Personal tools

Difference between revisions of "Integrated Information Processing"

From iis-projects

Jump to: navigation, search
(Audio Signal Processing)
(Nonlinear DSP)
Line 87: Line 87:
 
</DynamicPageList>
 
</DynamicPageList>
  
===[[Nonlinear DSP]]===
+
===[[Nonlinear Digital Signal Processing]]===
 
<DynamicPageList>
 
<DynamicPageList>
 
suppresserrors = true
 
suppresserrors = true

Revision as of 13:08, 12 November 2020

IIP overview nov 2020-v1.pngTheory, Algorithms, and Hardware for Beyond 5GPositioning with Wireless SignalsSimultaneous Sensing and CommunicationAll-Digital In-Memory ProcessingAnalog-to-Information Conversion for Low-Power SensingNonlinear Digital Signal ProcessingReal-Time OptimizationAudio Signal Processing

Integrated Information Processing Group

The Integrated Information Processing (IIP) Group carries out research in the following areas:

Theory, Algorithms, and Hardware for Beyond 5G

The main focus of the IIP Group is on theory, algorithm design, and hardware implementation of new technologies for beyond fifth-generation (5G) wireless communication systems. The projects in this area focus on emerging communication technologies including massive MIMO, millimeter-wave (mmWave) and terahertz communication, cell-free massive MIMO, intelligent reflective surfaces, ultra low-latency short-packet transmission, and testbed design for massive MIMO prototyping.

Positioning with Wireless Signals

Indoor positioning and outdoor positioning in urban scenarios of mobile phones is a notoriously difficult task. Recently, tools from machine learning have been used to perform positioning from channel-state information (CSI). The projects in this area focus on channel charting, a new technology developed in the IIP group that enables self-supervised positioning from CSI without the users' consent.

Simultaneous Sensing and Communication

Modern wireless systems are equipped with large arrays of parallel radio-frequency (RF) chains. Such RF chains are extremely accurate sensors that can be used not only for high-rate data transmission but also for sensing. The projects in the emerging area of simultaneous sensing and communication (SISCO) are on imaging the area next to the antenna array and on classification of user behavior using machine learning techniques.

All-Digital In-Memory Processing

Processing in memory (PIM) moves computation into memories with the goal of improving throughput and energy-efficiency compared to traditional von Neumann-based architectures. The projects in this area are in designing all-digital and semi-custom PIM accelerators (application-specific integrated circuits) that can be fabricated with conventional CMOS technologies and for emerging applications in machine learning, signal processing, and wireless communication.

Analog-to-Information Conversion for Low-Power Sensing

Always-on sensors that continuously monitor the environment for certain events must operate with energy-efficient classification and detection pipelines. The projects in this area build upon a novel classification pipeline developed in the IIP group called analog-to-feature (A2F) conversion that directly acquires features in the analog domain using non-uniform wavelet sampling (NUWS). Possible applications are real-time sensing and classification of EEG, ECG, RF, and audio signals.

Nonlinear Digital Signal Processing

Nonlinearities play a critical role in a large number of signal processing applications, including the areas of wireless communication, image processing, and machine learning. Unfortunately, analyzing the fundamental properties of nonlinear systems and estimating signals from nonlinear measurements are notoriously difficult tasks. The projects in this area focus on analyzing nonlinear systems and developing new algorithms that compensate nonlinear behavior or estimate quantities from nonlinear observation models.

Real-Time Optimization

Numerical optimization finds use in a large number of fields, including wireless communications, machine learning, imaging, physics, operations research, and control. In a growing number of embedded applications, convex as well as nonconvex optimization problems must be solved in real-time and with stringent latency constraints. The projects in this area focus on the design of novel algorithms that enable real-time numerical optimization at low latency and in a hardware friendly manner.

Machine-Learning-Based Audio Signal Processing

Machine learning and deep neural networks are currently revolutionizing a variety of applications, including the well-established field of digital signal processing. The projects in this area focus on the design of novel algorithms that enable real-time audio signal processing using emerging tools from machine learning and their implementation on digital signal processors (DSPs) or hardware accelerators (FPGAs and ASICs).

Available Projects

Theory, Algorithms, and Hardware for Beyond 5G


Positioning with Wireless Signals


Simultaneous Sensing and Communication


All-Digital In-Memory Processing


Analog-to-Information Conversion for Low-Power Sensing


Nonlinear Digital Signal Processing


Real-Time Optimization


Machine-Learning-Based Audio Signal Processing