In digital communication system, it’s important to measure the number of bit errors (bits corrupted by channels or bad synchronization) in the received bits over the total number of transmitted bits. The ratio between bit errors and total transmitted bits is usually called the bit error rate (BER). Pseudo-random binary sequences (PRBS) are commonly used for BER measurement in digital communication systems for several reasons, the main one being the randomized, balanced number of ones and zeros after a maximum sequence period.

Generation of PBRS sequences

PBRS sequences can be generated using linear feedback shift registers (LFSR) defined by two properties: length of shift registers and feedback taps configuration. An LFSR length of N bits generates a maximum sequence length as follows 

[1]:

L = 2– 1

Feedback taps configuration of the LFSR can be represented in finite field arithmetic as a polynomial modulo 2 [1]. Well-known LFRS polynomials can be found in many publications [1]. For example, PBRS9 has two taps at 9th and 5th can be represented as the following polynomial:

x9+ x+ 1

A feedback polynomial can be configured in either a Fibonacci or Galois style [1], as show in Figure 1and Figure 2, respectively.

Figure 1: PBRS9 with Fibonacci configuration

Figure 1: PBRS9 with Fibonacci configuration

Figure 2: PBRS9 with Galois configuration

Figure 2: PBRS9 with Galois configuration

Using PBRS in bit errors measurement

Figure 3 shows a simple digital communication chain with PBRS9 as a random bit generator with a maximum length of 511 bits.  Initial state of the LFSR of the transmit PBRS9 can be an arbitrary value but must be different from zero.

At the receiver, once the first bit is detected and demodulated by the demodulator, the receive PBRS9 generator uses the first 9 bits as an initial state for its  LFSR. As a result, the receive PBRS9 generates the same sequence delayed by 9 bits to that generated by the transmitter. An alignment of the demodulated bits from the demodulator and the receive PRRS9 generator is required to perform a measurement of error bits.

Figure 3: BER measurement using PBRS9

Figure 3: BER measurement using PBRS9

In order to avoid an invalid initial state of the receive PBRS9, an additional mechanism is required to detect burst errors between the demodulated bits and the PBRS9 generated bits. If there are too many consecutive bits corrupted by the channel or bad synchronization, the receive PBRS9 is required to re-initialize its initial state before generating a new bit sequence.

Conclusion

This blog post showed a simple method to measure BER in digital communication systems by using PBRS sequences. By using demodulated bits at the receiver as the initial state of the LFSR inside the PBRS generator, the receive PBRS can generate a delayed copy version of the PBRS sequence in the transmitter for BER measurement. In upcoming blog posts, I will demonstrate how to implement a PBRS generator for BER measurement of a simple digital communication chain using Xilinx System Generator [2] in conjunction with Nutaq’s Model-Based Design Kit (MBDK) [3].

References

[1]

Wikipedia. (2014) Linear feedback shift register. [Online]. http:/http://www.nutaq.com.wikipedia.org/wiki/Linear_feedback_shift_register

[2]

Xilinx Inc. (2012) System Generator User Guide. [Online]. http://www.xilinx.com/support/documentation/sw_manuals/xilinx14_4/sysgen_user.pdf

[3]

Nutaq Inc. (2014) Model-Based Design Kit (MBDK). [Online]. http://www.nutaq.com/software/model-based-design-kit