Phase shifters are at the heart of phased array antennas [23]. In simple terms, a phased array antenna consists of a set of phase shifters that control the amplitude and phase of the excitation to an array of antenna elements in order to set the beam phase front in a desired direction. While phase shifters that provide a lumped-element circuit phase shift as well as those providing a physical time delay phase shift may be employed in the implementation of a phased array, the true time delay approach enables frequency-independent beam steering, which permits the realization of phased arrays with wide instantaneous bandwidth [22]?
a highly desirable feature. Consider, for example, the conventional and true time delay phased array antenna schemes depicted . In both schemes, we find two antenna elements separated by a distance d, and driven through phase shifters in such a way that beams are set up in the direction q1 when the input frequency is w1, and q2 when the input frequency is w2. We notice, however, that whereas in Figure 4.19(a) the beam direction when the input frequency is w2 differs markedly from that when the frequency is q1, in Figure 4.19(b) the beam direction for the two frequencies is virtually identical. Let us examine this situation. To maximize radiation in the direction q1, the waves emitted from the adjacent antenna elements must interfere constructively in that direction, which requires that the path length difference between these waves (namely, k1, d sin q1) be equal to the phase difference with which the two elements are excited (namely, Df). Thus, we obtain the relation q1 = sin-1(Df ´ c/w1d), which gives the beam direction in terms of the difference in phase of the excitation between the two elements, the frequency, and the separation.
If the phase shift between elements Df varied linearly with frequency, then the ratio Df/w would be frequency-independent, and therefore, the beam direction would be independent from input signal bandwidth. This frequency-independence of the phase shift is difficult to achieve in lumped-element LC circuits; however, it is easy to obtain with the true time delay phase shifter approach, where Df = k(L2 – L1) = w(L2 – L1)/n and L and v are the physical length and the velocity of propagation in the delay lines, respectively. Inserting this value into the direction angle gives q = sin-1(c(L2 – L1)/nd ), which is frequency-independent. Clearly, since with a phased array antenna one is interested in directing the beam, possibly containing broadband signals, into a multitude of directions, it is necessary to employ true time delay phase shifters with not just two, but as many as practical phase shift states.
One phase shift topology, which is enabled by MEM switches, is shown. In this digital phase shifter, the overall phase shift is set by properly configuring the switches so that the RF signal is directed through one of 2N input-to-output path lengths and binary loop combinations. Clearly, the performance of the switches, particularly their insertion loss and isolation, is critical for the successful implementation of the scheme?and what better candidate than RF MEMS switches!
