A CONTRIBUTION TO LABORATORY PERFORMANCE

COMPARISONS OF IEEE 802.11 B,G POINT-TO-POINT LINKS

J. A. R. Pacheco de Carvalho

1,2

, P. A. Gomes

1,2

, H. Veiga

1,2

, C. F. Ribeiro Pacheco

, N. Marques

Centro de Informática,

U. de Det. Remota-Dept. de Física,Universidade da Beira Interior, 6201 Covilhã, Portugal

A. D. Reis

2,3

Dept. de Electrónica e Telecomunicações / Instituto de Telecomunicações, Universidade de Aveiro, 3810 Aveiro, Portugal

Keywords: WLAN, Wi-Fi, IEEE 802.11b, IEEE 802.11g, Point-to-Point Links, Wireless Network Laboratory

Performance Measurements.

Abstract: The importance of wireless communications has been growing. Performance is a very relevant issue,

leading to more reliable and efficient communications. Laboratory measurements are made about several

performance aspects of Wi-Fi (IEEE 802.11 b, g) point-to-point links. A contribution is given to evaluation

of this technology through performance comparisons, using available access points from Enterasys

Networks and Linksys. Detailed results are presented and discussed, namely at OSI levels 4 and 7, from

experiments involving TCP, UDP and FTP: TCP throughput, jitter, percentage datagram loss and FTP

transfer rate.

1 INTRODUCTION

Wireless communications are increasingly important

for their versatility, mobility and favourable prices.

It is the case of microwave based technologies, e.g.

Wi-Fi.

The importance and utilization of Wi-Fi have

been growing for complementing traditional wired

networks. Wi-Fi has been used both in ad hoc mode

and infrastructure mode. In this case an access point,

AP, is used to permit communications of Wi-Fi

devices with a wired based LAN through a

switch/router. In this way a WLAN, based on the

AP, is formed. Wi-Fi has reached the personal home,

forming a WPAN, allowing personal devices to

communicate. Point-to-point and point-to-multipoint

configurations are used both indoors and outdoors,

requiring specific directional and omnidirectional

antennas. Wi-Fi uses microwaves in the 2.4 and 5

GHz frequency bands and IEEE 802.11a,b,c,

included in IEEE Std 802.11-2007. Nominal transfer

rates up to 11 (802.11b) and 54 Mbps (802.11 a, g)

are permitted. CSMA/CA is the medium access

control. The 802.11 architecture has been studied in

detail, including performance analysis of the

effective transfer rate (Schwartz, 2005). An

optimum factor of 0.42 was determined for the

effective transfer rate in 11 Mbps point-to -point

links, giving an effective transfer rate of 4.6 Mbps.

Wi-Fi performance is available in indoor

environments (Sarkar & Sowerby, 2006).

Performance has been a very important issue,

resulting in more reliable and efficient

communications. New telematic applications are

specially sensitive to performance, depending on

application requirements. Application

characterization has been discussed (Monteiro &

Boavida, 2000). Several measurements have been

made for 2.4 GHz Wi-Fi (Pacheco de Carvalho et

al., 2008a), as well as WiMAX and high speed FSO

(Pacheco de Carvalho et al., 2008b), (Pacheco de

Carvalho et al., 2008c). In the present work further

Wi-Fi results arise, through OSI levels 4 and 7. Wi-

Fi (IEEE 802.11 b,g) is evaluated and performance

is compared in laboratory measurements of point-to-

point links, using available access points.

The rest of the paper is structured as follows:

Chapter 2 presents the experimental details i.e. the

measurement setup and procedure. Results and

discussion are presented in Chapter 3. Conclusions

are drawn in Chapter 4.

A. R. Pacheco de Carvalho J., A. Gomes P., Veiga H., F. Ribeiro Pacheco C., Marques N. and D. Reis A. (2009).

A CONTRIBUTION TO LABORATORY PERFORMANCE COMPARISONS OF IEEE 802.11 B,G POINT-TO-POINT LINKS.

In Proceedings of the International Conference on Wireless Information Networks and Systems, pages 95-99

DOI: 10.5220/0002189600950099

 SciTePress

2 EXPERIMENTAL DETAILS

Two types of experiments were carried out,

mentioned as Expa and Expc. Expa used Enterasys

RBTR2 level 2/3/4 access points (APa), with

firmware version 6.08.03 (Enterasys Networks,

2005), and 100-Base-TX/10-Base-T Allied Telesis

AT-8000S/16 level 2 switches (Allied Telesis,

2008). The radio cards were similar to the Agere-

Systems model 0118 type. The configuration was for

minimum transmitted power, micro cell, point-to-

point, LAN to LAN mode, using the radio card

antenna. Expc used Linksys WRT54GL wireless

routers (Linksys, 2005), with a Broadcom BCM5352

chip rev0, firmware DD-WRT v24-sp1-10011 (DD-

WRT, 2009) and the same type of level 2 switch.

The wireless mode was set to bridged access point.

In both Expa and Expc, interference free

communication channels were used. WEP

encryption was not activated. No power levels above

the minimum were required as the access points

were very close (30 cm).

Both types of experiments, Expa and Expc, were

made in point-to-point mode using the laboratory

setup shown in Figure 1. Measurements were made

using TCP and UDP communications at OSI level 4,

as mentioned in (Pacheco de Carvalho, 2008c),

permitting network performance results to be

recorded. For a TCP connection, TCP throughput

was obtained. For a UDP test with a given

bandwidth parameter, UDP throughput, jitter and

percentage loss of datagrams were obtained. TCP

packets and UDP datagrams of 1470 bytes size were

used. A window size of 8 kbytes and a buffer size of

the same value were used for TCP and UDP,

respectively. One PC, with IP 192.168.0.2 was the

Iperf server and the other, with IP 192.168.0.6, was

the Iperf client. Jitter, which can be seen as the

smooth mean of differences between consecutive

transit times, was continuously computed by the

server, as specified by RTP in RFC 1889. This

scheme was also used for FTP measurements, where

FTP server and client applications were installed in

the PCs with IPs 192.168.0.2 and 192.168.0.6,

respectively.

Batch command files were written to enable the

TCP, UDP and FTP tests. The results were obtained

in batch mode and written as data files to the client

PC disk.

3 RESULTS AND DISCUSSION

Both APa and APc access points were configured,

for each standard IEEE 802.11 b, g, with several

fixed transfer rates. For every fixed transfer rate,

measurements were made, for both Expa and Expc.

In this way, for each experiment type, data were

obtained for comparison of the laboratory

performances of IEEE 802.11 b and 802.11g links,

measured namely at OSI levels 4 and 7 using the

scheme of Figure 1.

At OSI level 1 in both Expa and Expc we have

monitored, for every one of the cases, the signal to

noise ratios SNR of the point-to-point links.

For both Expa and Expc, and for every standard

and nominal fixed transfer rate, an average TCP

throughput was determined. This value was used as

the bandwidth parameter for every corresponding

UDP test, giving average jitter and average

percentage datagram loss. The results are shown in

Figures 2-7. In Figures 2-3, polynomial fits were

made namely for each AP implementation of IEEE

802.11 g. It is seen that the best TCP throughput

performances are, by descending order, for 802.11g

and 802.11b. In Expc (Figure 3), the data for 802.11

b,g are higher than the corresponding data in Expa

(Figure 2). In particular for 802.11g and TCP

throughput APc shows, on average, a 10 % better

performance than APa. In Figures 4-7, the data

points were joined by smoothed lines. In Expa

(Figure 4) jitter is, on average, higher for IEEE

802.11b (2.6 ms). In Expc (Figure 5) jitter is also, on

average, higher for IEEE 802.11b (2.1 ms). In both

Expa (Figure 6) and Expc (Figure 7), generally, the

percentage datagram loss data agree reasonably well

for both standards. They are 1.3 % and 1.4 %, on

average, respectively.

At OSI level 7, FTP transfer rates were measured

versus nominal transfer rates configured in the APs

for the IEEE 802.11 b,g standards. Every

measurement was the average for a single FTP

transfer, using a binary file size of 100 Mbytes. The

results thus obtained in Expa and Expc are

represented in Figures 8-9, respectively. Polynomial

fits to data were made. It was found that in both

cases the best performances were, by descending

order, for 802.11g and 802.11b. The FTP transfer

rates obtained in Expc, using both standards, were

higher than in Expa. This means that APc presents a

better FTP performance than APa. FTP results have

shown the same trends found for TCP throughput.

WINSYS 2009 - International Conference on Wireless Information Networks and Systems

Figure 1: Laboratory setup scheme.

Figure 2: TCP throughput versus technology and nominal

transfer rate; Expa.

Figure 3: TCP throughput versus technology and nominal

transfer rate; Expc.

Figure 4: UDP – jitter results versus technology and

nominal transfer rate; Expa.

Figure 5: UDP – jitter results versus technology and

nominal transfer rate; Expc.

Figure 6: UDP - percentage datagram loss results versus

technology and nominal transfer rate; Expa.

A CONTRIBUTION TO LABORATORY PERFORMANCE COMPARISONS OF IEEE 802.11 B,G POINT-TO-POINT

LINKS

Figure 7: UDP - percentage datagram loss results versus

technology and nominal transfer rate; Expc.

Figure 8: FTP transfer rate versus technology and nominal

transfer rate; Expa.

Figure 9: FTP transfer rate versus technology and nominal

transfer rate; Expc.

4 CONCLUSIONS

In the present work a simple laboratory arrangement

was implemented that permitted systematic

performance measurements of available equipments

in IEEE 802.11 b, g point-to-point links. Through

OSI level 4 the best TCP throughputs were found,

by descending order, for 802.11g and 802.11b. TCP

throughputs were also found sensitive to AP type.

The lower values of jitter were, on average, found

for IEEE 802.11g. For the percentage datagram loss,

a reasonably good agreement was found for both

standards. At OSI level 7, the measurements of FTP

transfer rates have shown that the best performances

were, by descending order, for 802.11g and 802.11b.

FTP performances were also found sensitive to AP

type. FTP results have shown the same trends found

for TCP throughput. Additional measurements either

started or are planned using several equipments, not

only in laboratory, but also in outdoor environments

involving, mainly, medium range links.

ACKNOWLEDGEMENTS

Supports from Universidade da Beira Interior and

FCT (Fundação para a Ciência e a

Tecnologia)/POCI2010 (Programa Operacional

Ciência e Inovação) are acknowledged. We

acknowledge Enterasys Networks for their

availability.

REFERENCES

Allied Telesis 2008, AT-8000S/16 Layer 2 Managed Fast

Ethernet Switch (pp. 1-3, 617-000174 Rev. H),

retrieved December 20, 2008, from

http://www.alliedtelesis.com

DD-WRT 2009, What is DD-WRT?, retrieved February

15, 2009, from http://www.dd-wrt.com

Enterasys Networks 2005, Roam About R2 Wireless

Access Platform (pp. 1-4, Lit. #9012495-7 2/5),

retrieved December 20, 2008, from

http://www.enterasys.com

IEEE Std 802.11-2007, IEEE Standard for Local and

metropolitan area networks-Specific Requirements-

Part 11: Wireless LAN Medium Access Control (MAC)

and Physical Layer (PHY) specifications, retrieved

October 10, 2007, from

http://standards.ieee.org/getieee802

Linksys 2005, Wireless-G (WRT54GL) Broadband Router

User Guide (pp. 1-106), retrieved February 10, 2009,

from http://www.linksys.com

Monteiro, E & Boavida, F 2000, Engineering of Informa-

WINSYS 2009 - International Conference on Wireless Information Networks and Systems

tics Networks, 4th edn, FCA-Editor of Informatics Ld.,

Lisbon.

Pacheco de Carvalho, JAR, Gomes, PAJ, Veiga, H & Reis

AD 2008a, ‘Development of a University Networking

Project’, in Encyclopedia of Networked and Virtual

Organizations, eds GD Putnik, MM Cunha, IGI

Global, Hershey, PA, pp. 409-422.

Pacheco de Carvalho, JAR, Veiga, H, Costa, R, Gomes,

PAJ & Reis, AD 2008b, ‘A Contribution to

Experimental Performance Evaluation of Point-to-

Point WiMAX Links’, in Proceedings of the ISSPIT

2008-8th IEEE International Symposium on Signal

Processing and Information Technology, Sarajevo,

Bosnia and Herzegovina, pp. 150-153.

Pacheco de Carvalho, JAR, Veiga, H, Gomes, PAJ,

Ribeiro Pacheco, CFFP & Reis, AD 2008c,

‘Experimental Performance Study of a Very High

Speed Free Space Optics Link at the University of

Beira Interior Campus: a Case Study’, in Proceedings

of the ISSPIT 2008-8th IEEE International Symposium

on Signal Processing and Information Technology,

Sarajevo, Bosnia and Herzegovina, pp. 154-157.

RFC 1889, RTP: A Transport Protocol for Real-Time

Applications, retrieved January 15, 2009, from

http://www.rfc-archive.org

Sarkar, NI & Sowerby, KW 2006, ‘High Performance

Measurements in the Crowded Office Environment: a

Case Study’, in Proceedings of the ICCT’06-

International Conference on Communication

Technology, Guilin, China, pp. 1-4.

Schwartz, M 2005, Mobile Wireless Communications,

Cambridge University Press, Cambridge, UK.

A CONTRIBUTION TO LABORATORY PERFORMANCE COMPARISONS OF IEEE 802.11 B,G POINT-TO-POINT

LINKS