"Fixed" for what purpose? Full mobile wallets?
This would be unlikely, but I'm open to persuasion.
I don't know what you mean by "full" mobile wallets. Full node? If so, then only Moore's law will give us that, and it probably will. I don't know how far back you have to go for today's mobile devices (quad core CPUs, 64 GB of storage etc) exceed a typical desktop PC, but it isn't that far. You won't have to go far into the future for a mobile device exceed today's typical desktop computer, and the latter is certainly capable of running a full node.
If you mean some sort of method of operating a lightweight wallet on a phone that is short of a full node but doesn't rely on a trusted server, that is being worked on. I'm explicitly not promising this will succeed and there is no promised delivery whatsoever, so don't start calling this vaporware. I'm interested in a persuasive argument if there is one, that it "can't" be done.
Furthermore there is work being done to directly reduce the size of the blockchain itself, which is what I meant. I guess it could be argued that gets closer to there being a full node on a mobile device (i.e. trims some time off waiting for Moore's law).
Isn't the more significant issue the download time of the blockchain? Bitcoin is over 10GB and it isn't any where near the volume of Visa, not to mention the orders-of-magnitude increase for micro-transactions we need (to replace the Google monopoly ad model of funding the internet and to support Ethereum style crypto-economy).
And the one-time ring signatures (with transactions split into standardized factors to enable mixing) have another few orders-of-magnitude on top of that (at least is a constant multiple).
We are looking at block chain that exceeds a Terabyte easily. Might take a year to download it if we are talking about decentralization and accommodating slower connections.
The mini-blockchain can prune away most of that bloat, but one-time ring signatures are fundamentally incompatible with it as far as I can see.