Channel estimation for coded systems /
Three different problems dealing with parameter estimation for coded systems are examined in this research. The first focuses on estimating the frequency and phase of the carrier for systems employing M-ARY phase-shift-keying over additive white Gaussian noise channels. Assuming knowledge of the fra...
| Main Author: | |
|---|---|
| Format: | Thesis Book |
| Language: | English |
| Published: |
[Place of publication not identified] :
[publisher not identified] ;
2001.
|
| Subjects: | |
| Online Access: | http://proxy.library.tamu.edu/login?url=http://proquest.umi.com/pqdweb?did=725911611&sid=1&Fmt=2&clientId=2945&RQT=309&VName=PQD |
| Summary: | Three different problems dealing with parameter estimation for coded systems are examined in this research. The first focuses on estimating the frequency and phase of the carrier for systems employing M-ARY phase-shift-keying over additive white Gaussian noise channels. Assuming knowledge of the frame boundary, a maximum likelihood based algorithm is introduced that takes advantage of the underlying code structure. An enhancement is introduced that assigns a measure of reliability to decoded symbols such that erroneous symbols can be excluded from the estimation process. The second problem focuses on joint estimation of the signal-to-noise ratio and the symbol energy for coded systems employing M-ARY phase-shift-keying over additive white Gaussian noise channels. Again assuming knowledge of the frame boundary, a maximum likelihood based algorithm is introduced that takes advantage of the underlying code structure. It will be shown in this research that if the estimated codeword has few errors, the transcendental barrier of the likelihood function can be lifted. For systematic codes, a further enhancement is introduced that employs a test that is designed to reject erroneous parity symbols. The last problem deals with estimating the frame boundary for space-time coded systems. The analysis begins with the case of one transmit antenna and one receive for M-ARY phase-shift-keying over additive white Gaussian noise channels. Then, the general case of N transmit antennas and M receive over fading channels is examined. For decoders that can provide the log-likelihood ratio, the proposed algorithm will use the mode separation of the log-likelihood ratios to estimate the codeword phase. To reduce complexity, a list synchronizer is employed in conjunction with a test that determines whether the decision of the uncoded frame synchronizer is sufficient or is decoder processing required. This research will show that accounting for the code structure results in a higher performance algorithm. The proposed estimators have their origins in ML, but will require modifications. Monte Carlo simulations demonstrate that the performance of the proposed estimators are significantly better than the blind uncoded counter parts |
|---|---|
| Item Description: | Vita. "Major Subject: Electrical Engineering". |
| Physical Description: | xii, 102 leaves : illustrations ; 28 cm. Issued also on microfiche from University Microfilm Inc. |
| Bibliography: | Includes bibliographical references (leaves 89-92). |