Synchronization Signal Blocks (SSB) in 5G New Radio (NR)

This is a new episode of our series 5G Explained. In this video, we discuss the synchronization signal block or SSB in 5G New Radio.

We will look at its components: the synchronization signals and the broadcast channel, which carries the master information block. Finally, we will have a look at the block repetition pattern and organization as burst.

The synchronization block helps with initial synchronization. It consists of three components.

The first component of the SSB is the primary synchronization signal, which is one of three possible sequences. The PSS comes first and occupies the center 127 resource elements of the 240 block. The UE typically runs three correlators in parallel, one tuned to each one of the possible sequences and, when it detects one of them, it knows the timing of the SSB.

The second component is the secondary synchronization signal, one of 336 possible sequences. The SSS comes two OFDM symbols later and also occupies the center 127 resource elements. The second step identifies which SSS was sent. The combination of PSS and SSS yields one of three times 336 or 1008 possible physical cell IDs.

Finally, the broadcast channel or BCH, which carries the master information block with its essential set of information to get started, completes the SSB. DMRS symbols are sent along the PBCH and shown in yellow.

Altogether, the synchronization signal block is always 4 OFDM symbols long and 240 subcarriers wide, irrespective of the subcarrier spacing. Note that, when using the highest subcarrier spacing of 240kHz, the SSB is close to 60MHz wide.

The SSB is usually not sent just once, but in bursts of almost repetition of the SSB. We will talk about it in more detail later in this video. The main uses for the SSB are initial synchronization and cell search, as well as cell search of neighbor cells when connected. It also provides a first piece of information about suitable beamforming or, in other words, the relative position of the base station and the UE, as we will see later.

While the concept of PSS and SSS and two-step cell search is identical to LTE, there are a few differences worth noting:

·       The sequences are longer (127 instead of 62).

·       There are more cell IDs because there are more possible SSS compared to the number of SSS pairs used in LTE.

·       The scheduling is quite different, because the PSS, SSS and PBCH are always sent together as a block. The repetition interval for PSS is no longer 5ms. It can be longer. Also, only one SSS is sent, as opposed to a pair of 31 possible SSS in LTE.

·       Finally, the SSB can now be beamformed as explained in another episode of this series 5G Explained.

The purpose of the broadcast channel is to carry the master information block or MIB. The broadcast channel is mapped to the physical broadcast channel, which is transmitted as part of the synchronization signal block (SSB).

The main difference in the processing chain compared to LTE is the use of polar coding instead of tail-biting convolutional coding, as we saw in the “Downlink Control Information” episode of this 5G Explained series. The MATLAB code below shows how this chain is implemented in 5G Toolbox.

We now want to have a look at the payload itself or MIB.

The information carried by the broadcast channel consists of two parts. One part, the MIB, which is constant over 80ms, and another part that changes over 80ms and, therefore, is not really part of the MIB.

The MIB includes fundamental parameters and information for a UE trying to access the cell:

·       Is the cell accessible?

·       Information about where to find the next piece of information, the system information block 1 or SIB1

·       The location of the common resource grid, which carries the SIB1

·       The system frame number

·       Other pieces of information, which vary over 80ms include:

·       The SS block index. Actually, this piece of information is only present for FR2 which is mmW, and it only contains 3 bits of the 6 needed to identify an SSB. We will explain the reason for this content in our episode about initial acquisition procedures.

·       Other information such as the 4 LSBs of the system frame number as well as the CRC

Once encoded, the broadcast channel content undergoes scrambling and QPSK modulation before being mapped to the grid. This chain is similar to the downlink control channel chain.

Within one half frame, meaning 5ms, there are a number of occurrences of SS blocks. Note that, because the scrambling code for each block depends on the block index, these occurrences are not repetitions of each other.

The broadcast channel may appear in the first or second half of a frame. Its position is indicated by the half frame bit, which is part of the BCH content. Each one of these occurrences can be switched off, which means that the cell does not necessarily transmit all of them. One group of occurrences is called a synchronization signal burst, and it consists of one or several synchronization signal blocks or SSB.

We have already mentioned that the BCH content in different occurrences of an SSB is different. One additional important point to understand is that the DMRS in each occurrence is also different. It can be one of eight possible sequences. This will let the UE tell apart those occurrences, as will be explained in more detail in our episode about initial acquisition procedures.

The SSB can be transmitted with different subcarrier spacings, ranging from 15kHz to 240kHz. Note that this subcarrier spacing of 240kHz is available for the BCH but not for data or PDSCH, as explained in the 5G introduction episode of this 5G Explained series. Also, 60kHz is never used for BCH.

Remember that, whatever the subcarrier spacing, the SSB always occupies 240 subcarriers. This means that its bandwidth increases with increasing subcarrier spacing but, at the same time, its duration shrinks.

The table on this slide shows the maximum number of occurrences of an SSB in a synchronization signal burst. It is four or eight, depending on the carrier frequency, for FR1, but it can be as high as 64 for mmWave or FR2.

At those subcarrier spacings, the duration of the SSB is much shorter, and it is possible to transmit more of them in the same amount of time. This enables finer beamforming at mmWave frequencies.

This slide and the next show different configurations for BCH type A through E. For cases A, B, and C, two configurations are possible with a maximum number of occurrences of four or eight depending on the carrier frequency. In each case, all occurrences fit within one half frame.

For cases D and E, the maximum number of occurrences in one synchronization signal burst is always 64 and, here too, the burst fits within one half frame or 5ms.

In LTE, the PSS is transmitted every 5ms, while the broadcast channel is transmitted every 10ms. In 5G NR, the period is the same for both, but it can take a value as low as 5ms and as high as 160ms.

If you consider that the SSB is the only always-on signal in a cell, you can see that it is possible to have very low transmission power in some 5G cells with no or low traffic. This is quite different from LTE, with its always on cell-specific reference signals and frequent PSS/SSS and BCH.

As the standard says a UE can assume an SSB occurs every 20ms, there seems to be some conflicting data here. The reason is that, in ordinary cells, the SSB periodicity is likely to be 20 ms or less, but 5G NR allows for extra power savings in cells. Those cells may not be discovered by a UE, or at least not reliably, but they may be reserved for other purposes such as a secondary carrier component, which is not meant to be stand-alone.

Here, you can see how you can set up the SSB in MathWorks 5G Toolbox. Parameters include the block repetition pattern, as we just saw, the physical cell ID, which blocks get transmitted, where a 1 indicates a transmitted block, and the periodicity.

Then MIB content, such as information to access the SIB1, or whether the cell is accessible or barred. The last line generates the synchronization signal burst. 

Here, you can see two examples of settings, with the resulting time-frequency content of the SS burst. These pictures should be familiar as we have used them all along in this video. You can see that you can select to only send the first four blocks, as would be necessary under 3GHz, by setting the last four bits of the SSBTransmitted bitmap to zero. On the other hand, you can send all eight blocks when operating over 3GHz. Note that the carrier frequency at which we operate is not explicitly stated, and you could choose to send only a few blocks even above 6GHz.

This concludes this episode of the 5G Explained series on the synchronization signal block.