Low peak-to-average-power ratio (PAPR) transmissions significantly improve the cell coverage as they enable high power transmissions. A new modulation scheme, namely \pi/2- BPSK, was introduced in the Rel-15 3GPP 5G NR specifications to support low PAPR data transmissions in the uplink. However, in the existing 5G NR specifications, the reference signals employed for coherent demodulation of data symbols have higher PAPR than the data signals. This will potentially limit the cell coverage. It is, therefore, necessary to design reference signals that have low PAPR compared to the data signals. In this paper, we first present an architecture to minimize the PAPR of a binary sequence. We then present a systematic search procedure to obtain the best set of binary sequences, which, when employed for a reference signal generation, results in the best channel estimation performance. We show via simulations that the obtained binary sequences coupled with the architecture have a PAPR 2 dB smaller than the existing reference signals without compromising on the channel estimation performance. © 2020 IEEE.