FDM and CDM pilot (DM-RS) multiplexing among antennas

The link between mobile and network is referred as “Air Interface”. This interface is defined by specification bodies like 3GPP (Third Generation Partnership Project). The air interface defines basic rules or protocols for transmission and reception between mobile and networks. These rules include but not limited to: What frequency spectrum will be used, what modulation method will be used, what all stages information bits will go through before transmitted, what waveform will be used for transmission and reception. 

The “air interface” waveform for 5G-NR is based on Orthogonal Frequency Division Multiplexing (OFDM). This means frequency band over which information needs to be transmitted is firstly divided into multiple narrow bands. These narrow bands are called sub-carriers and are mutually orthogonal. Therefore, multiple modulated symbols can be transmitted in parallel on different sub-carriers.  

In 5G-NR, the smallest time-frequency resource is known as Resource element and is consist of one subcarrier in one OFDM symbol. The transmissions are scheduled in group(s) of 12 subcarriers, known as physical resource block (PRB).

A wireless “channel” is medium over which information is conveyed. This channel is unpredictable because of many factors like multipath and shadow fading, Doppler shift, and time dispersion or delay spread. These factors are all related to variability introduced by mobility of the user and the wide range of environmental conditions that are encountered as a result.

It is important that receiver has information about properties of the communication channel for reliable exchange and detection of information conveyed. These properties of the channel are estimated at the receiver. To facilitate estimation of channel, OFDM systems use reference signals (or pilot symbols). These reference signals are embedded in resource blocks. These are also known as DM-RS (Demodulation reference signal).

When M transmit and N receive antennas are deployed, dedicated pilots for each transmit antenna are required to estimate a total of MN channels. Different pilot modes possible:

• FDM (Frequency Division Multiplexing): where pilots for different antennas occupy different tones and thus orthogonal in frequency domain. Pilot sequences of different antennas can use the same Chu sequence. The length of the Chu sequence, or the number of pilot tones per antenna, is Np/M, where Np is the total number of pilot tones.

No alt text provided for this image

• CDM (Code Division Multiplexing): where pilots for different antennas occupy the same time/frequency resources but separated by different codes. Orthogonal pilot sequences can be constructed from shifted versions of the same Chu sequence (frequency domain CDM), possibly with additional Walsh code separation (time domain CDM). The number of pilot tones per block for any transmit antenna is always Np. The orthogonal pilot sequences can be constructed in two different ways, Chu sequence with different shifts, or Chu sequence with Walsh separation from two reference signal blocks.

  1. Frequency domain CDM: In the first construction, the orthogonality between pilot sequences of different antennas are achieved by exploiting the properties of shifted Chu sequences.
  2. Time domain CDM: In the second approach, each entry of the Chu sequence is further modulated by antenna-dependent Walsh codes.
No alt text provided for this image

Both the schemes has its own advantages and disadvantages as far as channel estimation performance in various scenarios is considered. For example for 4×4 MIMO systems:

  1. MIMO system with CDM pilots consistently outperforms that with FDM pilots.
  2. At higher speed, CDM-f4-t1 scheme outperforms others because it does not rely on the time domain coherence between two long RS to separate the antenna streams.


To Question or Not to Question? That Is the Question !

73256total sites visits.

René Descartes was a French philosopher and his best known philosophical statement is “I think, therefore I am”. Statement itself can be argued. However, whole idea is that — if you can question about your own existence, it is enough proof of your existence. If you didn’t exist, you wouldn’t be able to question it.

Questions are important. It may sound trivial but asking right scientific questions is not an easy task. objectives of asking questions may differ, one asks questions to discover something new, one asks questions to understand already well established facts. Questions and subsequently answers give clarity of thought and you get clarity about something you did not understand.

However, most obvious question really is: Can I understand something new in science or technology without asking a question. Almost always answer to this question is ‘NO’. And, then another question is: What really is a good question specially in scientific context. Some would say that a good scientific question is something that can be answered by direct observations, experiments and results of those experiments. These questions are different from questions that are based on people’s opinion or their belief.

In this blog post (and upcoming blog posts) I want to list some of the question that can be asked to understand 5G-NR physical layer.

Question-1: In LTE, Cell Specific Reference Signal (CRS) is an Always-On signal and transmitted in 4 or 6 OFDM non-continuous symbols of each sub-frame. LTE and 5G-NR can share the same DL frequency. So, what are design choices made in design of 5G-NR DL signal so that collision with LTE-CRS can be avoided.

Question-2: In 5G-NR, PSS/SSS/PBCH signals are structured in time and frequency in a very specific way and this structure is known as SS-PBCH block (or SSB in short).

  • What is the reason that this structure uses 240 sub-carries in frequency domain and is 4 OFDM symbol long.
  • Why SSS is sandwiched between two PBCHs ?
  • Were there other alternatives arrangements discussed, what were the disadvantages of other alternatives ?

Question-3: In LTE, Synchronization signals are always located at the center of the RF carrier. There is no decoupling between channel raster and synchronization raster. When UE detects PSS/SSS, it also knows center of the RF carrier.

  • What is the reason that 5G-NR decouples channel raster and synchronization raster ?
  • How does UE know exact SS block frequency-domain location on the carrier ?

Question-4: Multiple PUCCH formats are specified in NR to carry various UCI payload sizes on different formats.

  • Why sequence based design was adopted for PF0 in 5G-NR ?
  • Why DMRS based design with evenly distributed pattern and 1/3 overhead was adopted for PF2 in 5G-NR ?
  • Why specific DMRS based design was adopted for PF1 and PF3 in 5G-NR ?

Question-5: Why specific design choices are made for synchronization signal design?

  • Why does 5G-NR use m-sequence for PSS instead of ZC-sequence based as in 4G-LTE ? 
  • Why does 5G-NR use gold-sequence for SSS instead of interleaved concatenation of two m-sequences with different cyclic shifts as in 4G-LTE ? 

.. To be Continued



The Saga of Sync-up!

If I need to pick two things that are fascinating about nature, I would pick symmetry and synchronization. Mathematician Steven Strogatz in his famous TED talk explored how flocks of creatures (like birds, fireflies and fish) manage to synchronize and act as a unit — when no one’s giving orders. Similarly, The book “The Equation That Couldn’t Be Solved” explores symmetry in everything from biology and physics to music and the visual arts.

A lot of scientific and mathematical discoveries are based on concepts associated with symmetry and synchronization . Albert Einstein used symmetry as a guiding principle when he devised his General Theory of Relativity. James Clerk Maxwell, mathematical physicist, demonstrated symmetry between electric and magnetic fields. 

Initial synchronization is a very important process in communication systems. It comes into picture when a user equipment (UE) tries to connect to network very first time or it tries to handover to new cell. These procedures allow UE to synchronize with the network both in terms of time and frequency, identify itself with the network and acquire knowledge about the network to which it is trying to connect. To aid UE in this process, Synchronization signals are defined. These synchronization signals are transmitted by network and UE detects them to synchronize itself with the network. Keeping in mind importance of these signals, good amount of effort is given to design these signals. One of the design requirement for these signal is that synchronization signals should be based on mathematical sequences that have sharp (time/frequency offset detection) ambiguity function so that mis-detection at UE end does not happen due to time and frequency impairments.

This blog post tries to explore how these synchronization signals are changed with each generation of cellular communication standard starting from 2G-GSM to 5G-NR.

GSM uses TDMA technique for transmitting information and it uses
Frequency Correction Burst (FCB) signal as the frame synchronization information.  The FCB signal consists of a single Continuous-Wave (CW) tone (Pure Sine Wave) transmitted at a frequency 67.708 Hz above the nominal carrier frequency of the downlink signal. The tone is transmitted for 148 symbol intervals in the first time slot of every tenth frame, equivalent to 148/270833=546.4 microseconds. So, the task of the receiver is to detect this pure sine-wave for detecting the frame structure of down-link frame structure. This may look like a trivial signal processing task. However, detecting a sine wave that is deep buried under the noise and impaired by frequency offset and other impairments can be a daunting task.

In 4G-LTE, There are two types of synchronization signals: Primary Synchronization Signal (PSS) and Secondary Synchronization Signals (SSS). PSS is based on class of complex exponential sequences known as Zad-off Chu (ZC) sequence. Its waveform is defined by class of waveforms known as Constant Amplitude Zero-Autocorrelation (CAZAC). SSS is based on maximum length sequences (m-sequences). The sequence used for the SSS is an interleaved concatenation of two length-31 m-  sequences s0(m0) (n) and s1(m1) (n) (also known as “short codes”), where mand mare the indices for the short codes/sequences sand s1, respectively. An m-sequence is a pseudo random binary sequence which can be created by cycling through every possible state of a shift register of length resulting in a sequence of length.

In 5G-NR also there are same two synchronization signals as 4G-LTE. However, type of the signals used is different. PSS is based on m-sequence and SSS is based on Gold sequence. NR-PSS sequence is constructed based on a length-127 BPSK modulated m-sequence. In freq. domain 3 cyclic shifts (0, 43, 86) to get the 3 PSS signals carrying the index of cell ID in a cell ID group. NR-SSS sequence is generated by the multiplication of two BPSK modulated m-sequences with cyclic shifts determined by the cell ID.

Why does 5G-NR use m-sequence for PSS instead of ZC-sequence based as in 4G-LTE ? Given below is great article that explains the reason. In nutshell, The LTE PSS based on Zadoff-Chu (ZC) sequences has a time-frequency ambiguity problem, i.e., the time-frequency auto-correlation of each LTE PSS contains several significant side lobes besides the main lobe at zero time delay and zero carrier frequency offset (CFO). This may lead to false PSS/SSS detection. http://www.dpi-proceedings.com/index.php/dtcse/article/viewFile/26293/25707

Why does 5G-NR use gold-sequence for SSS instead of interleaved concatenation of two m-sequences with different cyclic shifts as in 4G-LTE ?
LTE cell searches can result in false alarms (FAs), where non-existent cells (also known as false cells or ghost cells) are detected. One cause of this is when two nearby cells have SSSs which share a short code. e.g., the cell ID NID(1)=1 has an m0=0, but so does NID(1)=30, NID(1)=59, NID(1)=87, NID(1)=114, NID(1)=160, and NID(1)=165. Accordingly, if a cell has NID(1)=30, and a nearby cell has NID(1)=114, they would have the same s0(m0) (n). This overlap in constituent short codes results in what is known as a “short code collision”, which, in turn, can result in a ghost cell being detected. Other causes of ghost cells include imperfect PSS/SSS cross-correlations between two cell IDs and random noise/interference. [Ref: System and method for cell search enhancement in an LTE system United States Patent 9432922]. In LTE, this issue can to a certain extent be alleviated by performing cell ID verification via the always-on common reference signal (CRS), which is scrambled by the cell ID. For NR, there is no always-on reference signal and the problem is solved inherently by proper SSS sequence selection.

~Peace ~~Dheeraj

Downlink Control Region: 4G-LTE Vs 5G-NR

NR introduced a new concept of CORESETs (COntrol REsource SETs). This concept is related to Downlink Control signaling. To understand this, let’s first revisit “LTE CORESET”. You may argue that there is no such concept in LTE. So, Let’s first define what could be LTE CORESET (or concept closest to CORESETs).
As it can be seen from figure below, In LTE, up-to 4 OFDM symbols are used for Downlink control signaling and it uses the full carrier bandwidth. This region is known as “Control Region”. Control region is situated a the start of the subframe and length can change dynamically and indicated by PCFICH channel. This region can be known as LTE CORESET.

Figure: LTE Control Region

Below given are some of the drawbacks of LTE CORESET

  1. There do not exist mechanisms to perform neither frequency domain scheduling nor Intercell Interference Coordination (ICIC) over the PDCCH and hence, low Signal to Interference plus Noise Ratio (SINR) levels at cell edges.
  2. Release-12 eMTC devices does not use full system bandwidth. This leads to the complexities.
  3. UE first needs to decode PCFICH channel and then only it knows the length of control region. This dependency makes pipelined processing complex.
  4. Resources assigned for control region can not be used for data channels.

Although LTE-A provides some mechanisms like introduction of ePDCCH to overcome some of the issues related to LTE PDCCH yet those are not enough. Therefore, a more flexible structure was introduced in NR.

  • In NR, CORESET is defined as time-domain and frequency-domain resources. In time domain, a CORESET is semi-statically configured with one or a set of contiguous OFDM symbols (up to 3 OFDM symbols) and it can be located anywhere in the slot. The configuration indicates the starting OFDM symbol and time duration. In frequency domain, a CORESET can be located anywhere in the frequency range of the carrier (However, not outside the active Bandwidth Part). And, is made up of multiple of resource bocks (i.e. multiples of 12 REs). This flexibility to blank-out certain CORESET or to configure overlapping CORESET provides greater flexibility and helps in avoiding intercell interference.
  • There are two types of CORESETs : Common CORESET and UE-Specific CORESET. After UE has detected the serving cell and decoded PBCH in SS- block to acquire the essential system information such as bandwidth, SFN and etc. Then the UE needs to decode the remaining system information to get the initial access related parameters. The remaining system information is scheduled by the NR-PDCCH, however, UE cannot expect any configuration information for the CORESET except for the information carried in the NR-PBCH at this stage. In this sense, the configuration for CORESET 0 (common CORESET) is indicated by MIB (Master Information Block). In the connected mode, UE can be configured with UE-specific CORESETS. There can be up-to 12 CORESETs configured (up-to 3 CORESETs in each BWP).
  • when the control resource set spans multiple OFDM symbols, NR support a control channel candidate to be mapped to multiple OFDM symbols or to a single OFDM symbol. (As shown in figure below). In this case, The gNB informs UE which control channel candidates are mapped to each subset of OFDM symbols in the control resource set.


73256total sites visits.