PicoChip offers 3G savings… eventually
- Article 7 of 25
- Infoconomist, August 2001
PicoChip Designs claims to have found a way of recycling base-stations used to create existing mobile phone networks. However, the savings for mobile operators might take some time to emerge.
After spending an estimated €115 billion to acquire their third-generation (3G) licences, mobile telecoms operators across Europe now face an €450 billion bill for rolling out the infrastructure. Understandably, those operators are now looking for ways to cut the cost of such roll-outs.
Step forward PicoChip Designs. The 'fab-less' semiconductor design start-up from Bath, England says that its mobile communications chip design will allow operators to deploy and upgrade new base-stations without having to rip-out and replace obsolete hardware.
Chief operating officer (COO) Peter Claydon says that, typically, the chips inside base-stations can only support one mobile-phone system. “With base-stations, it can be a three-year cycle from the initial design of a chip to having it in production. But the problem in wireless communications is that the standards change much faster and there are lots of them,” he says.
PicoChip has designed and patented a parallel processing architecture called PicoArray, which integrates thousands of processing elements for decoding mobile transmissions on a single chip. This will reduce the amount of space required to house a base-station, as well as 'future-proofing' it. Looking ahead, when a new protocol replaces 3G, the phone companies will be able to upload software to the chips to make them capable of processing it, says Claydon.
“One of the major issues with 3G is that there are going to be far more base-stations. Replacing them for upgrades will be very expensive,” says Claydon. So, although there will be some savings for companies if they replace their current GSM-standard base-stations with PicoChip-enabled 3G stations, they will still have to increase the number of stations.
The company received B#5 million (€8.1m) in venture funding from Atlas Ventures and Pond Ventures at the start of June 2001 and expects to have prototypes ready by February 2002. However, that funding will run out by mid-2002 and the company will have to persuade investors to stump up more funding in order to take the technology to market.
But operators are already purchasing new hardware to support their 3G roll-out plans and, consequently, PicoChip risks getting left behind in the frenzy. Furthermore, PicoChip's broad idea is not new. Intel's “wireless internet on a chip” – called Internet Client Architecture (ICA) – and its Internet Exchange Architecture are based on the same approach. Competing against an entrenched industry giant such as Intel will not be easy, to say the least.