IEEE Wireless Communications - April 2017 - page 116

IEEE Wireless Communications • April 2017
114
1536-1284/17/$25.00 © 2017 IEEE
Digital Object Identifier:
10.1109/MWC.2016.1500303WC
A
bstract
Some of the new trends emerging in future
wireless networks enable a vastly increased fluid-
ity in accessing a wide range of resources, thus
supporting flexible network composition and
dynamic allocation of resources to VNOs. In this
work we study a new resource allocation oppor-
tunity that is enabled by the cloud radio access
network architecture. In particular, we investigate
the relationship between cloud-based antennas
and spectrum as two important resources in virtu-
alized wireless networks. We analyze the interplay
between spectrum and antennas in the context of
an auction-based allocation mechanism through
which VNOs can bid for a combination of the
two types of resources. Our analysis shows that
the complementarity and partial substitutability of
the two resources significantly impact the results
of the allocation of those resources and uncovers
the possibility of divergent interests between the
spectrum and the infrastructure providers.
I
ntroduction
Extreme virtualization [1] sees future networks
as entities composed of consumable commodi-
ties, many of which will be shared and where the
boundary between pure software and hardware
resources is blurred. This pool of commodities
consists of highly heterogeneous resources: infra-
structure, remote radio heads (RRHs), spectrum,
storage, processing power, information content,
knowledge, backhaul, hypervisors, and so on.
In our vision, future virtual network operators
(VNOs) construct their networks from this shared
pool of resources, which enables them to dynam-
ically use and release resources according to their
demand [2].
In this work, we investigate the complex
relationships between network resources and a
mechanism to allocate resources to VNOs. In par-
ticular, we study the substitutability of spectrum
and massive cloud-based antennas. These two
types of network resources are partially substi-
tutable, which means that spectrum and cloud-
based antennas are interchangeable only after the
VNO has acquired a minimum required amount
of spectrum and number of antennas. The partial
substitutability of these resources and the dynam-
ic demand of VNOs for them introduce a new
and interesting resource management problem
that is different from the conventional methods of
spectrum and/or infrastructure management.
We consider cloud-based massive-MIMO (mul-
tiple input multiple output) antennas as a resource
that multiple VNOs can share at the same time.
The shared infrastructure could be provided by
the existing network operators to reduce their
operational costs. Deploying a public network is
another option that can contribute to the pool of
shared infrastructure [3]. Dynamic infrastructure
sharing also creates opportunities for new tech-
nology ventures and opens the market to small
size and local infrastructure providers.
On the other hand, in order to allow operators
to maintain some quality assurances in deploying
new services, we assume that spectrum resourc-
es should be orthogonally allocated to the oper-
ators. Spectrum can be made available through
the new frameworks that are envisioned in the 3.5
GHz Citizens Broadband Radio Service [4] and
licensed shared access (LSA) licensing scheme.
This approach to spectrum and infrastructure
management is in line with the radio access net-
work (RAN) sharing scenario proposed by the
3rd Generation Partnership Project (3GPP) in [5],
where the RAN is shared between the participat-
ing operators while each of them transmits on
orthogonal portions of the licensed spectrum.
In our approach, spectrum is auctioned to the
demanding VNOs, and its final price is determined
by the market. In particular, we propose a hybrid
auction model that is performed by a third party
auctioneer for the combined acquisition of spec-
trum and antennas. The results reveal a complex
interplay between antennas and spectrum in an
auction-based allocation, and show the presence
of potential divergent interests between the spec-
trum and the infrastructure providers. The auction
mechanism and the possible scenarios of having
collocated and distributed cloud-based massive-MI-
MO antennas are summarized in Fig. 1. We will dis-
cuss these scenarios and the auction mechanism
in the following sections. We begin with a brief
introduction to market-based resource allocation
in wireless networks. Major features of the Cloud
RAN architecture are then described. Following
that we describe the auction-based allocation
H
amed
A
hmadi
, I
rene
M
acaluso
, I
smael
G
omez
, L
inda
D
oyle
,
and
L
uiz
A. D
a
S
ilva
S
ubstitutability
of
S
pectrum
and
C
loud
-B
ased
A
ntennas
in
V
irtualized
W
ireless
N
etworks
Hamed Ahmadi is with
University College Dublin.
Irene Macaluso, Ismael
Gomez, Linda Doyle, and
Luiz A. DaSilva are with
Trinity College Dublin.
A
ccepted
F
rom
O
pen
C
all
1...,106,107,108,109,110,111,112,113,114,115 117,118,119,120,121,122,123,124,125,126,...132
Powered by FlippingBook