The use of higher data rates in applications
Even though the higher bit rates increase the transaction speed, in some cases the standard bit rate might be even more efficient than HBR. To get a better view on the relationship between bit rates and transaction speed, some timing measurements have to be made.
Transaction speed versus bit rates
So based on the Mifare DESFire card a timing measurement is done, and the overall timing versus bit rate is shown in Table 12: and Fig 3.

Table 12: Transaction time versus bit rateMifare DESFire example
bit rate 106 kbit/s 212 kbit/s 424 kbit/s 848 kbit/s
Read 1K Byte 130 ms 80 ms 52 ms 43 ms
Read 1k Byte encrypted 171 ms 116 ms 86 ms 76 ms
Read 1k Byte encrypted + PCD time (100ms) 271 ms 216 ms 186 ms 176 ms

This measurement shows the overall time for a read transaction, including the Card activation sequence, protocol activation and card deselection according to the ISO/IEC14443: REQA, Anticollision, Select7, RATS, PPS (for switching to HBR), Select Application8, Read, Deselect
A fast PICC was taken, and just a pure read of data was performed, which reduces the overhead times to a minimum9. In addition to that the amount of data was set to 1kbyte to reduce the impact of the card activation sequence (which is always done with 106kbit/s) to a minimum.
Remark: The measurement was done with a DESFire card, which has a frame size of 64 bytes. The HBRs are used with equal bit rates in both directions.
The first example (first line of Table 12:) shows a kind of “best case” scenario: the data is read in plain. This timing is a pure PICC related one, as the command times are measured from transmitting a command until the PICC answer is completely received. So no PCD timing is involved here. The second example (second line of Table 12:) shows an encrypted read of 1kbyte of data. This timing still is a pure PICC related one, as the command times are measured from transmitting a command until the PICC answer is completely received. So no PCD timing is involved here.

The third example (third line of Table 12:) shows an encrypted read of 1kbyte of data including a PCD timing. The PCD time is assumed to be 100ms, which is quite fast for such a transaction (decryption + data handling, and whatever the system does with the 1kbyte of data).
All three timings are shown in Fig 3.

Mifare 1K Printed Card,Mifare Full Color Offset Printing Cards,Mifare 1K Proximity Contactless Cards

Fig 3. Transaction time versus bit rate (example)

It turns out, that even though this measurement is based on a simple read transaction, the overall transaction time is only partly related to the bit rate. Doubling the bit rate from 106kbit/s to 212kbit/s reduces the transaction from 130ms to 80ms in best case, which gives a “gain” of approximately 40% (see Table 13:). When doing the same with encryption and considering the PCD – even with a very fast PCD – the result shows, that in a real application, doubling the bit rate from 106kbit/ to 212kbit/s just gives a “gain” of 20%. The gain is even less, if such an application does not use the “best case” scenario of a pure read, but requires some additional operation on the PICC or in the PCD. Doubling the bit rate from 424kbit/s to 848kbit/s in such an application only gains 5% (or less) in the overall transaction speed.