Internet of Things ( IoT ) applications often analyze collected data using machine learning algorithms. As the amount of the data keeps increasing, many applications send the data to powerful systems, e.g., data centers, to run the learning algorithms . On the one hand, sending the original data is not desirable due to privacy and security concerns.On the other hand, many machine learning models may require unencrypted ( plaintext ) data, e.g., original images , to train models and perform inference . When offloading theses computation tasks, sensitive information may be exposed to the untrustworthy cloud system which is susceptible to internal and external attacks . In many IoT systems , the learning procedure should be performed with the data that is held by a large number of user devices at the edge of Internet . These users may be unwilling to share the original data with the cloud and other users if security concerns cannot be addressed.

Broadband receivers for applications such as TV band tuners, multi-band cellular, or cognitive radio and software defined radio often employ a passive mixer harmonic rejection down-converter for high linearity and harmonic folding rejection [1-6]. Conventionally, a multi-phase passive mixer with either transconductance [1] or resistive weighting [2-5] is employed to approximate a synthetic sinusoidal down-converting clock. Such conventional harmonic rejection mixer (HRM) designs suffer from device and resistive noise limiting receiver sensitivity, as well as nonlinear distortion due to voltage-dependent transconductance or switch resistance, and finite input impedance of transimpedance amplifiers.  Noise-cancelling schemes compensate partially for the heightened noise levels due to resistive passive mixing [7] and incur greater circuit complexity.  Furthermore, the vast majority of HRM designs [1-7] operate at relatively high power consumption levels limiting their use for mobile applications operating under stringent battery size and weight constraints. The present invention fundamentally advances over the prior art in the use of charge-based (or capacitive) passive HRM, rather than transconductance-based and other active HRM (e.g., US Patent 7,738,851 [8]) or resistance-based passive HRM (e.g., US Patent 7,738,851 [9]). Distinct advantages of the charge-based HRM invention, overcoming the above shortcomings, are described below.

Brief description not available