I. Introduction
The continued push for increased speed in wireless communication systems has led to the emergence of evermore sophisticated and efficient modulation schemes and communication systems' architectures. One of the critical components of such systems is the transmitter. In particular, its power efficiency and linearity are two key parameters that must be maximized to optimize the communication system's performance. One of the techniques that has been touted as a potential candidate for providing both high linearity and high efficiency is the linear amplification with nonlinear components (LINC) technique. This technique was first introduced under the LINC name by Cox [1] in 1974, but has it roots in the much earlier work of Chireix [2] in 1935. In this technique, the transmitter is made of two branches having a high-efficiency amplifier each, operated with constant envelope signals, with a combining structure at the output. The average efficiency of the LINC power amplifier is, therefore, the product of the amplifiers' efficiency and that of the combiner. As the individual amplifier efficiency can be maximized for a constant envelope signal by operating it in class C or overdriven class B, the overall efficiency of the LINC amplifier remains dependent on the combining structure used. The combiner's efficiency, in turn, depends on the signal's dynamics, i.e., its crest factor and probability distribution function (PDF) [3]– [5], as well as on the combiner's topology [3], [6], [7]. There are two classes of combiners that can be used at the output stage. The hybrid combiner is a matched and lossy combining structure with high isolation between the combined paths. Typical structures of this type include the hybrid coupler and the Wilkinson combiner. Due to the isolation between the combined paths, these combiners yield perfect linearity at the output. However, their efficiency tends to degrade rapidly as the crest factor of the signal increases. This is so because the out-of-phase components of the combined signals end up being delivered to the isolated port load, in the case of a hybrid coupler, and the isolation resistor, in the case of a Wilkinson combiner [8]. A power recycling technique has been proposed in [8] whereby the power that would normally be delivered to the isolation resistor is converted to dc and used to bias the amplifier, thus yielding improvement in the overall efficiency.