EXPERIMENTAL RESULTS FOR SPACE-TIME CODING USING ARTM TIER-1 MODULATION

When using two antennas to transmit telemetry from an airborne platform, self interference results when both transmit antennae are visible to the receive antenna. This self interference can lead to link outages and severe distortion, especially as data rates increase above 5 Mbits/sec. Space-time coding can be used to provide transmit diversity to overcome this self interference problem. This paper describes the results of experiments (conducted at Edwards Air Force Base, California, USA) using FQPSK-JR waveforms coded with ARTM Tier-1 Space-Time Block Code. INTRODUCTION Historically, air-to-ground telemetry links were comprised of a transmit antenna mounted on the underside of the fuselage and a fixed ground station equipped with a tracking antenna as illustrated in Figure 1 (a). However, aircraft maneuvers can place the place the fuselage in between the transmit antenna and receive antenna thus blocking the line-of-sight propagation path as illustrated in Figure 1 (b). The traditional solution has been the use of two antennas to transmit the same signal. The typical configuration uses one antenna mounted on the bottom of the fuselage and a second antenna mounted on the top of the fuselage. As illustrated in Figure 1 (c), when the line-of-sight propagation from the bottom antenna to the ground station is blocked, line-of-sight propagation from the top antenna is unobstructed and the link is maintained. Since the two antennas are separated in space, the two signals arrive at the ground station with different phases. This is not a problem when only one of the signals has an unobstructed line-of-sight propagation path to the ground station. When both signals have an unobstructed line-of-site propagation path to the ground station, some unintended behavior is observed. For certain aspect angles, the phases of the two signals are such that the signals reinforce each other. However, for other aspect angles, the phases of the two signals are such that the signals cancel (or nearly cancel) out. As a consequence, the two-transmit antenna system behaves as a single composite antenna with an undesirable gain pattern as illustrated in Figure 1 (d). Several solutions to this problem have been considered. One obvious solution is to use different carrier frequencies for the two antennas. While simple, this solution requires twice the bandwidth and can claim the use of two ground station antennas. The reallocation of 60 MHz in the lower S-band fromaeronautical telemetryto commercial uses in 1997 has exacerbated the competing demands on the remaining spectral allocations. This situation, coupled with the increasing data rate requirements for new systems to be tested, makes the solution unworkable. Another solution to this problem is to equip the aircraft with a steerable antenna thereby allowing the signal to be “pointed” directly at the ground station. This solution requires high resolution TSPI and aircraft attitude information which are required to compute the proper “pointing angles.” These family of solutions require substantial on-board processing capability as well as sophisticated antenna technology to implement electronically steerable antennas. An alternate solution, proposed by the authors [1], is to transmit two different, but related, signals from the two transmit antennas. The relationship between the two signals is defined by a space-time code [2, 3] called the ARTM Tier-1 Space-Time Block Code. The two signals are transmitted at the same time and on the same frequency using any of the bandwidth efficient ARTM Tier-1 waveforms described in IRIG 106-04 [4]. The code is designed to that the instantaneous phase relationship of the two signals is adjusted to avoid destructive interference when averaged over the length of the block code. In addition, all data can be recovered when only one of the two transmitted signals is available at the receiver. In comparison with the other solutions described above, this solution does not require additional spectrum and places only modest complexity increases (to be described below) on the airborne platform. The complexity is concentrated in the ground station, where size and weight are less important. This concept was tested in an experimental setting at Edwards AFB in early 2004. This paper describes the experiment and the results of the experiment. It is shown that a dual-antenna space-time coded system operating in a real environment can eliminate the link outages caused by the self-interference associated with the traditional dual-antenna configuration. MATHEMATICAL MODEL The basic system is illustrated in Figure 2. The signal transmitted form antenna 0 is an ARTM Tier-1 signal, which may be expressed as an offset modulation as s0(t) = ∑ k [ a0(k)pI,k(t− kTs) + jb0(k)pQ,k(t− Ts/2− kTs) ] (1) wherea0(k) ∈ {−1, +1} andb0(k) ∈ {−1, +1} are the inphase and quadrature symbols, respectively, transmitted during thek-th symbol interval;Ts is the symbol time (twice the bit time); and pI,k(t) and pQ,k(t) are the data-dependent pulse shapes used for the inphase and quadrature components, respectively. Similarly, the signal transmitted from antenna 1 may be expressed as s1(t) = ∑ k [ a1(k)pI,k(t− kTs) + jb1(k)pQ,k(t− Ts/2− kTs) ] (2)