Received June 16, 2020, accepted July 6, 2020, date of publication July 20, 2020, date of current version

August 5, 2020.
Digital Object Identifier 10.1109/ACCESS.2020.3010500

On the Design Details of SS/PBCH, Signal

Generation and PRACH in 5G-NR
ARVIND CHAKRAPANI , (Member, IEEE)
Qualcomm Technologies, Inc., Bridgewater, NJ 08807, USA
e-mail: [email protected]
This work was supported by Qualcomm Inc., Technologies.

ABSTRACT The 3rd Generation Partnership Project (3GPP) specification of the fifth generation (5G) New
Radio (NR) allows for a highly scalable and flexible radio access technology to cater to network operators
with different requirements. Such scalability and flexibilities in network configurations inevitably translate
to complications in the design and implementation of 5G-NR systems. Radio access in 5G-NR is much more
complex and involved than its predecessor, 4G long term evolution (LTE) and LTE-Advanced technology.
Therefore, the 5G-NR specifications turn out to be quite dense. Specifically, the specifications are concise,
design motivations rarely explained, and the information can be convoluted or distributed across several
documents. Moreover, there are several key design details associated with the access layer procedures for any
given physical layer channel, which are often omitted in the specifications. For example, design motivation
aspects of initial access channels or signal generation can be quite difficult to follow or understand in
5G-NR. In this paper, all the design details associated with initial access channels and signal generation
in 5G-NR specifications are laid out. The contributions of the paper are three folds. First, the design details
and justifications associated with both downlink and uplink access channels are presented along with signal
generation details. Secondly, receiver design aspects of NR PRACH short formats are discussed. Lastly,
PRACH receiver implementation aspects and performance reports from different network operators are
presented and compared with 3GPP specified Radio Performance and Protocol aspect requirements for
millimeter wave (mmW) access. The work in this paper is of significant value to researchers and system
engineers looking to design and build initial access algorithms as part of 5G NR systems.

INDEX TERMS 3GPP, 5G-NR, 4G, LTE, OFDM.

I. INTRODUCTION accommodate NR-LTE coexistence; different patterns can be

With innumerous and radically diverse deployment scenarios configured for the time/frequency placement of SS/PBCH
in 5G-NR, the NR cell architecture should offer scalability and remaining system information (RMSI) blocks and so
and flexibility across an extreme variation in connectivity on. Further, SS/PBCH can be placed on a different sync
requirements [1]. Very high throughput (1 Gbps or more), raster than the regular channel raster (only option in LTE),
ultra-low latency (order of 0.5ms in some cases), ultra-high for faster downlink synchronization by making synchroniza-
reliability and mobility with low energy consumption are tion blocks sparser in frequency (i.e., SS/PBCH placement
the key driving factors of 5G-NR. Towards this end, all at only given frequency locations). This facility requires
the physical layers channels can be configured flexibly to calculations on the sync raster entries based on bandwidth
cater for different scenarios. Such flexibility has led to a configuration, SS/PBCH resource size, etc. For uplink initial
dense and complex specifications for physical layer proce- access/synchronization, the random access (PRACH) channel
dures in 5G-NR. For example, for initial downlink access, supports several long and short formats for different cover-
the synchronization block (SS/PBCH) can be transmitted ages and combinations of formats (short) with configurable
with a sub-carrier spacing (SCS) different than the carrier time/frequency placements. A PRACH preamble can have
SCS for faster synchronization; different time placement pat- varying number of sequence repetitions depending on the for-
tern of SS/PBCH can be adopted (only for 30kHz SCS) to mat to provide different coverage range. There could also be
several preambles within a slot to cater for multiple random
The associate editor coordinating the review of this manuscript and access occasions. Support of formats with large repetitions
approving it for publication was Larbi Boubchir . or multiple PRACH occasions within a slot, increases the

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
VOLUME 8, 2020 136617
 A. Chakrapani: On the Design Details of SS/PBCH, Signal Generation and PRACH in 5G-NR

receiver processing load at a 5G next generation-base-station

(gNB) and therefore the gNB feature support capabilities
should be carefully determined. Further, to help towards
beam correspondence, different SS/PBCH to PRACH map-
ping are defined so that single/multiple SS/PBCH map to
multiple/single PRACH occasions along with the regular one-
to-one mapping. These mapping options have implications on
the SS/PBCH and PRACH receiver processing complexity at
the gNB. Also, if a common digital front end (DFE) chain is
to be used for data and random access reception in the same
slot, the PRACH receiver at gNB will be required to perform
additional phase corrections due to cyclic prefix (CP) removal
on the data channel.
As 5G-NR specification allows for different SCS which
can be configured for initial access and data channels;
provides flexibility of placing the resources on separate band-
width parts (BWPs1 ), and therefore there are several fre-
quency offsets parameters defined in the 3GPP physical layer
procedures specification with complicated equations. These
offsets are necessary for correct placements of resources and
for correct OFDM signal generation and reception in 5G-NR.
To understand all the details on design goals and motivations
for SS/PBCH, PRACH channels and OFDM signal genera-
tion in 5G-NR, one must go through several 3GPP specifica-
tions and contributions. This could prove be a very difficult FIGURE 1. Synchronization signal structure in 4G-LTE and 5GNR.
task for anyone not familiar with 3GPP standardization pro-
cess and documentation. Moreover, information is scattered
across a vast number of documents with nested references, II. SS/PBCH DESIGN
making the task of connecting the dots a very tedious process. SS/PBCH block transmission allows UEs to synchronize or
The objective of this paper is to unwind all the design lock to a cell. In NR, synchronization signal blocks constitute
details associated with the initial access channels and illus- of a primary synchronization signal (PSS), a secondary syn-
trate them with examples wherever possible to help under- chronization signal (SSS), and a physical broadcast channel
standing better the overall design. Specifically, the following (PBCH). Unlike LTE, which has PSS/SSS transmitted every
contributions are made in this paper. First, all the design 5ms and PBCH transmitted every 10ms, the three SS/PBCH
details associated with downlink access channels and OFDM components in NR are always transmitted together (i.e., they
signal generation in 5G-NR are presented with examples. all have the same periodicity). As illustrated in Figure 1,
Secondly, design details of PRACH channel along with a given SS/PBCH is repeated within a set of SS/PBCH
important receiver design aspects are presented. Finally, transmissions, which can be used for gNB beam-sweeping
PRACH receiver implementation aspects and performance transmission [2]. One or multiple SS/PBCHs compose an
from different network operators are presented and com- SS/PBCH set as indicated in Fig 1. The SS/PBCH set is con-
pared with RAN42 recommended minimum requirements fined to a 5 ms window and transmitted periodically. For ini-
for millimeter wave (mmW) access. Rest of the document tial cell selection, the user equipment (UE) assumes a default
is organized as follows. In section 2, SSS/PBCH channel SS/PBCH set periodicity of 20 ms, but subsequently this peri-
design details are provided with illustrations and exam- odicity can be configured to be 5/10/40/80/160 ms depend-
ples. Section 3, details of the OFDM signal generation and ing on need to make SS/PBCH more frequent or sparser.
phase pre-compensation at the transmitter are presented. Synchronization during initial access is a two-step identifi-
Section 4 focuses on all the PRACH channel design aspects cation process (via PSS and SSS) to provide for both timing
and details. Lastly, important aspects of the PRACH receiver (only symbol and slot) and frequency synchronization. PBCH
design along with RAN4 performance from different network demodulation will reveal system frame number and enable
operators is presented and discussed. reception of control/data channels (PDCCH/PDSCH). Detec-
tion of RMSI (transmitted over regular PDCCH/PDSCH)
1 In 5G-NR, gNB can fragment the carrier bandwidth into smaller BWPs may be necessary during initial access for UE to perform ran-
with similar or different subcarrier spacing [4]. This flexibility allows users dom access (via PRACH). However, transmission of RMSI
with different bandwidth and device capabilities to operate within a smaller
bandwidth part (BWP) compared to the carrier bandwidth.
itself is optional to gNB (for non-standalone or NSA mode)
2 Radio Performance and Protocol aspect requirements are specified by the and the presence or absence of RMSI is indicated in the
3GPP Radio Access Network working group 4, also known as RAN4 PBCH block. Note that SS/PBCH transmission can also be

136618 VOLUME 8, 2020

A. Chakrapani: On the Design Details of SS/PBCH, Signal Generation and PRACH in 5G-NR

used for signal measurement purposes at the user equip-

ment (UE) and only those SS/PBCH associated with an RMSI
are referred to as a Cell-Defining SS Block (CD-SSB) (see
section 5.2.4 of [4]). PSS/SSS/PBCH can be transmitted with
SCS of 15/30kHz for frequency range 1 (i.e., FR1 or sub-6
bands) and with SCS of 120/240kHz for frequency range 2
(i.e., FR2 or mmW bands). However, RMSI is defined only
for SCS of 15/30kHz for FR1 and SCS of 60/120kHz for
FR2 [2]. This is mainly to cater for different channel condi-
tions and different coverages with different frequency ranges.
The SS/PBCH set contents, including the maximum number
of SS/PBCHs within an SS/PBCH set, SS/PBCH mapping
pattern, and SS/PBCH set mapping to slots in a radio frame
is also carrier-frequency-dependent.
Regardless of the SS/PBCH set composition, the trans-
mission of SS/PBCHs within an SS/PBCH set is confined
to a 5 ms window. The maximum number of SS/PBCHs FIGURE 2. Sync and channel raster.
within an SS/PBCH set (i.e., within 5ms period) is specified
to be 4 for frequency ranges up to 3 GHz, 8 for 3 to 6 GHz,
or 64 for 6 to 52.6 GHz in order to achieve a trade-off between LTE, the SS/PBCH block in NR need not be in a fixed
coverage and resource overhead. Furthermore, the number position within the configured bandwidth of the RF carrier
of actual transmitted SS/PBCHs is configurable and could but can be placed anywhere on the RF carrier bandwidth.
be less than the maximum number. This option is particu- This enables a sync channel raster that is sparser than the RF
larly useful towards reducing processing burden at the gNB, channel raster [5]. The advantage of having a sparser sync
especially in the case of requiring transmission of multiple channel raster is a reduced search time for the initial access
SS/PBCHs on multiple carriers within a slot. (less hypothesis to work with at the UE receiver). It relies
Within a broadcast channel (BCH) transmission time inter- on the SS/PBCH block bandwidth (i.e., PBCH bandwidth)
val (TTI) period of 80 ms, there are 16 possible positions being smaller than full channel bandwidth of the RF carrier
of an SS/ PBCH set if we consider the minimum period transmitted from the gNB and that it is not in a fixed position
for an SS/PBCH set is 5 ms. The 16 possible positions of within the configured bandwidth. The PBCH bandwidth of
an SS/PBCH set could be identified by the 3 least signif- 240 subcarriers corresponding to 20 resource blocks (RBs)
icant bits (LSB) of the System Frame Number (SFN) and defines the flexibility of placing the sync channel. A primary
1-bit half radio frame number index. The SS/PBCHs are cell (PCell) is always associated to a CD-SSB located on
repeated within an SS/PBCH set and when the UE detects the synchronization raster. Note that in 5G-NR the definition
an SS/PBCH, it can acquire the timing information from of a physical resource block (PRB) is the same as LTE
its PBCH, from which the UE can identify the radio frame with 12 subcarriers constituting an RB, but the bandwidth of
number, the slot index in a radio frame, and the orthogonal a PRB varies with the numerology or SCS being used.
frequency-division multiplexing (OFDM) symbol index in a The relation between the RF channel raster and sync chan-
slot. The timing information contains additional 6 bits for nel raster is demonstrated with an example in Figure 2. In the
SFN, 1 bit for half radio frame index, and 2, 3, or 6 bits example, two carrier positions 1 and 2 are shown leading to
for SS/PBCH time index for frequency ranges up to 3 GHz, two different placements of the PBCH. The PBCH (and the
3 to 6 GHz, and greater than 6 GHz, respectively. Within the SS) can be placed on a raster that is sparser than the RF carrier
SS/PBCH indices, two or three LSBs are carried by chang- raster and those sync channel raster positions are denoted
ing the demodulation reference signal (DMRS) sequence of by FSC,i . Position 1 is the highest (rightmost) position of
PBCH. Thus, for the frequency ranges below 6 GHz, the the configured bandwidth on the RF channel raster where
UE can acquire the SS/PBCH index without decoding the the PBCH can be related to the sync channel position FSC,1 ,
PBCH. It also facilitates PBCH soft combining over multiple thus the PBCH occurs as far left as possible on the carrier.
SS/PBCHs as these SS/PBCHs with different indices carry Position 2 is the next position on the RF channel raster, thus
the same content of PBCH payload. SS/PBCH can be placed offset by 1FCH ,Raster + 1FSC,Raster from Position 1. The
(in frequency) on channel raster (as in LTE) or the newly sync channel raster spacing 1FSC,Raster will be limited by
defined sync channel raster in 5G-NR. the following equation:
1FSC,Raster ≤ BWConfig − BWPBCH + FCH ,Raster
A. SYNC CHANNEL RASTER
Synchronization (or sync) channel raster identifies the set of where BWConfig is the configured transmission bandwidth,
possible frequency locations of the SS/PBCH, consisting the BWPBCH is the PBCH and 1FCH ,Raster is the channel raster
synchronization channels PSS/SSS and the PBCH. Unlike spacing.

VOLUME 8, 2020 136619

 A. Chakrapani: On the Design Details of SS/PBCH, Signal Generation and PRACH in 5G-NR

FIGURE 3. Alignment between SS block and channel RBs (see section 4.3.14 of [5]).

Sync raster is defined such that there is a minimum number 2. Offset in PRB units from Point A to the first usable PRB
of entries for each band [5] and raster entries are defined for - offsetToCarrier in SCS-SpecificCarrier
each band (see section 4.3.1 of [5]). Sync raster entries will 3. Carrier bandwidth in PRB units - carrierBandwidth in
be defined for initial system acquisition, but SS/PBCH blocks SCS-SpecificCarrier
can be transmitted by gNB in other frequency locations if the 4. Sub-carrier spacing to determine the size of PRB - subcar-
position is signaled to the UE explicitly3 using the parameter rierSpacing in SCS-SpecificCarrier
kSSB , which is derived from the frequency difference between
Note that point A is RE#0 of RB#0 used to gener-
the SS/PBCH block and Common Resource Block (CRB),
ate sequences for reference signals and scrambling. From
also referred to as Point A in [3] ( see section 7.4.3.1 of [3]
section 7.4.3.1 of [3], the quantity kssb is the subcarrier offset
and section 13 of [7]). The SS/PBCH block is not RB aligned SSB to subcar-
from subcarrier 0 in common resource block NCRB
with the data RBs in the channel as shown in Figure 3. Instead,
rier 0 of the SS/PBCH block, where the 4 least significant
there is an arbitrary offset between the edge of the SS/PBCH
bits of kssb are given by the higher-layer parameter ssb-
block RBs and the edge of the data RBs in the channel and
SubcarrierOffset and for SS/PBCH block type A the most
this offset can be up to 11 Resource Elements (REs) for
significant bit of kssb is given by aĀ+5 in the PBCH payload
mmW. Such placement enables multiple radio channels (with
as defined in subclause 7.1.1 of [3] (also see section 4,
different NR-Absolute radio-frequency channel numbers or
TS 38.212). The value kssb ∈ {0, · · · , 11} for FR2 and
NR-ARFCN) that are subcarrier grid aligned but not RB grid
kssb ∈ {0, · · · , 23} for FR1. For FR2, or when µ ∈ {2, 3}, the
aligned to use the same SS/PBCH block location. Hence, SSB are expressed in units of resource
quantities kSSB , and NCRB
radio channels with different NR-ARFCNs that are offset by
blocks assuming 60 kHz subcarrier spacing. For FR1, or when
up to 11 REs in frequency can re-use the same SS/PBCH
µ ∈ {0, 1} , the quantities kSSB , and NCRB SSB are expressed in
block frequency location.
units of resource blocks assuming 15 kHz subcarrier spacing
The gNB places the SS/PBCH block at SSref (one of the
(see section 4.2.2.2 of [3]). Figure 4 shows the computation of
synchronization raster’s). This is indicated by the red-line at
subcarrier offset kssb for different numerologies of SS/PBCH
the center of SS/PBCH block in Figure 3. The gNB also sig-
and RMSI in FR1 and why kssb ∈ {0, · · · , 23} for FR1. The
nals the offset (in number of subcarriers) between SS/PBCH
parameter kssb ∈ {0, · · · , 11} for FR2, since SS/PBCH is
block and the channel data RBs to help determine the control
defined only for 120/240kHz and RMSI is defined only for
resource set for Type0-PDCCH common. This serves as an
60/120kHz and that the configuration of SS/PBCH at 60kHz
indirection for RMSI. Following parameters are signaled in
and RMSI at 120kHz is not valid (and hence 12 values for
RMSI for each supported SCS (each SCS is usually a different
kssb suffice).
component carrier).
1. Absolute frequency of Point A - absoluteFrequencyPointA
B. CRB TO SS/PBCH OFFSET EXAMPLES
in FrequencyInfoDL, signaled as ARFCN NR
The common resource block and sub-carrier offsets to
SS/PBCH should be computed in terms of subcarrier spac-
3 This is handled by gNB by setting the Master Information Block (MIB) ing of SS/PBCH for correct placement of the resources in
parameter ssb-SubcarrierOffset(or the PHY parameter kSSB ) as kSSB > 11 frequency. In [3] however, for FR2, or when µ ∈ {2, 3},
for FR2 or kSSB > 23 for FR1 (see section 4.1 of [7]). SSB are expressed in units of
the quantities kSSB , and NCRB

136620 VOLUME 8, 2020

A. Chakrapani: On the Design Details of SS/PBCH, Signal Generation and PRACH in 5G-NR

C. SS/PBCH BLOCK
In order to provide enough deployment flexibility for NR,
the number of NR physical-layer cell identities (PCIDs)
is extended to 1008 (504 in LTE). Each NR-cell ID can
be jointly represented by a PSS/SSS. The SS/PBCH block
arrangement is shown in Figure 6. The PSSs consist of three
frequency-domain-based binary phase shift keying (BPSK)
m-sequences with length-127, and the SSSs correspond to
m-sequences of length-127 picked from 336 m-sequences.
Both PSS and SSS signals are mapped onto 127 contiguous
subcarriers. With very good cross-correlation properties of
m-sequences, NR SSs significantly outperform LTE SSs in
terms of both PCID detection probability and false detection
probability in the cases of initial and non-initial acquisition,
respectively, as shown in [11]. For each SS/PBCH block, the
PSS, SSS, and PBCH share the same single antenna port.
It should be noted that the physical beams applied to an
SS/PBCH block are transparent to the UE since the latter
only sees the equivalent SSs and PBCH signal after potential
precoding and/or beamforming operations that are up to the
network implementation. The unique physical-layer cell ID
is given by (see section 7.4.2 of [3]),
(1) (2)
cell
NID = 3NID + NID ,
(1) (2) (2)
where NID ∈ {0, 1, . . . ., 335} and NID ∈ {0, 1, 2}. NID is
(1)
carried by PSS, whereas NID is carried by SSS. PSS is a
frequency domain-based pure BPSK M-sequence with 1 gen-
erator polynomial (g (x) =x7 + x4 +1) and 3 cyclic shifts in
(2)
frequency domain (43N ID = {0, 43, 86}). The SSS sequence
FIGURE 4. Subcarrier offset ks sb in FR1.
is generated using two generator polynomials with cyclic
(1) (2)
shifts according to cell IDs NID and NID . For more details
(for implementation) on PSS/SSS sequence generation refer
resource blocks assuming 60 kHz subcarrier spacing. Let,
offset to section 7.4.2.2.1 and 7.4.2.3.1 of [3] respectively. The
SSBCRB represent the offset from point A to the closest
offset PBCH REs maps to 240 subcarriers in frequency and in
common resource block before SS/PBCH. Let, SSBsc be time spans over 2 full symbols and an additional 8 RBs
the subcarrier offset from subcarrier 0 in common resource on the SSS symbol. PBCH contains PBCH data REs and
offset
block SSBCRB to subcarrier 0 of the SS/PBCH block assum- PBCH DMRS REs. Energy Per Resource Element (EPRE)
ing sub-carrier spacing of SS/PBCH. The SSB CRB offset between PBCH-DMRS and PBCH-data shall be equal [3].
offset offset
(SSBCRB ) and subcarrier offset (SSBsc ) are illustrated for PBCH DMRS mapping will be frequency-first, time-second
FR2 in Figure 5. For FR2, SS/PBCH can only use either in increasing frequency order. Also, physical-layer cell ID
120kHz or 240kHz subcarrier spacing. The first tone of based frequency shift is used for PBCH-DMRS RE locations
SS/PBCH block (or the PBCH tone) can only begin at the (see table 7.4.3.1-1 of [3]). The 3 bits of SS block index are
center (or peak) of a 120kHz/204kHz subcarrier. This means carried by changing DMRS sequence within each 5ms period,
that kssb ∈ {0, . . . , 11} can only take values which are mul- half-frame information is also provided for max L = 4, with
tiples of 2 when SS/PBCH subcarrier spacing is 120kHz and the remaining bits of the timing information carried explicitly
values which are multiples of 4 when SS/PBCH subcarrier in the PBCH payload.
offset offset
spacing is 240kHz. Computing of SSBCRB and SSBsc using Note that the REs that are not used for SS/PBCH block in
SSB can be done as, SSB offset 1f ∗N SSB
kSSB , and NCRB = b CRB CRB
1fSSB c any data RB that partially or fully contain SS/PBCH block
 CRB
offset 1fCRB ∗NCRB
SSB REs are transmitted with zero power and other PHY chan-
and SSBsc = kSSB1f∗1f SSB
CRB
+mod 1fSSB
RB , N RB ,
∗ NSC SC nels are rate matched around such PRBs. On the PDSCH
where 1fCRB and 1fSSB are SCS of carrier and SS/PBCH carrying RMSI and the corresponding PDCCH CORESET,
respectively. Note, unequal tone amplitudes between different no SS/PBCH block is transmitted in the allocated resources.
SCS are illustrated in Figure 5 only to avoid clutter and need When the SS/PBCH and PDSCH are scheduled in the same
not be the case in practice. symbols, DMRS of data and SS/PBCH can be in the same

VOLUME 8, 2020 136621

 A. Chakrapani: On the Design Details of SS/PBCH, Signal Generation and PRACH in 5G-NR

FIGURE 5. CRB and subcarrier offset with different numerology in FR2.

information element (IE) consisting of 23 bits (shown below

in red font) defined in [S4] and the extra 8 bits related to
additional timing information (see section 7.1.1 of [3]). The
CHOICE bit along with the MIB payload is also known
as the BCH payload. Note that the ‘‘CHOICE Bit’’ (see
section A.3.2 of [S4]) is not part of the MIB but is part of the
BCH message which includes the CHOICE bit and the MIB
payload.

Note that the PBCH payload bits (denoted below with the
same notation as in section 7.1.1 of [3])
PBCH Payload
FIGURE 6. SS/PBCH block resource arrangement. z }| {
BCH Payload
z }| {
CHOICE BIT MIB payload Additional Timing bits
symbol, provided DMRS of data and SS/PBCH are not over- z}|{ z }| { z }| {
ā0 ā1 , ā2 , ā3 , . . . , āĀ−1 āĀ , āĀ+1 , āĀ+2 , āĀ+3 , . . . , āĀ+7
lapping in the frequency domain, spatially quasi co-located
(QCL’d) and have the same SCS (see section 10.1 of [3] and must be interleaved as defined in section 7.1.1 of [3], using
section 5.1.6.2 of [9]). the interleaver pattern defined in Table 7.1.1-1 of [3]. PBCH
payload is scrambled (1st scrambling) before polar encod-
D. PBCH PAYLOAD ing as per the method defined in section 7.1.2 of [3]. Note
The PBCH payload size is 32 bits and is contained within that the scrambling depends on fields that are already sent
the SS/PBCH command. The PBCH payload consists of in the PBCH command (NCell ID , ssbIndex) and header (SFN).
three parts, CHOICE bit (1bit), MIB payload or the MIB The scrambling sequence for the additional timing bits is 0.

136622 VOLUME 8, 2020

A. Chakrapani: On the Design Details of SS/PBCH, Signal Generation and PRACH in 5G-NR

Note that after the interleaving process, the position of the

additional timing bits will change according to the interleaver
pattern and care should to be taken to set only the scrambling
bits to zero which correspond to the additional timing bits.
III. OFDM SIGNAL GENERATION
In Figure 2, the spectrum allocation available for a given car-
rier for various subcarrier spacings with respect to reference
Point A was shown. The common RB numbering in Figure 2
(denoted as CRB in [3]) is for the whole carrier. The first sub-
carrier of the common RB of different numerologies coincide
at a reference point called Point A (see section 4.4.4.2 of [3]).
Different bandwidth parts (either for uplink or downlink)
can be defined by the network using different sub-carrier
spacings (i.e., with 1f , 21f , etc.). Irrespective of the differ-
ent bandwidth parts the OFDM signal generation equation
remains essentially common (with appropriate subcarrier-
based scaling for parameters involved) for a given carrier cen-
ter frequency. For successful demodulation on any bandwidth
part, a phase pre-compensation is applied for all channels
(except PRACH). The OFDM signal generation is described
below. The baseband OFDM signal (dropping antenna port p FIGURE 7. Center frequency mismatches at gNB and UE.
notation for convenience) in given by (section 5.3 of [3])
size,µRB −1
beginning of every symbol is described in the up-conversion
Ngrid NSC
related parts of the specifications (section 5.4 for [3]). The
µ
X
sl (t) = αk,l implementation is easier to perform at the baseband rather
k=0 than RF. To understand this, consider the baseband OFDM
µ µ µ
   
size,µ RB
j2π k+k0 −Ngrid NSC /2 1f t−TCP,l −tstart,l signal (dropping antenna port p notation for convenience) in
×e (1)
section 5.3 of [3].
µ µ µ
with TCP,l = NCP,l Tc being the duration of CP, tstart,l ≤ t < size,µ
RB −1
Ngrid NSC
µ µ µ
+
tstart,l + Ncp,l TC is the transmission time range of
Nu µ
X
sl (t) = αk,l
µ µ
symbol l, with Nu and Ncp,l being the symbol and CP sizes k=0
µ 
µ
 
µ µ

respectively; tstart,l is the symbol start time, as defined in [3]; size,µ RB
j2π k+k0 −Ngrid NSC /2 1f t−TCP,l −tstart,l
1f and µ are the subcarrier spacing and subcarrier spacing ×e (3)
configuration respectively, defined in section 4.2 of [3]; αk,l µ µ µ
with = NCP,l Tc being the duration of CP, tstart,l ≤ t <
TCP,l
is the l th transmitted data symbol on subcarrier index k; µ µ µ
µ tstart,l + Nu + Ncp,l TC is the transmission time range of
k0 is a term that enables single carrier frequency for all µ µ
numerologies symbol l, with Nu and Ncp,l being the symbol and CP sizes
µ
µ

start,µ size,µ
 respectively;TC is the basic unit of sample duration; tstart,l
k0 = Ngrid + Ngrid /2 Nsc RB
is the symbol start time, as defined in [3]; 1f and µ are

start,µ size,µ

RB µ0 −µ the subcarrier spacing and subcarrier spacing configuration
− Ngrid 0 + Ngrid 0 /2 Nsc 2 (2)
respectively, defined in section 4.2 of [3]; αk,l is the l th trans-
µ
In [3], µ0 is the largest µ value among the subcarrier mitted data symbol on subcarrier index k; k0 is a term that
spacing configurations provided to the UE for this carrier. enables use of a single carrier frequency for all numerologies
(see following sections). Assuming the center frequency of
A. PHASE RESET AT SYMBOL BOUNDARIES: DOWNLINK gNB
the gNB to be f0 , the transmitted signal is,
RECEPTION OF SS-PBCH/RMSI/BWP
gNB
During initial access UE will not know the downlink center xl (t) = ej2πf0 t
sl (t)
frequency and it may not be the same as the center frequency
Assuming the UE center frequency as f0UE , the received
of SS/PBCH block and RMSI. Similarly, when the UE is UE
configured to receive on a bandwidth part (BWP), there could signal can be written as, yl (t) = e−j2πf0 t xl (t). Expanding,
mismatch in frequencies used at gNB and UE. The mismatch µ
 
yl t + tstart,l
in center frequencies at transmitter and receiver during initial
access is illustrated in Figure 7. This is handled at the gNB gNB phase part UE phase part
z }| {
transmitter/UE receiver by resetting phase at OFDM symbol { −j2πf UE T µ +t µ 
z }|
j2πf0
gNB 
µ µ 0 CP,l start,l
boundaries at the RF center frequency. This phase reset at the =e TCP,l + tstart,l e

VOLUME 8, 2020 136623

 A. Chakrapani: On the Design Details of SS/PBCH, Signal Generation and PRACH in 5G-NR

size,µRB −1
Ngrid NSC
X
× hk,l ak,l
k
µ µ
   
size,µ RB
j2π k+k0 −Ngrid NSC /2+1k 1f t−TCP,l
×e (4)
gNB size,µ RB
f0 −f0UE µ Ngrid NSC
where 1k = 1f + k0 − 2 . Note that the
gNB
center frequency offset (i.e., f0 − f0UE ) is always an
integer number of subcarriers. Also, 1k is independent of l
µ
and hence will not impact reception. Therefore, the impact FIGURE 8. Relationship between k0 , f0 and other parameters.

of this center frequency offset is only the per-symbol phase

rotation caused by gNB center frequency as the UE phase is useful to do the following with only the knowledge of
part in Eq. (4 ) can be corrected as f0UE is known at the UE. the corresponding BWP parameters. 1) UE can receive SS-
The gNB part of the residual phase on every symbol can be PBCH, RMSI or data on a BWP 2) UE can transmit on any
compensated in the baseband (at gNB) by multiplying with BWP. After downlink synchronization, the center frequencies
gNB
the conjugate of the gNB part of the residual phase at the of gNB and UE will be the same, i.e., f0 = f0UE = f0 and a
transmitter. This is, single f0 is used to transmit and receive on a given BWP. Users
gNB µ µ
within a given BWP can center their FFT at the middle of their
µ (TCP,l +tstart,l )
sl (t) e−j2πf0 , (5) configured BWP, and while upconverting use a frequency
offset to the carrier center frequency f0 . The frequency f0BWP
which will result in the received signal not having terms
to be used during up-conversion is given by
dependent on gNB center frequency. The phase compensation 
can also be done at the up-conversion step (as described in the
.
BWP size,µ RB start,µ RB
f0 = f0 − Ngrid NSC 2 − NBWP NSC
spec [3]) as follows –

µ µ
   .
gNB size,µ RB start,µ RB
µ −NBWP NSC 2 + Ngrid NSC 1f .
gNB −j2πf0 TCP,l +tstart,l
Re sl (t) ej2πf0 t ∗ e
µ µ
  
gNB
µ j2πf t−TCP,l −tstart,l At the gNB receiver, a corresponding phase reset must be
= Re sl (t) e 0 (6)
applied at symbol boundaries in RF. Again, this is easier to do
in the baseband (just like in the transmitter). The per-symbol
This essentially implies the phase of the RF signal is reset
phase correction to be applied at the receiverfor OFDM sym-
at symbol boundaries. This is difficult to implement in RF µ µ

analog domain. The baseband signal can be pre-compensated bol index l is given by ejθl , where θl = 2πf0 tstart,l + TCP,l ,
so that there is no need for phase reset in RF. Note that, only where f0 is the receiver center frequency in Hz. If the receiver
for PRACH signal transmission, there is no common phase center frequency is a multiple of 1 kHz, it can be shown that
correction applied during the up-conversion at the transmitter the phase correction factors θl defined above are periodic with
(see section 5.4 of [3]) as accurate phase information may a periodicity of 1ms (i.e. at most 28 unique values for µ = 1,
not be necessary for PRACH demodulation. Common phase see [26]). If the receiver center frequency includes a 7.5 kHz
correction can be implemented in the IFFT/FFT modules of shift when UL is shared with LTE, the periodicity is 2ms (i.e.
transmitter/receiver respectively. This phase reset works only at most 56 unique values for µ = 1 [26]).
when the difference in the transmission and reception fre-
gNB B. FREQUENCY OFFSET PARAMETER
quencies is a multiple of subcarrier spacing. f0 is a multiple
gNB µ µ The locations of the RB grids of all numerologies are
of 1 kHz and hence the phase terms e−j2πf0 (TCP,l +tstart,l ) will start,µ
indicated to the UE using RRC parameters Ngrid,x and
repeat every 1 ms (see [26]). So, it is enough to pre-compute size,µ µ
Ngrid,x [8]. The parameter k0 is the subcarrier offset from
phase terms for 2µ ×14 symbols. The subcarriers in the range size,µ RB
  the middle subcarrier of numerology µ (i.e. the Ngrid,x Nsc /
start,µ start,µ size,µ RB
NBWP −N grid RB
NSC , 2 + 1‘th subcarrier in the Ngrid,x Nsc subcarriers) to f0 ,
 as illustrated in Figure 8. From Figure 8, one can write,

start,µ start,µ
 size,µ
RB RB
NBWP −N grid NSC +N NSC −1 start,µ
RB
f0 − fpoint−A = Ngrid,x Nsc 1f
BWP
size,µ RB µ
is of interest, where NBWP andNgrid
start,µ start,µ
are starting CRB +Ngrid,x N /2×1f − k0 1f (7)
sc
of BWP and carrier grid with sub-carrier spacing µ
Therefore, for any two numerologies µ and µ0 :
respectively (see Figure 8). So, a smaller FFT around
 size,µ
start,µ start,µ RB µ
 
N RB /2 − 1 should be suf- + Ngrid,x 0 N /2 − k0 0 × 2µ0
NBWP −N grid RB + N
NSC start,µ RB size,µ
Ngrid,x 0 Nsc
BWP SC sc
ficient. Note that the pre-compensation and IFFT operations 
size,µ RB µ

= Ngrid,x Nsc + Ngrid,x N /2 − k0 × 2µ .
start,µ RB
are interchangeable. The phase reset at symbol boundaries (8)
sc

136624 VOLUME 8, 2020

A. Chakrapani: On the Design Details of SS/PBCH, Signal Generation and PRACH in 5G-NR

TABLE 1. PRACH configuration for Frequency Range 2 (FR2).

FIGURE 9. PRACH preamble for different formats with short-sequence length.

µ
Rewriting the above equation for k0 , we have, length 839 and 139 respectively. Long formats can be con-
figured only for sub-6GHz frequency range (FR1), whereas
µ start,µ size,µ RB
k0 = Ngrid,x NscRB
+ Ngrid,x N /2 short formats can be configured for both sub-6 (FR1) and
sc

µ0 size,µ0 RB
 mmW (FR2) frequency range. Long formats option is mainly
+ k0 − Ngrid,x Nsc − Ngrid,x N /2 × 2µ0 −µ .
start,µ0 RB
(9)
sc aimed towards catering for LTE re-farming and large cell
Note that f0 does not have to be on the channel raster nor scenarios [2]. In this paper, we will consider only the short
the center of the RF filter. The channel raster or the RF preamble formats. The PRACH preamble consists of a cyclic
prefix of duration TCP (NCPRA samples) and sequence part of
implementations do not impose any restriction to the value
µ µ duration TSEQ (Nu samples and with or without repetitions
of k0 0 . Therefore, the value of k0 0 can be predefined and is
µ0 µ depending on the PRACH format) and guard duration of
set to 0 i.e., k0 = 0. Accordingly, k0 can be derived using:
TGP . Figure 9 shows the various PRACH preamble formats
µ
 
start,µ size,µ
k0 = Ngrid + Ngrid /2 Nsc RB
associated with short-sequence length, with sizes of cyclic

start,µ size,µ

RB µ0 −µ
prefix (CP), PRACH sequence, Guard duration (if any) in
− Ngrid 0 + Ngrid 0 /2 Nsc 2 . (10) number of samples for the configured parameters (i.e., µ, fs
and starting symbol shown on the right side of Figure 9)
IV. PRACH DESIGN and the bandwidth of each format. For PRACH formats with
In this section we will discuss PRACH channel and its no guard duration, the CP is aggregated for the number of
design principles in 5G-NR. Two preamble formats are repetitions of the sequence(s) and appended at the beginning
defined in NR, long and short preamble formats of sequence of the PRACH sequence(s). For PRACH formats with guard

VOLUME 8, 2020 136625

 A. Chakrapani: On the Design Details of SS/PBCH, Signal Generation and PRACH in 5G-NR

FIGURE 10. PRACH time frequency structure.

duration, the guard duration TGP extends from the end of the within the PRACH period. For a given SFN, a PRACH slot
preamble to the next symbol boundary. The preambles for all can be present only when nSFN mod x = y. Further, for a
formats always start and end at symbol boundaries within a given SFN, where PRACH slots can be present (i.e., when
slot (see Figure 12 in section IV-B). Also shown is the length nSFN mod x = y), PRACH can be only be on slots given by
of each preamble sequence in micro-seconds. The CP length, ‘Slot Number’ column in Table 6.3.3.2-4 of [3] for FR2.
sequence length and total preamble length (in samples) can As an example, let SFN = 1 (i.e., 2nd frame) and let PRACH
be calculated for short sequence length preambles using the config index = 14 (see Table 1). Then, for PRACH config
Table 6.3.3.1-2 of [3]. The time duration of different preamble index = 14, the parameters x = 1 and y = 0 (using
components (sequence, CP, guard etc) for different preamble Table 6.3.3.2-4 of [3] of FR2), shown below. PRACH slots are
formats shown in Figure 9 are with subcarrier spacing of on slot numbers 24,29,34,39 (using Table 6.3.3.2-4 of [3] of
1fRA =120kHz (µ = 3). Bandwidth for a frequency domain FR2), from the snippet shown in Table 1. Section 2 of [24] has
occasion with shortl sequence
m and SCS of 120kHz can be cal- more information on PRACH slot in a frame based on subcar-
culated as BW = 139 12 × 12 tones × 120kHz = 17.28MHz. rier spacing. Within any given PRACH slot (see Figure 10),
multiple PRACH occasions can exist consecutively in time
and frequency. Utilization of multiple time & frequency
A. PRACH TIME/FREQUENCY STRUCTURE resources is towards yielding NR-PRACH capacity as large as
Within a cell, preamble transmission can take place within a LTE [29]. This is because PRACH capacity would be limited
configurable subset of slots (denoted as the PRACH slots) compared to LTE with large subcarrier spacing and short
that repeats every PRACH configuration period as shown sequence lengths.
in Figure 10. The PRACH configuration period is given by Details on time domain occasions within a PRACH slot
the parameter x in Table 6.3.3.2-4 of [3] for FR2. PRACH is shown in Figure 10. There can be multiple frequency
periodicity can be configured to range from 10 ms up domain PRACH occasions jointly covering K ∗ M consec-
to 160 ms. A configurable set of PRACH slots are present utive resource blocks, where K is the preamble bandwidth

136626 VOLUME 8, 2020

A. Chakrapani: On the Design Details of SS/PBCH, Signal Generation and PRACH in 5G-NR

accommodate PDCCH on first two symbols (see section 2.1

of [25]). Start symbol 6 and 8 was chosen to avoid DL to
UL interference. A significant concern was the transmis-
sion point (TRP) to TRP, or DL-to-UL interference between
PRACH and a preceding PDCCH transmitted in a gap at the
beginning of the slot [32]. If the preamble format specified
in Tables 6.3.3.2-2 to 6.3.3.2-4 of [3] is A1/B1, A2/B2 or
A3/B3, then only the last TD occasion will have the pream-
ble format of B1, B2 or B3 respectively (see section 5.3.2
FIGURE 11. PRACH slot definitions for different µ.
of [3]). Note that, formats B2 and B3 are only supported
in combination with A2 and A3 respectively, i.e., B2 and
B3 cannot be configured by themselves. Also, with Ax/Bx
measured in number of resource blocks and M is the num- format, supporting Bx seems mandatory(see [30]), although
ber of frequency-domain PRACH occasions defined by the the 3GPP specification doesn’t explicitly state this. Here,
higher layer parameter msg1-FDM with M ∈ {1, · · · , 8}. Ax/Bx corresponds to A1/B1, A2/B2 or B3/B3 short format
PRACH configuration can be different across frequency pairs.
domain occasions, but within a given a frequency domain For preamble format A, guard duration is not explicitly
occasion, the PRACH configuration will same across all time- defined. Since Ax doesn’t have a guard band, Bx (which has a
domain occasions. guard band) was appended to Ax, when Ax/Bx is configured.
But again, having no Bx would have acted as a guard band
B. DEFINITION OF PRACH SLOT NUMBER for the last time domain occasion of Ax, but Bx was chosen
Column ‘Slot Number’ in Table 6.3.3.2-4 of [3] is defined instead [31] of defining a guard period for the last time
in terms of 60kHz SCS for FR2. Essentially, the number of domain occasion of Ax. The CP sizes of Ax are larger than
PRACH slots is defined as the number of PRACH slots within Bx. Given this, for example, consider a scenario where, UE-1
a subframe for FR1 (µ = 15, 30 kHz) and the number of is configured to use A2 format and UE-2 is configured to
PRACH slots within a 60kHz slot for FR2 (µ = 60, 120 kHz) use B2 format. Then if UE1 propagation delay is greater than
as shown in Figure 11. Therefore, the column ‘Slot Number’ B2 CP, then it may interfere with UE-2s PRACH (if UE-2 is
in Table 6.3.3.2-4 of [3] is always between 0 to 39 for FR2. close to gNB). The medium access control (MAC) scheduler
For a given PRACH config, when the number of PRACH should take care to avoid this scenario. Further, PRACH is
slots within a 60kHz is specified as 1 in Table 6.3.3.2-4 of [3] supposed to work at low SNR during interference, so UE-2
and when PRACH subcarrier spacing is 120kHz (with 2 slots performance may not be significantly impacted (see [30]).
in a 60kHz), PRACH is always transmitted on the 2nd slot One benefit of having PRACH preamble start and end at
of 120kHz SCS (see section 5.3.2 of [3]). For the example the user data OFDM symbols (as shown in Figure 12) is the
PRACH configuration in Table 1, within the allowable slots negligible inter-carrier interference (that too for the first and
numbers 24,29,34 and 39 (or slots {48,49}, {58,59}, {68,69} the last OFDM symbols) if the preamble is delayed. Note that
and {78,79} in terms of 120kHz SCS), PRACH can be trans- format C2 has the largest guard period and CP size across
mitted only in the 2nd slot corresponding to 120kHz numerol- all formats as it was intended for larger cell coverage (see
ogy, i.e., slots 49, 59, 69 and 79 slots of 120kHz. For a given cell dimensioning in section 4.C.2. Also, RAN4 performance
preamble type, corresponding to a certain preamble band- evaluations will be only for formats are A1, A2, A3, B4, C0
width, the overall available time/frequency PRACH resource and C2, see [6] and [23]. Coverages shown in Table 2 of
within a cell can now be described as follows (see illustration section 4.C, shows that support of formats A1, A2, A3 and
in Figure 10). Within a given frequency domain PRACH A3/B3 may suffice for small cells. Support of mixed format
occasion, there can be multiple TD occasions defined by the of format A and format B was agreed, in order to avoid the
parameter NtRAslot given in Table 6.3.3.2-4 in [3]. The duration inter-symbol interference to the following uplink/downlink
of each TD occasion is given by the parameter Ndur RA and data receiving/transmitting at gNB/UE due to the zero-guard
is in terms of PUSCH symbols. Within each TD occasion time of preamble format A (see [31]) and may need to be
the configured PRACH preamble is present, including CP, supported for a small cell scenario.
sequence(s) and guard (if any) of durations TCP , TSEQ and
TGT respectively, as shown in Figure 10. Note that all TD C. PRACH CHANNEL DESIGN
occasions are continuous in time (i.e., no gaps between TD As discussed in section 2.2.2.1, NR PRACH short sequence
occasions). Alignment of PRACH symbols for different for- formats are defined with different CP durations, and sequence
mats with OFDM symbol boundaries is shown in Figure 12. repetitions to cater for different size cells and PRACH capac-
For a given PRACH format/configuration, the time domain ity. In this section, brief details are provided on PRACH
occasion can start at different symbols as specified by the preamble generation/properties and cell sizes/cyclic-shift
‘‘starting symbol’’ (or l0 in section 5.3.2 of [3]) column of dimensioning. Understanding PRACH preambles genera-
Table 6.3.3.2-4 in [3]. Symbol start period of 2 was chosen to tion/properties will be useful towards understanding the

VOLUME 8, 2020 136627

 A. Chakrapani: On the Design Details of SS/PBCH, Signal Generation and PRACH in 5G-NR

FIGURE 12. Short preamble format timelines in 5G NR.

PRACH receiver detection/estimation. Understanding cell (i.e., with low or no Doppler), the cyclic shift is given by,
sizes/cyclic-shift dimensioning for all PRACH formats will   
LRA
be useful towards analysing receiver processing complexity CS v = 0, 1, . . . , −1,NCS 6= 0
vN
for any given PRACH format and defining gNB requirements. Cv = NCS (11)
0 NCS = 0

The basic PRACH preamble sequences (of duration TSEQ
and number of tones LRA ) are Zadoff-Chu (ZC) sequences The parameter Ncs is configured by the gNB which is
generated using a given root sequence and a given cyclic derived from the higher layer parameter zeroCorrelation-
shift. A gNB configures the set of preamble sequences (with ZoneConfig using Table 6.3.3.1-7 of [3] for FR2. The NCS
a maximum of up to 64 preambles) the UE can use, and UE parameter specifies how many cyclic shifts can be used
will select one preamble (randomly out of the 64 preambles) within a given root sequence. Using the start physical root
for transmitting msg1 (or PRACH preamble). sequence and NCS , the UE generates 64 ZC sequences (or
A PRACH receiver objective is to detect the transmit- preambles) for each time-frequency PRACH occasion, enu-
ted preamble at the correct root sequence and cyclic shift merated in first increasing order of cyclic shifts Cv of a logical
and estimate the timing offset (propagation delay) for the root sequence, and then in increasing order of the logical
given UE. The generation and transmission of preamble root sequence index (section 6.3.3.1 of [3]). A frequency
is briefly described below. A ZC sequence can be gener- domain (FD) version of the sequence is generated as yu,v (n) =
ated in time-domain (TD), using a physical root sequence PLRA −1 −j 2πmn
LRA
−j πun(n+1) m=0 xu,v (m) · e . All 64 preambles can be generated
u as xu (n) = e LRA , n= 0, 1, . . . ,LRA − 1, where using a single root sequence or using multiple root sequences
LRA is the sequence length with u∈ {0, 1, . . . ,LRA −1}. For depending on the sequence length LRA and the cyclic shift
short sequence preamble formats LRA = 139. The gNB con- parameter NCS . An example for each scenario is illustrated in
figures the UE to use a start logical root sequence i Figure 13 and Figure 14 for LRA = 139 (short sequence).
(higher layer parameter prach-RootSequenceIndex), which
is used to derive the start physical root sequence u, using 1) ZC PROPERTIES FOR PREAMBLE DETECTION
Table 6.3.3.1-4 of [3] for FR2. A cyclic shifted version (with Specifically, two properties of the ZC sequences (see chap-
cyclic shift value of Cv ) of the ZC-TD signal can be obtained ter 7 of [20] for all properties of ZC sequences) are use-
as xu,v (n) = xu ((n + Cv ) mod LRA ). For unrestricted set ful at the receiver for detection and timing estimation

136628 VOLUME 8, 2020

A. Chakrapani: On the Design Details of SS/PBCH, Signal Generation and PRACH in 5G-NR

where the number of available shifts depends on the maxi-

mum timing uncertainty to be considered (depends on cell
size). In general, for small cell sizes a relatively large number
of cyclic shifts are available compared to larger cell sizes.
Table 2 shows the maximum cell radius for different formats
of both long and short-sequence length PRACH preambles
for all values of µ. The supported cell size is determined by
the minimum of the zero-correlation zone length NCS and
the guard period length NGP . For low Doppler scenarios, all
possible cyclic shifts are allowed and are labelled as Unre-
stricted Sets. In the case of high Doppler scenarios, some of
the cyclic shifts are not allowed. This is because of a property
of ZC sequences according to which energy transmitted on a
given cyclic shift will appear at other cyclic shifts depending
upon the root sequence and the sequence length. These cyclic
shifts are not considered in forming the preamble sets and
are labelled as ‘‘Restricted Sets’’. For detection of preambles
FIGURE 13. Generating preambles from a single root sequence for LRA = from restricted sets, the correlation energy of these ‘‘co-cyclic
139 (short sequence). shifts’’ will also need to be accumulated. Restricted sets are
not applicable for frequency range 2, i.e., FR2 in 5G-NR.
The parameter NCS indicates the width of the
(for unrestricted sets). First, the normalized cross-correlation ‘‘zero-correlation zone’’, thus providing larger or smaller
between two unique ZC sequences (i.e., ZC sequences gen- ‘‘zones’’ in terms of timing error for which orthogonality
erated with two different root sequences) will be low and (i.e., zero correlation) is retained. The set of cyclic shifts
equal to √ 1 . Second property is that the cyclic shifts of a for a sequence are partitioned into groups of NCS shifts per
L RA
ZC sequence (i.e., generated with a given root sequence) are preamble index. Essentially, the gNB receiver looks for a
orthogonal (i.e., zero cross-correlation) to each other. Note strong correlation with one of the root sequences. Location
that the orthogonality between cyclic shifts is retained at the of the spike (i.e., which cyclic shift) determines the preamble
receiver side only if the relative cyclic shift between two index (within which NCS group it shows up in) and the
sequences is larger than any difference in their respective timing (where in the NCS group is shows up). Note that the
received timing. Therefore, in practice only a subset of the gNB is aware of the start root sequence and number of root
cyclic shifts can be used to generate orthogonal preambles, sequences used (depending on the parameter NCS gNB has

FIGURE 14. Generating preambles from different root sequences for LRA = 139 (short sequence).

VOLUME 8, 2020 136629

 A. Chakrapani: On the Design Details of SS/PBCH, Signal Generation and PRACH in 5G-NR

TABLE 2. NR PRACH Cell dimensioning.

configured) and hence needs to do sequence matching (or The maximum cell radius, number of cyclic shifts and root
correlation) for only the configured roots sequences. At the sequences required for all NR PRACH formats are shown in
receiver, the received signal is sequence matched with a Table 2 when un-restricted sets are used. Preamble indices
set of ZC-sequences starting from the starting root (and up not used for Contention based Random Access (CBRA) in
to number of roots required) and typically filtered using a PRACH occasion can be reserved for Contention Free
an IFFT for preamble detection and timing estimation (see Random Access (CFRA) as in LTE, so the overall preambles
section 5 and Figure 5.1 for more details). The IFFT filtering to be processed during a PRACH occasion could be less
operation will cover for all the cyclic shifts used within the than the maximum indicated in Table 2. Also, the quantized
given root sequence, i.e., preamble transmitted with a given Ncs values used with restricted sets are different (see [3])
root sequence and any cyclic shift within can be detected and therefore the maximum radius achievable for different
with a single IFFT. Therefore, the receiver complexity scales PRACH formats will be different for restricted sets com-
only with the number of root sequences used. The details on pared to the ones indicated in Table 2. The timing offset
receiver complexity is provided in section 6.A. estimate error tolerance is related to the SCS configured for
the PRACH channel [36]. While selecting the proper PRACH
2) CELL DIMENSIONING format and Ncs , the maximum timing offset allowable should
The cell radius, number of cyclic shifts and number of root be considered. The largest time offset 1τ , due to propagation
sequences required to generate 64 preambles can be com- delay cannot be larger than the length of the detection window
puted as below. Parameters Tcp , Tseq and path profile val- corresponding to the given v and Ncs (see [36]). That is,
ues (in samples) in Table 2 are assuming 15kHz SCS (see 
1

1

RA,µ
Table 6.3.3.1 − 2 of [3] and scaled accordingly for different 1τ < min TCP − Tdet , · Tdet − τd
2 2
SCS. Maximum radius in Table 2 for different SCS is com-
puted as follows. First, compute NCS as follows. Step 1: First where Tdet = LNRAcs
∗ 2048 ∗ κ ∗ 2−µ ∗ Tc . The timing offset
RA
NCp
compute N0CS = (2048) × LRA . Using 0 ,
NCS find the quantized ranges for different formats can be then be calculated.
value of NCS using Table 6.3.3.1 − 7 of [3] for F R2. E.g.,
for A1 format, NCS0 will be (288/2048)∗ 139 = 19.54 which D. RESOURCE MAPPING FOR PRACH
(p,µ)
will be quantized to 19.jStepk2: Compute number of cyclic The time-continuous signal sl (t) on antenna port p for
shifts required as Cv = NLRA . E.g., for A1 format, this will PRACH is defined (see [3]) by,
j k CS
(p,µ)
XLRA −1 (p,RA)
be Cv = 139 19 = 7. sl (t) = ak
k=0
Step 3: Compute the number l of m root sequences required

−
  
64 j2π k+K k1 + k 1fRA t−NCP,l
RA T −t RA
c start
to generate 64 preambles as, Cv . E.g., for A1 format, this ×e (12)
l m
will be 647 = 10. Step 4: The cell radius (see [20]) can The terms in the above equations are defined in section 5.3.2
be calculated as Cell Radius = 1fNRACSL − 2τµd × 300
 
of [3]. The PRACH resource mapping parameters of interest
2 (in
RA
1f
meters). Here, τd is the delay spread equal to the PUSCH CP are k1 and k̄. The factor K = 1f RA
, is required as k1 is
size (144 Ts or 4.69 µs@15kHz). defined in multiples of 1f . In this section we will discuss only

136630 VOLUME 8, 2020

A. Chakrapani: On the Design Details of SS/PBCH, Signal Generation and PRACH in 5G-NR

FIGURE 15. Location of PRACH transmission occasion within the resource grid.

the relevant terms used in PRACH signal resource mapping, (in Figure 16) are as follows, LRA = 839 with 1fRA =
which are required for correct tone extraction at gNB receiver. 1.25 kHz and 1f = 15kHz. PRACH bandwidth is then
839 × 1fRA = 1048750 Hz . 6 RBs at 1f = 6 × 12 ×
E. DETAILS ON THE SUB-CARRIER OFFSET k1 1f = 1.08MHz. Unused bandwidth is then 31250Hz (or
The term k1 (is in multiples of 1f ) is used to locate the 25 tones @1fRA ). Note that the PRACH resource offset k̄
lowest subcarrier of the lowest PRB of the PRACH trans- starts from the middle of the PUSCH subcarrier (i.e., 1f 2 )
mission occasion in frequency domain (nRA ) with respect to as in LTE. Therefore, 7.5kHz should be subtracted from
the start of the corresponding resource grid, as illustrated in 31250Hz on either side, which leaves 13 unused tones @1fRA
Figure 15 (also see [15]). k1 is given by (section 5.3.2 of [3]), ( = 16.25kHz). Out of these 7 tones are not used at the
µ start − N start,µ )N RB + nstart N RB + start, which is represented by the parameter k̄ and 6 tones
k1 = ko + (NBWP,i grid sc RA sc
size,µ RB are not used at the end as shown in Figure 16. PRACH
RA N RB − grid N Nsc µ
nRA NRB sc 2 . The parameter k0 is the sub- resource referencing is from the mid of the PUSCH sub-
carrier offset from the middle subcarrier of numerology carrier (i.e., from 1f
2 ) as in LTE (see [20]). Calculations
µ to the carrier center frequency f0 and as described in for short preamble (i.e., LRA = 139) with 1fRA = 120kHz and
µ size,µ
RB /2 represents 1f = 120kHz (see Table 3 above) are as follows. PRACH
section III-B. Here, the term k0 − Ngrid Nsc
bandwidth is LRA ×1fRA = 139 × 120×103 = 16.68MHz In
the start of the resource grid; the term nstart RB
RA Nsc repre- terms of PUSCH RBs (12 RBs from Table 3) this is
sents the start of the lowest PRACH transmission as a fre- RA × 1f ×12 = 17.28MHz. Guard tones = 600kHz, which
NRB
quency offset with respect to the start of the active uplink
RA N RB represents the start of the cur- is 5 tones @120kHz. With k̄ = 2 for this case, 2 zero tones
BWP; the term nRA NRB sc are in the beginning to PRACH start (i.e., with k̄ = 2 in the
rent (or target) PRACH transmission, and is the frequency
Figure 16) and 2 zero tones are at the end of PRACH (with
offset from the start of the lowest (or first) PRACH trans- 1f
mission; and the term NBWP,i start N RB is the frequency offset
sc 2 guard, there will be 2 tones at 1fRA =120kHz).
between Point A and the start of the active uplink bandwidth.
start,µ RB
Here, (N start
BWP,i − Ngrid )N sc is the frequency offset between
V. IMPORTANT ASPECTS OF PRACH RECEIVER DESIGN
resource grid and active uplink bandwidth part. Note that 1f In this section, all important aspects of PRACH receiver
is the subcarrier spacing of the initial active uplink bandwidth design will be discussed. Specifically, aspects which need to
part during initial access (for CBRA). Otherwise (for CFRA), be addressed while designing a PRACH receiver in 5G-NR
1f is the subcarrier spacing of the active uplink bandwidth are presented along with some real system use cases.
part. Also, nstart
RA cannot point to PRACH resources outside of
the active BW (see [33]). A. PRACH RECEIVER FLOW
A typical non-coherent PRACH receiver processing chain is
F. DETAILS ON THE SUB-CARRIER OFFSET k̄ shown in Figure 17 The data CP is removed from the received
k̄ (defined in table 6.3.3.2-1 of [3]) is the subcarrier offset time domain signal and NFFT point FFT is taken to obtain
(in multiples of 1fRA ) from PUSCH RB corresponding to the the frequency domain (FD) samples. The FD samples are
PRACH frequency domain occasion index nRA to the actual correlated with the configured root sequences. Since the UE’s
random-access subcarrier. This is illustrated in Figure 16 would have used random preamble indices for transmission,
(see [16]) with an example. The details of the below example several hypotheses, starting from the start root sequence and

VOLUME 8, 2020 136631

 A. Chakrapani: On the Design Details of SS/PBCH, Signal Generation and PRACH in 5G-NR

FIGURE 16. k̄ is offset to RA subcarrier from PUSCH subcarrier.

FIGURE 17. A typical non-coherent PRACH receiver processing chain [37].

up to number of root sequences configured will be used complexity here is due to use of an IFFT being used per
for detection purposes at gNB. Note that correlation in FD root/symbol/antenna/TD/FD occasion which could prove to
corresponds to complex multiplication of received FD sam- be a bottleneck for most systems. Coherent detection offers
ples with the root sequences. Cross correlation is followed lower complexity and better detection gains but will require
by a time domain energy analysis (using an IFFT) to detect combining of the received PRACH symbols (and/or anten-
correlation peaks and their exact location within the detection nas) before correlation and time domain analysis. Coherent
window. With a non-coherent PRACH detector, for each of combining of PRACH symbols will require a few additional
the received antennas and symbols, the received preamble processing steps before correlation and time domain analysis
can be processed independently, and time domain energies can be performed. Specifically, two additional steps will be
can be combined before peak detection. However, the per- required before coherently combining the received PRACH
formance and complexity of the non-coherent detector may symbols. First, the common phase correction which is typi-
not be desirable for PRACH formats with higher repetitions cally applied in the DFE for data channels, must be undone for
which are expected to operate in low SNR regimes. The PRACH processing. Second, the timing offset introduced due

136632 VOLUME 8, 2020

A. Chakrapani: On the Design Details of SS/PBCH, Signal Generation and PRACH in 5G-NR

TABLE 3. Subcarrier offset for PRACH resources. center frequency of the gNB in Hz. The phase offset for
PRACH signal due to the common phase correction should
be undone before coherent combining of the PRACH sym-
bols (repetitions). The phase de-rotation term can be com-
puted as follows. Let the PRACH start symbol (column 6
of Table 6.3.3.2-4 in [3]) be denoted as pstart ∈ {0, . . . 14}
and the number of PRACH repetitions as Nrep PRACH =
Nu
2048κ·2−µ
(with Nu defined in Table 6.3.3.1-2 of [3]). For the
PRACH symbol p ∈ {pstart .., pstart + Nrep PRACH − 1}, the
p,s
accumulated phase is tphase from the reference symbol (i.e.,
µ
symbol 0 of slot with mod n
s , 2µ == 0)) to symbol


µ
p for slot s = mod ns , 2µ . In slot s, for the jth FFT
output and PRACH symbol  p, the phase de-rotation term
gNB p,s
−2iπf t
to data CP removal must be compensated before combining. will be eiθ p =e 0 phase
. Since the phase term repeats
µ
The details associated with these processing steps are defined every 1ms (see derivation in [26]), tstart,l will repeat every 2µ
in the following sections. or 8 slots. The reference data symbol for phase compensation
will be symbol 0 of the slots where phase
term repeats, i.e.,
µ
B. TONE EXTRACTION data symbol 0 of slots with mod ns , 2µ == 0. Adding slot
µ µ,s
The first step though is to extract the required PRACH tones notation to tstart,l as tstart,l to denote the accumulated phase in
l,s
for processing, using all the sub-carrier offsets used in the slot s, the accumulated phase tphase from the reference symbol
OFDM signal generation. Consider the case when 1fRA = µ
to symbol l for slot s = mod ns , 2µ , is given by,


1fdata and where a common wideband FFT is used. The first
starting tone index for the nth
RA frequency occasion is given
l,s
tphase
by (see Figure 16), ( µ,s−1 µ,s µ µ
t +Tc (N cp,l=0 +(l − 1)×(N u +N cp,l )), l > 0
nRA
NtoneIndex = start µ,s−1 µ,s
tstart +Tc N cp,l , l =0
Guard tones
z }|  { (14)
size RB
NFFT − Ngrid NSC µ,s−1 µ
= − k0mu with tstart,l = 0 for slots with s = 0 and l = 0. Here, Ncp,l
2 is the number of samples in data CP for data symbol l Note
µ
z }|
tone offset to nRA freq.occasion
{ that Ncp,l=0 will be longer by 16κ, for slots s = 0 or 2µ−1 .

start,µ
start
+ NBWP,i −Ngrid RB
Nsc +nstart
RA scN RB
+ n N RA RB
N
RA RB sc
D. TIMING COMPENSATION (DUE TO DATA CP REMOVAL)
subcarrier offset
z }| { One of the design items discussed in 3GPP during the
+ k . (13) PRACH preamble design in NR was to enable the use of
the common wideband FFT (at gNB) for all uplink chan-
Simplifying we get (see derivation of k1 and k̄ in nels (see [18], [34] and [35]) when the sub-carrier spac-
nRA
section IV-E and IV-F), NtoneIndex = NFFT
2 +Kk1 + k̄, where ing between data and access channels is same as shown in
Kk1 and k̄ are defined in sections IV-E and IV-F and NFFT is Figure 18. This has a consequence on the PRACH receiver,
the wideband FFT size. Here, the factor K is required because which may have to account for the data CP removal before
k1 will be in units of 1f (see section IV-E). Starting from the common wideband FFT, especially when a coherent com-
nRA
NtoneIndex , LRA number of tones can be extracted for PRACH bining of symbols is being considered. The signal within
receiver processing. each FFT is a cyclic shifted version of the PRACH preamble
sequence and can be easily rotated. However, in case of long
C. UNDO COMMON PHASE CORRECTION
formats, a separate DFE chain needs to be present (to cater
During the up-conversion to the carrier frequency fUE to 1.25kHz and 5kHz subcarrier spacing) and a common
(or f0BWP , see section III-B), there is no common phase cor- wideband FFT cannot be used. The removal of data CP time
rection term applied at the UE for PRACH transmission (see samples results in a phase ramp on the frequency domain
section 5.4 of [3]). However, due to the use of common samples of PRACH and needs to be compensated for in the
DFE chain between PRACH and data channels, common PRACH receiver if coherent combining across symbols is
phase correction which is done for data channels (typically required. For a PRACH preamble with CP and sequence
within DFE) should be undone for PRACH processing if a being repeated several times, each OFDM symbol acts as
coherent detector is being used. This phase per-symbol √ for a cyclic prefix for the next OFDM symbol. However, the
OFDM symbolindex l is given by e−iθl , where, i = −1,
gNB µ µ

gNB
OFDM symbols which is repeated has much smaller length
θl = 2πf0 tstart,l + TCP,l , where f0 is the receiver as compared to LTE PRACH and equals the same length

VOLUME 8, 2020 136633

 A. Chakrapani: On the Design Details of SS/PBCH, Signal Generation and PRACH in 5G-NR

as adjacent user data OFDM symbols. The received signal minimum resolution of timing advancement, therefore the
within each FFT window will then act on a cyclic shifted peak detection index should be appropriately quantized. From
version of the PRACH preamble sequence [18]. section 4.2 of [7],
 the initial time alignment value NTA =
With the use of wideband FFT for PRACH, we see that TA ·16 × 64 × Tc 2µ , with index values of TA = 0, 1, 2, . . .,
each extracted PRACH symbol consists of two parts of sam- 3846. The lowest   TA = 1) for µ = 3 is
resolution (with
ples coming from adjacent two repetitions (see Figure 18). then 1 × 16× 64
8 × 1
480∗1e3∗4096 ≈ 65ns, which is basically
The timing offset compensation due to this operation needs to 0.52µs
2µ in terms of LTE resolution. Note that the detection
be done for proper PRACH detection/timing estimation. The
metric must be normalized with noise and compared with a
timing offset in the time domain is equivalent to the phase
threshold for a correct detection. The detection threshold for
ramping in the frequency domain. After tone extraction, the
each format and Ncs can be determined by using the method
phase ramping will be implemented for each of the symbols.
described in [42]. Noise estimation can be based on unused
After tone extraction, denote the frequency domain tone as
roots or cyclic shifts. It can also be made based on the IFFT
Xir [k] , 0 ≤k≤LRA − 1, 0 ≤l ≤ N for each symbol l, and
energy. In all estimations however, the estimated noise vari-
each received antenna r = {1, · · · N rx }. Here, N is the number
ance has a signal energy component bias, which seems to be
of repetitions of a PRACH sequence for a given PRACH
Nu the negligible while using unused roots. For all the parameters
format and given by N = 2048κ·2 −µ . Denote the phase
discussed so far, an example use case is as follows. For a
ramping factor as ρj for the jth FFT tick. The phase ramping
mmW system with BW of 100MHz and sub-carrier spacing
should start from the tone 0 of 1024 tones since coherent
of 120kHz for both PRACH and data channels, the common
combing will be done subsequently. Since PRACH is not
wide-band FFT size would be 1ffdata s
= 122.88Msps
120kHz = 1024.
located in the tone 0, the initial phase offset needs to take the
The IFFT size for PRACH detection and timing analysis
initial PRACH tone location into account. The initial phase
i2πk2 ρj i2π (n+k2 )ρj could be Nifft = 192 leading to a timing estimate resolution
is e NFFT . The phase ramping is X̂ir [n] = Xir [n]e NFFT , of ≈ 43.4ns. The timing estimate windows can be split into
where k2 = Kk 1 + k̄ + NFFT
2 with K , k̄ and k1 defined earlier. positive and negative windows (in some proportion) around a
The phase ramping factor can be computed as, given centre sample to be able to detect positive and negative
timing offsets respectively, depending on gNB requirements.
µ µ
  
ρj = mod NCP RA
− NCP,l − (j − 1) ∗ NCP,l>0 , NFFT .
(15) VI. PRACH RECEIVER IMPLEMENTATION ASPECTS
A. PROCESSING COMPLEXITY
Note that for formats Bx, which have guard bands at the The IFFT module used for time domain analysis in the
end of the PRACH preamble, some of the PRACH symbols PRACH receiver often dictates the receiver complexity. This
will be aligned to left of the PUSCH symbol boundaries is because, IFFT is typically implemented in hardware and
and some symbols are aligned to the right of the PUSCH while processing multiple symbols (combined or not) per
symbol boundaries (see Figure 12). The PRACH symbols antenna, per hypothesis (i.e., a given root), per TD/FD occa-
which
 are alignedto the left of the PUSCH symbol will result sion with a single IFFT engine could be prove limiting. One
RA − N µ µ
in NCP CP,l − (j − 1) ∗ NCP,l>0 < 0 and the phase can indeed limit the number of roots supported as a gNB
ramping factor will need to be circularly shifted by NFFT capability. Note that some formats may require roots greater
for coherent combining of symbols. Due to the possible fre- than what can be supported by gNB to derive 64 preambles (or
quency offset, it might not be possible to coherently combine whatever is configured by gNB) to obtain maximum coverage
all the repetitions without degrading performance. Let ξ is the offered by the format. In those scenarios, gNB will simply
residual frequency offset between receiver and transmitter, support up to max roots possible by gNB and Ncs should
i.e., residual frequency offset is the remaining frequency be configured appropriately (based on reduced coverage) so
offset after the downlink synchronization has happened at that 64 preambles (or the number of preambles to use as
the UE. In [23], for FR2, ξ = 4000Hz was assumed during configured by gNB) can be derived with the number root
simulations and the phase

rotation over 1 PRACH symbol will sequences which can supported.
µ ◦
be 2π × ξ × Ts × Nu = 6 . This angular rotation seems Coherent combining provides a better implementation
small enough to combined adjacent symbols. complexity with performance gains as the number of roots
to be processed is lower. However, there could be a limit
E. TIMING OFFSET ESTIMATION on symbols which can be coherently combined without
The IFFT size (≥ LRA ) determines the timing
j offsetk resolution degrading performance as noted previously (due to frequency
Nifft offset). Combining across the antennas before detection could
of the PRACH receiver. There will be LRA samples in
time domain to analyse for a jpeakkper root. The number of also be considered but will require precoding coefficients
Nifft information associated with the antennas. If precoding infor-
windows to analyse will be NCS . For 1fRA = 120kHz
mation is not available, coherent combining across anten-
and LRA = 139, the timing resolution with Nifft = 144 will be nas can be done in time domain after IFFT (as shown
1
120kHz
144 ≈ 57.87ns. This has to converted to some multiple of in Figure 17).

136634 VOLUME 8, 2020

A. Chakrapani: On the Design Details of SS/PBCH, Signal Generation and PRACH in 5G-NR

FIGURE 18. Common wideband FFT for PRACH with A1 short format and Data.

B. WIDEBAND FFT USAGE has multiple repetitions symbols to boost SNR, it may be
A note on the wideband FFT usage when 1fDATA 6 = 1fRA . For possible to tolerate such signal corruptions due to common
a given bandwidth and sampling rate, FFT size is computed wideband FFT usage. Alternatively, CP size worth 4.7µs
fs could be dropped before computing the FFT. But, 2.35µs
as NFFT = 1f , which for 100MHz bandwidth is 1024 for
120kHz SCS, 2048 for 60kHz SCS and so on. When worth of symbol information of higher SCS (PUSCH) will
1fDATA 6 = 1fRA and a narrowband chain is not available, FFT also be removed, which can significantly degrade uplink
size can be computed based on 1f = min(1fDATA , 1fRA ) throughput performance. This may not be acceptable. Also,
so that FFT grid is aligned to the smaller sub-carrier spacing with a common FFT approach, processing of PUSCH will
(see Figure 5 in section II-B for different tone alignments). incur extra delay of 1 symbol which may not be desirable.
However, when 1fDATA 6 = 1fRA , CP removal before wide- This is because we are buffering for at least 2.35µs+66.66µs
band FFT would introduce performance degradation on data before firing off the FFT. Whereas, PUSCH 1st symbol was
or PRACH. One way to mitigate the performance loss could already available by 2.35µs+33.33µs mark. Given the above
be as follows. While using a common FFT with resolution issues, it might be desirable to have separate DFE chains for
based on lower SCS, FFT window is from end of CP of higher PRACH and data.
SCS (longer CP) to end of symbol duration of lower SCS.
For example, consider fs = 30.72Msps for a 20MHz BW C. PRACH PROCESSING WITH DELAYS GREATER THAN
with 1fRA = 15kHz and 1fDATA = 30kHz. Assume, PRACH SYMBOL DURATION
and data are multiplexed in frequency. Then the normal CP Delays up to the length of one PRACH OFDM symbols can
size (in microseconds) of higher SCS is 4.7µs 2 = 2.35µs be detected by frequency domain matched filters as outlined
and symbol duration = 33.33 µs. For lower SCS, CP and in Figure 19 [37]. However, this receiver structure results
symbol sizes are 4.7µs and 66.66µs respectively. The FFT in a delay ambiguity when the delay exceeds the length of
size required if only PRACH was present is 30.72Msps15kHz = the PRACH OFDM symbol (i.e., > Nu ). This can happen,
2048. FFT size if only data was present is 30.72Msps
30kHz = 1024. for example, while using C2 format with start symbol 0 in
Apply a 2048-point FFT after removing 2.35µs worth of CP slots with longer CP duration. A modified PRACH preamble
samples on 66.66µs worth time samples. Basically, the FFT detector is then required. An example of a PRACH preamble
captures 1 symbol of 15kHz and 2 symbols of 30kHz. But detector for handling large delays (see [35]) is illustrated in
with this approach, for PRACH, 2.35µs of CP portion has Figure 19. Here, a detector is included which compares the
also been captured. These time samples (before FFT) may received signal in first and last FFT time windows resulting in
have corruptions based on multi-path fading which may affect a decision if delays are smaller or larger than the length of the
some portion of the spectrum (depends on the frequency PRACH OFDM symbol. This decision can then be combined
components of that corruption). with a delay estimator as in Figure 5-1 (see [35] and [37])
If this multi-path is from a previous PRACH signal, the resulting in a timing estimate with high time resolution and
corruption will likely impact PRACH channel. Further, note support possible propagation delays larger than one PRACH
that 2.5us worth of PRACH symbol end was cut off, because OFDM symbol. Such a large delay estimator may be required
of beginning the FFT earlier than 4.7us. The FFT window for when PRACH format C2 needs to be configured. See [35]
PRACH is basically mis-aligned. Since most PRACH formats and [36] for methods on resolving detection ambiguity.

VOLUME 8, 2020 136635

 A. Chakrapani: On the Design Details of SS/PBCH, Signal Generation and PRACH in 5G-NR

FIGURE 19. Small/Large delay PRACH preamble detector outline (see [35]).

TABLE 4. RAN4 performance requirements from [6] and [39] and Delay Line (TDL) channel with delay spread of 30 ns and
comparison with two different network operators reported performance.
maximum Doppler of 300Hz. The probability of correct
detection is defined in section 11.4.2.2.1 of [6] and the min-
imum requirements are listed in section 11.4.2.2.2 of [6].
PRACH test preambles used for RAN4 performance study
are defined in annex A.6 of [6] and delay profiles for each
of the channel model is defined in annex G.2.1.1 of [6]. The
minimum SNR (in dB) listed in Table 4 is the SNR at which
1% missed detection was achieved, timing error tolerance was
less than the values specified in Table 11.4.2.2-1 of [6] and the
false alarm was less than 0.1%. Table 4 also lists RAN4 per-
formance reported from Nokia and Ericsson where margins
≥ 3dB are seen for most scenarios. A simulation study of the
PRACH receiver outlined in this paper, revealed that perfor-
mance margins were comparable to [40]. Interestingly, these
margins could be achieved only with a PRACH receiver using
coherent combining of PRACH symbols before detection.

VII. CONCLUSION
In this paper, downlink and uplink initial access channels in
5G-NR was studied in detail. With the 5G-NR 3GPP specifi-
cations details being quite dense getting clear information can
prove to be a tedious task. Several design details associated
with the physical layer procedures of initial access channels,
which are skipped in the 3GPP 5G NR specifications were
explained in detail in this paper. Three important contribu-
tions were made in paper. First, the design details and jus-
D. RAN4 PERFORMANCE REQUIREMENTS tifications associated with both downlink and uplink access
Radio Performance and Protocol aspect requirements for base channels were discussed. Aspects of signal generation and
station [6] are specified by the 3GPP Radio Access Network phase correction were presented with mathematical analysis.
working group 4, also known as RAN4 and can be used Secondly, receiver design aspects of NR PRACH were dis-
to bench mark the performance of a PRACH receiver. The cussed in detail with illustrating examples. Lastly, implemen-
performance recommendations made by RAN4 committee tation aspects of a PRACH receiver and its performance com-
for both base-station performance [6] and base-station con- parisons with 3GPP defined Radio Performance and Protocol
formance testing [39] are shown in Table 4 for FR2. The aspect requirements or RAN4 requirements were presented
performance recommendations are for AWGN and Tapped for millimeter wave access. The work in this paper will be of

136636 VOLUME 8, 2020

A. Chakrapani: On the Design Details of SS/PBCH, Signal Generation and PRACH in 5G-NR

significant value to researchers and system design engineers [22] E-UTRA Random Access Preamble Design, document R1-060998, 3GPP
looking to design efficient initial access algorithms within the Contribution., Ericsson, Mar. 2006.
[23] Simulation Results for PRACH Requirements, document R4-1815361,
framework of 5G-NR systems. 3GPP Contribution, Ericsson, Nov. 2018.
[24] Remaining Details on PRACH Formats, document R1-17120277, 3GPP
ACKNOWLEDGMENT Contribution, Samsung, Dec. 2017.
[25] Remaining Details on PRACH Formats, document R1-1802945, 3GPP
The author would like to thank R. Bachu, for his inputs on Contribution, Ericsson, Mar. 2018.
system design aspects; L. Blessent and J. Li for supporting [26] NR OFDM Symbol Generation Analysis, document R1-1800296, 3GPP
the work in this paper. Contribution, Intel, Jan. 2018.
[27] On Numerology for NR PRACH, document R1-1711598, 3GPP Contribu-
tion, Samsung, Jun. 2017.
REFERENCES [28] NR BS PRACH Demodulation, document R4-1812592, 3GPP, Contribu-
[1] MWC. (Mar. 2018). Small Cells Taking Cellular to New Heights tion, NTT Docomo, Oct. 2018.
With IoT Tech and Global Deployments. [Online]. Available: [29] NR-PRACH Preamble Format Details for Capacity Enhancement and
https://www.qualcomm.com/news/onq/2018/03/16/mwc-2018-small- Beam Management, document R1-1711381 and R1-1712152, 3GPP Con-
cells-taking-cellular-new-heights-iot-tech-and-global-deployments tribution, Ericsson and Huawei, Jun./Aug. 2017.
[2] J. Liu, K. Au, A. Maaref, J. Luo, H. Baligh, H. Tong, A. Chassaigne, and [30] 6.1.1.4.1—Remaining Details on PRACH Preamble Formats for
J. Lorca, ‘‘Initial access, mobility, and user-centric multi-beam operation in Long/Short Sequence Length, document R1-1714040, 3GPP Contribution,
5G new radio,’’ IEEE Commun. Mag., vol. 56, no. 3, pp. 35–41, Mar. 2018. Ericsson, Aug. 2017.
[3] Technical Specification Group Radio Access Network; Evolved Uni- [31] On Remaining Details of PRACH Formats and Designs, document
versal Terrestrial Radio Access; Physical Channels and Modulation, R1-1720624, 3GPP Contribution, InterDigital, Nov. 2017.
document TS 36.211 V14.1.0, 3GPP, Dec. 2016. [32] Remaining Details of PRACH Formats, document R1-1802945 and
[4] Technical Specification Group Radio Access Network; NR; NR and R1-1800417, 3GPP Contribution, Ericson and Samsung, Mar./Jan. 2018.
NG-RAN Overall Description, Stage 2, document TS 38.300 V15.4.0, [33] NR-PRACH: Capacity Enhancements and Beam Management, document
3GPP, Jun. 2018. R1-1712065, 3GPP Contribution, ZTE, Aug. 2017.
[5] General Aspect for UE Radio Frequency (RF) for NR, document TR [34] NR PRACH Preamble Design, document R1-1611904, 3GPP Contribution,
38.817-01 V2.0.0, 3GPP, Jun. 2018. Ericsson, Nov. 2016.
[6] Technical Specification Group Radio Access Network; NR; Base Sta- [35] NR PRACH Design, document R1-1700298, 3GPP Contribution, Ericsson,
tion (BS) Radio Transmission and Reception, document TS 36.104 Jan. 2017.
V16.2.0, 3GPP, Dec. 2019. [36] Further Discussions on NR PRACH Requirements, document R4,1812770,
[7] Technical Specification Group Radio Access Network; NR; Physical Layer 3GPP Contribution, Ericsson, Oct. 2018.
Procedures for Control, document TS 36.213 V15.3.0, 3GPP, Mar. 2018. [37] PRACH Preamble Design for Capacity Enhancement, document
[8] Technical Specification Group Radio Access Network; Evolved Univer- R1-1711146, 3GPP Contribution, Qualcomm, Jun. 2017.
sal Terrestrial Radio Access; and Evolved Universal Terrestrial Radio [38] NR PRACH Design, document R1-1702127, 3GPP Contribution, Ericsson,
Access Network, Overall Description, document TS 36.331 V14.2.0, Feb. 2017.
3GPP, Mar. 2017. [39] Technical Specification Group Radio Access Network; NR; Base Station
[9] Technical Specification Group Radio Access Network; NR; Physical Layer (BS) Conformance Testing Part 2: Radiated Conformance Testing, docu-
Procedures for Data, document TS 38.214, V15.3.0, 3GPP, Sep. 2018. ment TS 38.141-2, 3GPP, Dec. 2019.
[10] On DL PTRS Design, document R1-1705906, 3GPP Contribution, Erics- [40] NR PRACH Design, document R4-1906366, 3GPP Contribution, Nokia,
son, Apr. 2017. May 2019.
[11] Summary of 7.1.1.1 Synchronization Signal, document R1-1807862, 3GPP [41] NR PRACH Design, document R4-1815361, 3GPP Contribution, Ericsson,
Contribution, Ericsson, May 2018. Nov. 2018.
[12] NR Synchronization Signal Design, document R1-1708719, 3GPP Contri- [42] A. Chakrapani, ‘‘NB-IoT uplink receiver design and performance study,’’
bution, Ericsson, May 2017. IEEE Internet Things J., vol. 7, no. 3, pp. 2469–2482, Mar. 2020.
[13] On k0 for OFDM Signal Generation, document R1-1805891, 3GPP Con-
tribution, Huawei, May 2018.
[14] Correcting NR OFDM Symbol Generation, document R1721601, 3GPP
Contribution, Intel, Dec. 2017.
[15] Remaining Issues on OFDM Baseband Signal Generation, document R1-
1809105, 3GPP Contribution, Sharp, Aug. 2018.
[16] Ways to Mitigate Frequency Offset With CAZAC Cyclic Shift, document ARVIND CHAKRAPANI (Member, IEEE)
R1-070227, 3GPP Contribution, LG Electronics, Jan. 2007. received the Ph.D. degree in electrical engineering
[17] Limitation of PRACH Sequence Allocation for High Mobility Cell, docu- from the University of New South Wales, Syd-
ment R1-073624, 3GPP Contribution, Panasonic and NTT, Aug. 2007.
ney, NSW, Australia, in 2011. He is currently a
[18] NR PRACH Preamble Design, document R1-1611904, 3GPP Contribution,
Systems Engineer with Qualcomm Technologies
Ericsson, Nov. 2016.
[19] Cyclic-Shift Dimensioning and Capacity Shortfall Solution for Agreed NR Incorporated (formerly Qualcomm Flarion Tech-
PRACH Formats, document R1-1716686, 3GPP Contribution, Huawei, nologies), where he is involved in product oriented
Sep. 2017. research in the area of the IoT and 5G NR. His
[20] S. Sesia, I. Toufik, and M. Baker, LTE, The UMTS Long Term Evolution: past research interests included resource allocation
From Theory to Practice. Hoboken, NJ, USA: Wiley, Aug. 2011. in LTE MAC schedulers, receiver algorithms for
[21] E. Dahlman, S. Parkvall, and J. Skold, 5G NR: The Next Generation V2V, and NB-IoT. His current research interest includes developing efficient
Wireless Access Technology. Amsterdam, The Netherlands: Elsevier, Aug. physical layer receiver algorithms for 5G NR.
2018.

VOLUME 8, 2020 136637