Wireless systems that have traditionally been designed using a layered approach have seen a recent paradigm shift to a cross-layered approach where the interactions between two or more layers are considered explicitly in an integrated framework. This dissertation presents new methodologies that aim to improve the performance of wireless systems through consideration of cross-layer based design.
The physical (PHY) and the medium access control (MAC) layers are the primary layers responsible for data transmission and user selection/control, respectively, in wireless systems. This dissertation begins with an analysis illustrating the use of multiple antennas and antenna arrays at. the PHY layer. A framework combining space-time block coding and beamforming for uplink in a wireless systems is con¬sidered for studying the trade-offs between antennas and antenna arrays at the receiver. Results indicate that in high noise environments the diversity achieved by using a large number of antennas combats bit error rate (BER) more efficiently than beamforming. On the other hand, in low noise environments beamforming plays an important role in reducing BER by minimizing the effect of interference from other co-channel users.
Two approaches of cross-layer design that are currently available are the bottom-up and top-down approaches. The bottom-up approach uses the PHY laver infor¬mation at the MAC and higher layers to make decisions that affect the system performance. Following a bottom-up approach, a new scheduling algorithm is designed that uses the channel state information and direction of arrival information of mobile users to efficiently schedule users for service. Both semi-analytical (based on the probability density and cumulative distribution functions) and numerical frameworks are used to compare the performance of this algorithm with the traditional round-robin and greedy scheduling algorithms. Both the numerical and semi-analytical frameworks which are shown to be consistent with each other yield improved system capacity for the proposed algorithm compared to the traditional algorithms. This is the result of explicitly considering the angular location of mobile users around the base station that results into the reduced interference between simultaneously served users on one other. The effect of channel availability on the scheduling algorithms is also investigated within a queuing framework and the results indicated that the system performance is also dependent on channel availability and traffic conditions.
A top-down approach is based on modifying the PHY layer to support the require¬ments or protocols used at the MAC layer to improve system throughput. Following a top-down approach, a new methodology is presented that reduces null depths of a given beam to address the hidden beam problem in IEEE 802.11 systems. The hidden beam problem occurs in carrier sensing multiple access (CSMA) systems when mobile users lying in deep nulls are unable to sense an ongoing downlink transmission and start transmitting data in an uplink. The modified beam with reduced null depths is compared with the original beam in terms of reducing the hidden beam problem when used in non-persistent CSMA systems. The modified beam is shown to improve the throughput of a slotted non-persistent CSMA system significantly when compared to original beam with relatively small changes to directivity and half power beamwidth.
The bottom-up and top-down approaches used in this dissertation illustrate that by jointly addressing the PHY and MAC layer issues in an integrated cross-layer framework the performance of wireless systems can be significantly improved.
| Identifer | oai:union.ndltd.org:uvic.ca/oai:dspace.library.uvic.ca:1828/2593 |
| Date | 13 April 2010 |
| Creators | Arora, Deepali |
| Contributors | Agathoklis, Panajotis |
| Source Sets | University of Victoria |
| Language | English, English |
| Detected Language | English |
| Type | Thesis |
| Rights | Available to the World Wide Web |
Page generated in 0.0015 seconds