ELECINF344/381

Partie interactive du site pédagogique ELECINF344/ELECINF381 de Télécom ParisTech (occurrence 2011).

Catégories

MB Led: Synchronization, algorithm reliability and improving IrDA transmission

Since yesterday, Guillaume worked on algorithm reliability, improving its performance and solving some problems such as information sharing (number of blocks in network, position…). Some loop-back which were over-charging the network have been deleted. He worked with Benjamin to improve the algorithm but some problems remain unsolved… In late afternoon, the algorithm was working rather well : in a 9 blocks network, blocks detected the good number of blocks, their position, the leader ID, the positions of the others in many cases. But sometimes, one block loses his IrDA connection so he has to start again the algorithm. When dividing a network in 2 networks, the blocks graph (developped with Benjamin) gives the blocks good informations.

Benjamin worked on improving the snake game, now it has a tail and if you cut it (meaning you disconnect blocks before the tail leaves) you lose. Some apples have shown their faces on the screen, but we must handle turn detection before evolving furthermore. He also worked with Guillaume on the implementation of a blocks graph in order to react faster when a block leaves the network. In fact, if a member leaves, blocks are able to calculate that every members of the branch also left. At last he tested synchronisation implemented by Guillaume displaying an animation on MB Leds, it is pretty quick and all blocks flash at once.

I have first worked on improving some aspect of IrDA like variable waiting time before a resend and size of packets in order to improve the  payload . I have continued  to implement firmware transmission and have begun testing it. Transmission seems a bit slow and tomorrow I will look for a way to improve its speed.

MB Led : Visual display of algorithm and snake first steps

Since our last post, Guillaume is still on the algorithm. Benjamin developed a display task for the old and new GLiPs. Now, we can see directly what’s going on in the network and try to debug it (see the picture). Yesterday, Guillaume tried to resolve bugs of freezing GLiP while the algorithm is running. He had to change many things in order to fix this and the only remaining freeze happens when some blocs leave the network.

There are some troubles when a block isn’t gone but don’t communicate with his neighbors. The biggest problem is when a block is rickety and that he communicates for time to time : he is considered as gone and soon after, he is here, etc. When this happens, the network can fail.

 

 

A little video to illustrate the visual debug:

 

Benjamin is develloping a special snake game concept for our system.

 

Guillaume and Benjamin.