Coding Theory and Software for WBANs and Portable Diagnostic Equipment.
Supervisors: Dr. Leo Creedon and Dr. Ian McLoughlin
Funding Body: IT Sligo
The aim of this project is to develop the error-correcting codes necessary for the software for wireless body area networks (WBANs) (also known as body sensor networks (BSNs)). This will allow the detection and prediction of damage in bone in real time by attaching several wireless acoustic sensors to the human body, for example in cases of severe osteoporosis. It will also be useful in the software applications needed by an SME in designing portable biomedical diagnostics.
The coding theory needs to be particularly efficient due to the small size of the devices comprising the WBAN. These devices typically have weak signals with lots of interference. In particular, it is desirable that the codes have no short cycles. Codes (in particular Low Density Parity Check Codes (LDPC) and Convolution Codes) can be constructed algebraically using group algebras.
The project will involve the design, selection and analysis of the group algebras suitable for the construction of Low Density Parity Check Codes and Convolution Codes. The codes produced using the group algebras in the project will have no short cycles and are known to perform exceptionally well.
The codes will be developed and software written to implement the codes.
The main question to be answered is:
How can finite group algebras be used to construct well behaved Low Density Parity Check (LDPC) codes with no short cycles for application in WBANs and portable diagnostic devices (in particular for predicting and detecting damage in human bone)?