ADPD107 datasheet(24/66 Pages) AD

24 / 66 page

ADPD105/ADPD106/ADPD107

Data Sheet

Rev. A | Page 24 of 66

LED Pulse and Sample

At each sampling period, the selected LED driver drives a series

of LED pulses, as shown in Figure 26. The magnitude, duration,

and number of pulses are programmable over the I2C interface.

Each LED pulse coincides with a sensing period so that the sensed

value represents the total charge acquired on the photodiode in

response to only the corresponding LED pulse. Charge, such as

ambient light, that does not correspond to the LED pulse is

rejected.

After each LED pulse, the photodiode output relating the pulsed

LED signal is sampled and converted to a digital value by the

14-bit ADC. Each subsequent conversion within a sampling

period is summed with the previous result. Up to 255 pulse values

from the ADC can be summed in an individual sampling period.

There is a 20-bit maximum range for each sampling period.

Averaging

The ADPD105/ADPD106/ADPD107 offer sample accumulation

and averaging functionality to increase signal resolution.

Within a sampling period, the AFE can sum up to 256 sequential

pulses. As shown in Figure 25, samples acquired by the AFE are

clipped to 20 bits at the output of the AFE. Additional resolution,

up to 27 bits, can be achieved by averaging between sampling

periods. This accumulated data of N samples is stored as 27-bit

values and can be read out directly by using the 32-bit output

registers or the 32-bit FIFO configuration.

When using the averaging feature set up by Register 0x15,

subsequent pulses can be averaged by powers of 2. The user can

select from 2, 4, 8 … up to 128 samples to be averaged. Pulse

data is still acquired by the AFE at the sampling frequency,

fSAMPLE (Register 0x12), but new data is written to the registers at

the rate of fSAMPLE/N every Nth sample. This new data consists of

the sum of the previous N samples. The full 32-bit sum is stored

in the 32-bit registers. However, before sending this data to the

FIFO, a divide by N operation occurs. This divide operation

maintains bit depth to prevent clipping on the FIFO.

Use this between sample averaging to lower the noise while

maintaining 16-bit resolution. If the pulse count registers are

kept to 8 or less, the 16-bit width is never exceeded. Therefore,

when using Register 0x15 to average subsequent pulses, many

pulses can be accumulated without exceeding the 16-bit word

width. This averaging can reduce the number of FIFO reads

required by the host processor.

Data Read

The host processor reads output data from the ADPD105/

ADPD107 via the I2C protocol on the ADPD105 or the SPI port

on the ADPD107. Data is read from the data registers or from

the FIFO. New output data is made available every N samples,

where N is the user configured averaging factor. The averaging

factors for Time Slot A and Time Slot B are configurable

independently of each other. If they are the same, both time slots

can be configured to save data to the FIFO. If the two averaging

factors are different, only one time slot can save data to the FIFO;

data from the other time slot can be read from the output

registers.

The data read operations are described in more detail in the

Reading Data section.

0.5

1.0

1.5

2.0

2.5

3.0

TIME (s)

OPTICAL SAMPLING

LOCATIONS

NUMBER OF LED PULSES (

nA OR nB)

LED

CURRENT

(ILED)

SHOWN WITH fSAMPLE = 10 Hz

Figure 26. Example of a Photoplethysmography (PPG) Signal Sampled at a Data Rate of 10 Hz Using Five Pulses per Sample

ADPD107 датащи(PDF) 24 Page - Analog Devices

ADPD107 датащи(HTML) 24 Page - Analog Devices

Html Pages

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

датащи скачать

ссылки URL