14-04-2014, 03:09 PM
Pulse compression for different types of radar signals
compression for different.pdf (Size: 1.43 MB / Downloads: 220)
Abstract
The aim of our project is to compare pulse compression of different waveforms using
correlation of real signals or matched filter for analytic signals. In this thesis a set of programs is
developed for this comparison. The characteristic of the matched filter is that it processes the
highest possible Signal to Noise ratio (SNR) under the assumption of white noise. The
implementation of the algorithm is made in Matlab.
Radar signal processors are usually carried out over a specified range window. Returns from
all targets within the received window are collected and passed through the matched filter to
perform the pulse compression. Because of the recent development of digital signal processors
(DSPs), this process is often performed digitally. The aim of this work is to get better quality in the
radar images, including SAR (Synthetic Aperture Radar) and to explain the characteristics of each
waveform. The thesis also includes an appendix, where the implemented programs and program
code are attached. The work also aims at illustrative and didactic purposes. The programs have
been developed so as to be easily understood and therefore useful for engineering students.
Introduction
1930s the radar scientists were not aware of the concept of matched filter and pulse
compression. They were still learning from experience how to maximize the output signal-to-noise
ratio for the simple pulse waveforms that were used at that time.
In almost all conditions, usually met in practice, maximizing the output-peak-signal-to-
noise ratio of a radar receiver maximizes the detectability of a target. A linear filter that does this
transformation is called a matched filter under the assumption of white noise. Thus a matched
filter, or a close approximation to it, is the basis for the design of almost all radar receivers
waveforms that have to be pulse compressed. Methods for the detection of desired signals and
the rejection of undesired noise, clutter and interference in radar are called radar signal
processing. The processes of matched filter and pulse compression, described next, are an
important example of a radar signal processor.
Analysis Matched Filter Response of Linear
Frequency Modulated Waveforms (chirp)
Preferably, most radar systems codes should permit long detection range and fine range
resolution. Therefore we have to transmit an extremely narrow pulse (high bandwidth) of
exceptionally high peak power if we use short pulses. However there are practical limits on the
peak power. To obtain long detection ranges for pulse delay ranging, very high power pulses must
be transmitted.
One solution to this dilemma is to use pulse compression. Which means, transmit
internally modulated pulses of sufficient bandwidth to provide the necessary average power at a
reasonable level of peak power (as we showed in chapter 2). Then, after reception, “compress”
the received echoes by decoding their modulation.
Linear frequency modulation (LFM) or often called chirp is the first and probably still the
most common method for transmitted pulse. It was developing during World War II, as can be
deduced from German, British and U. S patents (Cook and Bernfeld, 1967; Cook and Seibert 1988,
patent). The basic idea is to sweep the frequency band
linearly during the pulse duration .
Cross-Correlation
All transmitted and received waves are real. However a matched filter is based on the
assumption of complex signals. Therefore in many systems the sampled real signal is used to
match the signal. In this case both the received and the reference signals will be real and therefore
a correlation used to match the signals. There will be a difference to the complex form, what we
will illustrate in this chapter.