Electrical and Computer Systems Engineering

1
Optimizing Handovers in Next-Generation Wireless
Heterogeneous Networks Xiaohuan (Iris) Yan,
Department of Electrical and Computer Systems
EngineeringSupervisors Dr. Nallasamy Mani and
Dr. Ahmet Sekercioglu
Mobile high speed communications have become an
integral part of modern society Mobile
networks now cover 80 of the worlds population,
double the level in 2000. Future wireless
networks must be able to coordinate services
within a diverse network environment 1. Such
networks are very beneficial for disaster
recovery, security, navigation, e-learning and
etc.
Methodology
The flow chart of our handover decision model is
shown below.
We have built up an adaptive handover algorithm
to minimize the number of unnecessary handovers
from cellular networks to WLANs. Using the
received signal strength change rate information,
we predict the traveling distance d inside the
WLAN. A distance threshold L dependent on
various network parameters is introduced in our
algorithm. Only when d gt L is handover process
considered to be initiated.

• Research question
• Next-generation heterogeneous networks will
  include free-access WLAN hotspots as well as paid
  cellular access.
• Users want to enjoy wider variety of services
  and better quality of service, such as wider
  coverage for high speed multimedia applications.
• My work involves choosing the best network while
  maintaining full connectivity.

Key improvements
To analyze the handover failure probability and
unnecessary handover probability for different
handover decision methods, we assume that the
target pf and pu are 0.02 and 0.04,
respectively 4. For different velocities of the
mobile terminal (v) and expected distance
thresholds (C), results of the probability of
handover failures and unnecessary handovers for
Varma's, Mohanty's and our method are shown below.

• Previous studies
• In conventional handover decision methods such
  as Varma et al.'s method described in 3,
  handover from the cellular network to the WLAN
  was initiated once the mobile terminal entered
  the WLAN coverage.
• In scenario A, handing over to the WLAN is
  usually encouraged. However, in scenario B, a
  situation where the mobile terminal travels
  through the marginal area of the WLAN at a fast
  speed, whether handing over to the WLAN is
  necessary needs to be considered.
• Sometimes such handovers cause resource wastage.
  Furthermore, if the handover process has not been
  completed before the mobile terminal leaves the
  WLAN coverage, connection breakdown occurs.

As illustrated by the figures, for large values
of velocities and expected distance thresholds,
our method has lower probability of handover
failures and unnecessary handovers than the other
two methods. Moreover, we should note that with
our model, L can always change to get lower pf
and pu.
References 1 F. Zhu and J. McNair.
Multiservice Vertical Handoff Decision
Algorithms. EURASIP Journal on Wireless
Communications and Networking, vol. 2006, Article
ID 25861, 13 pages, 2006. 2 N. Nasser, A.
Hasswa and H. Hassanein, Handovers in fourth
generation heterogeneous networks,'' IEEE Commun.
Mag., vol. 44, pp. 96-103, Oct. 2006. 3 V. K.
Varma et al., Mobility management in integrated
Umobile terminalS/WLAN networks,'' in Proc. IEEE
ICC 2003, pp. 1048-1053.

• Recently in 4, Mohanty improves the
  conventional method by making the mobile terminal
  remain connected with both networks whilst
  staying in a boundary cell of the WLAN in order
  to avoid a ping-pong effect and connection
  breakdown. However, the wastage caused by
  unnecessary handovers was not considered.

4 S. Mohanty, A new architecture for 3G and
WLAN integration and inter-system handover
management,'' Wireless Networks, vol. 12, pp.
733-745, 2006.
Electrical and Computer Systems Engineering
Postgraduate Student Research Forum 2007