NOx production is a byproduct of high temperatures, to minimise NOx production you are really cutting the mechanical efficiency of the motor...
As far as charge tuning goes, the CS has relatively slow piston speeds (easy to work out with a bit of math mean and peak piston speed) and relatively low mean and peak gas induction speeds...
Tuning intake and exhaust works better at higher gas speeds, think japanese 2 stroke motorcycles, put a spanny chamber on the exhaust and at the right rpm range (don't forget piston valve 2 stroke = wads of valve overlap too) you can actually suck intake air in with the exhaust gases...
With a CS there is bugger all valve overlap, any gains you get are going to be marginal, and probably more than eaten up by the system providing them, eg intercoolers etc... with minimal valve overlap etc a long induction pipe can actually end up restricting intake flow...
personally if I was at altitude I'd use an centrifugal fan direct coupled to the crankshaft and supercharge a couple of psi that way, probably the most bang for your buck, lowest tech for reliability, and easy and cheap to do.
don't forget diesel is not stochastic, at less that full rack / power fuel air ratio is way low, so high altitude only limits peak BHP, even at constant RPM, which is an air pumping problem...
say for example at 10k feet you can only get 4 BHP, not 6, the simple solution is run a CS 8/1, eg 150 extra RPM, this changes your gearing requirements if driving an AC head and requiring a fixed frequency, but remember that in out example at 10k feet you 8/1 is only producing the same power as a 6/1 at sea level, eg it ain't working flat out.
Basically this whole problem falls neatly into the peak vs average demand problem, my original start-o-matic is rated at 2.5 KVA @ 50 Hz and 240 VAC from a 6/1, so it will produce 2.5 x 24 = 60 kWh in 24 hours, about 3 to 4 time my daily needs (guesstimating)
from this perspective, your actual problem lies in the area of peak vs average demand, yet you are looking for solutions in the wrong area, by tweaking the motor, so your solutions will never be optimal...
we used to get this ALL THE TIME on the boats, owners would size the genny to run the AC, battery chargers, lights, boil a kettle and run the cooker, so you'd get 50 foot boats with 17 or 20 kw gen sets, which meant buying 2 pole 3000 rpm units (small and compact to actually fit in the engine room, plus cheap per kw rating) which spent most of their lives running at 25% capacity of less, which meant they wore out quick and starting burning oil quicker....
DC is every bit as good as AC for everything except running synchronous rotating machinery, water heaters and bar fires work fine on DC, as do lights, and there is nothing in a washing machine that requires that the drum be driven by an AC motor, etc etc etc.....
MAny years ago I converted a washing machine from AC to DC motor power, and in doing so by passed all the programme electronics, so DC powered the heating elements, and DC powered the drum motor, necessity forced this on me at the time, DC was available in abundance and a DC servo motor was available..... you'd thing cuttiing out all the rotate clockwise for ten revolutions, pause for ten seconds then rotate anticlockwise for ten revolutions, programming with "flip the switch to start the drum, then reverse it when you pass by 20 minutes later" would utterly destroy the effectiveness of the washing machine, at least, that's what all the washing machine adverts would have you believe, fact is nobody could tell the difference, and I mean nobody.
"WHEN THE SHIT HITS THE FAN" scenarios, the last thing you want is 21st century crap powered by a 19th century diesel, it don't make sense and it ain't sutainable.
In the same way that the CS is simple, gut a washing machine of everything except the drum, run DC through the heating elements and a DC servo motor to spin the drum and you have something that will do EXACTLY the same job, JUST AS WELL, but you can fix that sucker if need be.
that is one notably lacking thing I see, survival types running generators that are 100% dependent of 2 cent electronic components, the analogy here is the modern auto / truck alternator vs the old dynamo and regulator, the dyno and reg you could fix, and I never saw anyone moaning about the lack of output from a basic 22 amp lucas dynamo longer than it took to go to the shop and fit a 50 amp dynamo, which was of course 110% electrically compatible with everything else, but oh no, we got marketed into the cheaper to make, disposable alternator, just like we got marketed into points instead of magneto because it was cheaper to make, but marketed as being superior....
back when I was running dynamo and magneto bikes (BSA A10) one of the top UK Honda bods was down at the new local Honda dealership going on about this stuff, I said "bollocks" he started to explain to poor stupid deluded me why he was right, I pulled a plug lead off the demo CBX (it was a new bike then) pulled a spare plug from my pack and hit the ignition while holding it, said "ow", and got him to do the same, then got him to hold it (first, I ain't stupid) while connected to one of the BSA HT leads, fucker dropped it like he'd been hit in the elbow with a pickaxe handle... I heard later that after I'd left he started using that as an example of why electronic ignition was superior, it was apparently "safer"..... bullshitting bastards the lot of them.
there is a moral here, the SOLE PURPOSE of making thinks like washing machines so complex is to place them beyond your reach, technologically, so they can control the source and implement new versions at will, the ACTUAL JOB done is no different....
A 5000 dollar programmable fancy oven doesn't cook sunday lunch or boil water for coffee ANY BETTER than a solid fuel Aga