CDMA Reference Blockset

Understanding Frame Contents and Padding

For some blocks in this blockset, the size of an input or output signal does not necessarily indicate the number of elements that the block processes while performing its communication task. This means that studying or troubleshooting a model requires a clear understanding of how to interpret the contents of a frame. Examining signal sizes alone is not sufficient.

The Problem.   Some blocks need to process different amounts of information in each frame, depending on their Rate input value, Rate set parameter, and/or Channel type parameter. As an example, the IS-95A Fwd Ch Convolutional Encoder library block encodes 192 bits when it simulates the encoder in a full-rate Paging channel, 96 bits when it simulates the encoder in a half-rate Paging channel, and 32 bits when it simulates the encoder in an eighth-rate Sync channel.

Furthermore, as the section Understanding Rates and Rate Sets indicates, some blocks (including the IS-95A Fwd Ch Convolutional Encoder library block) allow you to change the data transmission rate during a simulation by changing the value of the Rate input signal. However, Simulink does not permit a signal to change its vector size during a simulation. This means that a block cannot accept an input data frame of 192 bits when its Rate input indicates a full-rate channel, and 96 bits when its Rate input indicates a half-rate channel.

The Solution.   To accept or produce the correct amount of data while also keeping its input and output vector sizes constant throughout the simulation, some blocks simply accept or produce padded data. For example, the IS-95A Fwd Ch Convolutional Encoder always accepts an input of exactly 288 bits. This number is the frame size that the convolutional encoder for a full-rate Rate Set II Traffic channel uses. If you use the block to simulate the encoder for a full-rate Traffic channel, then your input signal is naturally a vector of size 288. If you use the block to simulate the encoder for a full-rate Paging channel, then you must pad the 192 bits that you want to encode with 96 zeros, so that the input vector to the block has a vector size of 288. Similarly, the block always produces an output of exactly 576 bits. It automatically pads its output vector with zeros if the coded data frame consists of fewer than 576 bits.

If a block uses this padded signal scheme to accommodate varying data amounts while keeping signal sizes constant, then its block reference page includes information such as:

• The vector size of input and output signals
• The number of elements of an input or output vector representing the actual data, as opposed to zeros used to pad the vector; columns of a table list the "Number of Relevant Input Bits" and/or "Number of Relevant Output Bits" for given channel types and rates.

You can access the block's reference page by pressing the Help button in its dialog box.

Hints for Using the CDMA Reference Blockset

Demo Models