Desmond P. Taylor

Performance enhancement of spectral-amplitude-coding optical CDMA using pulse-position modulation

Spectral-amplitude-coding optical code-division multiple-access (OCDMA) systems are limited by interference between incoherent sources. A detailed analysis of this limit for a system with a balanced receiver is presented. Additional pulse-position modulation (PPM) coding is proposed as a method to improve the system performance beyond this limit. A simple and robust PPM decoding structure is proposed, and the performance analysis of the whole PPM-OCDMA system is presented. The interference-limited performance of the PPM-OCDMA system is found to be superior to that of the original system when the number of simultaneous users is of the order of the PPM word length or larger. In particular, for a PPM word length of two, an increase in spectral efficiency of up to 100% is possible with no change in the signaling rate, data rate, or bit-error rate (BER).

Multilevel codes based on partitioning

Following V.V. Ginzburg (1984), a hierarchy of codes is proposed to match the geometric partitioning of a signal set. The authors show that coset codes (including Ungerboeck, lattice, and binary codes) and indeed any codes which rely on partitioning of the signal set are all subclasses of the proposed coding scheme. The combination of such codes in a multilevel scheme often leads to reduced complexity in comparison with previously published schemes. A variety of decoder structures is discussed. >

A bandwidth-efficient class of signal-space codes

A new class of codes in signal space is presented, and their error and spectral properties are investigated. A constant-amplitude continuous-phase signal carries a coded sequence of linear-phase changes; the possible signal phases form a cylindrical trellis in phase and time. Simple codes using 4-16 phases, together with a Viterbi algorithm decoder, allow transmitter power savings of 2-4 dB over binary phase-shift keying in a narrower bandwidth. A method is given to compute the free distance, and the error rates of all the useful codes are given. A software-instrumented decoder is tested on a simulated Gaussian channel to determine multiple error patterns. The error parameter R_{o} is computed for a somewhat more general class of codes and is shown to increase rapidly when mere phases are employed. Finally, power spectral density curves are presented for several codes, which show that this type of coding does not increase the transmitted signal bandwidth.

Maximum likelihood decoding of uncoded and coded PSK signal sequences transmitted over Rayleigh flat-fading channels

The problem of maximum likelihood (ML) detection for uncoded and coded M-PSK signals on Rayleigh fading channels is investigated. It is shown that, if the received signal is sampled at baud-rate, a ML receiver employing per-survivor processing can be implemented. The error rate performance of this receiver is evaluated by means of computer simulations and its limitations are discussed. In addition, it is shown that, on a fast fading channel, the error floor in the BER curve can be appreciably lowered if more than one received signal sample per symbol interval is processed by the receiver algorithm, Finally, a sub-optimum two-stage receiver structure for interleaved coded PSK systems is proposed. Its error rate performance is assessed for simple trellis-coded modulation schemes and compared to that provided by other receiver structures.

Trellis-coded continuous-phase frequency-shift keying with ring convolutional codes

Trellis-coding techniques are applied to continuous-phase frequency-shift keying (CPFSK). A new coding scheme based on convolutional codes on the ring of integers modulo-P is shown to be a natural way to apply trellis coding to CPFSK. Previous work has decomposed CPFSK into two parts: a linear encoder, with memory called the continuous phase encoder (CPE), and a memoryless modulator (MM), where the CPE often has a code structure defined over the ring of integers modulo-P. The combination of a modulo-P convolutional channel encoder (CE) and the CPE is a linear modulo-P encoder. Design examples are given for rate-1/2 coded quaternary CPFSK with modulation indexes 1/2 and 1/4, and rate-2/3 coded octal CPFSK with modulation index 1/8. Combinations are optimized in the normalized minimum Euclidean distance sense for a given total number of states in the overall MLSE receiver. Numerical results show that this new coding scheme consistently obtains better performance than previous schemes. >

An adaptive maximum likelihood receiver for correlated Rayleigh-fading channels

The paper develops a receiver structure for random Gaussian signals in additive noise based on the classic maximum likelihood (M-L), estimator-correlator derivation of Kailath [1960], and applies it to differential phase shift keying (DPSK) on the correlated Rayleigh-fading channel. It is shown to lower the error floor found in the performance of conventional DPSK receivers by orders of magnitude. In addition, the maximum-likelihood procedure is shown to make uncorrelated symbol decisions. The performance of both conventional and optimal receivers, which require knowledge of the channel statistics, is examined analytically. A recursive, channel-adaptive version of the optimal receiver, utilizing decision feedback to estimate the channel statistics, is developed. Its simulated performance shows no penalty compared to theoretical calculations which require explicit knowledge of the channel statistics. >

The effect of diversity on a burst-mode carrier-frequency estimator in the frequency-selective multipath channel

This paper describes a carrier-frequency offset estimator derived using maximum-likelihood techniques. The estimator is designed for a digital space-diversity receiver, operating in a wide-band frequency-selective multipath fading channel. The estimator is suited to a burst-mode time-division-multiple-access system, because the estimate is formed in an open-loop manner, and relies on a training sequence normally used in burst-mode systems. The main advantage is that it does not require channel state information; rather, it only requires knowledge of the auto-correlation of the channel. Simulation results show the estimator to be unbiased over a wide frequency range. The normalized error standard deviation is shown to be 0.0015 across the frequency offset range for a receiver with four diversity branches, when the normalized channel delay spread is 0.1, the signal-to-noise ratio (SNR) is 10 dB, and the training sequence length is 23 symbols. It is found that, for most parameter configurations, there is approximately a factor of two improvement in the estimate error standard deviation when the antenna diversity is doubled.

A comparison of reduced complexity decoding algorithms for trellis codes

Comparisons are made of a genie-aided sequential algorithm due to D. Haccoun and M.J. Ferguson (1975), the Viterbi algorithm, the M-algorithm, and the Fano algorithm for rate-1/2 and rate-2/3 trellis modulation codes on rectangular signal sets. The effects of signal-to-noise ratio and decoding-delay constraints on the choice of decoding algorithms for framed data are examined by computer simulation. Additionally, the genie-aided algorithm is used as a tool in estimating the asymptotic behavior of the M-algorithm. In general, the results conform closely to experience with convolutional codes due to the similar distance structure of the codes. The Fano algorithm produces good error performance with a low average number of computations when long decoding delay is permissible. The M-algorithm provides a savings in computation compared to the Viterbi algorithm if a small decoding delay is required. >

Multiple serial and parallel concatenated single parity-check codes

Single parity-check (SPC) codes are applied in both parallel and serial concatenated structures to produce high-performance coding schemes. The number of concatenations or stages, M, is increased to improve system performance at moderate-to-low bit-error rates without changing the overall code parameters (namely, code rate and code block length). Analytical bounds are presented to estimate the performance at high signal-to-noise ratios. The SPC concatenated codes are considered with binary phase-shift keying and with 16-quadrature amplitude modulation bit-interleaved coded modulation on the additive white Gaussian noise channel and the independent Rayleigh fading channel. Simulations show that the four-stage serial or parallel concatenated SPC codes can, respectively, outperform or perform as well as 16-state turbo codes. Furthermore, decoding complexity is approximately 9-10 times less complex than that of 16-state turbo codes. The convergence behavior of both serial and parallel concatenated SPC codes is also discussed.

Instantaneous Capacity of OFDM on Rayleigh-Fading Channels

For a power limited orthogonal frequency-division multiplexing (OFDM) system transmitting a large number N of subcarriers over a Rayleigh-fading channel, the distribution of the instantaneous capacity is shown to be approximately Gaussian. The mean and variance of the approximating distribution are derived. It is also shown that, in the limit as Nrarrinfin, the capacity approaches a constant value equal to the capacity of the infinite-bandwidth Gaussian channel