Intelligent cars with dumb drivers?

I wrote before about the idea of intelligent transport systems that allow cars to automatically drive people to destinations along highways that are excluded from human drivers.  I also mentioned about possible benefits of better packing density of cars on roads, fewer accidents and less congestion delay.  Much of this vision would be enhanced by inter-car communications, allowing the systems controlling the cars to talk to each other in a peer to peer type chain along a road.  

The EU has now agreed to reserve 30MHz of radio spectrum in the 5.9GHz band for precisely this function.  Of course they quote the same benefits, highlighting the number of road transport deaths due to accidents, and the amount of money lost due to congestion on European roads.  But can systems based on car to car communications provide any better information to drivers that they can act upon to attain these benefits?  Well I think it is marginal unless we take the bigger step of removing the driver from the control function of the car completely.  Making vehicles more intelligent without giving them control in place of humans is of little value.  This bigger step is much harder to achieve culturally.  It requires social change.  But one day we will.  

Set that screen loose!

I have written about my vision of devices splitting apart into individual functioning parts and being linked wirelessly in the future, once personal area networking (PAN) technology develops and standardises sufficiently.  One main benefit of this future will be the end of cables between devices and the display screens that they use.  Wireless display technology may be one of the first parts of devices to be split apart.  Televisions are typically becoming screens connected to set-top boxes that contain the tuner and other functions.  WiFi based connections to LCD projectors have been available for a couple of years but performance is slow and patchy so they haven't really taken on.  Wireless video needs more suitable protocols and increasingly more bandwidth.  To stream uncompressed high definition (HD) 1080p video wirelessly typically requires between 3 and 5 Gbit/s.  This is what a wireless replacement for the HDMI connections typically found on modern consumer video equipment would need to handle. 

Those looking forward to shipping video around their lounge without wires will be pleased to know that a couple of initiatives already exist, each backed by a number of major consumer electronics manufacturers.  "WiHD" was the first of the two, and operates in the 60GHz band providing a full 5Gbit/s but uses such a small wavelength radio that it is easily attenuated by walls and requiring line of sight between transmitter and receiver, meaning that a multi-room scenario would need extra nodes or repeaters.  The second initiative is known as WHDI, and uses the 5GHz band, has longer range but permits only 3Gbit/s bit-rate.   The danger is that the fight between these two initiatives could be reminiscent of the Betamax/VHS debate.  There is also a chance that UltraWideBand (UWB) technology may also intervene and help accelerate the pace of development of such initiatives.  The promise is that if standardisation and developments continue as per manufacturers' roadmaps, we could see this in consumer products in around five years time.  

It's in the mail ...

With 80% of email being spam these days, its hard to believe that anything of value is conveyed by this most basic and popular of internet services.  But it is possible to send money to an email address, and many websites use email addresses as usernames to access other systems that get even more done.  But in the future maybe we will be able to convey more complex things by email? 

Well, I'm sure that we'll invent a more suitable protocol than email but often the first idea of conveying stuff uses the most basic and best understood means available!  So what about when we begin to convey our emotions in real time over the net?  And the more awkward senses like smell and taste which are as yet not included in A/V?  Advances in haptics mean that we can already cope with touch.  Augmenting the sometimes very sterile online world with all of these additional characteristics will bring far more value to Internet communications, however much spam exists by then!

Wireless power again!

So Intel have now announced their latest results relating to increasing the efficiency of wireless induction systems for getting power to devices without cables.  The approach, originally termed WiTricity by MIT who first demonstrated a 45% efficient system, has now been improved by Intel researchers. Since then, MIT have managed 90% efficient systems, which is more like the numbers that would be demanded in today's energy conscious world.  However I tend to agree that it will be 5 years before such an approach could be commercialised for widespread use and probably substantially longer than this in practice. 

Meanwhile, the current crop of lithium-ion batteries that are common in consumer electronics products today are likely to be gradually replaced by the next chemical technology for batteries, based on silver-zinc, from 2009 onwards. 

And devices will continue to be smarter at using less power by selectively turning parts of their functionality off automatically.  

So the power consumption issue won't be addressed by any one silver bullet solution, but rather by developments in all three areas ... New power systems, new battery technology and smarter more intelligent power management in devices. 

Nanoscale sensors

I recently read about the combination of two different nanotechnologies which could result in more sensitive but lower power consuming sensor hardware.  The University of Southampton has been doing some work on this, through a European funded project, which would overcome the limitations of existing CMOS techniques.  The first nanotechnology used is nanomachining in order to actually construct such tiny sensor systems.  The second uses single-electron transistors consisting of a single bridge of airspaced silicon which confines individual electrons.  

This approach is being used to produce two types of sensor.  The first is able to sense the change in electrical conductance as a result of the charge transfer from a molecule captured on the bridge channel's surface.  The other sensor type uses the bridge gate in suspension to detect the small mass changes of captured molecules which signals an electrical change in the resonant frequency of the gate.

In a future world where sensors are a natural pervasive part of the environment, nanotechnology seems sure to have a number of applications ... the sensor approach described here being just the tip of the iceberg.  

Designing future chips

The current trend in computer processing chips is multi-core ... and I have written on this topic before.  But lets look at the design of these chips and the numbers involved.  First lets consider the designers - the skilled folk involved in this process.  Back in the 60's and 70's, there were probably around 5000 highly skilled engineers who could design chips, working for the main manufacturers of the first integrated circuits.  The advent of Application Specific Integrated Circuits (ASICs) probably took the numbers of designers to the next order (say 50000) and today's 'design' of Field Programmable Gate Array (FPGA) chips probably multiplies by the same factor again ... so about 500,000 people.  Further automation through hardware and software systems can extend the numbers to five million and it is conceivable in decades to come that altering the functions of chips will be possible by many people who have little or no familiarity with the technology itself!   Chips will eventually be self-analysing, self-repairing and programmed at extremely high abstraction levels.  

Secondly, lets look at the fabrication sizes.  Already the large manufacturers are talking about 32nm processes.  When it gets to 22nm, the smart money is on optical approaches.  By the time we reach 15nm it will be about economics even more ... (extreme ultraviolet may by then be less expensive than conventional optics).  And when the numbers get 'really small', at about 11nm, the wavelength of the light used will be bigger than the sizes of the features etched on the chip!   In such cases, the best guesses at present are the use of computational lithography to achieve the a profitable yield from the process.  

There aren't many companies on the planet today with the capability to build and run the manufacturing plant required for such processes. There may be even fewer who can cope with these future developments. 

HTC Diamond

Today I had the chance to play with HTC's Diamond phone for the first time.  It is the best HTC device I have seen.  It has a touch-screen and therefore the obvious comparison is the iPhone.  But that would be too superficial.  There is no multi-touch, no eco-system app store, and no deep level of integration between hardware, operating system and applications.  The latter means that the usual inconsistencies in user interface (fonts type and sizes, colour schemes, lack of shared behaviours between applications and hardware etc.) for Windows Mobile devices remain.  It is no iPhone.

However it is a very nice mobile phone, and the touch interface once practised is very usable.  It has some very nice hardware features (the touch wheel, accelerometers, etc.) but these are not exploited by all the applications which it offers.  It is slim, not too heavy and a nice form factor.  The screen size for such a small device is very good, and the resolution is excellent... video looks great on it.  HTC have done well with this device, and it is unlikely that such a phone would have come to market (along with all the other touch-screen phones) as soon as this had Apple not launched the iPhone.  We need more such innovation leaders so that very good phones like this are in the market for consumers to choose from.