Delaylines
Delaylines
Delaylines
SPIEDigitalLibrary.org/conference-proceedings-of-spie
ABSTRACT
Noise radars use random or pseudo-random signals to illuminate a target and coherent detection techniques in order to
process received noise signals. In the range of microwave frequencies the simple analogue systems of the correlative
detector may be used. It helps to overcome the fact that the real digital implementation of the autocorrelation function for
these frequencies is very difficult. A correlator receiver is a typical element of the noise radar. The important issue is that
coherent reception needs delay lines of constant or variable parameters to be applied in the receiving systems. To address
this issue the paper comprises comparison of available technologies of microwave analogue delay lines. The advantages
and disadvantages of presented solutions are presented. The paper comprises: the types of basic delay unit technologies,
tunable and controlled-delay lines.
Keywords: delay line; analogue correlation; noise radar
1. INTRODUCTION
Delay lines with ability to tune the physical or electric length are used in a number of applications in the field of
microwave engineering. The most interesting application seems to be a reference signal delay in noise radars [1-4]. The
principle of noise radar technology and applications, especially to determine a target distance and radial velocity are
presented in a number of papers [5-10]. Although the full digital receivers are under the interest [11-13] detection of
target movement, especially micro-movement (e.g. breathing and heart rate for vital signs detection) can be done with
the use of analogue signal correlator. In order to perform the coherent reception of a transmitted signal a noise radar
receiving system needs delay lines of constant or variable parameters.
To address this issue the paper comprises comparison of available technologies of microwave analogue delay lines. The
advantages and disadvantages of presented solutions are described. It is assumed that the scope of the interest is
frequency range for delay lines equal to form 1 to 10 GHz and working bandwidth is equal to 1 GHz. A digitally-
controlled line with switched reference sections is introduced. Next, a microwave analogue tapped delay line is
presented.
As far as radar is concerned, the time delay of transmitted signal should be regarded as round-trip length and real radar
range is half a distance corresponding to full delay. For example, when radar to target distance is 50 meters the round-
trip signal path is equal to 100 meters and whole propagation time delay (round-trip) equals about 333.33 ns.
Therefore, for radar range equal to 50 m the delay line in signal processing block should be set to value of delay equal to
333.33 ns. This delay line may be implemented in various technology therefore its physical length does not need to be
equal to 100 m.
Radioelectronic Systems Conference 2019, edited by Piotr Kaniewski, Jan Matuszewski, Proc. of SPIE
Vol. 11442, 114421J · © 2020 SPIE · CCC code: 0277-786X/20/$21 · doi: 10.1117/12.2565097
The design and structure of the whole line is optimized in order to obtain the possibility of changing the propagation
constant (thus group delay) in line by applying bias voltage to varactor diodes thereby causing the change of equivalent
artificial line shunt capacitance.
Figure 2. Schematic of a nonlinear transmission line (NLTL) with periodically connected varactors
varactor diode
OUT
3 dB / 90 deg
IN
bias
The hybrid coupler may be designed with the use of distributed elements i.e. transmission lines or lumped elements as
LC structure with specially optimized values.
Disadvantages:
limited frequency bandwidth of the coupler,
non-uniform coupler’s response in frequency,
small values of time delay,
coupler’s insertion loses,
overall circuit insertion losses dependent on varactor bias voltage (that means circuit phase shift).
Advantages:
good impedance matching to external network resulting from the use of 3dB quadrature coupler,
possibility of precise adjustment of phase shift value.
n n 1 2 n1 (1)
where n is the number of the bit, and 1 is the time delay difference between states (or in other words delay at lowest
state). For example, for N = 3 bits the basic time delays equal 1, 21, and 41 (Figure 4).
The delay lines: the references and the basics may be connected between each other by means of synchronically operated
SPDT switches. The overall number of available delay values (and states) is 2N, and the obtained time delay is between
N0 (the first state) and (2N-1) 1 (the last state). General scheme of digitally-controlled delay line with switched delay
sections, for N=3, is shown in Fig. 3. Assuming that the value of basic time delay n contains the amount of reference
delay 0 , the first state introduces the time delay equal to N0 and the last state (2N-1) 1 + N0. Hence, the time delay
difference with respect to the reference path is from 0 to (2N-1) 1.
1
1
1
IN OUT
0 0 0
SPDT SPDT SPDT SPDT SPDT SPDT
Figure 4. General scheme of digitally-controlled delay line with switched delay sections, example for number N of bits equal
to 3.
There is a number of advantages and disadvantages related to digitally-controlled delay lines. For cascaded line with
switched delay sections they are as follows.
Disadvantages:
insertion losses of used SPDT switches and their additive accumulation due to cascading,
necessity of use of 2N switches for N bits of control,
potential existence of transient-state during simultaneous driving of switches,
influence of non-flat switch response within the frequency band (i.e. S11 and S21),
existence of non-zero reference delay resulting from non-zero length of connections in reference path,
there is only one value of time delay available at the time of switched-on state, that results from principle of
operation.
Advantages:
for number N of stages there is 2N available states i.e. time delay values.
matched load
IN
0.5 1 0.5 1 0.5 1 0.5 1
Z0
Figure 5. General scheme of digitally-controlled delay line with digitally-added basic delay sections (trombone line).
As far as the trombone line is concerned, there is no additive accumulation of insertion losses of the switches. In this line
there is only one switch in conducting state during the state of delay. However, there are additional power losses due to
signal absorption in matched loads ending the signal pathways. Physical structure of the line results that for N available
delay states there are N basic delay stages required.
Both types of digitally-controlled delay lines mentioned above require properly designed basic delay units i.e. having
insertion and return losses as low as possible. The basic delay lines may be realized in any kind of technology as far as
the requirements indicated above are fulfilled.
5.3 Reflection-type line with switched delay sections
The delay line structure may be constructed with the use of reflective sections of transmission line [33]. The reflective
scheme means that one end of a section of transmission line is short-circuited and the second end serves as input and
output of delayed signal. In order to separate the input and output signals a microwave circulator is used. An outline of
such a circuit is shown in Figure 6.
The short-circuited section of a transmission line serves as a unit delay element. In order to obtain switching of this value
a SPST switch is used, which in open state allows signal to enter the delay section and in close state provides the short-
circuit at the input of delay section. The microwave circulator may be replaced by a microwave coupler (3 dB 90 deg), as
it is shown in Figure 7.
SPST L1
1
Figure 6. General scheme of one stage reflective delay line with microwave circulator.
IN OUT
3 dB / 90 deg
coupler
SPST L1 L1 SPST
1 1
Figure 7. General scheme of one stage reflective delay line with microwave 3dB quadrature coupler.
The following scheme may be multiplied to obtain multi-state solution. This situation is shown in Figure 8.
Disadvantages:
limited frequency bandwidth of the circulator or coupler,
non-uniform circulator’s or coupler’s response in frequency,
circulator’s or coupler’s insertion loses,
short-circuit is realized with accuracy limited by the (signal losses) resistance of SPST switch,
doubled number of delay sections in case of coupler compared to scheme with circulator.
Advantages:
small length of delay sections, a short-circuited line delays signal two times more with respect to its length.
L 1 1 L 1 1 L 1 1 L 1 1 Z0
3 dB / 90 deg
coupler
L 1 1 L 1 1 L 1 1 L 1 1 matched load
IN
Z0
Figure 8. General scheme of one stage reflective delay line with microwave 3dB quadrature coupler.
IN Z0
The analogue tapped delay line allows obtaining all time-gate signals almost simultaneously, as the signal propagates
down the cascaded structure. However, in this case a signal switch (SPDT or SPST) implemented to set the one desired
value of time delay is not needed.
The analogue tapped cascaded delay line offers several advantages:
multiple values of time delay quasi-simultaneously,
absence of unstable or transient states,
absence of microwave switches and their insertion losses,
absence of driving and bias signals.
6. DISCUSSION
There are several technologies for implementation of microwave delay line, both commercially available and in state of
development.
Delay lines in BAW technology offer high insertion losses it means that the switched delay line set using these lines will
have very high signal losses. This fact practically excludes BAW lines as complex digitally-controlled lines component.
In this case the insertion losses of microwave switches are far less than BAW lines insertion losses. Moreover, switching
the subsequent states in digitally-controlled line toward higher time delay causes in this case the stepped increase of
whole delay path insertion losses. This effect is not avoidable even though the trombone structure is used.
The sections of TEM lines used as the unit delay components offer low insertion losses and the time delay is proportional
to the physical length of the line. It means that to obtain high value of time delay the adequately long line is required.
When the delay line is realized in coaxial technology the whole delay lines set may occupy quite big volume.
The use of planar technology, especially with substrates of high permittivity, allows to shorten physical dimensions of
delay lines. TEM or quasi-TEM planar lines may be in that case substantially shorter, however when semiconductor
substrates are used (MMIC) high signal losses may occur.
Further decrease of unit delay element dimensions is possible by means of the use of special planar structures, for
example 2D and 3D meander lines or distributed interdigital capacitance lines in LTCC technology. The delaying
structures in the form of band-pass filters based on coupled planar lines allows obtaining compact dimensions and
relatively high group delays, however at the cost of frequency band. The group delay is approximately constant for a
frequency range lower than the pass band. Closely to the edges od pass band the group delay strongly depends on
frequency.
Comparison of existing delay line technologies allows concluding that in spite of constant development there is no ideal
solution still. It is extremely difficult to combine the features like: low loses, high delay, low cost and small dimensions
in one delay line design or technology.
The choice of solution to implement a delay line in a real system depends on the project requirements. There is a way to
combine components available on the market with specialized solutions designed on demand. In the case of noise radar
with analogue correlator one may consider the combined solution including a set of switched and tapped delay lines. The
technique of analogue correlation is especially useful in a case of noise radar when the center frequency of transmitted
signal bandwidth equals about a dozen of GHz. For such a case the is extremely difficult to apply direct technique of
analogue-to-digital conversion.
REFERENCES
[1] Narayanan, R. M.; Xu, X. Principles and applications of coherent random noise radar technology, Proc. SPIE.,
5113, 503-514 (2003).
[2] Theron, I.P.; Walton, E.K.; Gunawan, S.; Cai, L. Ultrawide-band noise radar in the VHF/UHF band, IEEE Trans.
Antennas Propag., 47(6), 1080–1084, (1999).
[3] Lukin K. A. The principles of noise radar technology, first international workshop on the noise radar
technology. Proc. NTRW’2002, Yalta, 13–22, (2002).