An application of multiple block codes (MBC) for realizing new digital transmission schemes that features in hunting-free reframing is investigated. First, the principles of MBC are overviewed to show the capability of filterless clock recovery. Then it is shown that modification of the simple frame structure of MBC line code can be used for attaining hunting-free refraining for multiplexing systems. Two types of MBC are developed to this end. While the one uses header blocks for hunting-free refraining, the other uses no headers. Finally, systematic design of MBC format for hunting-free refraining is investigated. It is shown that variable frame lengths can be accommodated by MBC with header type framing blocks.

Optical receivers for multichannel TV transmission in the VHF band are analyzed and tested from the viewpoints of linearity, sensitivity, and frequency response. Photodiodes and avalanche photodiodes are compared for use as a front-end detector. Optimization of the preamplifiers is studied in terms of both configurations and biasing conditions. Improvement of frequency characteristics, using a Percival coil, is also investigated. It is concluded that a receiver using a photodiode can attain as high a sensitivity as and a greater linearity (around 5 dB) than one using an avalanche photodiode.>

Analog transmission using laser diodes is quite attractive for transmitting multiplexed video signals, owing to the wide modulation bandwidth of laser diodes. Optimizations for transmitting color-TV signals (7 channel frequency division multiplexed) in the VHF band using buried heterostructure laser diodes are investigated. The major transmission objectives are a carrier to noise ratio of larger than 48 dB and intermodulation products of less than -55 dB. Design curves for minimizing distortion at the transmitter are demonstrated. A modulation index of 70 percent is shown to be capable of attaining the objective. Both the avalanche photodiode (APD) and the pinphotodiode (PD) are considered for use in the receiving amplifier. Minimum receiving power of -11 dBm obtained with the PD receiver is in good agreement with theoretical values. Deterioration of signal to noise ratio due to fiber guided transmissions is also reviewed in comparison with aerial transmissions.

An effort is made to devise pulse shaping methods for self-bit-synchronizing systems that minimize both intersymbol interference and pattern-dependent jitter. Attention is confined to minimum-phase shaping which is considered from the standpoint of baseband applications. A short review of jitter analyses is also provided. It lays stress on discerning jitter components that accumulate along repeater chains from those that do not. Pulse shaping functions for the signal path, i.e., Nyquist's problem under minimum-phase constraint, are investigated first and found to result in Bessel filtering as well as its steepest descent modifications. It is observed that reducing intersymbol interference leads to increased pattern-dependent jitter. Next, pulse shaping functions for the timing path are developed. A broad sense symmetrical waveform concept produces a method of synthesizing shaping networks that can drastically reduce waveform dependent jitter, at the sacrifice of increased intersymbol interference. Finally, simultaneous minimizations of intersymbol interference and timing jitter are attempted. A minimal oscillatory tail shaping method incorporating a predistortion technique is shown to be quite efficient for this purpose and thus attractive for simple implementations.

Repeater powering plans are reconsidered from the standpoint of fiber optic transmission. It is shown that conventional powering plans are quite inefficient. It is also shown that ordinary methods for analyses are inconvenient for studying powering network. A method that significantly simplifies the analyses is proposed. According to this method, the general solution can easily be obtained from a special solution. Repeater powering plans are classified and compared to show the efficiency of the method. The comparison proves to be quite convenient for choosing efficient powering plans in accordance with applications. In general, some classes of parallel powering plans receiving relatively high voltages (a few hundred volts) are preferable from the practical standpoint. Further considerations for realizing economical powering networks have been discussed. It is suggested that the development of optical devices and design of transmission equipment should carefully be based on these considerations. For example, development of photo diode (APD) with low-breakdown voltage is not required and the inclusion of high-voltage generators in repeaters is neither necessary nor desirable from the standpoint of repeater powering.

A new variable transfer function is proposed and its application to digital transmission is investigated. The realization of the new variable transfer function is developed for use in a digital transmission automatic line build out (ALBO) network. A simple inductorless ALBO network having ±15 dB variable range is demonstrated for use in 1.544 Mbit/s PCM transmission. Intersymbol interference and jitter increments due to equalization errors are around ten percent and ±2 degrees, respectively.

A new fiber optic baseband analog transmission plan is proposed. It arranges the original signal into a pulse train for driving the laser diode to generate baseband components at the transmitter. The receiver can be just the same as those used in conventional direct linear intensity modulation systems using light emitting diodes (LED). Advantages of this new plan are (i) potentially cheap fibers can be used, since the plan requires neither an extra wide bandwidth nor special light acceptance conditions, (ii) long repeater spacings can be attained, since abundant light can be coupled into fibers using laser diodes and also the simple structures required for fibers are suitable for realizing low transmission losses, and (iii) the transmitter and receiver can be quite simple. Some candidate pulse arranging plans for generating baseband components are investigated. A pulse frequency arranging plan and pulse width arranging plan seem to be promising for efficient picture transmissions. An experiment using pulse frequency arrangement proves the theory developed here to attain a color TV transmission using a 10 km long step index low-loss fiber for a signal to noise ratio of 44 dB. Future improvements are expected by redesigning the transmitter and receiver and also by improving the mode stability of laser diodes.

The optimizations of signal-to-noise ratios (SNR's) and pulse error rates in fiber optic receivers are investigated in terms of an avalanche photo diode (APD) model in which excess noise factors depend on the avalanche gains. The optimization of SNR results in an analytical solution for the optimum avalanche gain. A numerical example compares the result with those of conventional SNR optimization where excess noise factors do not depend on avalanche gains. Improvements of around 2 dB in the accuracy of the optimum SNR's are expected. It is shown that the optimization of pulse error rates for digital transmissions can not be attained analytically. This problem is solved by the use of approximate representations of excess noise factors. Two approximate representations are illustrated that provide very simple and accurate analytical solutions. Numerical examples demonstrate the efficacy of these approximate solutions.

Some new optical pulse formats are investigated for solving practical problems in fiber optic communication systems. These pulse formats provide many advantageous features such as error monitoring capability, abundant timing information, uniform optical power utilization, stable detection of optical input, and so forth. It is shown that a modification of Personick's receiver design theory can be used for comparison of various optical pulse formats. The comparison suggests that for state-of-the-art fiber systems with moderate fiber loss and moderate repeater spacing, where no pulse equalization is required, some new classes of 1 binary digit converted to 2 binary digits (1B2B) or 2B3B formats will permit the realization of very simple and reliable repeaters for fiber optic digital transmission. A future low-loss fiber system may permit a very long repeater spacing with the help of equalization. In this case, application of the correlative signal-processing technique is shown to be very promising. Experimental 6.3 Mbit/s and 100 Mbit/s transmissions demonstrate some advantageous features of these optical pulse formats.

Simple methods that compensate for low- and highfrequency delay distortions, which are the predominant distortions in a baseband hybrid digital transmision line, have been developed. Compensation for low frequency distortion is optimized by a rational minimum-phase function. Compensation for high-frequency distortion is attained by a combination of an optimum cutoff shape and an all-pass network. A local minimum problem in the optimization of the all-pass network is eliminated by a procedure that provides appropriate initial conditions. Combined realization of both low- and high-frequency delay compensation by a slightly active network results in very simple circuitry. The efficiency of this simple compensator is confirmed experimentally in a subline with 15 analog repeaters. In this experimental system, two subpredominant distortions, nonlinear distortion and equalization error, were experienced. Compensation for these distortions is also studied by the method reported in the literature and by a simple adaptive equalizer, respectively. Based on the results obtained above, the feasibility of a baseband hybrid digital transmission system is discussed. The simple compensator mentioned above will be very useful in realizing an economical hybrid digital transmission system with the help of the simple adaptive equalizer. A system that utilizes alternate mark inversion (AMI) pulse train seems to be feasible without the help of the adaptive equalizer.

A baseband hybrid digital transmission scheme is studied, and it is shown that the low- and high-frequency cutoffs in analog repeaters extremely degrade signal waveforms. Solutions studied are a combination of dc-constrained coding and modification of amplitude-frequency characteristics for low-frequency distortion, and a combination of flat amplitude equalization and simple phase equalization for high-frequency distortion. It is shown that these solutions reduce waveform distortions by a factor of ten. A solution of the control problem for automatic gain control (AGC) is also presented.

Timing extraction through nonlinearity in the presence of pulse overlaps is studied for baseband PCM or PAM systems. It is shown that the systematic jitter can be reduced by proper choices of nonlinearity, coding scheme, and pulse waveform. Discussions on choice of nonlinearity indicate that square-law rectification is desirable for elimination of jitter sources. This nonlinearity derives a frequency-domain expression for amplitude and phase of the timing wave that separates the effects of pulse pattern and pulse shapes and provides insight to their contributions to timing jitter. Proper choices of coding scheme and pulse waveform reduce amplitude and phase variation in the timing wave. For example, an optimum prefiltering (or correlative coding) reduces amplitude variation to one eighth for bipolar sequence. In the case of symmetrical reshaping by a Bessel (flat delay) filter, the maximum jitter exceeds 20 degrees if the order of a filter is less than ten. An asymmetrical reshaping makes the maximum jitter less than four degrees for the same order of filter. Some features of nonlinearities that approximate square-law rectification are discussed.