Trellis coding technique to increase adjacent channel interference protection ratio in land mobile radio systems under peak power constraints

A method of configuring a communication system, said system including a transmitter having an encoder with a memory length L and capable of being in a plurality of encoder states coupled to a mapper for mapping the encoder states to a phase angle signal, a pair of parallel branches coupled to an output of the mapper, each branch having a transmit filter coupled to the mapper for filtering the phase angle signal with a transfer function Q(f), a digital-to-analog (D/A) converter coupled to the transmit filter for sampling the filtered signal, a low pass filter and a modulator coupled in series with the D/A converter, a summer coupled to the modulator of each branch, and a transmit antenna coupled to the output of the summer, said system further including;

• a receiver having a receive antenna, a first down converter, a receive filter having a power spectrum H.sub.r (f), a second down converter, an analog-to-digital converter and a decoder all serially coupled;
• said method comprising the steps of:
• a) choosing a desired minimum acceptable value, (MPR), of power ratio P.sub.ib /P.sub.ab ;
• b) choosing a symbol period, T.sub.s ;
• c) choosing a number of input bits to be shifted into the encoder at each said symbol period thereby determining an encoder alphabet having M entries;
• d) choosing a number of bits to be output by the encoder during each said symbol period to determine length of an encoder output code and a number of constellation points;
• e) setting a probability associated with each said encoder alphabet entry uniformly to a single predetermined value thereby fixing a probability matrix X.sub.n ;
• f) setting a probability associated with each encoder output code uniformly to a single predetermined value thereby fixing a probability matrix Y.sub.n ;
• g) permutating all possible encoder states with all said encoder alphabet entries to result in a plurality of encoder sets U;
• h) choosing an encoder set as a present encoder set;
• i) calculating a P.sub.ib /P.sub.ab ratio for the present encoder set that has not been discarded according to the following equation:
• ##EQU18## the superscript H indicates Hermitian or conjugate transpose operator, M=number of input alphabet entries, N=number of encoder states, L=memory length of the shift register encoder, and T.sub.s is the symbol duration;
• j) adapting the encoder to provide a plurality of encoder output codes u from the present encoder set U according to the encoder's present state and a presently received encoder alphabet entry;
• k) comparing MPR with P.sub.ib /P.sub.ab, calculated in step "i" and discarding the present encoder set U if P.sub.ib /P.sub.ab does not exceed MPR;
• l) repeating steps "h"-"k" for all encoder sets U;
• m) determining a P.sub.DR value for each encoder set that has not been discarded according to the equation:
• ##EQU19## where P.sub.ave is an average power of a transmitted signal employing an encoder set, P.sub.peak is the peak power of the transmitted signal employing an encoder set, and d.sub.min.sup.2 is a minimum Euclidean distance associated with an encoder set;
• n) choosing an from the encoder sets that have not been discarded an encoder set having the greatest P.sub.DR is selected as the optimum coding set;
• o) adapting the transmitter to transmit the optimum coding set for all future transmissions;
• and
• p) adapting the decoder of the receiver to decode signals it receives according to the optimum coding set.