This article is more than 1 year old

Full duplex! Bristol boffins demo Tx and Rx on the same frequency AT THE SAME TIME

It's all down to silicon chippery crunching the waves

Thinking practically

The Bristol team demonstrated how the interference from the transmit can be reduced by 50dB, taking it down to the level of general background noise.

While this is a significant processing challenge, Laughlin believes that the route to doing this in a low-cost handset is reasonably simple, and claims that the technology developed in Bristol should be cheaper than rivals' offerings, which are being worked on and refined at Stanford University (PDF).

"Our prototype uses Electrical Balance Isolation, which requires just one antenna and can be implemented on chip. Kumu's system has some advantages over our own, particularly that it can easily cancel signal imperfections introduced by the non-ideal characteristics of the transmitter chain," said Laughlin.

"However, our prototype uses low-cost technologies which can be implemented on-chip, making it more suitable for implementing full duplex in mobile device form factors," he added.

One of the challenges for the technology is that 5G is almost certainly going to employ MIMO, using lots of different frequencies at the same time. So it will be necessary to 'shout and listen' on each of these. Laughlin told The Register that:

The radio channel is different at different frequencies, which means all of the channel estimation and signal processing involved in MIMO will need to be done separately for different frequencies.

In terms of hardware implementation, there is a problem with the duplication of RF components (transmitters, receivers, filters and antennas) for different frequency bands. Current handsets typically have separate transceivers and antennas for different bands, which in a sense is doing it separately.

To enable cost effective implementation of radios which can cover many different frequency bands still requires further advances in tunable transceiver technologies. This is not only important for 5G, but would also reduce cost and improve roaming capabilities in today's LTE systems by replacing duplicated transceiver components with a single tunable radio system.

By the time we deploy 5G, it is very likely that there will be tunable transceivers which can cover wide frequency ranges, and so in terms of hardware it wont all be done separately.

There is a lot of IP flying around, and while the lightweight parts of the industry gets into spats about rounded corners and hysteresis on scrolling, the serious radio guys seem to be good at accommodating optimal solutions.

"In the recent flurry of interest in full duplex many patents are being filed, Stanford/Kumu being among the first, but many companies including all of the big cellular players have patents which are now published (and who knows what other patents are as yet unpublished)," said Laughlin.

"I expect this wont be a barrier to standardisation any more than was the case with previous breakthrough technologies such as MIMO, and that has found its way into standards and products without any huge problems," concluded Laughlin. ®

More about

TIP US OFF

Send us news


Other stories you might like