Signal processing algorithms

Signal Processing for communications and array processing

This program has two related tracks:

• Signal processing for communications
  
• Array signal processing
  

Program chairmen: Alle-Jan van der Veen, Geert Leus
Researchers: Alon Amar, Shahzad Gishkori, Hadi Jamali Rad, Sina Maleki, Raj Thilak Rajan, Yiyin Wang, Tao Wang, Tao Xu, Siamak Yousefi, Mu Zhou, Dyonisius Ariananda.
Visitors: prof. Ubli Mitra, prof. Amir Leshem.

Mission

Signal processing for communications is an important area within signal prodcessing, as it covers around 20% of research. Among many questions, an important one is, how can multiple overlapping signals be separated? This problem is very relevant in wireless communications, where signals are overlapping in time, space and frequency (the 'cocktail party problem'). With multiple antennas and advanced separation techniques, the capacity and robustness of wireless links can greatly be improved.

Similar problems are relevant in radio astronomy, where the sky signals of interest are contaminated by man-made interference (e.g., communication signals), which need to be estimated and suppressed. The resulting algorithms are formulated in terms of linear algebra operations and provide interesting test cases for embedded system design.

Array signal processing refers both to parallel signal processing of an array of signals as it occurs in a multiple antennas/receivers situation, and to the mapping of such algorithms onto parallel hardware platforms. This topic has great relevance in a number of contexts: the mobile communication context, separation of airplane transponder signals using phased antenna arrays, and spatial filtering for radio telescope arrays.

Topics studied can be classified under "techniques" and "applications".

Techniques

• Estimation and detection theory; statistical signal processing
• Array calibration and beamforming algorithms; source localization
• Performance analysis and bounds
• Distributed processing; optimization techniques
• Sampling and reconstruction theory
• Adaptive signal processing
• Space-time coding; modulation; acquisition, synchronization, tracking

Applications

• CDMA and spread spectrum; multicarrier and OFDM; UWB
• Sensor networks
• MIMO communication; wireless networks
• Radar and sonar signal processing; radio astronomy; geophysics; remote sensing

Research

In our work, we combine "techniques" with "applications". We work on a variety of applications, such as sensor networks for the process industry, a distributed radio telescope in space, an underwater communication system. Typically, there is a receiver (antenna array) which receives a collection of transmitted signals, perturbed by a multipath channel that may even be highly time-varying. The challenge then is to derive algorithms that estimate the channel and detect the transmitted data.

Our work is theoretical: the development of new algorithms and the derivation of their performance, as well as practical: the development of experimental phased array measurement systems and the verification of the algorithms on the obtained data.

The focus of our work is as follows:

• Smart antenna technology for wireless communications: This includes research on algorithms for source separation, equalization and parameter estimation of communication signals, and application of blind source separation/equalization techniques to W-CDMA and OFDM.

• Signal processing over time-varying channels: If the transmitter and/or receiver is moving fast, a large Doppler spread makes the communication channel time-varying. This occurs in DVB-T systems (e.g., digital television received in high-speed trains), but is even more pronounced in acoustic underwater communication channels. The challenge is to estimate and compensate for these effects, especially for wideband (OFDM) signals.

• Signal processing for sensor networks: Distributed sensor systems consist of a large number of nodes with only local communication capabilities. Challenges include localization of the nodes, low-power communication protocols, and distributed estimation algorithms, where local estimates are combined to form global parameters estimates.

• Signal processing for radio astronomy: The trend in radio astronomy is to construct large arrays of small antennas. An example is the LOFAR system, where 13,000 antennas are distributed over 100 stations in The Netherlands and Germany. In the future, we will have SKA (square kilometer array, consisting of 1 million antennas) and OLFAR, a distributed radio telescope in space. Central issues for us are array calibration, interference cancellation, and image formation using array processing techniques.

Currently Running Projects

NOPTILUS: Autonomous, self-learning, optimal and complete underwater systems [2011-2014]

Current multi-AUV systems are far from being capable of fully autonomously taking over real-life complex situation-awareness operations. For this, significant advances are required, involving cooperative and cognitive-based communications and sonars (low level), Gaussian Process-based estimation as well as perceptual sensory-motor and learning motion control (medium level), and learning/cognitive-based situation understanding and motion strategies (high level). In the design of the NOPTILUS underwater system, robustness, dependability, adaptability and flexibility will be emphasised, especially when it deals with completely unknown underwater environments and situations "never taught before".

The synthesis of signal processing and radio astronomical calibration and imaging techniques [2011-2014]

The world radio astronomical community is planning some major new radio facilities. Foremost among these is LOFAR, constructed largely within the Netherlands. It is considered THE pathfinder for the SKA, the Square Kilometre Array, a project supported and led by the world radio community.

New, larger and more complex radio telescopes bring new challenges. Foremost among these is the calibration of the data in order to remove atmospheric and instrumental effects which corrupt the exceedingly faint signals from cosmic sources. Indeed, the scientific success of the new generation of radio telescopes will depend critically on the ability to calibrate the data, and to deliver 'thermal-noise-limited' performance. This project will study signal processing challenges in calibration, imaging and RFI mitigation, all tightly related.

VICI: Signal processing for self-organizing wireless networks [2009-2013]

The goal of this project is to shift from centralized communication networks to distributed self-organizing networks where nodes adapt their procedures (related to spectrum utilization, sensing, information processing, and localization) based on only local information. To develop large self-organizing networks we need cognitive radio devices that are capable of sensing the radio spectrum and adapt accordingly. We further require energy efficient distributed information processing and localization algorithms for large sensor networks. The mathematical tools we want to build on are compressed sampling, convex optimization, game theory, and linear algebra.

FASTCOM: Fast wireless network for sensor communication within a lithography machine [2010-2013]

Lithography machines have a fast-moving waferstage; also the mask is moving. These stages have many sensors used for accurate positioning. It is desired to connect these sensors wirelessly to a central controller. However, the aggregate data rate is very high (over 1 Gbps), and the latency requirements are very tight. Currently, there are no wireless standards that can accomodate this.

D2S2: Dependable distributed sensor systems [2010-2013]

The D2S2 project aims at developing a framework for programming and operating distributed sensor systems that can be depended on in practical application scenarios. To make an experimental approach feasible, the project focuses on localization and tracking systems in two scenarios that are very relevant to the Dutch society: traffic monitoring and control (static setup) and rescue operations by firefighters and policemen (dynamic setup). A key, innovative feature of the project is the development and use of an advanced miniaturized radar sensor that can operate under a wide range of "difficult" environmental conditions (smoke, fog, etc.) that cannot be handled by typical localization systems in operation today.

RACUN: Robust acoustic communications in underwater networks [2010-2011]

This project will develop and demonstrate the capability to establish an underwater ad hoc robust acoustic network for multiple purposes with moving and stationary nodes. Applications are e.g. surveillance and mine reconnaissance using autonomous underwater vehicles (AUVs). A network of underwater nodes should rapidly be deployed in littoral waters. In this project, we will develop modulation and receiver techniques for communication over highly time-varying channels.

OLFAR: distributed radio telescope in space [2010-2013]

One of the last unexplored frequency ranges in radio astronomy is the frequency band below 30 MHz. Because these frequencies are blocked by the ionosphere, earth-bound observations would be be severely limited or impossible. A radio telescope in space would not be hampered by the earth's ionosphere. In this project, we aim to design a distributed radio telescope in space, consisting of a swarm of about 50 nano-satellites flying in a cloud of 10 to 100 km.

SmartPEAS: Smart moving Process Environment Actuators and Sensors [2007-2011]

The quality of chemical reactions inside a tank can be improved if we can measure locally parameters such as temperature and flow. To do this, the idea is to throw small (tennis-ball size) units called PEAS into the tank, that make the observations and communicate their findings using UWB (either RF or acoustic) communication. Also the locations of the PEAS need to be measured.

Recently Ended Projects

MIMO for a mass-market [2006-2010]

This SenterNovem project studied how MIMO technology can be employed in WLAN equipment, now that the 802.11n standard makes WLAN with multiple antennas ubiquitous. We took an interdisciplinary approach: the multiple RF chains and ADC chains can become costly and power-hungry. Using feedback from the digital receiver part, can we design an analog preprocessing unit to compress the number of antenna signals into fewer receiver chains? The aim is to obtain better performance than simple antenna selection.

• More information
• Publication list

VICI: Signal processing for communications [2003-2009]

This NWO-sponsored VICI project studied multi-user, multi-antenna receiver algorithms for future generation communication systems, in particular multi-user CDMA (used in UMTS), OFDM (used in WLAN) and UWB radio. Specific topics are multi-antenna transmission and space-time coding, and advanced modulation techniques that facilitate multi-user detection. Applications to radio astronomy (the LOFAR telescope) were also studied.

• More information
• LOFAR signal processing
• Publication list

VIDI: Communication over time-varying channels [2005-2009]

The NWO-sponsored VIDI project considers the problem that Doppler shifts due to mobility and carrier frequency offsets introduce channel time-variations. As a result, some wireless systems can only provide low data rates at high mobility, e.g., the third generation wireless system UMTS, or even break down completely at high speeds, e.g., digital video broadcasting (DVB-T) applications. To solve these problems, the project proposes an innovative wireless system design that takes the time-varying nature of the channel explicitly into account.

• More information
• Publication list

AIR-LINK: UWB impulse-radio communication system [2002-2008]

The objective of AIR-LINK was to design a communication system using Ultra-Wideband (UWB) technology, which is using narrow radar-like pulses to transmit data. The advantage of this is that no carrier modulation/demodulation is needed, which enables cheaper single-chip solutions. Data rates are promised to be very high. Another feature is that a large part of the frequency spectrum below 10 GHz is used, on top of existing allocations. This is possible because the transmitted energy is very small. Key issues in the design are synchronization algorithms, source separation and multi-access interference mitigation. Our interest is specifically in signal processing for the physical layer of ad hoc networks that can be constructed using UWB devices.

• More information
• Publication list

UBROAD: Ultra wideband communication and ad hoc networks [2002-2004]

The U-BROAD project was a 6-th framework EU "STREP" project, on Ultra High bit rate over copper technologies for broadband multiservice access (VDSL). The main objective was to develop and integrate advanced access technologies of true broadband content over Ethernet based networks to the customer premises. It aimed at quadrupling the total bandwidth available to the end user. Project partners were Metalink, France Telecom, OTE, Bar-Ilan University (Amir Leshem), and the University of Crete (Nicholas Sidiropoulos).

• Publication list

Separation of airplane transponder signals [1999-2003]

The goal in this project was to separate received airplane transponder signals (secondary surveillance radar or SSR). Especially in Europe, the crowded airspace can lead to the reception of multiple return frames from different airplanes, partially overlapping in time and frequency. By employing a phased antenna array, these frames can be separated and the information detected. In addition, the direction of the airplanes can be estimated.
Apart from the derivation of new signal processing algorithms for this application, we have also developed a 4-channel recording system to obtain actual measurements using a small phased array mounted on the roof of the EWI department building.

NOEMI: Nulling obstructing interferers in radio astronomy [1998-2003]

Radio-astronomical observations are increasingly corrupted by RF interference, and online detection and filtering algorithms are becoming essential. Examples of interferers are GSM mobile phones, GPS satellites, and TV signals. In the STW project NOEMI (with ASTRON), we considered interference mitigation techniques for the Westerbork Synthesis Radio Telescope array. The approach is to formulate the astronomical problem in an array signal processing language, so that elementary algorithms from that field can be applied. Two topics are considered in detail: calibration of polarized telescope arrays using new signal processing techniques, and spatial filtering by subspace estimation and projection. This can be used to filter out continuously present interferers such as TV stations. Spatial filtering works better if a good estimate of the interfering signal is available. Although the telescopes themselves can be used for this, it is attractive to use a separate antenna for this. However, a simple omni-directional antenna will not have a sufficiently good signal to noise power ratio. To improve on this, we have started to use a 64-element adaptive phased array developed by ASTRON as reference antenna. The output of the array is a weighted sum of the elements. By adapting the weights (beamforming), the array can be pointed at any direction in the sky and give a much better reference signal.

• More information
• Publications

|  30 Apr 2010   |