Source-linked AI summary
Vertical Handover decision schemes using SAW and WPM for Network selection in Heterogeneous Wireless Networks
K. Savitha, C. Chandrasekar
TL;DR
In heterogeneous wireless networks, handover must maintain service as a mobile terminal moves between overlapping networks, and selecting a target network is a decision challenge. The paper compares SAW and WPM for that decision, reporting lower relative standard deviation for WPM than SAW and selecting WiMax cell A2.
Problem
Choosing a target network during handover in heterogeneous wireless networks is central to maintaining service continuity while reducing processing delay.
Source: §I · p. 2Method
The paper compares SAW and WPM for handover network selection using bandwidth, delay, jitter, and cost as decision parameters.
Source: a) Simple Additive Weighting (SAW)… · p. 4, b) Weighted Product Model (WPM) method · p. 5Results
Relative standard deviation was 12.64% for SAW and 35.75% for WPM; the authors report WPM as better and state that the mobile terminal connects to WiMax base station A2.
Source: iii. Ranking Order for Saw and WPM… · p. 6Takeaways & Limitations
The authors conclude that WPM is the better decision maker by relative standard deviation for selecting a visitor network in their handover setting.
Source: §VI · p. 6
Abstract
from arXiv · showhide
Seamless continuity is the main goal and challenge in fourth generation Wireless networks (FGWNs), to achieve seamless connectivity "HANDOVER" technique is used,Handover mechanism are mainly used when a mobile terminal(MT) is in overlapping area for service continuity. In Heterogeneous wireless networks main challenge is continual connection among the different networks like WiFi, WiMax, WLAN, WPAN etc. In this paper, Vertical handover decision schemes are compared, Simple Additive Weighting method (SAW) and Weighted product model (WPM) are used to choose the best network from the available Visitor networks(VTs) for the continuous connection by the mobile terminal. In our work we mainly concentrated to the handover decision phase and to reduce the processing delay in the period of handover. In this paper both SAW and WPM methods are compared with the Qos parameters of the mobile terminal (MT) to connect with the best network. Keywords: Handover, Vertical handover decision schemes, Simple additive weighting, Weight product method.
I. Introduction
Handover maintains service as a mobile terminal moves through overlapping coverage, including between heterogeneous wireless technologies. This paper focuses on choosing a target visitor network in the handover decision phase using SAW and WPM.
- Handover redirects a mobile user's service from one network, base station, or access point to another to support continuity and reduce processing delay in overlapping areas.Source: §I · p. 2
- In heterogeneous wireless networks, multiple network interfaces allow a mobile station to connect across different access technologies through vertical handover.Source: §I · p. 2
- The handover process comprises initiation, system discovery, decision, and execution; discovery gathers neighboring-network QoS information, while the decision phase compares it with the user's QoS.Source: §I · p. 2
- The paper applies SAW and WPM as decision makers to select a target visitor network, focusing on the handover decision phase.The proposed WPM method is used in a distributed manner and compared with SAW using bandwidth, delay, jitter, and cost.Source: §I · pp. 2–3
II. Related Work
The paper situates distributed handover decisions against centralized schemes and describes network scoring and trust checks. It compares D-VHD and T-DVHD, using MADM-based selection to address processing delay during handover.
II. Related Work
- The paper's related-work overview cites prior comparisons of SAW, MEW, TOPSIS, VIKOR, GRA, and WMC for vertical handover decisions.Source: §II · p. 3
a) Centralized vertical handover decision Schemes
- Centralized information exchange can increase processing delay and overall handover delay; D-VHD schemes were proposed to address this disadvantage.Source: a) Centralized vertical handover… · p. 3
b) Distributed vertical handover decision schemes
- D-VHD targets lower processing delay than C-VHD and uses jitter, cost, bandwidth, and delay to select a suitable visitor network through MADM.A target visitor network is the network the mobile node may connect to after handover.Source: b) Distributed vertical handover… · p. 3
c) Network Selection Function (NSF):
- The network selection function uses SAW or WPM to calculate each target visitor network's quality value, and the highest value is selected.Source: c) Network Selection Function (NSF) · p. 3
d) Distributed Decision scheme:
- T-DVHD extends D-VHD with a trusted handover decision before a request to connect to a new base station is sent.Source: e) Trusted Distributed Vertical… · p. 4
f) Level Of Trust (LOT) test function
- The LOT test gates handover on a threshold; after connection, offered QoS below required QoS lowers LOT, while meeting it raises LOT.Source: f) Level Of Trust (LOT) test function · p. 4
V. SAW and WPM
The section describes SAW as an additive weighted scoring method and WPM as a multiplicative alternative for network selection. A numerical example applies the methods to six candidate networks using voice-application criteria.
a) Simple Additive Weighting (SAW) Method
- SAW scores alternatives by summing normalized attribute values weighted by their relative importance.It requires comparable scales; benefit and worst criteria are normalized using separate equations.Source: a) Simple Additive Weighting (SAW)… · pp. 4–5
b) Weighted Product Model (WPM) method
- WPM replaces addition with multiplication, using attribute weights as exponents across candidate networks and criteria.WPM is also called Multiplicative Exponent Weighting (MEW); the passage identifies positive powers for benefit metrics.Source: b) Weighted Product Model (WPM) method · p. 5
c) Numerical Example
- The numerical example considers six WiFi and WiMax candidate networks using delay, bandwidth, cost, and jitter for a voice application.The user assigns weights to the handover criteria, and the section applies SAW and WPM to the resulting data.Source: c) Numerical Example · p. 5
i. SAW
The SAW procedure normalizes decision-matrix values before applying voice-application weights, while WPM applies weights as exponents. The SAW example selects A4 for service continuity with minimum processing delay.
i. SAW
- A4 is selected by SAW for service continuity with minimum processing delay.SAW first scales the decision-matrix entries to a comparable scale and applies voice-application weights to calculate the alternatives’ weighted averages.Source: i. SAW · p. 5
ii. WPM
- WPM applies the weight factor to alternative scores through multiplication, with weights acting as exponents and no necessary unit transformation.The passage describes WPM as dimensionless because its multiplicative structure eliminates units of measure.Source: ii. WPM · pp. 5–6
iii. Ranking Order for Saw and WPM methods
The section compares network rankings from SAW and WPM, then uses relative standard deviation (RSD) to compare the methods. It reports WPM as better by RSD and identifies WiMax cell A2 as the final connection.
iii. Ranking Order for Saw and WPM methods
- SAW and WPM rank A4 and A2 among the networks associated with less processing delay for seamless handover.The passage does not specify which method corresponds to each of these two ranks.Source: iii. Ranking Order for Saw and WPM… · p. 6, Table 1 · p. 6
- WPM is reported as better by RSD, with SAW at 12.64% and WPM at 35.75%; the final connection is WiMax cell A2.The ranking comparison is presented in Table 1.Source: iii. Ranking Order for Saw and WPM… · p. 6, Table 1 · p. 6
VI. Conclusion
The conclusion compares vertical handover decision schemes and identifies SAW and WPM as decision makers for selecting a visitor network. It reports WPM as the better decision maker by relative standard deviation, with the goal of decreasing processing delay.
VI. Conclusion
- WPM is identified as the better decision maker by relative standard deviation for selecting a visitor network in the handover decision phase.The stated goal is to decrease processing delay through different decision algorithms.Source: §VI · p. 6