Swept-tuned spectrum analyzer Gianfranco Miele, Ph.D www.eng.docente.unicas.it/gianfranco_miele g.miele@unicas.it
Video section Up until the mid-1970s, spectrum analyzers were purely analog. The displayed trace presented a continuous indication of the signal envelope, and no information was lost. However, analog displays had drawbacks. The major problem was in handling the long sweep times required for narrow resolution bandwidths. In the extreme case, the display became a spot that moved slowly across the cathode ray tube (CRT), with no real trace on the display. This problem was due to phosphor limited persistence.
Digital video section When digital circuitry became affordable in the mid-1970s, it was quickly put to use in spectrum analyzers. Once a trace had been digitized and put into memory, it was permanently available for display. It became an easy matter to update the display at a flickerfree rate without blooming or fading. The data in memory was updated at the sweep rate, and since the contents of memory were written to the display at a flicker-free rate, we could follow the updating as the analyzer swept through its selected frequency span just as we could with analog systems.
Digital video section Logarithmic amplifier Envelope detector Video filter ADC Buffer f S CPU Display Video section
What value should be displayed at each point? Digital displays had a very low horizontal and vertical resolution. As a consequence, we had to decide what value should be displayed for each display data point. This is also true nowadays where modern instruments adopt LC or LED displays. In fact even though they are characterized by a better resolution than old ones, the display data points are lower than the points that can be stored in a memory buffer.
What value should be displayed at each point? Assume that only N points can be depicted on our screen.
What value should be displayed at each point? each point must represent what has occurred over some frequency range and over some time interval (bucket).
What value should be displayed at each point? L L = span N 1 L = t sweep N 1
Detector A mathematical algorithm must be applied to the samples that are included in each bucket in order to select the bit of info to be displayed.
Sample detector The simplest algorithm that has been implemented is the sample detector. It chooses a particular sample among them that are included in a bucket. Some manufacturers prefer to select the sample at the center of each bucket, other prefer to consider the first or the last sample in each bucket.
Sample detector In this example each point represents the half division to its right. From AN 150 Agilent Spectrum Analysis Basics - Copyright Agilent Technologies
Sample detector The sample is taken at the end of the interval, and the point is plotted at the beginning (left side) of the interval. From AN 150 Agilent Spectrum Analysis Basics - Copyright Agilent Technologies
Sample detector While the sample mode does a good job of representing the randomness of noise, it is not the best algorithm for searching for unknown signals or for analyzing signals made up of discrete spectral components (sinusoidal signals). Unless a particular mixing product happens to be at the center frequency of the filter at the time a sample is taken, an incorrect value will be stored in memory and displayed. If RBW<<L the mixing product can be missed entirely.
Positive peak detector One way to insure that all sinusoids are reported at their true amplitudes is to display the maximum value encountered in each bucket. This operation is carried out by the positive peak detector or peak detector.
Positive peak detector From AN 150 Agilent Spectrum Analysis Basics - Copyright Agilent Technologies
Positive peak detector This detector is very good for analyzing discrete spectral components (sinusoidal signals) because none is missed regardless of the relationship between resolution bandwidth and the spacing between trace points. However, unlike sample detector, peak detector does not give a good representation of random noise because it only displays the maximum value in each bucket and ignores the true randomness of the noise. Moreover it is not good for analyzing noise-like signals.
Negative peak Negative peak detector displays the minimum value encountered in each bucket. It is generally available in most spectrum analyzers, though it is not used as often as other types of detectors. It is mostly used for comparing peak and negative peak results in electromagnetic compatibility (EMC) applications. This allows the user to differentiate CW from impulsive signals.
Negative peak From AN 150 Agilent Spectrum Analysis Basics - Copyright Agilent Technologies
Auto-peak detector Most spectrum analyzers provide this detector that allows the simultaneous display of maximum and minimum valuein each bucket. The two values are measured and their levels displayed, connected by a vertical line.
Auto-peak detector
Rosenfell (normal) detector It determines whether or not the signal rose and fell in the interval to be represented by the next trace point. If the signal both rose and fell in a bucket, it is classified as noise. In this case, the maximum value in the interval is plotted if the data point is odd numbered, and the minimum value is plotted if the data point is even numbered. Instead if the signal rose or fell in a bucket, it is classified as signal and the maximum value is plotted.
Rosenfell (normal) detector What happens when the resolution bandwidth is narrow, relative to a bucket? The signal will both rise and fall during the bucket. If the bucket happens to be an odd-numbered one, all is well. The maximum value encountered in the bucket is simply plotted as the next data point. However, if the bucket is even-numbered, then the minimum value in the bucket is plotted. The peak value of the this bucket is always compared to that of the next bucket (that is odd) and the greater of the two values is displayed.
Rosenfell (normal) detector Rosenfell detector can cause an error displaying two peaks when only one actually exists, especially if the tone is among three buckets and the first is odd. From AN 150 Agilent Spectrum Analysis Basics - Copyright Agilent Technologies
Rosenfell (normal) detector Even though it is the best detector in order to view signal and noise, it can produce a further measurement error. In fact, if the analyzing tone is between two buckets and the first is even the process may cause a maximum value to be displayed one data point too far to the right, but the offset is usually a small percentage of the span. From AN 150 Agilent Spectrum Analysis Basics - Copyright Agilent Technologies
Average detectors While spectrum analyzers typically collect amplitude data many times in each bucket, sample, peak and normal detector keeps only one of those values and throws away the rest. On the other hand, an averaging detector uses all the data values collected within the time (and frequency) interval of a bucket. Power (rms) Voltage Log-power (video)
Power (rms) averaging detector It takes the square root of the sum of the squares of the voltage data measured during the bucket interval, divided by the characteristic input impedance of the spectrum analyzer, normally 50 Ω. In this case, the samples of the envelope are required on a linear level scale. Power averaging calculates the true average power, and is best for measuring the power of complex signals. V rms = 1 N N i=1 v i 2 P = V rms 2 R
Voltage averaging detector It calculates the linear average of the voltage data measured during the bucket interval. Also for this calculation the samples of the envelope are required on a linear level scale. V av = 1 N N v i i=1
Voltage averaging detector From AN 150 Agilent Spectrum Analysis Basics - Copyright Agilent Technologies
Log-power (video) averaging detector It averages the logarithmic amplitude values (db) of the envelope signal measured during the bucket interval. Log power averaging is best for observing sinusoidal signals, especially those near noise. V av_db = 1 N N i=1 20 log 10 v i
Trace averaging Digital displays offer another choice for smoothing the displayed trace: trace averaging. In this case, averaging is accomplished over two or more sweeps on a point-by-point basis. At each display point, the new value is averaged in with the previously averaged data A avg,n = n 1 n A avg,n 1 + A n n where n is the current sweep A avg,n is the new averaged value A avg,n 1 is the prior averaged value A n is the measured value in the current sweep
Trace averaging As with video filtering, the degree of smoothing can be selected. We do this by setting the number of sweeps over which the averaging occurs. While trace averaging has no effect on sweep time, the time to reach a given degree of smoothing is about the same as with video filtering because of the number of sweeps required.
Trace averaging But we can get significantly different results from the two averaging methods on certain signals. For example, a signal with a spectrum that changes with time can yield a different average on each sweep when we use video filtering. However, if we choose trace averaging over many sweeps, we will get a value much closer to the true average.