The Development of Smart Antennas Randy L. Haupt Utah State Universitv Electrical and Computer En ’ineering 4120 Old Main Hilf Logan, UT 84322-4120 haupt8ieee.org ~
Introduction A smart or adaptive antenna is an antenna that modifies it’s receive or transmit characteristics in order to enhance the antenna’s performance. The performance modification results from purposely modifying the signal cancellations that occur. Smart antennas are important for reducing the deleterious effects of intentional jamming signals, unintentional co-channel interference, and multipath. These antennas were first known as sidelobe cancelers then adaptive antennas and more recently smart antennas. A smart antenna is composed of two or more antennas. The amplitude andor time delay (phase) of the signals received by all the antennas are modified then combined in such a manner as to improve reception of the desired signal. This paper presents the development of smart antennas from sidelobe cancelers in the 1950’s until today.
Sidelobe Cancelers The first type of smart antenna was the sidelobe canceler developed at GE in the late 1950’s [1]. A sidelobe canceler has a high gain antenna for receiving the desired signal accompanied by one or more small low gain, broad beam antennas for sidelobe cancellation (Figure 1).
low gain
$TIi
feedback
output Figure 1. Single Howells-Applebaumloop for a sidelobe canceler.
0-7803-7070-8/01/$10.00 02001 IEEE
48
The outputs of the antennas are weighted and combined until the jamming signal is eliminated. The general idea is that the low gain antenna amplifies the jamming signal about the same as the desired signal and close to the amplification of the sidelobe that the jammer is incident upon. Thus, when the low gain antenna signal is subtracted from the high gain antenna signal, the desired signal remains close to the same while the jamming signal is cancelled. The feedback process where the weights were calculated was developed by Applebaum [2]. Combining several of these loops together in an array antenna created the adaptive array.
Adaptive Antennas The first adaptive antennas were arrays employing the sidelobe cancellation technique. An error results when the received signal differs from the desired signal En = d,
- W'S,
(1.1)
where is &n the difference between the desired signal (d) and the weighted (w) receive signal (s).The mean square error is given by
E [ E.'] = E [ d,'] - w'Sw
-2
4 E [w's]
(1.2)
where the signal covariance matrix is s,sz
"'
S,SN
1 (1.3)
The optimal adaptive weights are found by taking the gradient of (1.2) with respect to the weights and setting it to zero. The resulting least mean square (LMS) optimal weights is given by the Weiner solution
w,,
= S-'s
Inverting the signal covariance matrix is not always easy. Thus, a myriad of algorithms has been developed to avoid the matrix inversion. One technique is the least mean square algorithm [3], and it is based on using the steepest descent method to find the minimum of (1.2). The weights are updated iteratively and are given by
49
Adaptive Nulling Based on the Output Power Adaptive nulling described in the previous section has a receiver at every element in the array. This approach is a very expensive proposition and requires a method that maintains calibration of all the receivers. A much simpler approach makes use of conventional phased array architecture and varies the phase shifters and attenuators and phase shifters to minimize the total output power of the array. Phase only adaptive nulling has the least amount of hardware requirements of any adaptive nulling approach [4]. antenna V ' elements
BV
V
phase shifters amplitudf weights
Figure 2. Diagram of an adaptive array using the total output power as the feedback mechanism.
Digital Beamforming The signal covariance matrix is easily formed when every element in the array has a receiver. Ideally, placing an AD converter at each element in the array feeds a digital signal to the computer where all the beamforming and beam steering is done. Multiple beams are important in spatial discrimination of signals in a multisignal environment. Adaptively switching beams as well as placing nulls in sidelobe becomes relatively easy with a digital beamformer. Unfortunately, calibrating the hardware and developing the hardware necessary to do the processing is difficult and expensive. antenna elements AD converters
I
I
computer
Figure 3. Digital beamforming array.
50
Biological Beamforming Living things constantly adapt to their environments. Sometimes the time scale is short, such as chameleons changing color. Other times the adaptation is over along time scale, such as evolution. Several recent efforts have used neural networks [SI and genetic algorithms [6] to perform adaptive beamforming. This area of research is relatively new and shows some promise. Smart Antennas All this historical development leads to smart antennas. It's a new buzzword for adaptive antennas. Since commercial communications technology is driving the need for this type of antennas, there is a different niche from the military drive for adaptive antennas. In any event, the explosion of wireless communications applications necessitates the need for antennas that optimally adjust their performance based upon the signal environment. References [7] through [9] are good references to learn more about smart antennas. Bibliography [l] P.W. Howells, "Explorations in fixed and adaptive resolution at GE and SURC," IEEE AP-S Trans., Vol. 24, No. 5, Sep 76,575-584. [2] S. P. Applebaum, "Adaptive arrays,'' IEEE AP-S Trans., Vol. 24, No. 5, Sep 76,585-598. [3] B. Widrow, P.E. Mantey, L.J. Griffiths, and B.B. Goode, "Adaptive antenna systems," Proceedings of the IEEE, vol. 55, no. 12, Dec 67, 2143. [4] C.A. Baird and G.G. Rassweiler, "Adaptive nulling using digitally controlled phase-shifters," IEEE AP-S Trans., Vol. 24, No. 5, Sep 76, pp. 638-649. [5] Southall, H.L., Simmers, J.A., and ODonnell, T.H., "Direction finding in phased arrays with a neural network beamformer," IEEE AP-S Trans., Vol. 43, No. 12, Dec 97, pp. 1369-1374. [6] R. L. Haupt, "Phase-only adaptive nulling with genetic algorithms," IEEE AP-S Trans., vol. 45, May 97. [7] R. A. Monzingo and T. W. Miller, Introduction to Adaptive Antennas, New York: Wiley, 1980. [8] R.T. Compton, Jr., Adaptive Antennas Concepts and Performance, Englewood Cliffs, NJ: Prentice Hall, 1988. [9] T.S. Rappaport, Smart Antennas :Adaptive Arrays, Algorithms, & Wireless Position Location, New York: Prentice-Hall, 1998.
51
