Engineering topics
Publications and source records attributed to Swanson, Laif.
Report surveys state of art of coding digital data for nearly errorless communication over long distances. Coding techniques described include mainly ones that have been or might be used to transmit imagery and/or other data from spacecraft to receivers on Earth.
Shortest sequence of bits likely to be filled with error bursts examined. Algorithm improves synchronization of frames of noisy binary-coded data signals after Viterbi decoding (recovery from "inner" convolutional code used in transmission channel) and before Reed-Solomon or other decoding (recovery from "outer" error-correcting block code). Based on comparisons of sequences of correct and erroneous Viterbi-decoded received bits with known marker sequence denoting beginning of frame of data. Does not require count of number of bits in received sequence disagreeing with corresponding bits in marker sequence.
Repetition preserves most critical data during severe attenuation. Added to existing system using Reed-Solomon and convolutional encoding by additional components.
The observations of Posner (1983) that on a wideband Gaussian broadcast channel ordinary time-shared coding performs almost as well as broadcast coding are investigated. A quantitative version of Posner's results is derived. A numerical example comparing the performance of broadcast coding and time-shared coding for a Gaussian broadcast channel model is presented.
Upper bounds on the decoder error probability for Reed-Solomon codes are derived. By definition, decoder error occurs when the decoder finds a codeword other than the transmitted codeword; this is in contrast to decoder failure, which occurs when the decoder fails to find any codeword at all. The results imply, for example, that for a t error-correcting Reed-Solomon code of length q - 1 over GF(q), if more than t errors occur, the probability of decoder error is less than 1/t. In particular, for the Voyager Reed-Solomon code, the probability of decoder error given a word error is smaller than 3 x 10 to the minus 14th power. Thus, in a typical operating region with probability 100,000 of word error, the probability of undetected word error is about 10 to the minus 14th power.