Antennas or Wireless Communications outline: mobile data growth and the need or adaptive antennas adaptive antenna concepts and aspects - cellular base station / microwave adaptive antennas. wide-angle azimuth beam steering. requency-selective elevation beam steering. broadbanding / requency adaptivity - point-to-point / millimeter-wave adaptive antennas. small-angle θ & ϕ beam steering. wide-angle azimuth beam steering. wide-angle θ & ϕ beam switching conclusion < 2 >

3 exponential growth o wireless data traic volume mobile data volume grows with CAGR o ~ 40 70% [various sources] i.e., increase by actor o until 2020 [Cisco, Feb 2013] < 3 >

4 solutions or tackling the problem o wireless data growth more requency bandwidth more clever coding less power per user can millimeter-wave be an option? (LOS!) or inrared? but: not much let unused between 0.7 and 2.8 GHz cognitive radio (requency sharing) is promising but: today s algorithms are close to Shannon s limit lower power, i.e., shorter range, i.e., small cells & het nets possible capacity increase or the next decade [my guess]: x 2 x 2 x 2 x adaptive antennas come into play here < 4 >

5 wireless mobile versus nomadic not all cellular traic is mobile : [Detecon 2012 ] o-loading o data hot spot traic in small cells makes sense. consequences or signalling, Doppler, pricing the resulting network consists o dierent kind o wireless installations:. wide-area coverage using macro-cells and micro-cells. hot-spot (nomadic) secondary coverage using pico-cells. hot-spot backhaul using mm-wave point-to-point links. hot-spot (ixed) GBps-coverage using LOS-hubs (mm-wave, inrared) < 5 >

6 the wireless network macro cell mm-wave mesh backhaul micro cell Gbit hotspot Gbit hotspot Gbit hotspot. wide-area coverage using macro-cells and micro-cells. hot-spot (nomadic) secondary coverage using pico-cells. hot-spot backhaul using mm-wave point-to-point links. hot-spot (ixed) GBps-coverage using LOS-hubs (mm-wave, inrared) < 6 >

7 dierent needs or antenna adaptivity wide-angle azimuth beam steering requency-selective elevation beam steering broadbanding / requency adaptivity small-angle θ & ϕ beam steering wide-angle azimuth beam steering cellular point-to-point [wikipedia] low gain high gain (< 20 dbi), (> 20 dbi), wide band narrow band (> 15%) (< 15%) antennas antennas /GHz wide-angle θ & ϕ beam switching low req backhaul? probably not??? mm-wave cellular? probably yes < 7 >

8 wide-angle azimuth beam steering or cellular: - or large-area coverage using macro-cells - results in large orm-actor antennas - RF issues: orm actor, weight weight issue addressed: 4 column dual-pol array MHz (-14 db) metalized plastic + carbon composite structures [Huber+Suhner] < 8 >

9 requency-selective elevation beam steering or cellular: - or large-area coverage using macro-cells - requires active radios behind each radiator in the antenna column - RF issues: bandwidth, phase-ront calibration, lexibility, weight complete TX/RX-chain behind each radiator ( LightRadio ) dierent tilt/ootprint or dierent channels/ requencies/ul-dl [Alcatel-Lucent] < 9 >

10 broadbanding / requency adaptivity or cellular: - multi-band cellular covers up to 4:1 requency range ( MHz) - stacked dipoles / stacked crossed dipoles / stacked patches can cover multiple bands λ 1 / 2 λ 2 / 2 - however element spacing in an array should scale with requency, too - interlaced arrays are geometrically complex and prone to high crosspolarization, particularly or non-integer requency ratios solution: connected array ront view side view [Nortel, US 6,211,841 B1, 2001 ] < 10 >

11 broadbanding / requency adaptivity or cellular: adaptive impedance transormation coupling capacitance use o connected array principle: - optimum usage o aperture area in terms o directivity and beam steering capability - beam steering using λ/2 aperture connected array, cont d: - ull requency lexibility - adaptive impedance match needed or the antenna but can be used or the ampliier at the same time < 11 >

12 small-angle θ & ϕ beam steering or mm-waves: long-distance mm-wave backhaul requires high-gain parabolic dish antennas and very careul alignment cost driver due to required manpower dishes with switched ocal plane array or small-angle electronic beam alignment based on inexpensive eed-horn array and numbers o switches easible but cost is an issue switched eed horns mirror < 12 >

13 wide-angle azimuth beam steering or mm-waves: mm-wave wide-angle beam steering is an enabler or GBps wireless adaptive mesh backhaul phased arrays: eeding each array element with a separate transceiver is too expensive. beam orming networks: are very lossy or high requencies and/or or reasonably large number o beams??? examples < 13 >

14 wide-angle azimuth beam steering or mm-waves: beam orming networks: - example 1: 10 GHz Rotman lens (9:9) avg. 50% dissipative loss dummy ports split dielectric Rotman lens [G. Tudosie, 2009] : simulation measurement < 14 >

15 wide-angle azimuth beam steering or mm-waves: beam orming networks: - example 2: 60 GHz Butler matrix (8:8) 5 layer LTCC (Ferro AS6-S, 0.2mm) avg % dissipative loss in the LTCC Butler matrix (and another 80 85% loss in the eed circuitry) LTCC Butler matrix [G. Tudosie, 2009] : AF measurement (blue), simulation (black) < 15 >

16 wide-angle azimuth beam steering or mm-waves: phased arrays, Rotman, Butler do not work - too expensive, too lossy requirement or high eiciency results in - optical space eed beam orming or - multiple-eed parabolic mirrors or lenses primary need or wide-angle steering in azimuth will simpliy the problem (1D mirror or lens) 30 GHz planar TE mode air/metal Luneburg lens : [ C. Hua et al., IEEE Trans. MTT, vol. 61, no. 1, January 2013, pp ] < 16 >

17 wide-angle θ & ϕ beam switching or mm-waves: GBps at the wireless UE requires short range, directed mm-wave beams LOS likely to be very helpul. On the UE-side: GBps on-body multihop network needed - mm-wave? -UWB? < 17 >

18 wide-angle θ & ϕ beam switching or mm-waves: mm-wave multi-beam hotspot - with hemispherical coverage, - with switched-beam pattern, e.g., beams o 32 dbi : - on the surace: sphere 280 λ - using a graded lens: sphere 14 λ multitude o patch arrays on a hemispherical surace principle o Luneburg lens Luneburg lens modiied Luneburg lens allowing planar eed array < 18 >

19 conclusion: - adaptive antennas will ind various applications on the inrastructure side o wireless networks - only quite speciic orms and eatures o antenna adaptivity makes sense rom a point o view o perormance, orm actor, cost - system design needs to take into account adaptive antennas at a very early stage < 19 >