Hello,
I have some questions regarding PowerTOSSIM (I am using the version that
comes bundled with tinyos 1.1.15).
I set up a very simple experiment using two routing protocol
implementations that I created so I could measure the amount of energy
required by both protocols. The experiment is run on a specified
network topology containing 36 nodes (in a grid). After running both
simulations for an equal period of time, I ran the PowerTOSSIM script on
the output using the command:
/opt/tinyos-1.x/tools/scripts/PowerTOSSIM/postprocess.py --sb=1 --em
/opt/tinyos-1.x/tools/scripts/PowerTOSSIM/mica2_energy_model.txt myapp.trace
The average values I get the for the energy consumed by the radios on
the nodes are:
Protocol A: ~3600 mJ
Protocol B: ~3700 mJ
The strange thing about this is that, during the simulation, Protocol A
sends twice as many messages over the radio as Protocol B. If that is
the case, why does Protocol A have a lower radio energy consumption???
Both protocols use the same comm component (GenericCommPromiscuous),
Does the size of the data structure used to transmit the message have
anything to do with radio energy consumption? I have included the
message data structures used by Protocol's A and B below just in case.
I would really appreciate it if someone has any insight into this matter.
Thanks,
Brett P
Protocol A Message Data Structure
--------------------------------------
typedef struct ProtocolAMsg {
uint8_t type;
uint16_t interest_id;
uint16_t reading;
uint16_t seqNum;
uint16_t destaddr;
uint16_t source;
uint16_t neighbour;
} __attribute__ ((packed)) ProtocolAMsg;
Protocol B Message Data Structure
--------------------------------------
typedef struct ProtocolBMsg {
uint8_t type;
uint8_t interest_id;
uint16_t reading;
uint8_t destination;
uint8_t source;
int8_t routeRecord[15];
} __attribute__ ((packed)) ProtocolBMsg;
_______________________________________________
Tinyos-help mailing list
[email protected]
https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-help
