A simple yet high performance time-to-digital converter (TDC) architecture is proposed in this paper. Its key advantage is its ability to sample-and-hold a time interval and thereafter oversample the stored quantity to provide sub-gate delay resolution and high linearity. The converter is fully digital, synthesizable from standard logic cells, and owes its properties to the time storage mechanism which relies on injecting more than one signal edge into a ring oscillator and tracking their relative angle. Results from a prototype on FPGA reveal excellent noise suppression by achieving a single-shot precision of 0.05 times the unit inverting logic cell delay in the ring oscillator by using an oversampling ratio of 64.

Digital pre-distortion is a powerful technique for combating nonlinearity in power amplifiers. Nonlinear response identification of the amplifier is one of the primary objectives in all pre-distortion schemes. If the estimation performance of the adaptive algorithm is not fast and accurate enough, the system will not be able to satisfy the stringent timing and spectral requirements of modern standards. This problem is exacerbated in modern communication systems which are rapidly converging towards a multi-carrier scheme. In such schemes, due to high peak to average power ratios, the convergence of pre-distorters worsens significantly. This paper highlights the problem of slow convergence in multi-carrier systems, and proposes a novel solution, which consists of implementing a twin feedback loop, one including the power amplifier and the other including the pre- distortion system itself. The goal is to leverage the learning curve of the pre-distorter itself and build on what is already known from the partially converged characteristics. Simulations with standard multi-carrier schemes indicate that such a method can deliver vastly improved performance over conventional systems.

Power amplifiers (PAs) are inherently nonlinear devices and are used in virtually all communications systems. Digital baseband predistortion is a highly cost-effective way to linearize PAs, but most existing architectures assume that the PA has a memoryless nonlinearity. For wider bandwidth applications such as wideband code-division multiple access (WCDMA) or wideband orthogonal frequency-division multiplexing (W-OFDM), PA memory effects can no longer be ignored, and memoryless predistortion has limited effectiveness. In this paper, instead of focusing on a particular PA model and building a corresponding predistorter, we focus directly on the predistorter structure. In particular, we propose a memory polynomial model for the predistorter and implement it using an indirect learning architecture. Linearization performance is demonstrated on a three-carrier WCDMA signal.

Power amplifiers (PAs) are inherently nonlinear devices and are used in virtually all communications systems. Digital baseband predistortion is a highly cost effective way to linearize PAs, but most existing architectures assume that the PA has a memoryless nonlinearity. For wider bandwidth applications such as WCDMA, PA memory effects can no longer be ignored, and memoryless predistortion has limited effectiveness. In this paper, instead of focusing on a particular PA model and building a corresponding predistorter, we focus directly on the predistorter structure. In particular, we propose a memory polynomial model for the predistorter and implement it using an indirect learning architecture. Linearization performance is demonstrated on a 3-carrier UMTS signal.

Power amplifiers (PAs) are important elements in communications systems and they are inherently nonlinear. Nonlinearity generates spectral regrowth (i.e., spectral broadening) in digitally modulated signals which causes adjacent channel interference. In this paper, we present a closed-form expression for the auto-covariance function of the PA output, whose Fourier transform yields the output power spectral density (PSD). The PA input does not need to be Gaussian and the PSD calculation can be carried out using common PA descriptions such as the AM/AM and AM/PM characteristics. We assume that the input is narrow-band, thus PA memory effects, on the order of a symbol period, are not present. The analytical results allow us to predict spectral regrowth without running expensive or time-consuming time-domain simulations; they also lead to a variety of optimization possibilities in transmitter design that includes the PAs.