Uplink Multiple Access For IMT-Advanced

Abstract: The wireless communication industry has always been under tremendous improve-ments in the past few decades, but will the needs for improvement reach an end?Not at all! The challenge for the future wireless systems is the ability to operate inwidely di®erent deployment scenarios supporting di®erent tra±c rates. As part ofthe International Telecommunication Union Radiocommunication Sector (ITU-R)International Mobile Telecommunication (IMT) Advanced capable Wireless WorldInitiative New Radio (WINNER) system concept, a diversity based multiple ac-cess scheme for robust uplink transmission denoted as Block Interleaved FrequencyDivision Multiple Access (B-IFDMA) was proposed to be used in scenarios wheretransmit channel state information is not readily available due to the imposed over-head, e.g. as with high speed or low data rate and for short control packets.

With its dispersed allocation of multiple blocks with equidistant spacing in fre-quency, where each block consists of a few consecutive subcarriers in a few consecu-tive Orthogonal Frequency Division Multiple Access (OFDMA) symbols (Svenssonet al. 2007), this scheme obtains its robustness and takes advantage of the wirelesschannel characteristics.

Moreover, the dimension of the blocks allocated enables a tunable degree of fre-quency diversity and low allocation signaling overhead. In addition, it allows thehigh power ampli¯ers (HPA) in the uplink to operate at higher e±ciencies becauseof the use of a Discrete Fourier Transform (DFT) precoding step. In addition, the°exibility of slot allocations enables robust and e±cient transmission even for smallpackets while improving the battery life in user terminals.

In this thesis, end-to-end analysis of B-IFDMA and the other candidate multipleaccess DFT-precoded techniques Interleaved FDMA (IFDMA), Localized FDMA(LFDMA), together with the non DFT-precoded Block Equidistant FDMA (B-EFDMA) was carried under di®erent deployment scenarios. This was followed byoptimal parameter extraction since the results were highly dependent on all themodules involved ranging from the equalization schemes, modulation techniques,channel coding methods, to the channel estimation performance at the receiver.

With perfect channel state information at the receiver, Minimum Mean Square Error(MMSE) equalization and, no pulse shaping, IFDMA provides high frequency diver-sity and low peak-to-average power ratio (PAPR), compared to LFDMA. However,in more realistic scenarios, LFDMA has a much better channel estimation perfor-mance at the receiver, the di®erence in PAPR after pulse shaping is rather small, andcombined with frequency hopping rather good frequency diversity can be obtainedwith a penalty on an increased packet transmission delay. The B-IFDMA schemeaims to provide a trade-o® between the strengths of IFDMA and that of LFDMA forvery wideband scenarios and in situations where the delay requirements are strin-gent, e.g. due to multi-hop transmission. On the other hand, the non DFT-precoded scheme B-EFDMA collects as much frequency diversity as in B-IFDMA with veryclose bit error rate performance, but su®ers from high back-o® requirements on theHPA at the transmitter (user terminal). Despite its worst PAPR, it performs thebest in the presence of Zero Forcing (ZF) equalization. Lastly, all these schemeswere bene¯cial in collecting diversity even in the presence of robust channel codes.