Design Article
Automotive ultrasonic ranging: Increasing gain may not improve detection distance
Arun T. Vemuri, Texas Instruments
6/11/2012 1:33 PM EDT
As ultrasonic-based distance ranging becomes more prevalent in applications such as blind spot detection, objects at distances greater than six meters (20 ft) have to be detected. The amplitude of the echo signal reflected by objects at far distances is very small. So there is a temptation to increase the amplifier gain, K, in order to detect objects at such distances. In this article, we show that increasing the amplifier gain, K, may not always result in the ability to detect objects at farther distances.
Background
One application for advanced driver assistance systems (ADAS) in a passenger car is ultrasonic-based distance ranging. Ultrasonic sound wave time-of-flight (TOF) is used to calculate distances to objects to assist the driver in parking the car, identifying parking spots, or detecting objects in the driver’s blind spot.
In ultrasonic-based ADAS, piezoelectric transducers typically are used to convert the ultrasonic waves into electrical signals. The receiver sensitivity of piezoelectric ultrasonic transducers usually is small, resulting in very small voltages. Figure 1, below, shows a typical signal chain used to process the echo voltage. (For an example of an integrated automotive ultrasonic signal conditioner for automotive park assist systems, see TI’s PGA450-Q1)
This echo signal, which is an AM signal, is corrupted with noise. The noise in Figure 1 is input-referred noise and is the sum of noise from external environment and from all components in the signal chain. This corrupted signal is then amplified by an amplifier with gain K. The amplified signal is digitized using an analog-to-digital converter (ADC). The digitized AM signal is bandpass-filtered.
The bandpass filter (BPF) primarily is used to improve the signal’s signal-to-noise ratio (SNR). The filtered signal level is compared against a threshold, L, to detect the presence of an object. Bandpass filters typically are followed by an amplitude demodulator. However, for the purpose of this article, the demodulator is not relevant.
Threshold analysis
Next: Observations
Background
One application for advanced driver assistance systems (ADAS) in a passenger car is ultrasonic-based distance ranging. Ultrasonic sound wave time-of-flight (TOF) is used to calculate distances to objects to assist the driver in parking the car, identifying parking spots, or detecting objects in the driver’s blind spot.
In ultrasonic-based ADAS, piezoelectric transducers typically are used to convert the ultrasonic waves into electrical signals. The receiver sensitivity of piezoelectric ultrasonic transducers usually is small, resulting in very small voltages. Figure 1, below, shows a typical signal chain used to process the echo voltage. (For an example of an integrated automotive ultrasonic signal conditioner for automotive park assist systems, see TI’s PGA450-Q1)
Figure 1: Using ultrasonic-based echo processing to detect objects deals with noise—both external (shown) and internal.
This echo signal, which is an AM signal, is corrupted with noise. The noise in Figure 1 is input-referred noise and is the sum of noise from external environment and from all components in the signal chain. This corrupted signal is then amplified by an amplifier with gain K. The amplified signal is digitized using an analog-to-digital converter (ADC). The digitized AM signal is bandpass-filtered.
The bandpass filter (BPF) primarily is used to improve the signal’s signal-to-noise ratio (SNR). The filtered signal level is compared against a threshold, L, to detect the presence of an object. Bandpass filters typically are followed by an amplitude demodulator. However, for the purpose of this article, the demodulator is not relevant.
Threshold analysis
Next: Observations
|
|
|
|
|
Navigate to related information
T.ORR
6/13/2012 1:16 PM EDT
Hi
Did I miss something - like how do you make it better.
A bigger TX signal is about the best method.
Regards
Tim Orr
Sign in to Reply
cdhmanning
6/14/2012 9:40 PM EDT
There are other ways too.
As an example, consider process gain achieved by coded signals. That's used in GPS signals to give many dB of gain and noise rejection.
Just yelling louder doesn't help. Remember there are many vehicles out there and if they're all yelling, you still end up not hearing properly.
Sign in to Reply
Arun V
6/19/2012 1:03 PM EDT
Agreed and that is the point of this article. A stronger echo (i.e., yelling louder) or large gain in the signal chain may not help detect objects at longer distance. One has to address the noise aspects also to detect objects at longer distances.
Sign in to Reply
WKetel
6/15/2012 9:32 PM EDT
The minimum detectable signal (MDS) determines the maximum range, and the MDS in turn is determined by the signal to noise ratio,(SNR). Of course the most obvious way to raise the SNR is to increase the transmitted signal. Unfortunately there are often a number of challenges to that approach. Of course there are quite a fewq other ways, but they are mostly not simple and not cheap, and sometimes they don't work very fast, either. Of course, speed is always relative, so it may be that they work fast enough.
Sign in to Reply
Arun V
6/19/2012 1:05 PM EDT
Agreed that this is a challenge. To your point, one has to consider many aspects and not just the gain of the signal chain to improve MDS.
Sign in to Reply
GREAT-Terry
7/9/2012 9:43 PM EDT
This is the same in almost all kind of signal chain processing, increasing gain sometimes won't help if the noise in front of the gain stage (including the noise from the amplifier itself) is dominating.
Sign in to Reply