The VNA stands for vector network analyzer. An instrument analyzing the s-parameters s11, s22, s21, and s12 parameters. It will measure the input and output impedance, gain, phase, and group delay.
It will show cable loss, oscillation, shorts, and impedance discontinuities. It will show phase shift. It is a very useful device.

The s-parameters math was developed during 50' and 60's. The instrument use theory about incident and reflected wave to calculate s-parameters. You can flip between the s-parameters, impedance, admittance, and polar plot.
Complex numbers are in Cartezian or Polar form. Have fun.
The VNA needs calibration. Calibration kit is 50 ohm, open, and short. Calibration kits with specific SWR are useful (SWR = 1.2). Measurement errors come from calibration kit, cables, finite directivity, and the software. Then set the delay, extentions. It is a good idea to read User Manual.
Without calibration, you do not know what you measure.

Market:
Multiple HF vna devices are on the market. Some measure single port VSWR (voltage standing wave ratio). Some devices spit out off-note numbers. Some devices are power thirsty. Some require carrying SoundBlaster card to the house roof top. Some are good as a flowerpot. Once connected to the antenna, it becomes a haywire device. "Borrow one" to learn what works, what doesn't work. Some products are prone to provide fake numbers with strong signal present in different band. Low ADC resolution can deliver fake phase angle.

My opinion:
I had 26GHz handheld FieldFox from HP (Agilent, Keysight). The software was user friendly and it gave transparent repeatable results. I can recommend it.
Then I had handheld Rohde&Schwarz. The user interface was pain in the ass.

Resonant antennas can be tuned with GDO and matched with tuner box.
A SWR meter will answer quickly how much energy flows to the load, and how much energy is reflected. SWR meter won't tell if the impedance is 42 ohms or 100 ohms. It is peppered by the transmission line, cable. Slight antenna mismatch can be worse than expected at the cable end. Along the transmisison line, the mismatched impedance moves in cyclic circle curve on the Smith Chart (i.e. 42 to 100 ohms).
Any antenna is tricky to tune, and tricky to match. The antenna works as an impedance transformer. You match it one way, the impedance moves little bit different way. The improved radiation change the parameters.
For HF, tune at least 0.2 Lamba from the metal fence. (4 meters on 20m band). Metal end-wires make difference. use Nylon. When you are done, the stuff really works. :-)

What I have seen around
- after few attempts to adjust two-band sloper HF antenna, the people realized the antenna simulation was completely wrong. Then real antenna adjustment with vna went nowhere. Lack of ideas? The vna lands at the bottom shelf. Then somebody propose to buy multi dipole antenna with shortened radiator sections from China. Then the story starts over again. This is not the right way.

The target is to get 50 ohms at the antenna feed point (with 50 ohm cable). When antenna is matched, the antenna and cable have both 50 ohms. Then the SWR, cable losses, and cable current nodes are minimized. The target is to force antenna to radiate, instead of the cable. Resonant antennas are probably best.
Antenna tuning and measurement is a two-man process. One person will measure, the second person makes notes while holding the ladder and providing important comments. Have a good time and sunny weather.

The MiniVNA was introduced few years ago
People ask questions and discuss what do you think about this and that.
Try it, and you will see.
MiniVNA is a single board hand held vna device for HF bands. The Minivna schematic diagram is drawn with Altium Designer - BS software from China. Thinking about the Minivna, I got some thoughts. Put it on paper.
The thoughts were proofed to fully correlate with people's experience after purchase.
If somebody ask me, I get a "lender device" and provide full evaluation. I Will.
Purchasing devices sealed in plastic, without full testing and evaluation option, is a waste of time.


MiniVNA, Very politely said - "There is a space for improvements"

1) Is the MiniVNA device with USB-connector the right concept, while ballancing roof top?
2) The chip Bandwith is above 1GHz. 10 MHz BW kills dynamic range by 70dB or more.
(System architecture problem).
3) There is no filtering. RF, IF, baseband, DSP.
4) There is missing Phase Sign for reactance.
5) It needs better S/N.
6) It needs reliable calibration kit.
7) It needs port extentions.
8) How is the broadband four-quadrant multiplier chip balanced?
9) How long will your batteries last?
10) It was a good try.   Now you can dispose it. Values with 100% Error have no meaning.

The N2PK' architecture is right.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
It is new whole generation of Euro-BS-Architecture Products (EBAP).
Ideology with PR. Marketing is more important than performance? then you sell polished farts.

My colleague bougth-and-sold six of these vna devices. The vna are all same. With (40% to 100%) ERRORS or more. I admire his patience and persistance.

One of three vna from eBay.com works. They measure high harmonic content.
(Hewlett Packard bridge precision is +-10% with +-20% on the range edges.

With the left oriented staff at EU and American universities, staff that never worked in any industry!, you have the whole picture. Universities teach Marxism for 15+ years. Then you get one generation of human dung.

It is "Debilisation of western society". Smart Phones, Internet, emails about nothing, videos, fuckbook.
Just for entertainment, look at Analog.com and Maxim's linearization appnotes of Wheatstone's Bridge. Nearly all formulae are wrong. One copies from another. It is math, fractions from elementary school grade 5?. The document dating is missing. These idiots sit there, well paid (Walt K.) and emit fake application notes full of mistakes. Philips is same. The application notes are edited by people without adequate qualification. Laugh?

The above applies to the semiconductor sector.
Compare ST Micro voltage regulator with two times higher thermal resistance Rth, to genuine National Semiconductor product. ST Micro product does not fit any commercial, industrial, or military thermal specs. ST arrogance.

When I saw vacuum cleaner ROWENTA (French, EU) with second air-filter placed under 6 screws by the motor (accessible only from inside the guts), you must take the whole thing apart. My wife was daring, can't believe it.! My thought was - YOU MUST BE KIDDING. These French designers are assholes.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

It is always about the people.
Self-education is one of the ways, how to became professional.