thz channel properties - tut presentation major focus: 0.1 – 3 thz primary: 3 mm – 100 µm ....

Department of Electronics and Communications Engineering

THz Channel Properties

Presented by: Vitaly Petrov, Researcher

Nano Communications Center Tampere University of Technology


q  Growing interest towards the THz band §  Suitability for bandwidth-oriented applications §  Feasible for micro- and nano-scale devices

Devices miniaturisation trend

)DU�EH\RQG��

7LPH

'HYLFHV�VL]H�

'HYLFHV�TXDQWLW\

0DLQ�IUDPHV

3&V

&DUULDEOH�HOHFWURQLFV

,R7

,R17

��

§  Adaptation of communication techniques is required §  Novel research challenges raise


Ubiquitous connectivity

Converged infrastructure for Personal Area Networking


Industry 2008: IEEE 802.15 THz Interest Group (IG)

2013: IG upgraded to a Study Group on 100G

2014: Task Group .3d has been established

Over 300 contributions

Interest growth in numbers

Academia

§ Workshops at INFOCOM and ICC

§  Symposiums at GLOBECOM and ICC

§  IEEE Transactions on THz, 2 Special Issues in JSAC

More than 500 articles

Several proofs of concept for elements


Definition of the THz band

Frequency range Wavelengths Industry, IEEE 802.15.3d 0.3 – 3 THz 1 mm – 100 µm

Academia 0.1 – 10 THz 3 mm – 30 µm

Smart academia 0.06 – 10 THz 5 mm – 30 µm

Current presentation Major focus: 0.1 – 3 THz Primary: 3 mm – 100 µm


Part 1. THz Channel Properties


q  Spatial loss §  E.g. free-space loss for

omnidirectional antennas q  Molecular absorption loss

§  Due to internally vibrating molecules on frequencies similar to signal ones

§  Feature of the THz Band §  Coefficients è from HITRAN database

Propagation and path loss

LP f ,d( ) = 4π fdc0

!

"#

$

%&

2

( )( )df

dfLA ,1,

τ= ( ) ( )dfkdfk IGIGeedf ,,)(, ∑==

−−τ

LT f ,d( ) = LP f ,d( )+ LA f ,d( )


Numerical estimation of losses

0 2 4 6 8 1020

40

60

80

100

120

d = 0.01m.

d = 1m.

d = 1m.

Path loss, dB

f, frequency, THz0 2 4 6 8 10

1 104

0.01

1

100d = 0.01m.

d = 1m.

Absorption loss, dB

f, frequency, THz

0 2 4 6 8

50

100

150

200 d = 0.01m.

d = 0.1m.

d = 1m.

Overall loss, dB

f, frequency, Hz

The most beneficial range is 0.1 – 1 THz

Range of minimal losses


q  Feature of the THz band §  Molecules convert part of the

absorbed energy into kinetic energy

Noise in the THz band (2) Molecular absorption noise

PN f ,d( ) = kBNM f( )= kBT 1−τ f ,d( )"# $%=

= kBT 1− e−k f( )d"

#$%2 4 6 8 10

260

240

220

200

d = 0.01m.

d = 0.1m.

Molecular noise, dB

f, frequency, THz

Noise highly fluctuates through the frequencies

Range of minimal noise level


1 104 1 10

3 0.01 0.1

1

10

100SNR, dB

d, distance, m.

q  Tx/Rx hardware è SNR threshold q  Application requirements è Capacity th. q  (SNR + Capacity) vs distance è Estimation

of effective communication range

SNR and Capacity

1 104 1 10

3 0.01 0.1

1 1011

1 1012

1 1013

1 1014

C, Capacity, Bits/s.

d, distance, m.


q  Distance vs frequency for §  SNR = 10 dB – good performance of

predicted MCSs (discussed further) §  C = 2 Gpbs – sufficient for target

applications §  300 MHz bandwidth

Effective communication range

0 1 2 3 4 5 6 7 8 9 101 10

5

1 104

1 103

0.01

0.1

1SNR = 10dB (C = 1.99Gbps)

d, distance, m.

f, THz

1.  THz channel is highly-frequency selective

2.  Utilisation of so-called “transparency windows” is proposed

0.1 – 3 THz Best performance


Part 2. Transparency Windows and sub-channels


Frequency-selective channel

Window Frequency range Bandwidth Half pulse duration 1 0.10 – 0.54 THz 440 GHz 1.48 ps 2 0.63 – 0.72 THz 95 GHz 6.53 ps 3 0.76 – 0.98 THz 126 GHz 4.92 ps 4 7.07 – 7.23 THz 160 GHz 2.59 ps 5 7.75 – 7.88 THz 130 GHz 3.88 ps

0 2 4 6 8

50

100

150

200 d = 0.01m.

d = 0.1m.

d = 1m.

Overall loss, dB

f, frequency, Hz

q  First transparency window is the most promising

Study 0.1 – 0.54 THz in-depth


q  ~20dB gain over 0.1 – 3 THz (!) §  (10 times in amplitude, 100 in power) §  Sufficient for decoding with major MCS §  Suggested for transmission over

“longer” distances: ≥1 cm

First transparency window, 0.1 – 0.54 THz

0.2 0.4 0.6 0.820

40

60

80

100

120d = 0.01m.

d = 0.1m.

d = 1m.

Overall loss, dB

f, frequency, Hz


Range/capacity trade off. Small channels

1 103 0.01 0.1 1

1

10

1000.10-0.54Thz

0.44-0.54Thz

0.49-0.54Thz

0.10-0.20Thz

0.10-0.15Thz

SNR, dB

d, distance, m.

0.01 0.11 10

8

1 109

1 1010

1 1011

1 1012

1 1013

0.10-0.54Thz

0.44-0.54Thz

0.49-0.54Thz

0.10-0.20Thz

0.10-0.15Thz

C, capacity, bits/s.

d, distance, m.

q  For 10 cm distance: Frequency range (bandwidth) SNR Capacity

0.1 – 0.54 THz (440 GHz) 20 dB 500 Gbps

0.1 – 0.2 THz (100 GHz) 33 dB 300 Gbps

0.1 – 0.15 THz (50 GHz) 35 dB 200 Gbps

Enabling complex MCSs


Range/capacity trade off. Tiny channels

0.01 0.1 1 10 1001

10

1001MHz

10MHz

100MHz

1000MHz

SNR, dB

d, distance, m.0.1 1 10

1 106

1 108

1 1010

1MHz

10MHz

100MHz

1000MHz

C, capacity

d, distance, m.

q  For SNR = 10 dB, Smart metering case Frequency range (bandwidth) Range Capacity (at 1 m)

~0.1 THz (1000 MHz) 2 m 8 Gbps

~0.1 THz (10 MHz) 6 m 0.1 Gbps (100 Mbps)

~0.1 THz (1 MHz) 15 m 0.01 Gbps (10 Mbps)

Applicable for sensing applications


Part 3. Modulation and Coding


q  Limitations of continuous-wave MCS: §  Generating carrier at 1-2THz and higher §  Filtering at higher frequencies §  Energy efficiency Advances in physics are needed

q  On/Off keying q  Transmitting s(t):

§  s(t)=1 è Pulse §  s(t)=0 è Silence

“Low-complex” hardware

On/Off Keying simple modulation

1 1 0 01

tt t t t t

...

v v v v


q  Asymmetric channel: pE1 ≠ pE0 q  Set of threshold è optimisation problem q  BER can be lower than 0.001

BER and throughput estimation for OOK

FEC codes are applicable

0 0.2 0.4 0.6 0.8 10.16

0.18

0.2

0.22

0.24

0.26

Throughput, v=0.0ps

Throughput, v=0.4ps

Throughput, v=0.8ps

Throughput, v=1.2ps

T, thoughput, Tbps

TP, relative energy detection threshold0 0.2 0.4 0.6 0.8 1

1 104

1 103

0.01

0.1

1

v=0.0ps

v=0.4ps

v=0.8ps

v=1.2ps

pE, bit error probability

TP, relative energy detection threshold


Part 4. Summary, open challenges and future research directions


Immersive terahertz networks

System-level performance analysis is required

Link level characteristics: q  Extensive

bandwidth: 0.1–10 THz q  Theoretical

capacity: Tbits/s q  Effective

communication range: 1-5 m

Truly 5G (Beyond 5G) technology

Illustration from Akyildiz et al. “Terahertz band: Next frontier for wireless communications”, 2014

thz channel properties - tut presentation major focus: 0.1 – 3 thz primary: 3 mm – 100 µm ....

Documents