Explicit closed-form expressions of the bit error probabilities are obtained for space-time block codes based on generalized orthogonal designs with differential encoding and differential detection utilizing 2/sup b/-ary phase shift keying mapping. The frequency non-selective, block-wise constant Rayleigh fading channel is considered here. The results are applicable to any arbitrary Doppler spectrum and any number of transmit and receive antennas, where the number of transmit antennas is dictated by the available coding schemes.

Explicit closed-form expressions of the bit-error probabilities are obtained for space-time block codes based on generalized orthogonal designs with differential encoding and differential detection using 2/sup b/-ary phase-shift keying mapping. The frequency-nonselective, block-wise constant Rayleigh fading channel is considered here. The results are applicable to any number of transmit and receive antennas, where the number of transmit antennas is dictated by the available coding schemes.

This paper presents a fundamental approach for deriving the bit error probability of BDPSK and QDPSK over the nonselective Rayleigh fading channel for a receiver with an arbitrary IF filter, and for a fading process with an arbitrary Doppler spectrum with arbitrary Doppler bandwidth. The results generalize those published earlier which were restricted to matched filter reception and to a fading process with a small Doppler bandwidth compared to the symbol rate. This allows the error probability to be studied in the presence of varying degrees of ISI due to the bandlimitation of the received signal by the IF filter, and in the presence of fading fluctuations of various rates. The analytical approach presented is simple, and yet powerful in that it can handle the case of diversity reception. This is a great advantage over the alternative approach of using the distribution of the differential phase of the received signal over a symbol interval. The bit error probability results apply to both conventional BDPSK and QDPSK, as well as /spl pi//2-2DPSK and /spl pi//4-4DPSK, and allow the irreducible bit error probability as well as the SNR at which this irreducible value sets in to be studied as a function of the Doppler bandwidth and IF filter bandwidth. The computed results are applicable to the design of digital cellular mobile communication systems.