Frequency Error Measurements Basic Rate

Frequency error measurements are performed on Basic Rate packets as specified in test cases RF/TRM/CA/BV-08-C and RF/TRM/CA/BV-09-C in section 4.5 of the Bluetooth Test Specification RF.TS.p33.

  1. For each averaging count j (one packet), the following operations are performed.
    1. Frequency Error is computed on Preamble. This is recorded as Initial Frequency Error, denoted by ƒ0 j

    2. Frequency error is computed on each 10-bit group in the payload starting from the 2nd bit. These frequency errors are denoted by ƒ kj with k = 1, 2, …, n is the index of the 10-bit group. The maximum difference between ƒ kj and ƒ 0j , for all k , is recorded as Peak Frequency Drift.

    3. The maximum difference between ƒ k+5j and ƒ kj , for all k , is recorded as Peak Frequency Drift Rate

  2. The value of the Initial Frequency Error that results into the maximum absolute value across all averaging counts is returned as the Initial Frequency Error Maximum.

  3. The value of the Peak Frequency Drift that results into the maximum absolute value across all averaging counts is returned as the Peak Frequency Drift Maximum.

  4. The value of the Peak Frequency Drift Rate that results into the maximum absolute value across all averaging counts is returned as the Peak Frequency Drift Rate Maximum.

Frequency Error Measurements Enhanced Data Rate

Frequency error measurements are performed on Enhanced Data Rate Packets as specified in the test case RF/TRM/CA/BV-11-C in section 4.5 of the Bluetooth Test Specification RF.TS.p33 .

  1. For each averaging count j (one packet), the following operations are performed.
    1. Frequency Error is computed on the packet header. It is recorded as Header Frequency Error and denoted as ωi j .

    2. The Enhanced Data Rate portion of the packet is compensated for ωi j . The frequency error is then computed on every 50 µs block of the EDR portion (sync, payload Header, EDR payload, trailer and CRC) of the EDR packet. This frequency error is denoted by ω0 k,j , where k = 1,2,…, n is the index of the 50 µs block.
    3. The sum of frequency error ωi j and ω0 k,j , for all k , is recorded as Frequency Error ωi j + ω0 k,j .

    4. The maximum value of Frequency ω0 k,j , for all k , is recorded as Peak Frequency Error ω0 k,j .

    5. The maximum value of Frequency Error ωi j + ω0 k,j , for all k , is recorded as Peak Frequency Error ωi j + ω0 k,j .

  2. The value of the Header Frequency Error ωi j that results into the maximum absolute value across all averaging counts is returned as the Header Frequency Error ωi Maximum.

  3. The value of the Peak Frequency Error ω0 j that results into the maximum absolute value across all averaging counts is returned as the Peak Frequency Error ω0 Maximum.

  4. The value of the Peak Frequency Error ωi j + ω0 j that results into the maximum absolute value across all averaging counts is returned as the Peak Frequency Error ωi + ω0 Maximum.

Low Energy (LE)

Low Energy Uncoded without Constant Tone Extension (CTE)

Frequency error measurement are performed on LE packets of symbol rate 1Ms/s and 2Ms/s as specified in section 4.6.4 of the Bluetooth Test Specification RFPHY.TS.p23 .

  1. For each averaging count j (one packet), the following operations are performed:
    1. Frequency Error is computed on Preamble. The result is recorded as Initial Frequency Error, denoted by ƒ0 j .

    2. Frequency error is calculated from the second payload bit. LE packets use 10-bit groups at 1 Ms/s and 20-bit groups at 2 Ms/s. These frequency errors are denoted by ƒ nj , where n=1, 2, …, k is the payload group index of the LE packet.
    3. The maximum value of ƒ nj , for all n , is recorded as Peak Frequency Error.

    4. The difference between ƒ 1j and ƒ 0j is recorded as the Initial Frequency Drift.
    5. The maximum difference between ƒ nj and ƒ 0j , for n=2, 3, … k , is recorded as Peak Frequency Drift.

    6. The maximum difference between ƒ nj and ƒ n-5j , for n=6, 7,…k, is recorded as Peak Frequency Drift Rate.

  2. The value of the Peak Frequency Error that contains the maximum absolute value across all averaging counts is returned as the Peak Frequency Error Maximum.

  3. The value of the Initial Frequency Drift that contains the maximum absolute value across all averaging counts is returned as the Initial Frequency Drift Maximum.

  4. The value of the Peak Frequency Drift that contains the maximum absolute value across all averaging counts is returned as the Peak Frequency Drift Maximum.

  5. The value of the Peak Frequency Drift Rate that contains the maximum absolute value across all averaging count is returned as the Peak Frequency Drift Rate Maximum.

Low Energy Uncoded With Constant Tone Extension (CTE)

Frequency Error measurement on constant tone extension (CTE) field is performed on LE uncoded packets of symbol rate 1 Ms/s and 2 Ms/s as specified in section 4.6.5 of the Bluetooth Test Specification RFPHY.TS.p23.

  1. For each averaging count j (one packet), the following operations are performed.
    1. Frequency Error is computed on the Preamble. The result is recorded as Initial Frequency Error, denoted by ƒ0 j

    2. For LE packet of symbol rate 1 Ms/s, Frequency error is computed on 16 bits of the payload. This computation starts from the (n+1)th bit of payload where n=((Max_Tx_Length*8)-20). For LE packet of symbol rate 2 Ms/s, Frequency Error is computed on 32 bits of the payload. This computation starts from (n+1)th bit of payload, where n=((Max_Tx_Length*8)-36). The first n bits and last 4 bits are discarded. The remaining bits are recorded as average center frequency, denoted by ƒ pj . This recorded value is used in the computation of the Initial Frequency Drift. Initial Frequency Drift is the RFmx frequency error result property.
    3. The average frequency deviation is computed on each 16 us unit of CTE, starting from 1st bit of reference period of 16 µs CTE. The average frequency deviation is denoted by ƒ3 maxij , where i=1, 2, …, k is the index of the 16 µs unit. These frequency errors are denoted by ƒsi j .

      For computation of dƒ1 avg , refer to the df1 and df2 concept topic.

    4. The maximum difference between ƒ3 maxij and dƒ1 avg j , for all i , is recorded as Peak Frequency Error.

    5. The difference between ƒs1 j and ƒ pj is recorded as Initial Frequency Drift.

    6. The maximum difference between ƒsi j and ƒ0 j for all i , is recorded as Peak Frequency Drift.

    7. The maximum difference between ƒs1 j and ƒs(i-3) j , for i=4,5,..k, is recorded as Peak Frequency Drift Rate.

  2. The value of the Peak Frequency Error that contains the maximum absolute value across all averaging counts is returned as the Peak Frequency Error Maximum.

  3. The value of the Initial Frequency Drift that contains the maximum absolute value across all averaging counts is returned as the Initial Frequency Drift Maximum.

  4. The value of the Peak Frequency Drift that contains the maximum absolute value across all averaging counts is returned as the Peak Frequency Drift Maximum.

  5. The value of the Peak Frequency Drift Rate that contains the maximum absolute value across all averaging count is returned as the Peak Frequency Drift Rate Maximum.

Low Energy Coded (S=8)

Frequency error measurement are performed on Low Energy Coded (S=8) packets of symbol rate 125Kbps as specified in section 4.6.4 of the Bluetooth Test Specification RFPHY.TS.p23.

  1. For each averaging count j (one packet), the following operations are performed.
    1. Frequency Error is computed on each group of 16 symbols in the preamble field starting at the third symbol. The last 14 symbols of the preamble are discarded. These frequency errors are recorded as ƒ0 j , ƒ1 j , ƒ2 j , and ƒ3 j , with ƒ0 j being the Initial Frequency Error.

    2. The computation skips the first 26 symbols of the payload, then starts computing frequency error on each remaining 16-symbol group of the payload. This computation starts starting at the 27th symbol in the payload. These frequency errors are denoted by ƒ nj , where n=4,5,…,k is the index of the 16-symbol group.
    3. The maximum value of ƒ nj , for all n , is recorded as Peak Frequency Error.

    4. The maximum difference between ƒ nj and ƒ 0j , for all n , is recorded as Peak Frequency Drift.

    5. The maximum difference between ƒ nj and ƒ n-3j for n=3 & 7,8,…,k, is recorded as the Peak Frequency Drift Rate.

  2. The value of the Peak Frequency Error that contains the maximum absolute value across all averaging counts is returned as the Peak Frequency Error Maximum.

  3. The value of the Peak Frequency Drift that contains the maximum absolute value across all averaging counts is returned as the Peak Frequency Drift Maximum.

  4. The value of the Peak Frequency Drift Rate that contains the maximum absolute value across all averaging counts is returned as the Peak Frequency Drift Rate Maximum.

Low Energy High Data Throughput (LE-HDT)

Frequency error measurement are performed on format0 and format1 packet format for all data rate configurations.

  1. For each averaging count j (one packet), the following operations are performed:
    • Frequency error is computed across all preamble samples and is recorded as Preamble Frequency Error ω0,j. All packet samples are compensated for ω0,j.
    • After compensating for ω0,j, frequency error is computed over the payload samples and is recorded as Payload Frequency Error ω1,j. Payload samples of the packet are compensated for ω1,j.
    • The sum of frequency error ω0,j and ω1,j is recorded as Frequency Error ω0,j + ω1,j.
  2. The value of the Preamble Frequency Error ω0,j that contains the maximum absolute value across all averaging counts is returned as the Preamble Frequency Error ω0 Maximum.
  3. The value of the Payload Frequency Error ω1,j that contains the maximum absolute value across all averaging counts is returned as the Payload Frequency Error ω1 Maximum.
  4. The value of the Frequency Error ω0,j + ω1,j that contains the maximum absolute value across all averaging counts is returned as the Frequency Error ω0 + ω1 Maximum.