|
From: Ast D. (astdom01) <ast...@st...> - 2014-09-22 20:39:59
|
Dear all I have a question about routing with rpl. I simulate a very simpel case with cooja where every sender sends one message to a hard coded receiver every 30 min. As long as I do not move a mote in the simulation I have a success rate of about 90% with the selected medium. When I move motes (2 motes of 50) this rate decreases to less then 75%. I dont delete motes and after the move all motes still have a well connectivity to other nodes - but to others then before. I expected the success rate to increase back to 90% again after a while but after 5 to 6 simulation hours it is still less then 80%. I use the RPL configuration below. I tried different values of RPL_CONF_DEFAULT_LIFETIME_UNIT, RPL_CONF_DEFAULT_LIFETIME and RPL_CONF_DIO_INTERVAL_DOUBLINGS to achieve this behaviour but without success. Can I achieve healing routes by changeing RPL parameters or is there somethings else to consider? Thank you for any hint. Best regards, Dominic #undef NETSTACK_CONF_RDC_CHANNEL_CHECK_RATE #define NETSTACK_CONF_RDC_CHANNEL_CHECK_RATE 8 #undef RPL_CONF_DIO_INTERVAL_MIN #define RPL_CONF_DIO_INTERVAL_MIN 12 #undef RPL_CONF_DIO_INTERVAL_DOUBLINGS #define RPL_CONF_DIO_INTERVAL_DOUBLINGS 9 #undef RPL_CONF_DAO_LATENCY #define RPL_CONF_DAO_LATENCY (CLOCK_SECOND * 4) #undef RPL_CONF_DIO_REDUNDANCY #define RPL_CONF_DIO_REDUNDANCY 10 #undef RPL_CONF_DEFAULT_LIFETIME_UNIT #define RPL_CONF_DEFAULT_LIFETIME_UNIT 0xffff #undef RPL_CONF_DEFAULT_LIFETIME #define RPL_CONF_DEFAULT_LIFETIME 0xff #undef UIP_CONF_ND6_SEND_NA #define UIP_CONF_ND6_SEND_NA 0 |