Search

Satellite communications service for Space Based Broadcasting has been suggested by the National Research Council's Technical Operation's Committee for the Voice of America [I]. The service is referred to as Direct Broadcast from SpaceAudio (DBS-A). This paper presents a concept incorporating low bit rate source coding for voice and music, digital modulation, forward error correction coding and Viterbi decoding, multiple satellite beams, and multiple carrier time division multiplexed (TDM) transmission to provide a system capable of world wide coverage. The system provides the capability t o narrowcast to specific geographica l ly c lus te red audiences . Emphasis is on the Space Segment ( T r a n s m i t t e r ) and Ea r th Segmen t ( R e c e i v e r ) u s i n g the t e c h n i q u e s mentioned above.

This paper describes an advanced multibeam communications satellite system using on-board processing and hopping beams in conjunction with low-cost earth-stations to derive a significant advantage over fibre optic submarine cable. It serves thousands of small earth-stations located near the origins of traffic. An economic analysis is presented for both this satellite system and competitive fibre optic cable for the 1995 to 2007 epoch. Results indicate the satellite service to be significantly lower in cost, and in addition it delivers the service near to the origin of the traffic, providing additional savings in terrestrial connections.

The continuing need for increased traffic capacity for telephony over satellite links combined with the availability of low cost and accurate LSI compander chips has caused renewed interest in SSB satellite transmission using companders to reduce the impairments caused by thermal noise, interference and intermodulation encountered on the transmission link. This type of transmission is designated as CSSB/AM and may be operated in a multicarrier per transponder (FDMA) manner. The capacity achieved is a function of the average power during the occurrence of speech and because this value has decreased in modern telephone systems, the advantages to be gained from CSSB/AM are increased. This paper presents the development of expressions for channel capacity of CSSB/AM links as a function of carrier-to-noise ratio and speech power level while active. These are used to estimate the capacity achievable in an INTELSAT IV type transponder.

This paper describes the basic control features of a typical satellite switch system, such as INTELSAT VI. Factors are addressed pertaining to the data format that is used to express the switch plan, the timing control of the switch states comprising the satellite switch frame, the transmission of the switch plan to the satellite, storage and execution of the switch plan on board, and the coding of switch plan data for protection from radiation-induced bit errors. Also, the implementation used to achieve precision control of the on-board clock is discussed, and the influence of clock control variation on all the terrestrial interfaces that participate in the system is demonstrated. Finally, the method for achieving network-wide coordinated switch state time plan/TDMA burst time plan changes, which are essential in an operational satellite-switched (SS)-TDMA system, is discussed.

Measurements were made of the random variation produced in the amplitudes of sonic pulses propagated through a water medium in which was induced a nearly homogeneous random perturbation in temperature. The measurements were carried out at frequencies of 480 kHz and 1.42 MHz for transmitter‐to receiver separations ranging from 1638 to 7658 in. The amplitude variations of the received pulses were analyzed to determine their time autocorrelation and variance as functions of range, i.e., transmitter‐to‐receiver separation and frequency. It was found that the time autocorrelation of the sonic‐pulse amplitudes was equivalent to that measured for the temperature microstructure. The variance in sonic‐pulse amplitude was determined as a function of range at both operating frequencies. These findings are compared with the theoretical findings of Mintzer and Bergmann. An equivalence of the time autocorrelation of sonic‐pulse amplitude and temperature microstructure predicted by theoretical considerations is indeed s...

This paper presents basic network control concepts for the Advanced Communications Technology Satellite (ACTS) System. Two experimental systems, called the low-burst-rate and high-burst-rate systems, along with ACTS ground system features, are described. The network control issues addressed include frame structures, acquisition and synchronization procedures, coordinated station burst-time plan and satellite-time plan changes, on-board clock control based on ground drift measurements, rain fade control by means of adaptive forward-error-correction (FEC) coding and transmit power augmentation, and reassignment of channel capacities on demand. The NASA ground system, which includes a primary station, diversity station, and master control station, is also described.

An investigative program conducted at Melpar under U. S. Air Force sponsorship has resulted in the development of a speech band‐width compression system capable of transmitting speech within a total band width of 150 cps (as compared with 3000 cps for speech transmission by conventional means). This system consists of an analyzer which extracts from the acoustical speech signal six vocal analog control signals, a suitable transmission link, and a synthesizer which employs the control signals to reproduce the essential features of the original speech. The vocal analog control signals vary slowly in time because of natural physical constraints, and thus require considerable less transmission band width than the original acoustical signal. Articulation tests conducted on the system have resulted in scores of 67% for PB (phonetically balanced) word lists and 88% on Spondee word lists. A breadboard model of this system will be described and tape recordings of processed speech will be presented. [This work was supported by Contract AF33 (600)‐3.]

Method and results are presented for the application of statistical pattern‐recognition techniques to automatic classification of words and talkers using formant‐tracking data as the input. The method employs the expression of an entire formant pattern as a single n‐dimensional vector. Each vector component represents a sample of one of the formant functions at a giventime. Thus, the dimensionality of the n space is determined by the product of the number of formant parameters sampled and the number of observation epochs. A computer program processes these data in two steps. In the first step, called a “learning” program, separate statistical categories are established by the computation of the mean vector and covariance matrix describing a normal distribution function for the set of vectors that define each category. In the second step, called the “recognition” program, an unknown is recognized by computing the statistical distance to all possible categories and selecting the category for which this distance is minimum. Results, in the form of confusion matrices, are presented for word recognition of the first ten letters of the alphabet as spoken by 10 talkers and for recognition of the talkers.

A formant‐tracking method based on the measure of majority period is discussed. This tracker employs as input data the instantaneous period measures of formants that have been selected by appropriate formant‐range filters. These periods are subjected to an electronic circuit that, by a unique feedback arrangement, suppresses periods that deviate beyond a fixed threshold distance from the mean period. The resulting circuit output very nearly tracks the majority period of the set of periods that originally comprised the input data. The method also incorporates a means of weighting the majority period formant measure toward the median of the formant range. By this means, the tendency to be captured by adjacent formants is significantly reduced. The paper describes the method of implementation and the application of the method to formant tracking in the second‐formant range. [Research supported in part by U. S. Army Electronics Research and Development Laboratory.]

This paper presents the results of recent experiments into the baseband excitation of the formant vocoder for use at data rates of 2400 and 4800 bit/sec. Among the various techniques that have been considered are the use of 1‐, 2‐, and 3‐bit amplitude sampling of the voice‐excitation baseband from 200 to 700 cps at rates compatible with the fulfillment of the total‐data‐rate objectives of 2400 and 4800 bit/sec, respectively. The paper is accompanied by tape recordings of system performance at the above data rates and sampling conditions.

The formant vocoder has now been under investigation for use as a speech‐compression system for more than a decade. During that time, significant advances have been made in the technology of the analysis and synthesis portions of the formant vocoder, and in the encoding of the parametric control signals for communication of the speech information at data rates normally of 1200 bit/sec and as low as 600 bit/sec. The consequence has been considerably improved vocal quality and articulation. Also, a voice‐excited operation mode has been developed for the formant vocoder that is designed to provide high‐quality speech transmission at a data rate of 4800 bit/sec. In this paper, some of the more important aspects of the formant‐vocoder system design are to be discussed. These include adaptive techniques for formant‐frequency tracking the use of the serial vocal analog for synthesis, encoding techniques for transmission of the vocoder parameters in binary form, and the voice‐excited mode of operation of the form...

Measurements were made of the random variation in the amplitude of sonic pulses at frequencies of 480 and 1420 kc/sec in a water medium perturbed in refractive index by thermally induced temperature microstructure. The received pulse‐amplitude variation was analyzed to determine time autocorrelation and variance over the transmitter to receiver separations ranging from 1633 to 7655 in. It was found that the time autocorrelation of sonic pulse amplitude was the same as that observed for the temperature microstructure, and that the variance varied as the square of separation distance. The above findings are compared with the theoretical findings of Mintzer and Bergmann. The observed equality in temperature microstructure and pulse amplitude autocorrelation indeed is predicted by Mintzer. However, neither Mintzer's nor Bergmann's theoretical developments predict the dependence of variance on the square of range. The observed departure of variance from the theoretical predictions is explained in terms of the ...

This paper discusses design and performance of a speech‐compression system that incorporates a formant‐tracking vocoder for the analysis‐synthesis of the vocalic utterances in speech, and a channel vocoder with 5 fixed‐frequency channels and 1 tracking channel for the analysis‐synthesis of the nonvocalic utterances of speech. The information is multiplexed and transmitted over eight 25‐cps low‐pass analog channels. For vocalic utterances, the parameters transmitted are the first 3 formants, their respective amplitudes, the frequency of pitch, and a separate vocalic nonvocalic decision signal. For nonvocalic utterances, the information transmitted is the amplitude of 5 vocoder channels. and the frequency and amplitude of 1 tracking channel covering the range from 2300 to 6000 cps, used to control a variable‐frequency pole‐zero filter in the synthesizer. For male talkers, this system demonstrates a mean articulation score of 87%. Confusion matrices and a demonstration tape recording are presented.

One of the formant vocoder's principal promises lies in the more efficient encoding of the word information content of speech than can be achieved by the use of the more familiar channel vocoder. The consequence of this more efficient encoding manifests itself in the capability for encoding speech for binary‐encoded formant data at binary‐data rates potentially as low as 150 bits/sec. In this paper, various avenues of approach for achieving such low‐binary‐data‐rate transmission are explored for system operation at various data rates ranging from 150 to 1200 bits/sec. Results of recent laboratory tests are presented.