Institutional Repository
Technical University of Crete
EN  |  EL

Search

Browse

My Space

Fast synchronization of FSK signals

Konstantinidis Konstantinos

Full record


URI: http://purl.tuc.gr/dl/dias/0D73A4AE-2135-4EFB-81D3-5B388D707534
Year 2016
Type of Item Diploma Work
License
Details
Bibliographic Citation Konstantinos Konstantinidis, "Fast synchronization of FSK signals", Diploma Work, School of Electrical and Computer Engineering, Technical University of Crete, Chania, Greece, 2016 https://doi.org/10.26233/heallink.tuc.67205
Appears in Collections

Summary

In this diploma thesis, we address the problem of noncoherent synchronization, channel estimation, and detection for offset quadrature phase-shift keying (OQPSK) and binary frequency-shift keying (BFSK) modulations. We develop a gradient-descent algorithm which exploits some characteristics of the concentrated likelihood function (CLF) of the time delay as well as some observations about the transmitted sequences and performs noncoherent synchronization, channel estimation, and detection for the aforementioned types of modulation. The proposed algorithm is based on the efficient noncoherent detection of the transmitted symbols, under channel and timing uncertainty. The development of our algorithm is based on the maximum-likelihood (ML) criterion and utilizes state-of-the-art noncoherent detection techniques from the literature. After we reproduce the results of a recent work on joint noncoherent synchronization, channel estimation, and detection for OQPSK, we design an algorithm that concentrates on the data detection problem and is applicable for both OQPSK and BFSK. We are mainly interested in the transmitted sequence but a clock estimate is obtained as byproduct. The latter is achieved by converting the likelihood function to a form that depends only on the estimates of the transmitted sequences and the actual channel delay. The method used involves the first-order derivative of the likelihood with respect to the delay as well as some useful properties of the estimated sequences.

Available Files

Services

Statistics