Thermal Stability Analysis on Pattern-Dependent BER and SNR Decay

Drive-level pattern-dependent bit error rate (BER) and signal-to-noise ratio (SNR) degradation during thermal decay were studied. A 511 bit pseudorandom binary sequence was used to help reveal the fundamental relationship between degradation trends of SNR and BER. Degradation profiles of both SNR and BER within specific patterns were plotted and analyzed. In the short-term thermal decay, SNR mostly decayed at the nontransition part, while BER mostly decayed at the transition part. It was found that transition BER degradation was caused by the degradation of nontransition SNR, while nontransition bits hardly have any error until nontransition SNR degradation reached certain levels. It was difficult to observe nontransition BER degradation in thermal stability experiments, which were normally limited by time. Simulation is necessary to assist the prediction of thermal decay performance.