Communications & Signal Processing Research

CSP Research Overview

The field of Signal Processing is concerned with the design and implementation of algorithms for manipulating, forecasting or classifying signals containing information. Examples include classification of objects (e.g., faces) in an image, speech recognition and synthesis, video compression, medical diagnosis from biological signals such as the heartbeat in wearable devices, or learning a detailed map of the area surrounding an autonomous vehicle to allow safe navigation in unknown environments.  Signal processing techniques are ubiquitous in engineering systems, including but not limited to communications devices (cell phones, WiFi), digital music/video players, televisions, GPS receivers, radar and sonar systems, the Internet of Things (IoT), and medical imaging and monitoring systems.

Communications is as essential to modern society as electricity, providing technologies for both high speed connectivity (cellular, WiFi, Internet) and efficient storage (hard drive, flash memory). The field of communications spans signal processing and error control coding for transceiver design, information theory to identify fundamental performance benchmarks, and the design of communication network protocols.

The communications and signal processing faculty at UCSB engage in cutting-edge research in communication theory and networking (with current focus on next generation wireless communication and sensor networks), novel machine learning and optimization techniques (spanning a range of fundamental questions and practical applications), image informatics and reconstruction (extracting and organizing information from images for biological research, diagnosis, surveillance, production monitoring). The emergence of a well-developed software infrastructure for machine learning and large-scale optimization, along with the availability of large amounts of data in a variety of fields, is feeding a number of exciting research initiatives.  CSP faculty are leveraging these advances to attack problems that are difficult to solve with classical techniques, as well as bringing a CSP perspective to explore the fundamentals of robust, interpretable machine learning. In addition, research continues to be directed at advancing core knowledge in CSP in areas such as information theory, estimation theory, compression, and harmonic analysis. Much of the research in the CSP group involves interdisciplinary scientific and industrial collaborations, including technology transfer via startup companies. Close connections with industry are maintained through faculty interactions and student internships, leading to an understanding of emerging trends and bottlenecks in technology.

For more detailed descriptions of faculty research and activities, please visit the CSP Research Activities accordions in the box.

CSP UCSB Affiliations