Chapter 6
Detection of Radio Signals and Its
Parameters Measuring
6.1 Detection of Radio Signals
Issues on detection of radio signals and its parameters measurement will be examined
in this chapter on example of detection of radar signals (detection of radar targets).
Decision making on target detection can be done based on threshold principle; i.e.,
a target is considered detected if output signal of optimum system y out exceeds some
specified level—threshold y thr . In Fig. 6.1, an output signal of processing system
y out by module during measuring of reference signal delay t s at presence of several
targets and in noise conditions effect is depicted as an example.
It is obvious, that at no-noise case, any signal increase y out above zero level will
testify on target presence in observation zone. However, noises availability extremely
complicates a situation since a noise component of output signal y out makes difficult
to answer for sure a question on a target presence. From Fig. 6.1, it follows that a
lot of “bursts” of output signal is observed, which could be mistaken for a sign of
target presence. In this case, it is reasonable to set some (threshold) y thr level and
consider for a sign of target presence only those bursts, which exceed a threshold.
It is clear, that in Fig. 6.1 a false burst of “target” is available, which is stipulated
by a noise signal. It is generally accepted to consider such burst as “false alarm.” A
question arises, how to select a y thr level to provide a reliable detection of a target and
decreasing of threshold probability increase by noise bursts or other types of noises?
To solve this problem we use methods, developed in theory of statistical hypothesis
testing.
At each element of delay in range t s and in Doppler frequency f s an output signal
y out can exceed thresholds or not exceed it. With this connection, in simplest case
of binary selection of statistical events we adopt conventionally that in observation
zone (at each t s and f s position) a target can be presented (first hypothesis) or a
target is absent (second hypothesis). As an output signal y out is a random, then four
statistical events are possible at decision making on target detection.
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021
D. A. Akmaykin et al., Theoretical Foundations of Radar Location
and Radio Navigation, Springer Aerospace Technology,
https://doi.org/10.1007/978-981-33-6514-8_6
127
Detection of Radio Signals and Its
Parameters Measuring
6.1 Detection of Radio Signals
Issues on detection of radio signals and its parameters measurement will be examined
in this chapter on example of detection of radar signals (detection of radar targets).
Decision making on target detection can be done based on threshold principle; i.e.,
a target is considered detected if output signal of optimum system y out exceeds some
specified level—threshold y thr . In Fig. 6.1, an output signal of processing system
y out by module during measuring of reference signal delay t s at presence of several
targets and in noise conditions effect is depicted as an example.
It is obvious, that at no-noise case, any signal increase y out above zero level will
testify on target presence in observation zone. However, noises availability extremely
complicates a situation since a noise component of output signal y out makes difficult
to answer for sure a question on a target presence. From Fig. 6.1, it follows that a
lot of “bursts” of output signal is observed, which could be mistaken for a sign of
target presence. In this case, it is reasonable to set some (threshold) y thr level and
consider for a sign of target presence only those bursts, which exceed a threshold.
It is clear, that in Fig. 6.1 a false burst of “target” is available, which is stipulated
by a noise signal. It is generally accepted to consider such burst as “false alarm.” A
question arises, how to select a y thr level to provide a reliable detection of a target and
decreasing of threshold probability increase by noise bursts or other types of noises?
To solve this problem we use methods, developed in theory of statistical hypothesis
testing.
At each element of delay in range t s and in Doppler frequency f s an output signal
y out can exceed thresholds or not exceed it. With this connection, in simplest case
of binary selection of statistical events we adopt conventionally that in observation
zone (at each t s and f s position) a target can be presented (first hypothesis) or a
target is absent (second hypothesis). As an output signal y out is a random, then four
statistical events are possible at decision making on target detection.
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2021
D. A. Akmaykin et al., Theoretical Foundations of Radar Location
and Radio Navigation, Springer Aerospace Technology,
https://doi.org/10.1007/978-981-33-6514-8_6
127