ADF7021BCPZ_06 View Datasheet(PDF) - Analog Devices

Part Name

Description

Manufacturer

ADF7021BCPZ_06

High Performance Narrowband ISM Transceiver IC

Analog Devices 

ADF7021

Preliminary Technical Data

Binary Frequency Shift Keying (2FSK)

Two-level frequency shift keying is implemented by setting the

N value for the center frequency and then toggling it with the

TxDATA line. The deviation from the center frequency is set

using Bits R2_DB[19:27]. The deviation from the center

frequency in Hz is:

Direct RF output:

FSK DEVIATION [Hz] =

PFD ×TX

_

FREQUENCY

216

_

DEVIATION

Divide-by-2 enabled:

FSK DEVIATION [Hz] = 0.5 ×

PFD ×TX _ FREQUENCY _ DEVIATION

216

where:

TX_FREQUENCY_DEVIATION is a number from 1 to 511

(R2_DB[19:27]).

4R

FSK DEVIATION

FREQUENCY

PFD/

CHARGE

PUMP

PA STAGE

VCO

÷N

–FDEV

+FDEV

TxDATA

THIRD-ORDER

Σ-Δ MODULATOR

FRACTIONAL-N

INTEGER-N

Figure 16. 2FSK Implementation

Three-Level Frequency Shift Keying (3FSK)

In three-level FSK modulation (also known as ternary FSK), the

binary data (Logic 0 and Logic 1) is mapped onto three distinct

frequencies:

• the carrier frequency (FC),

• the carrier frequency minus a deviation frequency

(FC − FDEV), and

• the carrier frequency plus the deviation frequency

(FC + FDEV).

A Logic 0 is mapped to the carrier frequency while a Logic 1 is

either mapped onto frequency FC − FDEV or FC + FDEV.

The bits to frequency mapping result in a reduced transmission

bandwidth as energy is removed from the sidebands and

transferred to the carrier frequency. This increases the spectral

efficiency in comparison to 2FSK.

To ensure there is no loss in SNR at the receiver due to the

ternary nature of the modulation, a simple convolutional

encoder is used. A block diagram of the transmit hardware used

to realize this system is shown in Figure 17. The convolutional

encoder polynomial used to implement the transmit spectral

shaping is P(D) = 1 − D2. A precoder with transfer function

1/P(D) is used to undo the encoding process of the

convolutional code P(D) at the transmitter side.

Tx DATA

0, 1 PRECODER 0, 1

1/P(D)

CONVOLUTIONAL

ENCODER

P(D)

0, +1, –1

FSK MOD

CONTROL

AND

DATA FILTERING

FC

FC + FDEV

FC – FDEV

TO

N DIVIDER

Figure 17. 3FSK Encoding

The signal map of one-zero binary data to the 3-level convolu-

tional output follows. The convolutional encoder restricts the

maximum number of sequential +1’s or −1’s to two and also

delivers an equal number of +1’s and −1’s to the FSK modulator,

thus ensuring equal spectral energy in both 3FSK sidebands.

3-Level Signal Mapping of the Convolutional Encoder

Tx DATA

1 01 1 001 001

Precoder O/P 1 0 0 1 0 1 1 1 1 0

Encoder O/P +1 0 −1 +1 0 0 +1 0 0 −1

3FSK is selected by setting bits in R2_DB[4:6]. It can also be

used with raised cosine filtering to further increase the spectral

efficiency of the transmit signal.

Four-Level Frequency Shift Keying (4FSK)

Four-level frequency shift keying differs from binary FSK in

that four possible frequencies are used to represent each pair of

data bits to be transmitted instead of two possible frequencies

for each bit of data. This halves the symbol rate because a

frequency change is only required after every second bit. It also

doubles the number of symbols because there are four possible

arrangements of a pair of consecutive NRZ data bits. By

minimizing the separation between symbol frequencies, 4FSK

can have high spectral efficiency.

Tx DATA 0 0 0 1 1 0 1 1

F

+3FDEV

SYMBOL

FREQUENCIES

+FDEV

–FDEV

–3FDEV

t

Figure 18. 4FSK Encoding

Oversampled 2FSK

In oversampled 2FSK, the data is sampled at 32 times the

programmed rate, allowing odd data rates.

Rev. PrI | Page 16 of 44

Share Link: