The primary objective of this thesis is to design advanced interference resilient schemes for asynchronous slow frequency hopping wireless personal area networks (FH-WPAN) and time division multiple access (TDMA) cellular systems in interference dominant environments. We also propose an interference-resilient power allocation method for multiple-input-multiple-output (MIMO) systems.
For asynchronous FH-WPANs in the presence of frequent packet collisions, we propose a single antenna interference canceling dual decision feedback (IC-DDF) receiver based on joint maximum likelihood (ML) detection and recursive least squares (RLS) channel estimation. For the system level performance evaluation, we propose a novel geometric method that combines bit error rate (BER) and the spatial distribution of the traffic load of CCI for the computation of packet error rate (PER). We also derived the probabilities of packet collision in multiple asynchronous FH-WPANs with uniform and nonuniform traffic patterns.
For the design of TDMA receivers resilient to CCI in frequency selective channels, we propose a soft output joint detection interference rejection combining delayed decision feedback sequence estimation (JD IRC-DDFSE) scheme. In the proposed scheme, IRC suppresses the CCI, while DDFSE equalizes ISI with reduced complexity. Also, the soft outputs are generated from IRC-DDFSE decision metric to improve the performance of iterative or non-iterative type soft-input outer code decoders.
For the design of interference resilient power allocation scheme in MIMO systems, we investigate an adaptive power allocation method using subset antenna transmission (SAT) techniques. Motivated by the observation of capacity imbalance among the multiple parallel sub-channels, the SAT method achieves high spectral efficiency by allocating power on a selected transmit antenna subset. For 4 x 4 V-BLAST MIMO systems, the proposed scheme with SAT showed analogous results. Adaptive modulation schemes combined with the proposed method increase the capacity gains. From a feasibility viewpoint, the proposed method is a practical solution to CCI-limited MIMO systems since it does not require the channel state information (CSI) of CCI.
| Identifer | oai:union.ndltd.org:GATECH/oai:smartech.gatech.edu:1853/7647 |
| Date | 24 August 2005 |
| Creators | Kim, Kihong |
| Publisher | Georgia Institute of Technology |
| Source Sets | Georgia Tech Electronic Thesis and Dissertation Archive |
| Language | en_US |
| Detected Language | English |
| Type | Dissertation |
| Format | 1016982 bytes, application/pdf |
Page generated in 0.0014 seconds