Computer Simulation and Hardware Implementation of a pi/4-DQPSK Demodulator

碩士 === 國立中正大學 === 電機工程學系 === 85 === TDMA(time-division multiple access) architecture has been proposed for portableradio application to provide efficient digital services including voice and dataapplications. The advantages of a TDMA channe...

Full description

Bibliographic Details
Main Authors: Shyu, Albert, 徐永健
Other Authors: Thomas T. Fang
Format: Others
Language:zh-TW
Published: 1997
Online Access:http://ndltd.ncl.edu.tw/handle/04736084765533806744
Description
Summary:碩士 === 國立中正大學 === 電機工程學系 === 85 === TDMA(time-division multiple access) architecture has been proposed for portableradio application to provide efficient digital services including voice and dataapplications. The advantages of a TDMA channel architecture are often offset by the requirement of a significant overhead for syn-chronization. Two key elements in the design of a demodulator are symboltiming synchronization and offset frequency synchronization . The applicationof digital techniques to demodulation processing for short TDMA bursts canimprove TDMA efficiency by removing the need for overhead for symbol timingand frequency offset synchronization. This paper studied the demodulatordesign to a set of specifications code-named PACS (Personal Access Communi- cation Service) proposed by Bellcore (Bell communication research) in theUnited States. Bellcore has proposed a method of implementing the PACS demo來dulator which has been patented. We investigate an alternative methodof implementation which will perform as well as the Bellcore method. The Bellcore metdulator obtains (1)symbol timing synchronization, (2) an estimate of the offset frequency, and (3)the "figure of merit",i.e., an estimate of the channel (path) quality simultaneously after a predetermined set of computations. The new method developed in this paper obtains symbol syn-chronization and a frequency offset estimate through separate calculations,with symbol synchronization preceding frequency offset estimation. Howeverthe total of the two calculations does not exceed the one data burst limit. Furthermore, the channel quality estimation is obtained as a byproduct ofthe frequency offset estimation. The amount of computation in this new method is comparable to that of the Bellcore method.