I am exploring the world of RTLS system and trying to produce one on my own. I chose 433Mhz frequency for its penetration capabilities and I produced a PCB mounting CC430 TI family MCU-transceiver, two quarter-length PCB antennae to reduce heading problems in RSSI measurements and a 250mA battery. Right now the tag is transmitting a 30 bytes packet (using the SimpliciTi communication protocol from TI) with 0dBm every minute at 1.2kbaude data rate. The receivers use pretty much the same circuitry except that they mount a half length dipole rubber duck antenna and they are wired at the power line.

The beacon sent by the tag serve for a reliable trilateration in a 30 meters range indoor with sporadic and slim (10 cm) concrete walls

My experiments show that the battery gets completely drained just after 3 days of transmitting a beacon every minute, which is probably due to the extremely low data rate used (which I suspect it has an impact on the readings' accuracy due to small fading phenomenon and multipath).

I would like to have some advice on best practices for extending battery life in an RF application such as the one I described without compromising too much the accuracy of the RSSI readings and suggestions on how and what instruments I can employ to test such drainage. The desired battery life for the tag is 6 months.


2 Answers 2


It's hard to say what exactly is going wrong in your setup, but I've design 434 MHz active RFID tags that sent a packet every 10 seconds and ran for a whole year on a CR2032 battery.

30 bytes is way more than necessary for basic identification. Our tags sent around 80 to 90 bits per packet, which was somewhat data dependent. That was enough for a 32 bit unique ID, a few bits of status information, a 20 bit CRC checksum, and the usual preamble and stuff.

Using a complete RF communication module is also inappropriate if total power is a priority. We used a bare transmitter chip and toggled it on/off directly by the firmware. This isn't that hard. In one case this was all done by a PIC 10F202, and it only used about 3/4 of the code space.


Assuming that's a 250mAh battery, that's quite small ..

Strategies I would look at:

  • what's the standby power consumption when the unit is not transmitting? It should be possible to get this down to microamps.
  • what's the transmit power consumption? Can this be reduced? (eg by turning down TX power, shorter transmissions)
  • Can the frequency of transmissions be reduced? Can they be triggered on demand?
  • Can you just get a bigger battery?
  • \$\begingroup\$ Good questions. The standby power is roughly 3.5mAh. The frequency of transmission can't be reduced over once every minute. Triggering transmissions on demand would mean that the chip should be actively listening all the time and I'd assume it is better to just wake up and send packets. \$\endgroup\$
    – Autholycos
    Jan 10, 2013 at 17:31
  • \$\begingroup\$ That's quite high for standby - that gives you 71 hours standby, before even allowing energy for transmission. \$\endgroup\$
    – pjc50
    Jan 10, 2013 at 22:20
  • \$\begingroup\$ My bad, I meant 0.035mAh \$\endgroup\$
    – Autholycos
    Jan 10, 2013 at 23:49
  • 1
    \$\begingroup\$ Er - mAh (milli amp hours) is a measure of battery capacity, not current, which should be mA - now I'm confused, because 0.035mA is very good for standby. What's the power consumption when transmitting? \$\endgroup\$
    – pjc50
    Jan 11, 2013 at 9:23
  • \$\begingroup\$ indeed the current is 0.035mA. The power when transmitting is incredibly higher (in the ballpark of 84mA but it the value fluctuates quite a bit). I suspect that I have screwed up with the circuit layout somehow. Still the request for best practices stands \$\endgroup\$
    – Autholycos
    Jan 13, 2013 at 12:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.