organization-based (OGB) communications hold a
massive share of data- and telecommunication.
Organization members often connect others by
cellular phones on the move. Many mobile
communications happens between organization
members. Furthermore, organizations usually have
regular and frequent contact-targets, mobile calls
generated from an organization usually have locality.
For the properties of moderate amount of users and
the access locality of dialed numbers, OGB caches is
another approach for enhancing mobile NP service.
In this paper we introduce the IP-based iNetwork
(Cheng & Chung 2003) as an example to illustrate
the operation of OGB caches in mobile NP service,
and to demonstrate that OGB caches can efficiently
alleviate the traffic load of NPDB and remarkably
enhance the efficiency of mobile NP service.
The rest of this paper is organized as follow.
Section 2 gives an overview of related work. Section
3 and 4 introduce the operation model of OGB
communication systems and OGB caches. Section 5
investigates the performance of OGB caches in NP
service. A brief conclusion is given in section 5.
2 RELATED WORK
For providing mobile NP service, mobile
communication networks must be able to identify
portable numbers, and to obtain the routing
information of the number. NPDB which keeps the
mapping of portable numbers and routing
information is maintained in the operator network.
The approaches for providing mobile NP service
can be classified into SRF-based and IN-based.
Signal relay function (SRF)-based solutions enhance
the switch functions and utilize the MAP (mobile
application part) protocol to enable the portable
number translation. Depending on the
implementation, the translation can be performed in
the NRH (number range holder) network, which first
issued a telephone number, or in the subscription
network, which a subscriber registered to. The SRF
is typically implemented on signaling transfer points
(STP) in the SS7 communication model, and the
interrogation of NPDB and HLR are processed via
GMSC (gateway MSC).
The simplified NP call process is illustrated in
Figure 1. The origination network, which the calling
party connects to, receives a call initiation request
(step 1). It identifies the callee’s NRH network by
the prefix of the dialed number (MSISDN), and
issues an ISUP IAM message to the NRH GMSC to
initiate a call (step 2). The NRH GMSC consults
HLR and identifies the number was ported (step 3).
It queries NPDB by MAP sending routing
information message (step 4) to determine the
routing address of the callee’s subscription network,
and forwards the IAM message to the subscription
network (step 5). The subscription GMSC queries
HLR for the address (MSRN) of the termination
network (step 6), and routes the request by the
MSRN to set up the call (step 7).
Origination
switch
NRH
GMSC
Subscription
GMSC
Termination
MSC
HLR
HLR
2
4
5
6
7
3
NPDB
SRF
Origination network
Termination network
Number range holder
(NRH) network
Subscription network
1
8
Figure 1: SRF-based NP call routing
The IN-based solutions are implemented on the
service control point (SCP). The IN-based solutions
differ form the SRF-based solutions in the way to
access NPDB. In the SRF-based solution only
GMSC can query NPDB, but in the IN-based
solution every switch equipped with the IN protocol
can access NPDB. The call initiation processes of
IN-based and SRF-based systems are similar.
When a number was ported, the whole group of
numbers were taken as portable numbers. Every NP
call requires a NPDB query for the routing
information to reach the subscription network. When
the amount of users grows rapidly, the numerous
data of users incurs long NPDB query delay. The
considerable amount of queries burdens the load of
NPDB and HLR. NPDB searching becomes the
bottleneck of NP service. The bandwidth reserved
for the caller is occupied during the call setup
process, but operators rarely make profit on it.
The survey of Carpenter et al showed that caches
can offload a substantial amount of traffic at the
database, even with relatively small cache
(Carpenter et al. 2000). However, the effect of a
cache varies with the component and the hierarchy
the cache was applied. Caches can be implemented
to operator switching centers or to the user ends. In
the former case, a large cache is required to
accommodate the enormous dialed numbers and the
corresponding routing information. Mobile
subscribers move in and off a service realm of a
switching center, the numbers dialed by the
extensive amount of subscribers bring about almost
randomly access to the cached data, and result in a
poor cache hit rate. This approach confronts the
problem that either the cache size is too large to be
practical or the cache hit rate is poor to be
AN ORGANIZATION-BASED CACHE MECHANISM FOR SUPPORTING PCS NUMBER PORTABILITY SERVICE
81