CC2510 RSSI Calculation

Chris,

The difficulty may be that you're not waiting long enough for the RSSI to settle after going into RX mode.  Please see DN505 (http://focus.ti.com/general/docs/lit/getliterature.tsp?baseLiteratureNumber=swra114) and the RSSI example code (rnd_rssi.c) in the CC2510 code examples (http://www.ti.com/litv/zip/swrc117), which has this snippet:

/* Radio is now in IDLE.
       Strobe command SRX takes radio to RX. */
    RFST = RFST_SRX;

    /* Wait until RX is entered. */
    while ((MARCSTATE & MARCSTATE_MARC_STATE) != MARC_STATE_RX);

    /* Wait a little longer than expected RSSI response time
     * (see design note DN505). Note that the below loop is inaccurate,
     * so the user must implement a proper delay according to DN505.
     */
    for(rssiValidCnt = 0; rssiValidCnt < RSSI_VALID_TIME; rssiValidCnt++);

Regards, GrantC

Hi Grant,

Thanks for the help. This definately solves the slope problem. However, now we are reading an RSSI value a few dB higher than expected. We are using a signal generator to insert  signals from -20dBm to -103dBm (compensating 1dB of cable loss verified with spectrum analyzer). The RSSI readback values closely follow this slope at -24dBm to -105dBm. We are using a datarate of 9600 baud and an rssi_offset value of 72, which was the recommended value from TI support for all data rates (9.6k to 230.4k). Does the rssi_offset need to be changed? Is there a formula for finding this value?

Thanks again,

Chris

Hello Support,

Any ideas for the RSSI offset issued mentioned above?

Thanks,

Chris

Chris,

Table 68 of the CC2510 data sheet recommends an RSSI offset of 74 dB at 10 kbaud.  I guess that's 2 dB -- are you off more than that?

-GrantC

Hi Grant, thanks for the response.

With an offset of 72, we are -4dBm across the band (i.e. expect -20dBm, get -24dBm). Changing the offset to74 would make it -6dBm lower across the band.

~Chris

Chris,

I guess the next step is to separate hardware and software.  Have you tried the RSSI measurement using SmartRF Studio for comparison?

Hi Grant,

Here's the results of plugging our register settings into SmartRF Sudio and using the TI evaluation board with RSSI offset set to 74.

Sig.Gen.          RSSI

-20dBm           -28dBm
-30dBm           -37dBm
-40dBm           -48dBm
-50dBm           -58dBm
-60dBm           -68dBm
-70dBm           -78dBm
-80dBm           -86dBm
-90dBm           -96dBm
-100dBm        -105dBm

The radio settings are:

  FSCTRL1   =     0x08;       // Frequency Synthesizer Control
  FSCTRL0   =     0x00;       // Frequency Synthesizer Control
  FREQ2     =     0x5C;       // Frequency Control Word, High Byte
  FREQ1     =     0x5C;       // Frequency Control Word, Middle Byte
  FREQ0     =     0xEC;       // Frequency Control Word, Low Byte
  MDMCFG4   =     0x78;       // Modem configuration
  MDMCFG3   =     0x83;       // Modem Configuration
  MDMCFG2   =     0x13;       // Modem Configuration
  MDMCFG1   =     0x23;       // Modem Configuration
  MDMCFG0   =     0x8A;       // Modem Configuration
  CHANNR    =     0x4A;       // Channel Number
  DEVIATN   =     0x44;       // Modem Deviation Setting
  FREND1    =     0x56;       // Front End RX Configuration
  FREND0    =     0x10;       // Front End TX Configuration
  MCSM0     =     0x14;       // Main Radio Control State Machine Configuration
  FOCCFG    =     0x16;       // Frequency Offset Compensation Configuration
  BSCFG     =     0x6C;       // Bit Synchronization Configuration
  AGCCTRL2  =     0x43;       // AGC Control 
  AGCCTRL1  =     0x40;       // AGC Control 
  AGCCTRL0  =     0x91;       // AGC Control 
  FSCAL3    =     0xA9;       // Frequency Synthesizer Calibration
  FSCAL2    =     0x0A;       // Frequency Synthesizer Calibration
  FSCAL1    =     0x00;       // Frequency Synthesizer Calibration
  FSCAL0    =     0x11;       // Frequency Synthesizer Calibration
  TEST2     =     0x88;       // Various Test Settings
  TEST1     =     0x31;       // Various Test Settings
  TEST0     =     0x09;       // Various Test Settings
  PA_TABLE0 =     0xFE;       // PA Power Setting
  PKTCTRL1  =     0x04;       // Packet Automation Control
  PKTCTRL0  =     0x05;       // Packet Automation Control

Thanks,

Chris 

Chris,

The RSSI offset value can vary significantly from device to device (and over temperature and power supply), which probably explains your results.  The RSSI measurement is meant to be an indicator of received signal strength, but the variation makes it not the best absolute RSSI indicator.

-GrantC

OK, so you're saying this variation is expected. I just want to make sure I'm setting up the radio correctly.

Thanks,

Chris