An automatic smart shopping cart deployment framework based on pattern design

An automatic embedded software generation framework that can create and evolve Zigbee applications is proposed. The framework consists of two major modules, pattern extraction and code generation. Pattern extraction and development are designed to provide Zigbee application with model reuse and modification. SysML serves as a medium between pattern development and code generation. State diagrams, class diagrams and sequence diagram help describes a specific application scenario. A smart shopping cart application has been implemented using this pattern based software framework. The current location of a shopping cart has been monitored constantly by the WSN network through a signature sensing technique implemented by our lab. Both the electronic map on the server console and the PXA255 on board embedded system find a shortest path from the current location to the target product location assigned by user on the LCD screen. This evidence proves the success of the framework process. This software framework allows user to store the whole smart shopping application scenario in the design pattern repository. Further extension of this pattern can be exercised for creating other Zigbee applications such as patient localization in a hospital environment, item retrieval in large scale storage ware house, and green power energy saving applications etc.