Performance of a novel selection diversity technique in an experimental TDMA system for digital portable radio communications

The authors present the measured average bit error and block error ratio performance of a coherent burst time-division multiple-access radio link in a simulated flat Rayleigh fading environment using a unique technique for implementing two-branch selection diversity. The link consists of a 500-kb/s quadrature amplitude modulation burst transmitter and a coherent receiver with a fast carrier recovery circuit. The receiver is switched between two signals which are corrupted by uncorrelated Rayleigh fading. If the transmitter is on before the actual data burst, received signal measurement and selection can be performed just before the reception of a data burst. This way, when data is to be demodulated, the receiver is already switched to the branch with the higher power. With this technique, the performance of selection diversity is achieved with only one receiver chain. Experimental results are presented showing link performance as functions of Rayleigh fading rate and the time delay between signal measurement/diversity selection and the actual data burst