An Analysis of the factors which determine signal/noise discrimination in pulsed-carrier systems

Abstract

DWIGHT 0. NORTH?, FELLOW, IEEE Summary-The smallest signal discernible through background noise is formulated in terms of the pdse energy, its repetition rate, the receiver design, and the choice of integrating and indicating means.The smallest signal visible on a Type-A scan can be improved upon by the use of electromechanical integrators. Integration before detection is, in theory, ultimately the most effective, but runs into serious practical difficulties. These are avoided when the integration is performed after detection.Optimum predetector selectivity is formulated. The optimum detector (square law) is found, and other detectors compared. A transmitter criterion provides a basis for comparing the effectiveness of transmitters, and shows how, at the expense of range resolution, longer pulses can increase visibility. I. ELEMENTS OF THE PROBLEMK H I N "TARGET" lying within a segment of range Ar reflects a pulsed carrier, producing a t the terminals of a receiving antenna an opencircuit voltage, e(t) = F ( t ) COS (00t + e).(1)With a few exceptions, the pulsing function will be supposed rectangular, duration r , and recurrent, period T , so that in a sample period, -T / 2 <t < T / 2 , IO7 F ( t ) = 1 v, -7 / 2 < t < 7 / 2(2) -T / 2 < t C -~/ 2 (0, 7 / 2 < t < T / 2 .The total system noise may be described by a statement of the equivalent open-circuit mean-square noise voltage at the same terminals, -en2 = X (operating) .4kToR,,Aj (3) where T o =room temperature, degrees Kelvin, R, =antenna resistance, Af = predetector noise bandwidth, k = Boltzmann's constant, N(operating) =operating noise factor [ 11 of receiver. (Note: This term refers to total system noise including that arriving via the antenna. It is related to the ordinary noise factor N of the receiver via N(operating) = N + T o / To -1, where T, denotes the noise temperature of the antenna. DON, 1963.) * RCA Laboratories, Princeton, N. J., Tech. Rept. No. PTR-6C;

Ultrasonic Flaw Detection Using Signal Matching Techniques

“…Theoretically, if the noise has a constant power spectral density over the frequency range occupied by the signal, then the improvement in signal-to-noise ratio caused by the matched filter is the best possible that can be achieved with any linear filter; and in the restricted case of Gaussian noise, it is the best attainable with any filter, linear or nonlinear. The matched filter technique has been used in a variety of application areas including radar [2], sonar [3], seismology [4], as well as in the field of NDE [1], [5]. In addition to being the optimal filter for detecting signals in colored noise, flaw size estimation can be simultaneously carried out in an M-ary system of multiple matched filter banks.…”

Application of Two-Dimensional Matched Filters to X-Ray Radiographic Flaw Detection and Enhancement

“…It originated in 1943 in pulsed radar detection [1] and has recently been extended to two-dimensions and applied to images in blood vessel detection, cloud clutter suppression and fingerprint enhancement [2][3][4]. It is essentially a crosscorrelation filter in which the correlation functions are the measured noisy signal or image and the expected ideal signal.…”

“…The one-dimensional matched filter is given by Eq. (1) j=l wheref(i) is the measured noisy signal and h(i) is the expected signal to be detected. This filter can be generalized to the two-dimensional image and is given by…”

An interpretation of Woodward's ambiguity function and its generalization

“…The measurement problem of radar is to design a radar waveform to be broadcast by a radar or sonar, so as to maximize the receiver response to the signal which has interacted with an object. The means to accomplish this was drawn from work by North [2] during WW II who invented the "matched filter". A matched filter is obtained by correlating a known signal template with an unknown signal to detect the presence or absence of the template in the unknown signal.…”