Heart rate variability (HRV) denotes the natural variation in the time gaps between the consecutive heartbeats due to physiological changes. The HRV analysis is useful in the diagnosis of different clinical and functional conditions. Computation of HRV parameters from high-resolution electrocardiogram (ECG) signals on resource-constrained wearable internet of things (IoT) devices poses challenges due to the substantial computational load and increased energy consumption. Compressed ECG sensing for R-peak detection emerges as an approach for fast processing. Frequency domain HRV (FDHRV) analysis based on the power spectrum is capable for quantifying various clinical and functional conditions through the examination of ultra-short-term (1 minute) segments of ECG data. This paper presents a comparative study of FDHRV analysis using compressed ECG sensing for R-peak detection and FDHRV analysis through conventional high resolution R-peak detection. For this comparative study, 48 ECG records of MIT-BIH arrhythmia database and 18 ECG records of MIT-BIH normal sinus rhythm (NSR) database are used. During comparative FDHRV analysis, we have observed an average deviation of 0.003 Hz in respective band-wise peak frequencies. For both databases, the observed peak-frequency locations are 99.25% and 99.23% similar, respectively. Also, compared methods have provided relative power with 99.64% and 94.68% similar values for both databases, respectively. These findings support the suitability of compressed ECG sensing based FDHRV analysis for fast health monitoring using an ultra-short duration ECG signal. © 2023 IEEE.