Smart Optical-Wireless In-Home Communication Infrastructure
Project Manager: Prof. dr. ir. Gerard Smit
Faculty of Electrical Engineering, Mathematics and Computer Science
Tel.: +31 53 489 3734
Today’s society is confronted with immense challenges like climate change, depletion of resources, aging society, etc. This has triggered a change in mindset from continuous growth to sustainability, smart buildings, smart mobility, etc. Today’s homes and buildings are responsible for 41% of the energy consumption in the European Union. They contain many functionalities like domestic appliances, heating, air-conditioning, lighting, electronics, multimedia communication networks, wireless systems, that are not responding adequately to actual situations: e.g., they do not adapt to actual energy needs, to roaming of the users, and mostly they are not aware of each other due to a lack of overall coordination and control. This leads to unnecessary energy consumption.
Saving energy using ICT
Especially in homes and buildings, energy can be saved by using modern technology with situation-aware devices that are able to communicate. Home automation encompasses the increased automation of appliances in residential dwellings through electronic means, to meet the specific needs of the inhabitants. The term “home automation” is used in contrast to the more mainstream "building automation", which refers to similar technology for general needs, particularly the automatic or semi- automatic control of lighting, doors and windows, heating, ventilation, air conditioning, etc. Both home and building automation heavily rely on sensors and actuators that interchange information via reliable communication links with control units.
For efficiency reasons, the communication for home/building automation should preferably be integrated with the communication system used for data exchange and multi-media entertainment, which is not the case today. Needless to say that the integrated system should be reliable, flexible, scalable, and of course it should consume minimal power. It should support fixed and mobile devices as well as low and high bit-rate devices efficiently. Finally, the wireless communication should work reliably with a minimum of electro-magnetic emission for energy-efficiency and to eliminate potential health hazards due to exposure to EM radiation in a world full of wireless devices.
New energy-efficient wired/wireless network concept
To meet the above wish list, we propose a novel energy-efficient in-building optical/wireless communication network. This network features intelligent routing and transmission functionalities for wireless services between ever-growing numbers of sensors and actuators, and of fixed and mobile wireless broadband communication devices (laptop computers, data servers, printing/scanning devices, multimedia entertainment terminals, ...). The network enables Home/Building Automation and Communication for saving energy (quantified in section 6c).
The network is based on an adaptively routed optical fiber backbone which feeds transparently its wireless end-points with radio signals. These wireless pico-cells operate in the 60GHz frequency band. Each pico-cell covers a single room, as 60GHz waves do not penetrate walls. Hence the distance between a radio end-point and a mobile terminal is small (e.g., less than 5 m). Because the cells only cover a single room, frequencies can be reused in neighbouring rooms and thus the network capacity for the whole building is a multiple of the capacity of a single antenna system2 (such as WLAN). Moreover, as the range of an antenna in our system is much smaller than in the WLAN case, the emitted radio power is reduced considerably, contributing to a better power efficiency and a reduction of potential health hazards due to prolonged exposure to electro-magnetic radiation. Literature  shows that the reduced cell sizes result in a gain in wireless transmission capacity of 1600x at RF power levels 2 orders of magnitude smaller than in traditional radio networks. Calculations show that a reduction of the overall power consumption of the hybrid fiber-radio system itself with one order of magnitude is possible compared to traditional full-wireless or copper/wireless solutions. Because of the increased capacity of the network, a power reduction per bit of three orders of magnitude is achievable. This figure is in line with the goals of the recently established “GreenTouch” consortium3  that stimulates research to increase the efficiency of communication networks with a factor of 1000 in year 2015. Many of the research items which we address fit in the GreenTouch goals. The network integrates a wide range of functionalities into a single, power-efficient future-proof infrastructure, which can accommodate a plethora of services.
In addition to the 60GHz pico-cell radio-over-fiber techniques, we propose that each antenna site does not cover the whole room (omni-directional), but sends out confined radio beams (‘pencil beams’), directed by means of adaptive radio beam steering (either or not via reflective surfaces) towards the fixed and mobile wireless devices in the room. In addition to the reduced power constraints of the pico- cell architecture, beam-steering contributes further to a more efficient use of power at improved communication quality. The well-directed radio power implies less mutual interference among beams, thus increasing the system performance with less power consumption.
Furthermore. we propose multiple beam-steered antennas per room: 60GHz waves require unobstructed (‘line-of-sight’) paths between antenna and the user terminal. With multiple antennas, obstructions can be avoided, and using MIMO (multiple input multiple output) antenna techniques the throughput and reliability of the system is improved. By deliberately using reflective surfaces available in the room, obstructions can be also circumvented in ‘non-line-of-sight’ conditions.
Project duration: 1-6-2011 / 1-6-2015
Project budget: 768.9 k-€
Number of person/months: 3 fte / yr
Project Coordinator: TU/e
Participants: TU/e, TU Delft, UT
Project budget CTIT: 256 k-€ funding
Number of person/years CTIT: 1.2 fte/year
Involved groups: Computer Architecture for Embedded Systems (CAES), Design and Analysis of Communication Systems (DACS)