LTE Initial Access (All about Cell Search, Cell Selection, PSS, SSS)

A UE must perform certain steps before it can receive or transmit data. These steps are-

– cell search

– cell selection

– Derivation of system information and

– Random access.

The complete procedure is known as LTE Initial Access.

Although there can be many algorithms defined for this procedure the basic sequence is as follows:

1.  A list of preferred earfcn is maintained in UE sim. UE measures RSSI  of each element of this list.
2. From the list of 1. all channels with RSSI > a threshold value is determined..
3. UE decode sync/reference signals and find Physical Cell Id of each candidate from 2.
4. Of all the cells of step 3. UE decode MIB and SIB. Now, UE has a list of frequency, PCI and PLMN of the filtered cells.
5. From all of the above information, UE makes a decision of on which of these cells it would camp on.

Now, let us get into details.

1. UE tune to every channel it supports and measure RSSI:-

In the equipment, operator stores all the frequency bands it supports. For each band, preferred earfcn values are also stored in order. Based on that information UE will scan all those channels and measure the RSSI. RSSI is Received Signal Strength Indicator. It is total signal power of each resource element including interference or noise. This is included here in measurement since UE currently does not have any information about network and measuring this parameter doesn’t require any channel coding process.

Suppose an enodeb corresponds to one operator, say airtel, working in band 3, earfcn 1400 and operating bandwidth 10 Mhz (1820 to 1830 Mhz). Another enodeb in same cell corresponds to different operator, say vodafone, in the band 3 with earfcn 1500 and operating bandwidth 10MHz(1830 to 1840). Now, suppose the UE supports band 3 (1805 to 1880 Mhz) which has a overall bandwidth 75Mhz and band 7 of overall bandwidth 70 MHz.. It starts RSSI scanning from band 3 on a 100khz raster basis so overall it will scan 750 Earfcn in Band 3.  The preferred earfcn list is orderly maintained with UE as- 1400, 1500…. This list can contain multiple earfcn for 1 operator but the list is maintained according to preference. Out of these, earfcn that have a RSSI value greater than a certain threshold value will be shortlisted and maintained in order. Earfcn would give the carrier frequency, according to formula-

F_DL = F_{DL_Low} + 0.1(N_DL – N_{offs_DL}) , for downlink

F_DL = F_{UL_Low} + 0.1(N_UL – N_{offs_UL}) , for uplink

F_{DL_Low  –} Lowest Dowlink frequency in a given band.

F_{UL_Low –} Lowest Dowlink frequency in a given band.

N_DL- Downlink EARFCN

N_UL- Uplink EARFCN

N_{offs_DL –} EARFCN DL Offset

N_{offs_DL –} EARFCN UL Offset

2. From the list of 1. all channels with RSSI > a threshold value is determined.

The thresholds are implementation dependent

3. UE decode sync/reference signals and find Physical Cell Id of each candidate from 2:-

There are 504 Physical Cell Identities(PCI). These identities are divided into 168 unique cell layer identity groups in the physical layer, in which each group consists of 3 physical layer identities.
This information is transmitted using two different signals. The two signals, carrying the physical layer identity and the physical layer cell identity group, are the primary and the secondary synchronization signals respectively

The UE first looks for the primary synchronization signal (PSS) which is transmitted in the last OFDM symbol of the first time slot of the first subframe. This enables the UE to acquire the slot boundary independently from the chosen cyclic prefix used in this cell. PSS is transmitted twice per radio frame, so it is repeated in subframe 5 (in time slot 11). This enables the UE to get time synchronized on a 5 ms basis, This was done because if a UE starts reading cell from between a subframe, it can get time synchronized.

In FDD, PSS is broadcast using the central 62 subcarriers.

The PSS is used to:
– Achieve time syncronization.
– Identify the center of the channel bandwidth in the frequency domain
– Finds which 1 of 3 Physical layer Cell Identities (PCI), cell belongs
PCI are organised into 168 groups of 3 so the Primary Synchronisation Signal identifies the position of the PCI within the group but does not identify the group itself.

Secondary Syncronization symbol (SSS) is located in the symbol before PSS, transmitted twice per subframe. The two transmissions of the SSS are different so the UE can detect which is the first and which is the second. This sequence alternates in even and odd subframes, ex In subframe1 sequence A is transmitted in symbol a and sequence A’ is transmitted in symbol b, then in subframe 2 sequence A is transmitted in symbol b and sequence A’ is transmitted in symbol a. This way subframe syncronization is achieved.

SSS is also transmitted in central 62 subcarriers in the symbol before PSS.

The SSS is used to:
– achieve radio frame synchronisation
– Find which 1 of 168 Physical layer Cell Identity (PCI) groups is used,
Hence,the PCI could be deduced when combined with the pointer from the PSS. Following formula is used:-

PCI  = 3*physical layer id group(from SSS)+phy layer cell id(from PSS).

4. Of all the cells of step 3. UE decode MIB and SIB. Now, UE has a list if frequency, PCI and PLMN of the filtered cells:-

Here, I am not explaining the structure of MIB and SIB. At this point UE knows the PLMN from SIB1 and can select the cell to camp on.

5. UE finds a suitable cell. A suitable cell is one that fulfills cell selection criteria.

Criteria for Cell Selection are:-

– Cell must transmit power strong enough to be detected by UE.

i.e  S_rxlev > 0
Where,
S_rxlev = Q_rxlevmeas – (Q_rxlevmin+Q_{rxlevminoffset}) – P_compensation

Q_rxlevmeas     = RSRP measured by UE in dBm
Q_rxlevmin       = min. required RSRP signalled within SIB1
Q_{rxlevminoffset}  = usually included in SIB1. If not included than value of zero is set.
P_compensation = MAX(P_EMAX – P_UMAX, 0)

These values are decoded from SIB.

– Cell must not be barred.

– PLMN saved in UE sim must match with cell’s PLMN.

There are 2 types of cell selection:-

– Initial cell selection,

– Stored procedure cell selection.

In the initial cell selection procedure, as described above, no knowledge about RF channels is available at the UE. In that case the UE scans the supported E-UTRA frequency bands to find a suitable cell. Only the cell with the strongest signal per carrier will be selected by the UE.

The second procedure uses information about carrier frequencies and optionally cell parameters received and
stored from previously-detected cells. If no suitable cell is found using the stored information the UE starts with the initial cell selection procedure.