OverviewThe Center for Wireless Systems and Technology at TU Delft combines the expertise of 7 research groups in Electrical Engineering and Computer Science. The complete field of wireless communication is covered, including some related areas such as radar systems and radio astronomy signal processing. The research capacity is about 20 fte faculty, and over 80 fte scientific staff.
Sensor network communicationA specific research theme focuses on sensor networks. Seen as a distributed system, communication will often be point-to-point. This leads to questions on scalable spectrum management, distributed sensing, detection, estimation and control, low-power communication systems. Some applications are underwater monitoring, distributed radar, and indoor localization.
Communication is the bottleneckMoore's law predicts that every two years computers get twice as powerful, a trend that has been valid since the 1970s. Even more impressive is the growth in density of computer data storage: in the past decade it has been over a factor of 3 every two years. With this exponential growth in compute power and the amount of stored data, communication between compute nodes becomes essential. Here, a bottleneck arises: communication data rates have not kept pace. This is in particular visible in the area of mobile communication. Due to the rise of powerful smartphones, ubiquitous wireless internet and extensive sensor networks, communication capacity is under considerable strain, and this will become worse in future.
What can be done to increase this capacity? First of all, we can increase the available bandwidth to a service, which then also requires higher carrier frequencies. Secondly, we can use smaller transmission distances, so that more "cells" can be accomodated within the same area. These factors can provide a substantial increase in capacity, although there are some practical issues that pose limits as well: higher frequencies do not transmit well through walls nor bend around corners.
Beyond this, we can consider the use of multiple antennas (MIMO technology). Especially at higher frequencies, antennas become small and can be placed at close distances or even be integrated on-chip, which enables massive arrays of antennas. Theory predicts that with 50 antennas, we can transmit up to 50 overlapping signals simultaneously and thus obtain 50 times higher data rates. However, this is only possible if the receiver can separate those 50 messages. If it can, then this technology will ultimately lead to terabits per second data rates over short distances.
Research challenges that relate to improving data rates are thus:
- Spectrum reuse, which involves source separation techniques, massive MIMO arrays, and cognitive radio. Applications could be spectrum reuse for WIFI stations, and disaster relief networks (e.g., the TETRA system).
- 60 GHz technology, as higher frequencies offer more bandwidth and admit small form-factor antenna arrays. Ultimately, the antennas may be integrated on-chip.
- Complexity reduction. With current physical layer techniques for multiple-access, every doubling of the data rate results in at least a fourfold increase in complexity, not only numerically (computations), but also in techniques to compensate for all kinds of unwanted physical effects, e.g., multipath and time-varying propagation. The number of operations per received bit already has gone up to order 10,000 and is still growing. With current approaches, communication rates cannot keep up with the pace set by Moore's law but will increasingly lag behind.
EducationThe Center is tightly connected to the MSc Track on Telecommunication.
Electronics circuit design, with a focus on low-power RF frontends and 60 GHz technology.
THz sensing and communication
THz UWB front-ends and advanced antenna architectures for THz sensing instruments and ultra-high-speed wireless communication.
Signal processing for communication, array processing
Statistical signal processing (estimation and detection), beamforming and array processing, distributed processing and optimization, sampling and reconstruction.
Embedded software for distributed systems
Software development for wireless sensor networks, networking and Internet-of-Things.
Complex networks, with a focus on robust and energy-efficient network design.
Wireless systems for remote sensing and navigation
Microwave wireless systems (both analog and digital parts) and radar applications (including air-vehicle navigation and synthetic vision).
Radio-frequency engineering for space
- Tue, 5 Mar 2019
- lecture hall Boole, EWI building
Signal Processing Seminar
Sensor and Machine Learning at The Arizona State University
- Thu, 7 Mar 2019
- HB 17.150
PRORISC and SAFE 2019
PRORISC and SAFE 2019
- 4 -- 5 Jul 2019
- Aula Conference Centre of TUDelft