A ntenna modules provide system integrators with a means of saving weight and volume in an airborne avionics platform. As mentioned in Part 1 of this article, combination Traffic Collision and Avoidance System (TCAS) and Transponder suites used on many aircraft rely on four different antennas, ten cables, and separate receivers and transmitters to provide protection against and awareness of other aircraft. Because of the use of discrete components, a TCAS system can be large in size and heavy in weight. But integration can provide considerable savings where savings in size and weight are required.

A conventional TCAS antenna array is passive. A TCAS transmitter with pulse power of about 200 W provides a transmit range of 45 nautical miles given typical directional antenna gain. During receive operation, signals received by the antenna elements are conveyed at very low power levels, because there is no power boosting from the passive antenna array. Receiver systems incorporating passive antenna arrays have an inherent receive range that is dependent, among other things, upon the characteristics of the passive components provided along the receive path, as well as noise, loss, sensitivity and other factors. The limited receive sensitivity is due in part to cable losses and noise within the receive signal path. In certain applications, it is desirable to provide a TCAS receive range of 100 nautical miles, which may not be possible with conventional passive antenna systems.

Figure 5 shows a block diagram of an active antenna module.

A transmit signal passes through one antenna port (6) and then splits and switches between four antenna BFN inputs (1, 2, 3, and 4). During directional transmit mode, each BFN input is activated alternately to provide directional antenna patterns for one quadrant (forward, right, aft, or left). During omnidirectional transmit mode, all four BFN inputs (1, 2, 3, and 4) are activated simultaneously. During directional receive mode, receive signals from the BFN pass through lowpass filters, switches, lownoise amplifiers (LNAs), and cables to the transmit-receive block. In the one common receive-transmit channel of the antenna module, the receive signal passes through the four-way splitter, two SPDT switches, and LNA2. The active antenna module provides greater receive sensitivity than a passive module, because the lossy cables and the transmit-receive block, placed after the LNAs, have a negligible effect on the overall receiver noise figure. The noise figure contribution of the transmission cables can be effectively eliminated, about 3 dB or a substantial portion of the receiver's noise budget. Mismatches at antenna module interfaces (cables, transmit-receive devices) also have a negligible effect on antenna module performance.

Several aspects of an antenna module adversely affect the performance of an amplitude monopulse system in estimating the bearing of surrounding aircraft11: (1) manufacturing tolerances, (2) different RF frequency, (3) antenna pattern shape, (4) sector to sector gain variation, (5) physical shape of the antenna, (6) variance of coupling between antenna ports, (7) antenna interface mismatching, (8) environmental conditions, and (9) component time-varying errors. The first five aspects can be corrected for by the antenna LUTs, and the next four aspects (69) can be corrected by the special calibration procedure. The amplitude monopulse system requires a strong amplitude balance between four receiver channels, including cables between the antenna module and the front-end receiver, because the aircraft bearing is measured by comparing the responses of the four receive channels. Difference in insertion loss and gain arise between these channels as a result of variation in parameters of physical components. The investigation results showed11 that the receiver channel loss/gain variance should be less than 0.5 dB to provide the acceptable bearing accuracy. Table 2 illustrates the influence of the destabilization antenna factors on the bearing error without an amplitude calibration.

To provide acceptable bearing accuracy,17 the following is recommended: (1) provide dynamic amplitude calibration; (2) use antenna LUTs; (3) the antenna pattern side/backlobe level should be less than -10 dB; (4) the antenna pattern gain variance from sector to sector should be less than 0.5 dB; (5) the antenna interface matching should be under control (return loss less than -15 dB); and (6) the variance of the differential coupling between antenna ports should be minimized.

Compensation of the destabilization factors can be realized by dynamic (during flight) amplitude calibration.10,11 For the amplitude monopulse system, the calibration should be done during a system operation called "on-line calibration," because the antenna and the receiver transfer functions can vary with environmental conditions (temperature, vibration, humidity, etc.). Usually, calibration is performed by the system manufacturer to compensate for the effects of manufacturing tolerances. However, imbalances due to changing conditions during flight require a dynamic calibration. There are two types of antenna array calibration. The first, a "radiative" calibration, employs free space as the calibration path between antenna monopoles. The second is the "non-radiative" approach, where calibration is performed by means of receiver channels and there is no radiation from the antenna array in the process of calibration. The second method of calibration does not correct for errors induced by the mutual coupling of antenna elements.

Consider a non-radiative calibration. Figure 2(a) (reproduced here from Part 1) illustrates the antenna SBFN with the calibration SPST switch.

The switch uses diode D3, which should be in the closed position during calibration mode, when the output power of hybrid H3 is reflected to the SBFN.10 Calibration of the system is achieved using a two-step process. In the first step, the calibration signal is applied to port 1 and reflected from the calibration switch to ports 2 and 3 of the SBFN. Information about the signal levels in these two channels is recorded in computer memory. During the second step, the calibration signal is applied to port 2 of the SBFN and passed to ports 1 and 4 of the SBFN. Finally, all four signals are compared to provide correction of the four output signals used for the bearing estimation.

In a radiative calibration, the calibration signal arrives at one antenna input, passes through the BFN, is radiated by the four antenna monopoles, and is received by the same monopoles due to mutual coupling between them. This calibration method uses the mutual coupling between antenna monopoles and does not require the special calibration switch. The mutual coupling is significant in small antenna arrays where monopoles are closely spaced. The amount of mutual coupling depends on the separation between antenna monopoles, antenna array geometry, and frequency. During the alternate activation of four channels by the calibration signal when, for example, channel 1 is activated, the three non-activated channels (2, 3, and 4) receive calibration signals due to mutual coupling between the antenna monopoles. The transmit/receive switches of these three channels are in receive mode and the receiver/ transmitter channels in these channels are in receive mode.