Exploiting Heterogeneity in Mobile Networks Anna Zakrzewska Ph. D. Thesis December 2013

136 Pages·2014·3.62 MB·English

4G and Beyond
- Exploiting Heterogeneity
in Mobile Networks
Anna Zakrzewska
Ph. D. Thesis
December 2013
(cid:105) (cid:105)
(cid:16)main(cid:17) (cid:22) 2014/4/7 (cid:22) 18:32 (cid:22) page 1 (cid:22) #1
(cid:105) (cid:105)
4G and Beyond
- Exploiting Heterogeneity in Mobile Networks
Anna Zakrzewska
December 2013
Networks Technologies and Service Platforms
DTU Fotonik
Technical University of Denmark
(cid:216)rsteds Plads 343
2800 Kgs. Lyngby
Denmark
(cid:105) (cid:105)
(cid:105) (cid:105)
(cid:105) (cid:105)
(cid:16)main(cid:17) (cid:22) 2014/4/7 (cid:22) 18:32 (cid:22) page i (cid:22) #3
(cid:105) (cid:105)
T.T.T.
Jeg har skrevet et sted,
hvor jeg daglig m(cid:229) se,
det manende tankesprog:
T.T.T.
N(cid:229)r man fłler hvor lidet
man n(cid:229)r med sin (cid:29)id,
er det nyttigt at mindes, at
Ting Tar Tid.
Put up in a place
where it’s easy to see
the cryptic admonishment:
T.T.T.
When you feel how depressingly
slowly you climb,
it’s well to remember that
Things Take Time.
Piet Hein
(cid:105) (cid:105)
(cid:105) (cid:105)
(cid:105) (cid:105)
(cid:16)main(cid:17) (cid:22) 2014/4/7 (cid:22) 18:32 (cid:22) page iii (cid:22) #5
(cid:105) (cid:105)
Preface
This thesis presents a selection of the research work conducted during
my Ph.D. study from June 15th, 2010 until December 6th, 2013.
TheprojectwasmainlydoneintheNetworksTechnologyandService
Platforms group, at the Department of Photonics Engineering, Tech-
nical University of Denmark under supervision of Associate Professors
Michael S. Berger and Sarah Ruepp.
A 6 months research stay took place at Bell Laboratories, Alcatel-
Lucent in Dublin, Ireland, under supervision of Dr. David L(cid:243)pez-PØrez
and Dr. Holger Claussen.
This Ph.D. study, the external research stay and participation in
international conferences was (cid:28)nanced by the Technical University of
Denmark,TheDanishAdvancedTechnologyFoundationthroughproject
SAIRS "Standard Agnostic Intelligent Radio Systems for the High Ca-
pacity Wireless Internet", as well as travel grants from Otto Młnsteds
Fond, Oticon Fond, P. A. Fiskers Fond and ACM N2 Women.
iii
(cid:105) (cid:105)
(cid:105) (cid:105)
(cid:105) (cid:105)
(cid:16)main(cid:17) (cid:22) 2014/4/7 (cid:22) 18:32 (cid:22) page v (cid:22) #7
(cid:105) (cid:105)
Abstract
Currentandfuturemobilenetworkswillconstituteofmultiplecoexisting
RadioAccessTechnologies(RATs),cellsofdi(cid:27)erentsize(macro-,metro-,
pico-, femtocells) forming a Heterogeneous Network (HetNet), and Base
Stations (BSs) of various architectures. This thesis addresses di(cid:27)erent
aspects of mobile networks and focuses on the main challenges resulting
from their heterogeneity.
To e(cid:27)ectively manage this diversity, a novel hierarchical approach
considering all types of resources, spectral, optical and computational is
proposed. Thegenericframeworkcoversallkindsofnetworkheterogene-
ityandcanbebene(cid:28)cialatanystageofnetworkdeploymentoroperation.
Furthermore, adedicatedstudyofRATselectionandresourceallocation
is performed and a new optimisation model is introduced and evaluated.
The results show a signi(cid:28)cant reduction of number of handovers at the
cost of a slight throughput degradation, which leads to a more stable
connectivity in multi-RAT environments.
A new HetNet architecture with control and user plane separation is
thoroughly evaluated. The study focuses on the soft-pilot requirements
and assignment problem, which is solved using a proposed optimisation
model and a number of heuristic approaches. The analysis for a number
of deployment scenarios demonstrates the feasibility of this architecture.
Furthermore, the obtained results show the potential bene(cid:28)ts of the new
architecture in terms of mobility management and energy e(cid:30)ciency.
Network sharing has become a common strategy among network op-
erators and it is important to evaluate the bene(cid:28)ts of such cooperation.
Ateletra(cid:30)ctheoryanalysisforvariousnetworkdimensioningapproaches
is provided and based on the results clear recommendations are given.
v
(cid:105) (cid:105)
(cid:105) (cid:105)
(cid:105) (cid:105)
(cid:16)main(cid:17) (cid:22) 2014/4/7 (cid:22) 18:32 (cid:22) page vi (cid:22) #8
(cid:105) (cid:105)
All the presented concepts are standard agnostic, meaning that they
canbeappliedtoanyRAT,includingtechnologiesthathavethepotential
to become part of the next generation mobile networking. This makes
the proposed solutions universal and in line with the current trends to
design future mobile networks.
(cid:105) (cid:105)
(cid:105) (cid:105)
(cid:105) (cid:105)
(cid:16)main(cid:17) (cid:22) 2014/4/7 (cid:22) 18:32 (cid:22) page vii (cid:22) #9
(cid:105) (cid:105)
ResumØ
Nuv(cid:230)rendeogfremtidigemobilnetvilbest(cid:229)afmultiple, sameksisterende
RadioAccessTeknologiers(RAT),forskelligecellestłrrelser(makro-,me-
tro-, pico- og femtoceller) udgłrende et heterogent net (HetNet), samt
forskelligebasestationsarkitekturer. Denneafhandlingbehandlerforskel-
lige aspekter af mobilnet med fokus p(cid:229) de vigtigste udfordringer skabt
af denne heterogenitet.
Til e(cid:27)ektivt at styre denne diversitet, foresl(cid:229)s her en nyskabende
hierarkisk metode, der tager alle ressourcer, spektrale, optiske og pro-
cessorkraft, ibetragtning. Dettegeneriskeframeworkd(cid:230)kkeralleformer
for netv(cid:230)rksheterogenitet og kan v(cid:230)re gavnlig i alle faser af netv(cid:230)rks-
udrulning og drift. Ydermere er et dedikeret studie af RAT-selektion og
ressourceallokering blevet udfłrt og en ny optimeringsmodel introduc-
eres og evalueres. Resultaterne viser en signi(cid:28)kant reduktion i antallet
af handovers p(cid:229) bekostning af et mindre fald i throughput, hvilket fłrer
til mere stabile forbindelser i multi-RAT miljłer.
En ny HetNet-arkitektur med adskilt control- og user-plane bliver
undersłgt grundigt. Studier her fokuserer p(cid:229) problemet med krav til og
tildeling af soft-pilot, og der foresl(cid:229)s en optimeringsmodel samt et antal
heuristikker. En analyse af en r(cid:230)kke udrulningsscenarier demonstrerer
anvendeligheden af denne arkitektur. Ydermere viser resultaterne de
potentielle fordele ved denne arkitektur med hensyn til mobility man-
agement og energie(cid:27)ektivitet.
Netv(cid:230)rksdeling er efterh(cid:229)nden blevet en udbredt strategi for oper-
atłrer og det er vigtigt at f(cid:229) undersłgt fordelene ved s(cid:229)dan et samarbe-
jde. Her foretages en teletra(cid:28)kteoretisk analyse af forskellige tilgange til
netv(cid:230)rksdimensionering, hvilket udmunder i konkrete anbefalinger.
vii
(cid:105) (cid:105)
(cid:105) (cid:105)
(cid:105) (cid:105)
(cid:16)main(cid:17) (cid:22) 2014/4/7 (cid:22) 18:32 (cid:22) page viii (cid:22) #10
(cid:105) (cid:105)
Alle de pr(cid:230)senterede koncepter er standard-uafh(cid:230)ngige, hvilket vil
sige, at de kan anvendes p(cid:229) enhver RAT inklusive teknologier, der har
potentialet til at blive brugt i n(cid:230)ste generations mobilnet. Dette głr de
foresl(cid:229)ede lłsninger universelle og i tr(cid:229)d med aktuelle tendenser i design
af fremtidens mobilnet.
(cid:105) (cid:105)
(cid:105) (cid:105)
(cid:105) (cid:105)
(cid:16)main(cid:17) (cid:22) 2014/4/7 (cid:22) 18:32 (cid:22) page ix (cid:22) #11
(cid:105) (cid:105)
Acknowledgements
The adventurous journey of my Ph.D. study is about to end. Those
who have met me, quickly recognize that I love travelling. The desire to
travel,meetnewpeople,visitnewplacesandlearnaboutotherculturesis
not much di(cid:27)erent from research. Both require curiosity and dedication,
sometimes careful planning or more intuition and spontaneity. Hereby,
I would like to acknowledge those who have helped me along the way.
I am very much grateful to my supervisors, Associate Professors
Michael S. Berger and Sarah Ruepp for their guidance, encouragement
and trust to let me explore and (cid:28)nd my path. They provided all the
necessary support so that I did not drown while diving in the ocean of
wireless communication research.
With a great delight I thank Dr. Villy B. Iversen who took me for
a trip to unknown regions of teletra(cid:30)c theory. His knowledge and re-
search spirit made it truly explorative.
IthankalsoDr.FabioD’AndregiovannifromZuseInstitutBerlinwho
showed me new better paths to climb the operations research mountain.
I am deeply indebted to Dr. David L(cid:243)pez-PØrez and Dr. Holger
Claussen from Bell Laboratories, Alcatel-Lucent in Dublin who invited
me and guided during my external research stay. I acknowledge their
strong dedication, creativity and enthusiasm that are a great source of
inspiration for me. This adventure was like a balloon (cid:29)ight, showing me
the wireless landscape from an industrial perspective. I am thankful to
all colleagues from Bell Labs for a very warm welcome and all the help.
This external stay would not have been possible without the involve-
ment of Prof. Ejner Nicolaisen and everyone dealing with its formal
aspects. I extend my appreciation to the administrative personnel of
DTU Fotonik. Thank you very much for your help and assistance.
ix
(cid:105) (cid:105)
(cid:105) (cid:105)
(cid:105) (cid:105)
(cid:16)main(cid:17) (cid:22) 2014/4/7 (cid:22) 18:32 (cid:22) page x (cid:22) #12
(cid:105) (cid:105)
I am grateful to my collaborators, colleagues and o(cid:30)ce mates from
theNetworksgroupforfruitfulteamwork, especiallyonthemanagement
framework,valuableadviceandgreatatmospherethatmadethisjourney
much more joyful.
I acknowledge those, who helped (cid:28)xing my ’Ph.D. bike’. Dr. Lars
Staalhagen has been of a great help with all OPNET related ques-
tions. Aleksandra Che¢ko, Dr. (cid:138)ukasz Brewka, Dr. Henrik Christiansen,
Dr.DavidL(cid:243)pez-PØrezandmysupervisorsdedicatedtheirtimetoreview
partsofthisthesis-Ihighlyappreciateallthecommentsandsuggestions.
Mange tak til Dr. Henrik Christiansen for translating the abstract.
I would like to thank Dr. Mirjam Godskesen, who gave me a valuable
Ph.D. driving lesson- I made it through without crashing.
I thank my fellows from IEEE Student Branch DTU for side trips
from the main Ph.D. track and all the events we organised together.
This thesis is a result of slightly more than three years of work but
thejourneystartedmuchearlierandIwouldliketotakethisopportunity
to thank all those that had an in(cid:29)uence on my personal development.
Dr. Leszek Kosza“ka and Prof. Ryszard Zieli«ski from Wroc“aw Uni-
versity of Technology, Poland once upon a time showed me a map of
research and assisted when I was making the (cid:28)rst steps. I greatly ac-
knowledge their support to all my initiatives throughout the years.
IowewordsofgratitudetoHalina(cid:138)api«skaandWojciechJa–kiewicz,
whohelpedmegoalongsomebumpyroadsandshowedthatdreamscome
true through persistence.
I would also like to thank my friends from all over the world who
always (cid:28)ll my suitcase with lots of smile and encouragement. Thanks to
Ola and Artur for taking care of me in Dublin.
I am grateful to aunt Nora and uncle William for support, keeping
their (cid:28)ngers crossed miles away and emergency English consulting.
It is my heart full of feelings to thank Grzegorz, who has been pa-
tiently waiting for my return to begin a new journey.
Finally, I express my enormous gratitude to my mom for her com-
mitment and understanding. I would never reach this point if not her
brave decision nearly 20 years ago. Dziƒkujƒ!
KongensLyngby,December6th,2013 AnnaZakrzewska
(cid:105) (cid:105)
(cid:105) (cid:105)