** co-install with factory ECM is a very complex task (harder than PlugAndPlay, which itself is harder than a normal standalone engine-control - which already requires attention and patience)

Does this car have ?

* CAN-throttle, like InTake/DriveByWireThrottle

* cruisecontrol

* ABS? obviously yes

** ABS connection with ECM ? via CAN?

** ESP via CAN ?

* AC ? likely yes

** AC connection with ECM ? Simple on/off signal from ECU ? or via CAN ?

* MAF sensor ?

* MAP sensor ?

* lambda sensor ?

** wideband ?

** narrowband ?

* immobilizer

* boostcontrol ?

* traditional fuel pressure regulator with return line ? Or some fancy fuel-vaporizer type pressure-regulator ?

----

Hijack what?

We have to think about the work splitted between factory and piggy ECU. Min +18..32% bigger injectors needed, in any case. However, there are functions that can be controlled by either factory ECU or piggy.

* fully mapped cheat signals for the factory ECU: MAP (and/or MAF, and possibly MAT and CLT as well)

* fuel injectors controlled by factory ECU or piggy (piggy might need to use MAPdot acceleration enrichment, if TPS data not available)

* ignition controlled by factory ECU or piggy

* boost controlled by factory ECU or piggy

* IAC most certainly controlled by factory ECU (not piggy).

* other things, like fuel evaporator canister, ABS-connection and AC clutch handled by fatory ECU

----

For example, if all outputs are controlled by factory ECU, it would think MAP=190kPa, pumping 270 HP of air (injecting that much fuel), while in fact MAP=250kPa (and 350 HP of air), still injecting sufficient amount of fuel, due to the bigger injectors. (MAP/HP values just for the example, not real)

Concerns:

* ignition (if controlled by factory ECU) is not as retarded as you might like

** maybe use 98 or 99 octane fuel instead of 95 octane

* something might rail inside the factory ECU. This might trigger CEL to light up, or might cause some other unexpected result

** the ECU might not understand, why MAP is so low at idle

** there might be artificial limits, like for the maximum boost solenoid PWM duty. This might be circumvented by adjusting the wastegate mechanically.

* ECU might try to force stoich mixture up to 90kPa or so (rather using retarded ignition than richer mixture). With the cheating, ECU might try to force stoich upto MAP=120kPa real (when ECU thinks it's only 90kPa), which does not sound too good.

Involving the MAT and CLT signals in the cheating can help the final result, but make the system and mapping more complex, more error prone and riskier.

likewise, controlling more functions (like ignition) by the piggy adds more freedom, but adds complexity and other risks too. Life might produce more surprises than one might even think about, so just an example: controlling function from piggy might accidentally disable some safety mechanisms. Eg. if electronic throttle-plate stucks wide open, while the throttle pedal is released, the factory ECU might cut power with fuel and or spark. However, the piggy, not knowing about the malfunction, could happily continue adding power. Even with a manual gearbox, that is an unpleasant surprise. Dangerous.

If the car is FWD, than the whole project is certainly too expensive.

Even if the car is 4WD, than it only makes sense to experiment, if an experienced engine-tuner and an experienced programmer are available very cheap (like friends, who help for some beer+pizza+fun) and you are prepared to waste months, and take the risks (including having the car out of operation for some time, and possible damage to engine, car, or worse).