A High Performance Class of DSTBC for HAPs

Benigno Rodr´ıguez and Juan Pablo Gonz´alez

Institute of Electrical Engineering, Faculty of Engineering, Universidad de la Rep´ublica, Montevideo, Uruguay

Keywords:

Project Management, Blended Learning, Communication, Integration.

Abstract:

High Altitude Platforms (HAPs) has been an interesting topic for years, but today it has a renewed interest

for several companies, organizations and governments. In this paper a high performance class of Differential

Space Time Block Codes (DSTBCs) is analysed for HAPs. The results show that this particular class of

DSTBCs is a good alternative to consider for HAPs communications.

1 INTRODUCTION

HAPs is not a new topic, but it has renewed atten-

tion from the moment that Google

and some gov-

ernments, like the Brazilian one

pay attention to it.

In Google’s web page for “Project Loon” a descrip-

tion of Google project can be read. The idea is to

provide Internet access without using traditional in-

frastructure, in this way places where Internet is still

not available can be served. For this purpose, strato-

spheric balloons powered by solar energy at an alti-

tude of 20 km approximately are considered.

If it is true that there are several organizational

tasks to solve in this particular project, probably

global and giant companies like Google are the ones

who have the best conditions to cope with them.

On the other hand, as it was said, it is not a new

idea

3 4 5

, there is an accumulated amount of know

how over this technology, just by reference Skynet,

Skystation projects are two Japanese projects in these

area. HeliNet and CAPANINA (Eurpean projects)

and ETRI and KARI (Korean projects) are other ex-

amples of research in this area.

Provided that Orthogonal Frequency Division

Multiplexing (OFDM) is a robust and well proved

system in several different applications is natural to

consider it as an alternative and evaluate it. Some dif-

http://www.google.com/intl/es419/loon/?utm

source=google

&utm

medium=cpc&utm campaign=Global semBK es.

http://www.telesemana.com/blog/2013/10/15/al-igual-que-go

ogle-brasil-llevara-internet-a-zonas-remotas-a-traves-de-globos/

http://www.haps.cl/tour

arte5.htm.

http://www.wtec.org/loyola/wireless/0d

14.htm.

http://www.stanford.edu/class/msande237/viewgraphs/Ohmo

ri3.pdf

ferences between HAPs and other wireless systems is

the particular channel and high frequencies used for

them (40 GHz)

The robust Space Time Block Coding (STBC)

technique has already been considered for HAPs ap-

plications

. In this work a particular Differential

Space Time Block Code (DSTBC) technique, pro-

posed for OFDM systems will be evaluated for HAPs

applications.

For more information about the use of MIMO-

OFDM in HAPs communications, the reader can

check Chapter 20 in (Jiang et al., 2010). Just to show

the potential of this technology, a recent trial in Swe-

den can be mentioned, there, operating in the range

of millimeter waves, at 24 km of altitude, a cover-

age area of 60 km in diameter was achieved. Three

important advantages of HAPs compared with satel-

lite systems are that the ﬁrst ones are much more

cheaper, faster to deploy and are placed at less alti-

tude (less distance for the radio link) than the second

ones. Among the applications tested over HAPs are:

Internet access via WiFi, High Deﬁnition Television

(HDTV), mobile communications, etc.

2 CHANNEL MODEL AND

SCENARIOS

The stratospheric channels may cause strong attenua-

tion, Doppler shifts and multipath fading, apart form

atmospheric attenuation due to rain when millimeter

waves band is used. To model this kind of wireless

http://www.haps.cl/documentos/12.pdf.

http://www.haps.cl/documentos/14.pdf.

Rodríguez B. and González J..

A High Performance Class of DSTBC for HAPs.

DOI: 10.5220/0005507200670070

In Proceedings of the 12th International Conference on Wireless Information Networks and Systems (WINSYS-2015), pages 67-70

ISBN: 978-989-758-119-9

 2015 SCITEPRESS (Science and Technology Publications, Lda.)

communications three scenarios can be considered,

“Parking on land”, “Takeoff and landing” and “Fly-

ing platform” (as in footnote 6). The evaluation of the

HAPs in these three scenarios is representative of its

general performance during operation.

The system is considered to work around a central

frequency ( f

) of 40 GHz. For “Parking on land” sce-

nario a velocity of 3 m/s was considered, for “Takeoff

and landing” scenario the velocity was of 15 m/s and

for the “Flying platform” scenarios a velocity of 30

m/s was used (as in footnote 6).

As channel model, a Wide Sense Stationary with

Uncorrelated Scattered (WSSUS) channel was con-

sidered. In order to estimate the maximum time delay

for multipath fading (τ

max

) the decision to consider

reﬂections inside a radius of 200 meters was made.

At 40 GHz the electromagnetic waves are atten-

uated by rain falls, but in this work this natural phe-

nomena is not considered. To have a raw idea of how

millimeter wave are attenuated with rain falls (ITU-R,

2005), (Barabino and Rodr´ıguez, 2013) and (El-Disi,

2010) can be consulted.

On Table 1 the simulation parameters used for

these three scenarios are shown.

Table 1: Simulation parameters.

Parameter Value

Carrier Frequency f

= 40 GHz

Bandwidth B = 16 MHz

Number of subcarriers N

FFT

= 128

Subcarrier spacing ∆f =

FFT

= 125 KHz

Symbol Duration T

= 8 µs

Guard interval T

= 1 µs

Symbol interval T

S+G

= T

+ T

= 9 µs

Number of paths P = 30

Number of clusters (groups of paths) N

= 1

Maximum time delay τ

max

= 0, 8895//0, 6313//0, 6067 µs

HAP velocity v = 3//15//30 m/s

Maximum Doppler shift f

Dmax

= f

≈ 400//2000//4000 Hz

using f

= f

Time Delay distribution b =

max

ln(1000)

= 0, 1288//0, 0914//0, 0878 µs

3 TRANSMISSION SYSTEM

As a transmission system, a DSTBC OFDM sys-

tem with 2 transmit antennas and 1 receive antenna

(2x1) is considered. It has been extensively proved

as an efﬁcient and very robust system. Particularly

in this work the DSTBC class deﬁned in (Rodr´ıguez

and Rohling, 2006) (“4A16PSK PCM2”) and exten-

sively analysed in (Rodr´ıguez and Rohling, 2007),

(Rodr´ıguez, 2007), (Rodr´ıguez, 2012) is evaluated

for HAPs applications. The description of 4A16PSK

PCM2 can also be read from Chapter 10 in (Jiang

et al., 2010).

4 RESULTS

To estimate the performance of 4A16PSK PCM2

technique for HAPs applications, three representative

scenarios were considered as it was previously ex-

plained. In this section the results obtained for each

scenario are discussed.

• PARKING ON LAND

The simulation parameters considered for this sce-

nario were: f

= 40 GHz, v = 3 m/s and τ

max

0,8895 µs.

In Fig. 1, the performance of 4A16PSK PCM2

technique is compared with conventional 64-PSK

modulation used in DSTBC. By comparing the aver-

age performance obtained for 4A16PSK PCM2 and

for 64-PSK a good advantage for the ﬁrst one can

be observed. In particular for BER = 1x10

−1

improvement of 4,6 dB can be obtained by using

4A16PSK PCM2 instead of 64-PSK. These results

were compared without using any channel coding

technique in order to compare directly the perfor-

mance difference between these two techniques (the

proposed in (Rodr´ıguez and Rohling, 2006) and the

reference technique -conventional 64-PSK-). How

things change when channel coding is used is anal-

ysed in (Rodr´ıguez, 2007). There the reader can

see that the advantage for 4A16PSK PCM2 is main-

tained but a little bit diminished; how much can de-

pend on the channel coding technique considered, in

(Rodr´ıguez, 2007) convolutional coding and Viterbi

decoding was considered.

• TAKEOFF AND LANDING

At this scenario the simulation parameters were: f

40 GHz, v = 15 m/s and τ

max

= 0, 6313 µs.

In Fig. 2 an improvement of 10 dB can be ob-

tained for BER = 1x10

−1

. For this scenario the im-

provement is much higher than for the previous one.

The convenience of considering 4A16PSK PCM2 is

very clear. Further more, considering the simplicity

of this technique, this should be had in account for

these kind of applications.

15 20 25 30 35

−5

−4

−3

−2

−1

SNR [dB]

BER

64−PSK average

4A16PSK PCM2 realizations

4A16PSK PCM2 average

Figure 1: BER versus SNR results for “Parking on Land”

scenario.

15 20 25 30 35

−2

−1

SNR [dB]

BER

64−PSK average

4A16PSK PCM2 realizations

4A16PSK PCM2 average

Figure 2: BER versus SNR results for “Takeoff and Land-

ing” scenario.

• FLYING PLATFORM

The parameters considered for this scenario were:

= 40 GHz, v = 30 m/s and τ

max

= 0, 6067 µs.

In Fig. 3 is also very evident the advantage of us-

ing 4A16PSK PCM2 instead of 64-PSK in DSTBCs.

In this ﬁgure, something that can take the reader’s at-

tention is that BER do not improves enough when the

SNR is increased. To explain it, the results in Fig.

4 are shown, there the same scenario is analyzed but

this time with a central frequency of 4 GHz instead of

40 GHz, then a more natural behaviour for BER ver-

sus SNR is observed. What happens is that for central

frequency of 40 GHz, a velocity of 30 m/s is very

high one, and it deteriorates the performance deﬁni-

tively, then the improvement in SNR do not play a

perceptively role in the general performance. When

the central frequency is diminished to 4 GHz, then for

this frequency 30 m/s is not a so high velocity and

does not deteriorate the performance in a so severe

way than previously; then for this case when SNR is

increased BER is improved in a much more natural

way.

15 20 25 30 35

−2

−1

SNR [dB]

BER

64−PSK average

4A16PSK PCM2 realizations

4A16PSK PCM2 average

Figure 3: BER versus SNR results for “Flying Platform”

scenario.

The parameters considered in Fig. 4 simulation

were: f

= 4 GHz, v = 30 m/s and τ

max

= 0, 6067 µs.

15 20 25 30 35 40 45 50 55 60 65

−5

−4

−3

−2

−1

SNR [dB]

BER

64−PSK average

4A16PSK PCM2 realizations

4A16PSK PCM2 average

Figure 4: BER versus SNR results for “Flying Platform”

scenario (f

= 4 GHz).

5 CONCLUSIONS

For the three considered scenarios, 4A16PSK PCM2

demonstrated to be a much better option to conven-

tional 64-PSK modulation technique, when DSTBC

OFDM systems are considered for HAPs applica-

tions. In (Rodr´ıguez, 2007) the reader can check that

the difference in complexity between these two tech-

niques are not high.

At 40 GHz, even a speed of 30 m/s is high speed

distorting very much the performanceof the transmis-

sion. Apart from that 40 GHz is a frequency that can

be signiﬁcantly affected by rain falls. These are two

challenging aspects of working in this range of fre-

quencies. By considering lower frequencies these two

aspects could be improved.

It is also interesting to remark that by using

receive diversity in the downlink, the performance

can be signiﬁcantly improved (see Section 7 in

(Rodr´ıguez, 2007)). Which is still more interesting

considering that at 40 GHz is not a problem to have

several antennas in the receiver. Also is possible to

have receive diversity in the uplink by using several

receive antennas in the HAP.

Considering the present interest in HAPs, and

what the state of the art of the OFDM techniques can

offer nowadays, HAPs seems to be a very promising

alternative in near future.

ACKNOWLEDGEMENTS

The authors thank to Comisi´on Sectorial de Inves-

tigaci´on Cient´ıﬁca (CSIC). This work was partially

supported by CSIC, Universidad de la Rep´ublica.

REFERENCES

Barabino, N. and Rodr´ıguez, B. (2013). Performance

evaluation of FSO and MMW for the uruguayan

weather conditions. Wireless Personal Communica-

tions, 73:1077–1088.

El-Disi, A. Y. (2010). A thesis on utilizing high alti-

tude platforms (HAPs) to provide wireless communi-

cations coverage to close coverage gaps - case study:

Providing UMTS service to the non-radar coverage

area in the gulf of mexico region (GOMEX).

ITU-R (2005). International telecommunication union ra-

diocommunication sector, speciﬁc attenuation model

for rain for use in prediction method. Recommenda-

tion ITU-R P.838-3.

Jiang, T., Song, L., and Zhang, Y. (2010). Orthogonal Fre-

quency Division Multiple Access Fundamentals and

Applications. CRC Press, Taylor and Francis, Florida,

USA.

Rodr´ıguez, B. (2007). Differential STBC for OFDM based

Wireless Systems. PhD thesis, Technische Universit¨at

Hamburg-Harburg (TUHH), Department of Telecom-

munications.

Rodr´ıguez, B. (2012). Differential space time block codes

for high mobility scenarios. In Proc. of the Interna-

tional Conference on Wireless Information Networks

and Systems (WINSYS 2012), pages 377–381, Rome,

Italy.

Rodr´ıguez, B. and Rohling, H. (2006). A new class of

differential space time block codes. In Proc. of the

International OFDM-Workshop, pages 106–110,

Hamburg, Germany.

Rodr´ıguez, B. and Rohling, H. (2007). Receive diversity in

DSTBC using APSK modulation schemes. In Proc. of

the 12

International OFDM-Workshop, pages 90–

94, Hamburg, Germany.