Klasifikacija linearnih diskretnih modulacija temeljena na statistici višeg reda

scientific master's theses

• Title (english): Linear Discrete Modulations Classification Based on Higher-order Statistics
• Author: Ema Čular
• Mentor: Borivoj Modlic (mentor)
• Committee member: Vjekoslav Sinković (predsjednik povjerenstva)
• Committee member: Borivoj Modlic (član povjerenstva)
• Committee member: Miljenko Huzak (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; (Department of Communication and Space Technologies); Zagreb
• Defense date and country: 2002-10-29, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Electrical Engineering; Radio Communications
• Universal decimal classification (UDC): 621.3 - Electrical engineering

Abstract

U ovom radu istražene su mogućnosti klasifikacije linearno diskretno moduliranih signala metodama oslonjenim na statistiku višeg reda. U tu svrhu postavljen je jedinstveni matematički model signala. Analizirane su korelacijske funkcije momenta i kumulanta višeg reda kompleksnog modulacijskog signala u diskretnom modelu u njegovom ciklostacionarnom opisu. Postavljen je klasifikacijski postupak koji se temelji na izravnoj procjeni kumulanata četvrtog reda digitalnog modulacijskog znaka. Postupak je osobit po tome što ne zahtijeva poznavanje valnog oblika modulacijskog simbola. Eksperimentalni rezultati na sintetičkim signalima sa superponiranim uskopojasnim Gaussovim šumom za četiri različita valna oblika modulacijskog simbola potvrđuju da je klasifikator primjenjiv čak i u slučaju interferencije među simbolima. Procjena momenata vršena je usrednjavanjem preko polifaznih komponenti signala. Točnost klasifikatora bolja je za modulacijske postupke s manje stanja u dijagramu, a degradira se s pogoršanjem odnosa signal/šum. Klasifikator vrlo dobro raspoznaje osnovne grupe klasifikacijskih postupaka (MPSK, MASK, MQAM s kvadratnim i MQAM s križnim dijagramom stanja).

Abstract (english)

We consider classification of linear digitally modulated signals based on the higher order statistics. The uniquely model of signal is established. Moment and cumulant based higher order correlation functions of complex envelope in cyclostacionary description are analyzed. We propose the method for modulation classification based on the direct estimation of fourth-order cumulants of digital modulation symbols. We hold this method original because the pulse shape is assumed to be unknown. Simulation results with various types of modulation corrupted by additive narrowband Gaussian noise and with four various pulse shapes confirms that classifier is usable even when ISI is presented. Estimation of moments is performed by averaging through polyphase components. The classifier performances are better for modulation types with simpler constellation diagrams and degrade with signal to-noise ratio. The high discrimination between main modulation groups (MPSK, MASK, square constellation QAM and cross constellation QAM) is achieved.

• Keywords: Diskretna modulacija; klasifikacija; statistika višeg reda; ciklostacionarnost; moment; kumulant; korelacijska funkcija; procjenjivanje
• Keywords (english): Digital modulation; classification; higher-order statistics; cyclostationarity; moment; cumulant; correlation function; estimation
• Language: croatian
• URN:NBN: urn:nbn:hr:168:162810
• Study programme: Title: Postgraduate master programme in electrical engineering; Study programme type: university; Study level: postgraduate; Academic / professional title: Magistar znanosti elektrotehnike (Magistar znanosti elektrotehnike)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-04-30 07:47:57