FPGA-based Vector Processors for Compute Acceleration

Modern systems are increasingly being made “digital” and interconnected to the rest of world through computers. Many types of sensors, such as image sensors or digital radios, can produce vast amounts of data in a short amount of time. Rather than sending all of this raw data back to a central server, it is usually more efficient to deploy an embedded system close to the sensor that can inspect, analyze, compress, and otherwise reduce the data into a more compact, meaningful form. This allows the device to not only be smart and aware of its function, but also to save on energy and communications costs. This research project involves the design of a new type of computer processor that is optimized to perform this type of data processing in the embedded environment. The objectives are to provide higher performance, expanded capability, and easier methods of programming the processor.

Faculty Supervisor:

Dr. Guy Lemieux

Student:

Aaron Severance

Partner:

VectorBlox Computing Inc.

Discipline:

Engineering - computer / electrical

Sector:

Information and communications technologies

University:

University of British Columbia