ADREM: Ad-hoc Dynamic Radio-spectrum Exploitation via Multi-phase Radio

Wireless communication at radio frequencies up to a few GHz has become ubiquitous, and acronyms like GSM, WLAN and Bluetooth sound familiar now to many people. To avoid interference, radio communication is subject to regulations, which assign dedicated parts of the radio frequency spectrum to dedicated services. Due to this static frequency planning, large parts of the spectrum remain temporarily locally unused. To better exploit the scarce available spectrum, new radio devices could monitor the locally available spectrum, and use unused or "white space", but “step aside” if traditional radio services enter the picture. This asks for a smart agile radio transmitter/receiver, aware of its radio environment, referred to as a “Cognitive Radio”.

As spectrum is scarce (remember the huge amounts paid for UMTS licenses!), regulatory organisations worldwide are now considering more dynamic scenarios to exploit the radio spectrum more effectively. Examples can be found by regulatory measures to allow Ultra-Wideband (UWB) equipment in the 3.1 – 10.4 GHz range. Although the Federal Communications Commission (FCC) in the US has set very relaxed requirements, the European Union is considering Detect-And-Avoid (DAA) schemes in the bands below 6 GHz in order to protect incumbent services like WiMAX. Although it is unclear what will be the exact future standards, it is quite clear that more dynamic scenarios with more flexible radio devices are highly desired.

Cognitive radio requires a very flexible radio, a kind of "software radio" with highly flexible hardware programmable by software. However, state-of-the-art radio functionality is still largely defined in hardware and is highly dedicated for one specific radio standard, optimized for cost and power consumption, but certainly not for flexibility. In this project we aim for innovations in the radio architecture, building on our software radio expertise.

As CMOS IC technology is the mainstream technology for smart signal processing, we aim for a CMOS Cognitive Radio. Towards this aim, we see three major challenges to be addressed:

  1. A frequency scanner which quickly analyzes the locally available spectrum for white spaces
  2. An agile radio transmitter using a modulation technique that can effectively use white spaces
  3. A flexible radio receiver that can demodulate cognitive radio signals in the presence of strong interferers (incumbent services)

These are scientifically challenging goals, because fast frequency scanning over a wide band conflicts with accurate sensing. An agile radio transmitter is difficult because the transmitter should not cause interference in other frequency bands, while traditional fixed-frequency band-pass filters cannot be used (are not programmable). The receiver is difficult as it has to cope with very strong radio signals of traditional radio devices, putting high requirements on the linearity of the cognitive radio receivers.

In recent publications, we proposed a new multiphase multipath radio transmitter architecture without any dedicated filters, which only causes very low interference. This can be seen as a breakthrough on the path towards dynamic spectrum access, and an overview paper on this subject was accepted for a special issue of IEEE Communication Magazine on dynamic spectrum access. Also, our recent findings on sampling jitter and software radio architectures can help to make cognitive radio feasible. Finally, Jaap Haartsen (inventor of Bluetooth) joined the Integrated Circuit Design (ICD) recently, so that we can bridge the traditional gap between radio system design and CMOS circuit design.

Free Joomla Templates designed by Web Hosting Top