dc.creator | Noraini, Azmi | |
dc.date | 2016 | |
dc.date.accessioned | 2023-02-21T07:56:42Z | |
dc.date.available | 2023-02-21T07:56:42Z | |
dc.identifier.uri | http://dspace.unimap.edu.my:80/xmlui/handle/123456789/77905 | |
dc.description | Master of Science in Computer Engineering | en_US |
dc.description.abstract | Wireless Sensor Networks (WSN) technology is rapidly deployed in applications such as robotic, healthcare, military, environmental monitoring, and other low-power large scale monitoring that requires high data accuracy with possibly minimal latency and data losses. The Quality of Service (QoS) of WSN is often compromised by the interference from other wireless technologies that are high in transmission power and bandwidth such as Wireless Local Area Network (WLAN), Bluetooth, microwave oven, cordless phone and wirelessUSB. The ubiquitous increase in the number of wireless devices leads to the frequency spectrum occupancy issues as various wireless technologies are forced to share the free and unlicensed 2.4GHz frequency band.
Compared to other wireless technologies, the interference from the WLAN devices
caused a significant packet loss experienced by WSN. This study focuses on
investigating the effect of WLAN interference on WSN performance through
experimental study. This research provides an empirical study on the effect of WLAN
on WSN performance in terms of Energy Detection (ED) and packet loss. A WLAN
access point (AP) or router, laptops as end-client and IxChariot software are used to
emulate WLAN traffic while Waspmote is used as WSN nodes. Tests had been
conducted in two different environments, which is in controlled and uncontrolled
environment. Preliminary test found that scan duration need to be set to 3 in order to
achieve the best ED value after considering the tradeoff between accuracy and false
detection. Result from this study demonstrates that ED is around -84dBm with no
packet loss for test conducted without WLAN interferer. Similarly, test conducted with
one WLAN interferer (without traffic) shows that there is no packet loss but there is
high increase in ED reading that is approximately -44dBm. Besides that, this research
found that the density of traffic yield from WLAN network does not significantly affect
the ED (around -41dBm) by WSN in comparison to when the WLAN AP is simply
turned on without traffic (ED value is around -44dBm). However, the WLAN traffic
does affect the WSN packet loss where packet loss increases from 14% to 36% when
traffic increased from 10% to 30% in controlled environment. Despite that, further tests
revealed that the frequency offset between WLAN and WSN centre frequency did affect
the ED by the WSN. The ED value when the frequency offset between WLAN and
WSN is less than or equal to 3MHz is approximately -42dBm and the ED value for
frequency offset is 8MHz and 13MHz is approximately -54dBm and -68dBm
respectively. In addition, this research also studies the performance of WSN for fixed
channel allocation. Besides that, this research also proposed a technique to improve
WSN performance by performing dynamic channel selection. The technique reduces the
WSN packet loss from 7% to approximately 0% packet loss. | en_US |
dc.language.iso | en | en_US |
dc.publisher | Universiti Malaysia Perlis (UniMAP) | en_US |
dc.rights | Universiti Malaysia Perlis (UniMAP) | en_US |
dc.subject | Wireless LANs | en_US |
dc.subject | Personal area networks (Computer networks) | en_US |
dc.subject | Wireless Sensor Networks (WSN) | en_US |
dc.title | Performance analysis of IEEE802.15.4 wireless sensor network to mitigate IEEE802.11 wireless local area network interference | en_US |
dc.type | Thesis | en_US |
dc.contributor.advisor | Latifah Munirah, Kamarudin, Dr. | |
dc.publisher.department | School of Computer and Communication Engineering | en_US |