University of Cape Town

Amit Kumar Mishra

Alan Wilson-Langman

Michael Raymond Inggs

Systems and methods for object identification are provided. In a method, primary data is received. The primary data is generated by a primary sensor that receives signals that are human perceptible and records a scene. Secondary data generated by a secondary sensor that simultaneously records the same scene is received. The secondary sensor receives signals that are not human perceptible. The primary data is processed to identify object signatures relating to objects present in the scene. The processed primary data is used to train a secondary data-based object identification model configured to identify, in the secondary data, object signatures relating to objects present in the scene. A method includes using the secondary data-based object identification model to process the secondary data to identify object signatures relating to objects present in the scene. A method includes augmenting the processed primary data with the processed secondary data.

Object identification in data relating to signals that are not human perceptible

A symbiotic radar and communication system is disclosed. The system includes a plurality of base stations in communication with a wider communication network, each base station being configured to transmit and receive communication signals to and from a plurality of user terminals. The system performs communication data processing and radar data processing on received signals so that target object reflections contained therein can be resolved. Preferably, initial radar data processing is performed at each base station which embeds time and position information into the data and communicates it to a central radar server, the central server combining the received data from several of the base stations into a single range resolution profile.

Growing community of inventors

Cape Town, South Africa

Amit Kumar Mishra