Getting Started with the Communications Blockset (Communications Blockset)

Communications Blockset

Learning More About the Example

After you understand how the model works, you can learn even more about it and about how the Communications Blockset works by modifying the model or by reading other sources of information. This section suggests modifications and reference sources that might be helpful.

Modifying the Model

The table below describes some changes that you can make in the model and how to implement such changes. Changes that require a greater understanding of the blocks or techniques involved appear towards the end of the table.

Table 1-2: Ideas for Modifying the Example Model
Description of Change
Modification of Model

Change sequences of random numbers
Change the Initial seed parameter in the Bernoulli Random Binary Generator block and/or the AWGN Channel block to another positive integer.

Change distribution of random numbers in message signal
Change the Probability of a zero parameter in the Bernoulli Random Binary Generator block to another number between 0 and 1.

Change amount of noise in channel
Change the Es/No parameter in the AWGN Channel block to another real number. Alternatively, change the Mode parameter to Signal to noise ratio (SNR) or Variance from mask and set the associated SNR or Variance parameter, respectively.

Change traceback depth in decoder
Change the Traceback depth parameter in the Viterbi Decoder block to another positive integer, and also change the Receive delay parameter in the Error Rate Calculation block to that integer plus one.

Change convolutional code
Change the Trellis structure parameters in both the Convolutional Encoder and Viterbi Decoder blocks. For example, change it to poly2trellis(9,[753 561]).
Note that if the new code rate is not 1/2, then you might need to change other parts of the model to ensure that the signal sizes and sample times are appropriate throughout the model.

Change number of soft-decision bits
Change the Number of soft decision bits in the Viterbi Decoder block, and also change the Quantization partition and Quantization codebook parameters in the Sampled Quantizer Encode block. For example, change these three parameters to 2, [-.5 0 .5], and [3 2 1 0], respectively.
More generally, the length of Quantization partition must be one less than the length of Quantization codebook, and the values in Quantization codebook must be between 0 and 2^b-1, where b is the Number of soft decision bits parameter.

**Table 1-2: Ideas for Modifying the Example Model**
Description of Change	Modification of Model
Change sequences of random numbers	Change the Initial seed parameter in the Bernoulli Random Binary Generator block and/or the AWGN Channel block to another positive integer.
Change distribution of random numbers in message signal	Change the Probability of a zero parameter in the Bernoulli Random Binary Generator block to another number between 0 and 1.
Change amount of noise in channel	Change the Es/No parameter in the AWGN Channel block to another real number. Alternatively, change the Mode parameter to Signal to noise ratio (SNR) or Variance from mask and set the associated SNR or Variance parameter, respectively.
Change traceback depth in decoder	Change the Traceback depth parameter in the Viterbi Decoder block to another positive integer, and also change the Receive delay parameter in the Error Rate Calculation block to that integer plus one.
Change convolutional code	Change the Trellis structure parameters in both the Convolutional Encoder and Viterbi Decoder blocks. For example, change it to `poly2trellis(9,[753 561])`. Note that if the new code rate is not 1/2, then you might need to change other parts of the model to ensure that the signal sizes and sample times are appropriate throughout the model.
Change number of soft-decision bits	Change the Number of soft decision bits in the Viterbi Decoder block, and also change the Quantization partition and Quantization codebook parameters in the Sampled Quantizer Encode block. For example, change these three parameters to `2`, `[-.5 0 .5]`, and `[3 2 1 0]`, respectively. More generally, the length of Quantization partition must be one less than the length of Quantization codebook, and the values in Quantization codebook must be between 0 and 2^b-1, where b is the Number of soft decision bits parameter.

Other Blocks For Further Study