For reference and to protect from possible future edits, here is the circuit as you show it:
There are a number of obvious problems:
- You have to provide power to both the PIC and the attenuator.
- Just giving each power isn't enough. The grounds have to be connected.
- The power voltages must be the same, or you have to be really sure that a high output from the PIC is enough for the attenuator to interpret it as high, but to still be within its maximum operating spec.
The easiest would be to connect both chips to the same power supply, if there is a common voltage both can operate at.
- Each power pin should have its own bypass cap to ground, as physically close to the chip as reasonably possible.
- There are no such pins as Pn on PICs that I remember seeing. I haven't used the PIC32 family much, so maybe there is some special "P" function available in them. However, it looks like all you want are regular digital outputs. Those are called R<port><bit>, like "RB3", "RC7", etc.
If you are using some peripheral, as apposed to just programmed output pins, you should explain what it is and what it does for you that plain output pins can't.
I see you have edited the schematic to show separate 5.5 V power and common ground to each IC. That only addresses problems 1-3 above, not the rest.
This also causes new problems. In the datasheet, see section 29.0 Electrical Characteristics, subsection Absolute Maximum Ratings, third line, on page 151:
I can't even guess by what hallucination 5.5 V could possibly be acceptable.
Then in section 29.1 DC Characteristics, Table 29-1 Operating MIPs Vs Voltage, parameters DC5 and DC5b on page 152, it clearly shows:
So not only are you attempting to run the PIC well above its operating voltage, but also well above the voltage that can damage it.
Naturally anything the PIC does at this point can no longer be considered a surprise. Once absolute maximums are violated, all other promised made in the datasheet are null and void.