5G New Radio: Physical Downlink Control Channel(NR-PDCCH)
- Published on November 15, 2019
Anil Kr. Shaw
Research Assistant @IITH || Working on Physical Layer of 5G-NR at IIT Hyderabad
Similar to the previous radio access systems, e.g., LTE, 5G new radio(NR) uses the Physical Downlink Control Channel(PDCCH) to perform physical layer control functions such as Scheduling the Downlink(DL) broadcast and DL/uplink(UL) unicast data transmission and signaling triggers for periodic and aperiodic transmission/reception.
NR supports a Large variety of frequency bands ranging from sub-GHz to 100 GHz. Due to Path Loss and Channel condition corresponding to different transmission environment, the design should ensure good coverage for NR-PDCCH transmission. This becomes a significant consideration especially in mmWave bands wherein narrow beamforming is typically employed to achieve coverage and additional design elements to incorporate beamforming and beam-sweeping implementation at the gNodeB(Base Station) become essential.
Note: For both TDD and FDD duplex schemes, as well as in both downlink and uplink, a group of 12 sub-carriers contiguous in frequency over one slot in time form a Resource Block (RB).
Qs. Why we go for PDCCH?
Ans: PDCCH is used to decode DCI(Downlink Control Information ) which is of minimum 12bits coming from the upper layer. Decoding of which give scheduling allocation to the UE on the PDSCH or PUSCH. For example, if the UE has the data in the PDSCH it needs to know where the data is located. If the UE is unable to decode PDCCH then the UE canโt read the PDSCH in that sub-frame and consistent decoding failure leads to RLF(Radio Link Failure). Hence the decoding of PDCCH is extremely important.
RLF(Radio Link Failure): Itโs a total connection failure between UE and gNB(Base Station).
one โOut of Sync Indicationโ in this diagram means โ20 subframes of consecutive PDCCH decoding failure. one โIn Sync Indicationโ in this diagram means โ10 subframes of consecutive PDCCH decoding success.
Letโs go for Understanding of NR-PDCCH briefly:
First We will go for the technical terms used in the PDCCH-
1.CCE(Control Channel Element): PDCCH consists of numbers of CCEs defined by aggregation level(maximum=16) which is of power of two. Itโs a group of bundles of REGs
2. Aggregation Level: Which denotes how many CCEs have used for data transmission. Based on user location we go for different aggregation levels. If the user is near to the gNB(Base Station) we go for a lower aggregation level and vice-versa.
3. Bandwidth Part: There was no flexibility in LTE in terms of bandwidth usage. We have to use the whole bandwidth for data transmission. But in the case of NR, we divide the whole bandwidth maximum of 4 as per the usage. And each part we say BandWidthPart(BWP).
Note: In one BWP we have 273 PRBs.
BandWidthPart in NR
4. CORESET(Control Resource Set): Itโs a set of time-frequency resources. it is the same as the control region in LTE whole system bandwidth is used for the DCI transmission and PCFICH is used to determine how many time-domain region{1,2,3} is used but in NR, the exact duration is configured to the UE by higher layer such as SI or UE-specific RRC depending on whether itโs common CORESET or UE-specific CORESET.
But in NR, CORESET is not defined for whole bandwidth for some set of PRBs we are using in NR CORESET and the time-domain region is defined by RRC (Radio Resource Control is the protocol used for signaling between 5G radio network and UE) signaling.
Note: We can have a maximum of 3 CORESET per BWP.
NR-CORESET Structure
5.Search Space: Itโs the place where UE goes for Blind Decoding. CORESET and Search Space are related to each other. The following equation is used to calculate CCE indices in the search space. It gives the different PDCCH candidate(M) values based on the nRNTI (Which is UE-specific) for the given slot and sub-frame number.
CCEs indices calculation
Note: In one BWP maximum 10 search space exists.
CORESET to Search Space(SSP)mapping is done by Base Station. Letโs say in one DLBWP 3 SSP, to coreset0, 3ssp to coreset1 and rest in coreset2 (total 10 SSP). TS 38.331 we find inside SearchSpaceId, ControlResourceSetId exists which tell about in that search space which time domain symbol and frequency domain resource blocks we have to look for to decode DCI.
6.RNTI(Radio Network Temporary Identifier): There are different types of RNTI is used. RNTIs are used to differentiate/identify a connected UE in the cell.
Flow Chart for the PDCCH Tx
Transmitter Chain of PDCCH
Flow chart for PDCCH Receiver
Receiver Chain of PDCCH
Resource Mapping to the frame of PDCCH
Qs. Why we are going for rate matching?
Ans: In simple words to fit the data into the number of resources available for PDCCH transmission we do rate matching but for in-depth we have also heard about the Shannon Channel Capacity Theorem. Shannon showed that it is, in fact, possible to communicate at a positive rate and at the same time maintain a low error probability as desired. However, the rate is limited by a maximum rate called the channel capacity. If one attempts to send data at rates above the channel capacity i.e R>C, it will be impossible to recover it from errors. This is called Shannonโs noisy channel coding theorem and it can be summarized as follows:
โ A given communication system has a maximum rate of information โ C, known as the channel capacity.
โ If the transmission information rate R is less than C, then the data transmission in the presence of noise can be made to happen with arbitrarily small error probabilities by using intelligent coding techniques.
โ To get lower error probabilities, the encoder has to work on longer blocks of signal data. This entails longer delays and higher computational requirements.
Thatโs one of the reasons for Rate Matching (RโคC).
Qs. Why we go for CCE to REG interleaving?
Ans: To achieve Diversity without placing more antenna is one of the reasons we go for time and frequency diversity and CCE to REG interleaving function does the same job. It interleaves the sequence in a first-time manner then frequency. We allow data to get interleaved beyond coherence time and coherence bandwidth to achieve diversity.
Qs. Why we go for Scrambling with Pn sequence?
Ans: The scrambling process encodes every bit within every code-word depending on the scrambling sequence. The purpose of scrambling is to reject the inter-cell interference. Interference from other cells is de-scrambled as noise when the received data is de-scrambled with a known cell-specific scrambling sequence at the receiver. The scrambling process is to use input bit sequence to carry out a bit-wise XOR operation with a cell specified pseudo-random sequence generated by length-31 Gold sequence generator. The input bit sequence is the block of bits within the code-word and the output bit sequence is the scrambled sequence with the same length as the input sequence.
Thanks a lot for reading, if you really like my article you can follow me on ๐ท๐๐๐๐๐๐ด๐, ๐ฟ๐๐๐๐๐๐, ๐ด๐๐๐๐๐๐๐๐, ๐ฝ๐๐๐๐๐๐๐๐ฒ๐๐๐ & ๐ผ๐๐๐๐
