Blind subspace multiuser detection based on intersymbol correlation in DS-CDMA systems

Abstract

<P>DIRECT-SEQUENCE CODE-DIVISION MULTIPLE ACCESS (DS-CDMA) IS A VERY POPULAR MULTIPLE ACCESS TECHNOLOGY FOR PERSONAL COMMUNICATION SYSTEMS. IT HAS MANY ATTRACTIVE FEATURES SUCH AS SOFT CAPACITY, SOFT HANDOVER AND MULTIPATH REJECTION. HOWEVER, IN A DS-CDMA SYSTEM, A MAJOR FACTOR THAT LIMITS THE SYSTEM CAPACITY IS THE MULTIUSER INTERFERENCE. SIGNIFICANT EFFORTS HAVE BEEN FOCUSED ON MULTIUSER DETECTION TECHNIQUES TO SUPPRESS MULTIUSER INTERFERENCE IN ORDER TO INCREASE THE CAPACITY OF CDMA SYSTEMS.</P><P> THE MAJOR CONTRIBUTION OF THIS THESIS IS THE DEVELOPMENT OF A NEW BLIND SUBSPACE MULTIUSER DETECTION SCHEME TO JOINTLY SUPPRESS THE INTERSYMBOL INTERFERENCE (ISI) AND MULTIPLE-ACCESS INTERFERENCE(MAI). HENCE, THIS SCHEME IS SUITABLE TO WORK IN A HIGH RATE DISPERSIVE DS-CDMA SYSTEM WHERE ISI AND MAI CONSTITUTE A MAJOR IMPEDIMENT TO THE SYSTEM PERFORMANCE. BY EXPLOITING THE ISI INFORMATION, THE DISPERSIVE CHANNEL PARAMETER OF THE DESIRED USER CAN BE IDENTIFIED THROUGH A SUBSPACE METHOD. BA