What frequency range do you want to work over?
Your 'RHO' device seems to be a 4-port hybrid with a detector built in on one port. It isn't difficult to make another hybrid, unless you want particularly wide bandwidth. I made one that covers the MF and HF bands using a couple of toroidal ferrite cores and a bit of wire. A signal source feeds one port (HF/MF transmitter) and a load (antenna) goes on the associated through line port. I happened to build it for 1/144th power sampling. 1/2 would have been possible and any other perfect square ratio, but I was intending this one to work to a few hundred watts. The two other ports produce separated samples of the the forward wave and of the backwards wave on the transmitter to antenna path. I put a pair of attenuators on the 'sample' ports and fed their signals to a pair of Analogue devices logarithmic detector chips.
The voltage from the logger on the forwards port drives a meter with a power scale on it... a linear scale of dB(Watts) which ought to make Ofcom feel happy
The other logger drives a meter showing dB(Watts) of reflected power. A meter reading the two logger voltages has a scale of return loss linear in dBs which is handy. So I have a computing VSWR meter with a usefully expanded scale.
Going the whole hog, the AD logger chips I used also had limiter outputs just itching to be used, so I added a bit more fast gain with some ECL line receivers, and built a phase detector in ECL logic. So I have a phase meter.
This amounts to having made a vector network analyser that monitors my transmitter to antenna path. I get return loss and phase, ready to plot on a Smith chart as polar coordinates.
Setting up a three-knob ATU is normally fun as you have to explore a bit to find the best VSWR dip. But if you try it with a phase meter working, it becomes a lot easier to do. The phase meter tells you which way to go.
R&S made an impedance monitor that compared the vector forwards and backwards sample voltages on a CRT. Linear voltages, no logging needed or wanted, but a scaling function needed. The CRT had a Smith chart directly on it. They called it the 'Diagraph'
The 'which direction do I go?' uncertainty also affects simple things like LCR bridges. HP made a neat little one with an added phase detector that lit one of two neon arrows above the main knob saying 'this way to the null'
You can have a lot of fun with these things.
I made my vector VSWR meter 30 years ago with a couple of sample AD640 loggers courtesy of ADI's rep. Their current versions are a LOT cheaper and far less current hungry there are also some decent phase detectors available as integrated parts.
One problem with phase detectors is that they will be somewhat iffy near the wrap-around point, so you have a blind spot that can be awkward to avoid.
Oh, and my big goof was that I built the whole thing into a monster ATU of a type which is completely unsuited to the HF antennae I can fit in at home. Ooops!
David