The first part of my week was largely focused around the finishing of our slides for the design review. This consisted of a lot of work with the other members choosing/finalizing what components we would be using for this project as well as talking to Nate about what we thought the specs should be for the software side of the project (as well as how we would test/validate these subsystems).

 

After this was finished however I got back onto working on the Python backend. This has largely consisted of thinking about how to handle the 2 process design present now and refine it into either a single process (which I foresee issues with) or researching how shared memory (that has to be able to be read-write accessible for both processes) is best handled in python. Currently, instead of launching the processes from a shell script like I had planned, I’m looking to leverage Python’s multiprocess module to dispatch the 2 processes from a master file launched by a shell script on start up. These 2 processes should share a global Manager() list object that should act as the shared memory between the 2. If this can be made to work then I’ll have the majority of this subsystem complete as the thread dispatch is in place as well as the reconstruction queue.  These require testing of course but I feel that these will be easier parts to test and debug than the multiprocessing parts of the system so I’m optimistic there. The data is currently being stored in a list “post-processing” (or where it would be post processing) so this will have to be changed as we figure out how we need to format this data in order to compress it for saving to the SD card. Also critical points in the program, such as where input from the Nano’s GPIO will be checked for recording/not recording, are given placeholder comments and alternative snippets of code to provide similar functionalities in the short term. This should allow for initial testing of this system to be done hopefully within the next week once all changes and redesigns have been satisfactorily implemented.


0 Comments

Leave a Reply

Your email address will not be published. Required fields are marked *