Noncoherent Coded Modulation

Citation

Raphaeli, Dan (1994) Noncoherent Coded Modulation. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/fexw-ex94. https://resolver.caltech.edu/CaltechETD:etd-12112007-082807

Abstract

In this thesis, we are concerned with the transmission of data over noncoherent channels (the carrier phase is random). We consider a receiving system which does not attempt to estimate the carrier phase from the received data. Instead, the transmitter and receiver will be designed so that the data transmission is robust with respect to the unknown phase variations of the channel. For the transmitter, we propose new combined coding and modulation, specifically designed to match the noncoherent channel. For the receiver, efficient decoders to noncoherently decode the coded modulation are developed. As a result, we are able to show, both analytically and by computer simulations, that Noncoherent Coded Modulation (NCM) approaches the performance of coded coherent modulation. NCM achieves almost the same power efficiency without bandwidth expansion or an extensive increase in complexity.

We consider the problem of the Uniform Error Property (UEP) for a broad class of transmitters and receivers. A sufficient condition for a general linear code to satisfy the UEP is presented and a structure of a trellis-coded modulation that satisfies this condition is offered. We call these codes "linear noncoherent trellis-coded modulation" since they apply to noncoherent detection. The problem of noncoherently catastrophic codes, which can result in noncoherent detection of trellis codes, is discussed and a general solution which does not rely on differential encoding of the code output is offered.

High performance noncoherent detection is achieved using multiple symbol observations. Unlike previous approaches, a sliding window is used for the observations, with each observation covering several branches of the trellis, so that the observations are time-overlapped. We define a new type of noncoherent maximum likelihood sequence estimator, and analyze its performance over the Additive White Gaussian Noise (AWGN) channel by numerical calculation of the union bound. We perform a computerized search and present new high performance coded M-ary Phase Shift Keying (PSK) modulations for noncoherent detection and their performance. The new codes cover many useful rates and complexities and achieve higher performance than existing codes for noncoherent detection. We evaluate the performance of NCM in the presence of phase jitter in the channel. The method can also be used for multiple symbol demodulation of M-ary Differential PSK (MDPSK) and of Continuous Phase Modulation (CPM). We provide results for both, full and partial response CPM schemes as well as convolutionally coded CPM. The complexity and power efficiency of this new method is superior to all past schemes known to the author for noncoherent detection.

The optimal implementation of the decoder, using the Viterbi Algorithm (VA), is given. For L-symbols observation, it requires a number of states that grows exponentially with L. Three novel sub-optimal algorithms are presented, whose number of states is the same as the original code so their complexity has a relatively weak dependence on L. For practical values of L, these algorithms are substantially less complex than the optimal algorithm.

The first suboptimal algorithm to be described is called the Basic Decision Feedback Algorithm (BDFA). In this algorithm, the symbols from the decisions are fed back to be used in the subsequent decisions. This algorithm suffers from increased error event probability and from error propagation. However, by a small modification of the BDFA, we obtain another improved algorithm, which will be called Modified Decision Feedback Algorithm (MDFA).

To obtain close to optimal performance, the third algorithm, the Estimated Future Decision Feedback Algorithm (EFDFA) is offered. This sophisticated algorithm, which uses the BDFA as a basic building block, is based on a novel concept called "estimated future." Performance analysis and simulation results are given.

Thesis Files