Effects of noise and jitter on algorithms for bandpass sampling in radio receivers

Abstract

Bandpass sampling (BPS) is an undersampling technique by intentional aliasing. Conventional uniform discrete sampling within an f/sub s/ band normally results in a bad signal-to-noise ratio (SNR) due to signal spectrum aliasing. The noise combined in each of the f/sub s/ bands below the highest frequency of the signal (the so called noise spectrum aliasing) and timing jitter are two causes of performance degradation in BPS system. Nonuniform BPS has the potential to suppress signal spectrum aliasing due to the aperiodic property of nonuniform sampling (NUS). In this paper, the frequency spectra of uniform sampling (US) and NUS are analyzed, signal spectrum aliasing, noise spectrum aliasing and jitter effects in BPS are studied. Finally, the performance of reconstruction algorithms (RAs) for nonuniform BPS in the presence of sources of performance degradation are discussed based on simulations.