00000ctm a22000003i 4500 UP-99796217612880638 Buklod 20230215105418.0 m |o d | ta 181017s xx d r |||| u| (iLib)UPD-00396664523 DENGII rda eng LG 995 2018 E64 S25 Santos, Christopher G. author. Optimal oversampling using probabilistic error modeling in low-power digital energy detection receivers Christopher G. Santos ; Louis P. Alarcon, adviser. Quezon City College of Engineering, University of the Philippines Diliman 2018. 129 leaves illustrations (some color) 28 cm text rdacontent unmediated rdamedia volume rdacarrier Thesis (Master of Science in Electrical Engineering)--University of the Philippines Diliman July 2018. F - no patentable invention or creation, not for personal publication and no confidential information. Yes - available to the general public. On-Off Keying (OOK) modulation has been a popular choice for communications in energy-constrained systems. For the receiver side, energy detection is preferred because of the error-reduction it offers. Parts of the energy detection process for OOK demodulation can be easily implemented in digital and oversampling makes the system more tolerant to erroneous inputs to the digital circuitry. Thus, with increased oversampling, the performance requirements of the preceding analog blocks can be relaxed, resulting in lower power consumption for the analog blocks. However, this comes at the cost of larger power consumption for the digital blocks. Thus, tradeoff in performance and power consumption of the analog and digital blocks must be considered in identifying the optimal oversampling for minimum power consumption. The study outlines and demonstrates the necessary steps to determining the optimal oversampling rate. Four error components were identified and modeled using the probability of transitions in the thresholding block output at specific sampling instances. The behavior of all four error components were found to be vary approximately as ko · (ERi)k. Using the error model of the system and given a fixed target bit error rate, the required input sample error rate E Ri for different oversampling rates can be identified. With this, the tradeoff in performance in terms of ERi and oversampling rate is established. On the other hand, oversampling was shown to be less effective when ERi increases and that oversampling cannot relax the ERi requirement beyond 0.5. Thus, the SNR at the output of the analog blocks can only be decreased to some SN Rinin and the power consumption cannot be further reduced. With the power of the digital block increasing at an approximately linear rate and with power consumption of the analog blocks decreasing with a lower limit as the oversampling rate is increased, the tradeoff in power consumption between the two blocks is established and the optimal oversampling rate can then be identified. Digital modulation. Energy detection. Alarcon, Louis P. adviser. Thesis FI UP UPD DENG-II LG 995 2018 E64 S25 Thesis