Two level aggregation scheme with optimization in subframes headers for IEEE 802.11n wireless networks
Frames aggregation is a means of utilizing channel efficiency at MAC of WLAN device, through amortizing the MAC overheads over multiple frames. IEEE 802.11n define two standards of the scheme: A-MSDU and A-MPDU. They proved efficient but introduced yet another headers overhead due to the aggregation...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2014
|
| Subjects: | |
| Online Access: | http://psasir.upm.edu.my/id/eprint/39369/1/FSKTM%202014%203%20IR.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Frames aggregation is a means of utilizing channel efficiency at MAC of WLAN device, through amortizing the MAC overheads over multiple frames. IEEE 802.11n define two standards of the scheme: A-MSDU and A-MPDU. They proved efficient but introduced yet another headers overhead due to the aggregation, which parts with large part of the efficiency gained. Furthermore, investigations revealed that
while one of the schemes shortfalls in one situation the other is efficient in that same condition, vice versa: depending on the PHY rate, aggregation size, frames sizes
and/or channel conditions. We are motivated by the efficiencies of the schemes; considered that the overhead can be minimized when the frame structure is well reformulated and that their complementary efficiencies are exploitable to arrive at a more efficient scheme. This work, therefore, is carried out with the aim of minimizing
the overhead due to the aggregation so as to maximize the throughput gain, and also to utilize the efficiency of both of the scheme by systematically merging them to co-work together in what is referred to as two-level aggregation.
We extended Markov’s discrete even chain model and arrived at generic model for the aggregation throughput. A model is also derived for the headers overhead. The duo proved to accurately predict and comparatively analyze various aggregations behaviours. Based on the study of the aggregations conducted, we designed an enhanced A-MSDU (eA-MSDU), with optimized header overhead. By both analytical
and simulation, the eA-MSDU improved channel’s utilization by up to 10% and 20% in terms of throughput increase and overhead reduction respectively. Furthermore, a new two-level aggregation model is developed by integrating the eA-MSDU into A-MPDU and designing suitable operating algorithm. This results to further improvements in respect of both the overheads reduction and throughput gain, by 18%
and 7% respectively. It also show consistent channel utilization of around 90% and an average of 4% overhead, irrespective of the subframes or aggregation size.
With the newly designed two-level aggregation scheme; overhead due to aggregation is contained; channel utilization improved; and the aggregation efficiency is consistent, even with varying conditions mentioned. When applied therefore, WLANs could compete with other technologies in serving throughput-intensive applications
such as grid computing and on-demand multimedia streaming. With proper schedulingand adaptation on top of the scheme, it will serves best in real-time applications. |
|---|