There are at minimum 20k planes, but possibly up to 100k. Let's estimate that half of 20k planes have this installed, at an expenditure of one trillion dollars.
You're off by a factor of a thousand here -- 10k * $100k is $1 billion, not $1 trillion. According to the article, ongoing costs would be in the hundreds of millions per year per airline, so a rough order-of-magnitude estimate might be $1 billion per year.
Worth it? Maybe. I'd like to see some examples of where a system like this would have actually saved some lives first -- the article doesn't give any. After all, there's an awful lot of things you can do for $1 billion a year that are more likely to save lives than betting it on the off chance of an airliner crash.