A Parameters Tuning Algorithms in Wireless
Networks
Hua-Ching Chen
1
, Ben-Bin Chen
3
, Dong-Hui Guo
4
Department of Electronic Engineering
Xiamen University
Xiamen Fujian 361005; China
galaxy.km@gmail.com, chenbenbin@163.com
dhguo@xmu.edu.cn
Hsuan-Ming Feng
2
Department of Computer Science and Information
National Quemoy University
No. 1 University, Rd., Kin-Ning Vallage Kinmen, 892,
Taiwan
hmfeng@nqu.edu.tw
Abstract—The arbitration inter frame space, Contention
window minimum and Contention window maximum are some of
the most important parameters of 802.11e, and the enhanced
parameters tuning algorithm is applied for their adjustment. To
achieve the high quality of service (QoS), priority combinations
strategy with simpleness and effectiveness is proposed. In such a
strategy, the internal competition of business analysis methods is
used to detect the channel busy probability. Via different settings
of the above parameters, the EPT reduces the conflict probability
to complete the performance analysis while retreating the traffic
business to the idle and zero states. Simulation environments are
built for test and validation the better adapted regulation
mechanism with the parameters.
Keywords—802.11e; EDCA; Wireless Network;
I. INTRODUCTION
Different from the wired access, the wireless network
platforms today is widely used to offer the better QoS and
solve the complex and unexpected problems. However, the
fact is that the traditional IEEE 802.11 standard is not able to
guarantee the real-time business model under heavy traffic
loads. The improvement on the accessing mechanisms of the
IEEE 802.11e protocol provides several voice and video
applications in the WLAN (Wireless Local Area Network [3]
and both non-real-time and real-time dataset are offered a
high-speed accessing rate in such a network.
IEEE802.11e standard protocols distinguish the priority of
various types’ applications by the evaluation of arbitration inter
frame space (AIFS), contention window minimum (CWmin),
contention window maximum (CWmax) and internal collision
mechanism. However, it is not guaranteed that the traffic
service is able to achieve the maximum channel utilization in
the standard of IEEE802.11e [4], and the performance analysis
of saturated throughput is presented by combining the three
parameters. A Markov chain model is proposed in Bianchi [1,
2], which uses the re-analysis to approach the suitable
saturation throughput within a generic time slot and finds the
solution of the Markov equilibrium probability for packet
transmission chain.
Lots of related work has been done in [9], the mathematical
model of the average replace is proposed, which can be used to
calculate the probability of the divided packet collisions and
solve the maximum collision probability problem of the traffic
throughput; the modified Bianchi model is applied in [10] to
solve the retry limit testing problem; the detail of the IEEE
802.11e EDCA mechanisms is described in [5, 8] to simulate
and further evaluate the methods in the literature; reference [11]
proposes the EDCA priority to obtain the effects of transferring
delay conditions when changing the queue buffer size.
Reference [14] presents the EDCA learning-based network
analysis of 802.11e model under the condition that AC is
divided into four grades and applies the machine of priority
channel competition.
II.
EDCA MODEL AND MARKOV CHAIN
A. EDCA MODEL
In many analytical WLAN modes, the enhanced distributed
channel access (EDCA) has been approved [12]. The hybrid
coordination function (HCF) is adopted with 802.11 e by the
isolation of various traffic business to offer the higher QoS
requirements in the WLAN system. The HCF controlled
channel access (HCCA) and EDCA methods are proposed by
the important part of the improvement of the distributed
coordination function (DCF) and PCF in the 802.11 e machine
[7]. In study experiments, voice and image data are assigned
to obtain the higher priority for reducing the package collision
in the traffic flow channel. The EDCA competition of the
network accessing channel is similar to the DCF way. In the
EDCA transmission model, it first waits for the fixed IFS
(Inter-frame Space) at the default interval and goes into the
chaotic-backoff timer cycle when the idle state of wireless
medium is detected. It is noted that the transmitted frames will
start to resend the package when the countdown cycle is
completed. The traffic businesses are divided into four access
categories (AC), they includes voice (AC_VO), video
(AC_VI), best effort (AC_BE), and background (AC_BK) in
the EDCA machine. These categories contain various
priorities in the competed buffer of network channel. Traffic
orders of various data types from high to low are declared as
voice (AC[3]), video (AC[2]), best effort (AC[1]) and
background (AC[0]). The proposed model analysis is divided
into two stages in this research. The algorithm makes sure the
analysis of the internal events within the network site.
2014 Eighth International Conference on Software Security and Reliability - Companion
978-1-4799-5843-6/14 $31.00 © 2014 IEEE
DOI 10.1109/SERE-C.2014.49
257
2014 Eighth International Conference on Software Security and Reliability - Companion
978-1-4799-5843-6/14 $31.00 © 2014 IEEE
DOI 10.1109/SERE-C.2014.49
257
2014 Eighth International Conference on Software Security and Reliability - Companion
978-1-4799-5843-6/14 $31.00 © 2014 IEEE
DOI 10.1109/SERE-C.2014.49
257