Sunday, May 10, 2015

Cell Reselection explained!!

While handover controls the mobility of UE in Connected (EMM-Registered, ECM-Connected and RRC-Connected) state, cell reselection controls the same while in Idle (EMM-Registered, ECM-Idle and RRC-Idle) state. During a handover, it is the network (MME or source eNB) that decides which cell to handover to. During cell reselection, however, it is UE that decides which cell to camp on2.

A cell reselection procedure can be one of the two types as seen below and in Figure 1. In Figure 1, the UE is camping on Cell 5 that belongs to the TA list of {TA1, TA2} previously assigned by the MME at the time of network attach, and is staying in Idle state.
  • Cell Reselection without TAU (: EMM Case 7): UE moves to a TA that IS registered at MME (i.e. listed in the TAI list of the UE), for example TA2 in Figure 1. Cell reselection is performed but no Track Area Update (TAU) is required.
  • Cell Reselection with TAU (: EMM Case 9): UE moves to a TA that is NOT registered at MME (i.e. not listed in the TAI list), for example TA3 in Figure 1. After cell reselection, a TAU procedure is performed.

Figure 1. Two Cell Reselection Types

UE may reselect a cell i) that uses the same LTE frequency as its current serving cell that it is camping on (intra-frequency), ii) that does not use the same frequency (inter-frequency), or iii) that uses other Radio Access Technology (RAT3) (inter-RAT). In the example cases of cell reselection shown in Figure 1, we only discuss intra-frequency cell reselection in a single LTE network environment that uses only a single LTE frequency.

This document describes cell reselection case only, and the other case () will be covered in our next document “EMM Procedure 8 & 9. Handover & Cell Reselection with TAU”[6]. In Chapter II below, we will review the preliminary information required to understand the cell reselection procedure. In Chapter III, a detailed description of the cell reselection without TA procedure will be given.


II. Cell Reselection: Required System Information and Criteria

Before we learn the detailed cell reselection procedure in Chapter III, we will provide a brief overview of a cell reselection procedure in Section 2.1 first. Then, we will look into what system information is required for the procedure and the criteria for the cell reselection in Sections 2.2 and 2.3, respectively.

2.1 Overview of Cell Reselection Procedure

It is UE who is in control of cell reselection. UE obtains information needed for cell reselection (e.g. threshold values used to decide whether to measure the signal strength of neighbor cells or not, parameters used for calculating rank of the serving cell and neighbor cells, etc.) from the system information broadcasted by eNB. In case of intra-frequency cell reselection, required information is delivered through System Information Block (SIB) 3 and SIB 4, which will be discussed further in Section 2.2.

 Cell Reselection Triggering
Serving Cell Measurement: UE, in Idle state, wakes up at the end of every DRX cycle to measure the signal of its serving cell (Qrxlevmeas) and calculate the received signal level (Srxlev) of the serving cell to decide whether it should stay or move to another cell. Here, the UE’s transmission and reception conditions are reflected in the calculation, for example by applying minimum received signal level Qrxlevmin, allowed maximum TX power level PEMAX, etc., (see Section 2.2 for details).
Cell Reselection Triggering: If the received signal level of the serving cell (Srxlev) is greater than the specified threshold value (s-IntraSearch), the UE stays in the current serving cell. If not, it triggers a cell reselection procedure. The threshold value that works as triggering criterion is delivered through SIB 3, and defined as s-IntraSearch in Release 8 and as s-IntraSearchP and s-IntraSearchQ in Release 9.

UE in Idle state wakes up at the end of every DRX cycle to measure the received signal level of its serving cell (Srxlev) when it has stayed in the same location for a while. Let’s assume a UE that has camped on the same serving cell for a while because the received signal level of the cell has remained lower than the set threshold (s-IntraSearch). If it leaves the serving cell, the received signal level of the cell decreases gradually. Finally when the received signal level becomes lower than the threshold (s-IntraSearch), a cell reselection procedure is triggered. Then the UE begins to measure the signal strength of the neighbor cells (i.e. non-serving cells).  

 Cell Reselection Criteria
Cell-Ranking Criterion: The UE ranks each cell (Rs, Rn) based on the measured signal strength of the serving cell (Qmeas,s) and neighbor cells (Qmeas,n). Parameters required for cell ranking are delivered through SIBs 3 and 4 (see Section 2.2). The serving cell is ranked using the hysteresis (q-Hyst) value stored in SIB 3 while the neighbor cells are ranked based on the offset (q-OffsetCell) value specified for each cell in SIB 4.
Cell Reselection: Once the serving cell and neighbor cells (non-serving cells) are ranked, the UE checks whether the cell reselection criterion is satisfied (Rn > Rs) or not. If there are neighbor cell(s) that satisfy the criterion, the UE selects the best satisfying cell, and then camps there. Cell reselection is performed only when the criterion is satisfied for a certain period of time (t-ReselectionEUTRA).

Mobile operators can prevent too frequent cell reselection and make sure reselection is performed in accordance with the cell status by controlling the UE’s dwelling time on the serving cell, based on the hysteresis and cell-specific offset values. In addition, they can control the q-Hyst and t-ReselectionEUTRA values by applying appropriate scaling factor (q-hystSF, t-ReselectionEUTRA-SF) depending on the traveling speed of the UE.

2.2 System Information

System Information (SI) refers to the information broadcasted by eNB and consists of MIB (Master Information Block) and SIBs (System Information Blocks; SIBs 1 ~ 16) [4]. MIB, SIB 1 and SIB 2 are mandatory, but the rest SIBs are optional. All SI is delivered to UE through an RRC message like MIB, SIB 1, or SI message4. An SI message consists of a group of SIBs (SIBs 2 ~ 16), excluding MIB and SIB 1.

UE performs a cell reselection procedure based on the SI broadcasted by eNB. MIB, SIB 1 and SIB 2 are applied to all the UEs, either in Connected state (EMM-Registered, ECM-Connected, RRC-Connected) or in Idle state (EMM-Registered, ECM-Idle, RRC-Idle). Conversely, SIBs 3 ~ 8 are only used in cell reselection by those in Idle state. Table 1 describes different types of SI and their parameters (see our One-Shot Gallery for more information about for a selected list of SIB information5).

Table 1. Cell Reselection-related System Information

Let’s assume two mobile operators (A and B) who operate the network as follows:

 A:
-
 has a LTE-only nationwide network
   
-
 LTE frequency: only one channel of 10 MHz in 1.8 GHz band (lteFA1)
       
 B:
-
 has nationwide 3G (UMTS) AND LTE networks
   
-
 3G frequency: six channels of 5 MHz in 2.1 GHz band (3gFA 1/2/3/4/5/6)
   
-
 LTE frequency: one channel of 10 MHz in 1.8 GHz band, and another channel of 10 MHz in 850
       MHz band (lteFA1, lteFA2)

In case of Operator A, its eNB needs information relating to intra-frequency for cell reselection (eNB broadcasts SIBs 3 and 4, but not SIBs 5, 6, 7 and 8. On the other hand, Operator B’s eNB needs information relating to all three types of reselection: intra-frequency, inter-frequency, and inter-RAT (3G UTRA) (eNB broadcasts SIBs 3, 4, 5 and 6).

The SI broadcasted by eNB is commonly applied to all the UEs, but each UE may receive different type of SI depending on its capacity (e.g. Release 9 UE or Release 11 UE).

Figure 1 displays a UE camping on Cell 5 at eNB 2 along with its neighbor eNBs. Figure 2 however shows eNB 2 only, along with its mandatory and cell reselection-related SI.

Figure 2. Cell Reselection-related SIBs Broadcasted by eNB2 

The network used in Figure 1 is an LTE-only network that uses a single frequency (lteFA1) with home PLMN only. This document is about the intra-frequency cell reselection procedure. The SI related to intra-frequency cell reselection is broadcasted through SIBs 3 and 4. Table 2 below lists the parameters used in the two SIBs, but not those related to other frequency, RAT or VPLMN.6

Table 2. Cell Reselection Parameters (SIBs 3 and 4)

2.3 Cell Reselection Criteria

Though not within the scope of this document, below we will review briefly the criteria used in cell reselection after UE is turned on. One of the following criteria is applied (i.e. criterion (1-1) for Release 8 UE and (1-2) for Release 9 or later UE). Table 3 provides a description of parameters used in the criteria.

 Cell Selection Criteria

 Release 8:
Srxlev > 0
(1-1)
 Release 9:
Srxlev > 0  and  Squal > 0
(1-2)
   
where, Srxlev = Qrxlevmeas - Qrxlevmin - PCompensation [dB]
 
   
  Squal = Qqualmeas – Qqualmin [dB]
 

Table 3. Cell Selection Criteria Parameters (TS 36.304 [5])

During the initial cell selection, a cell must have the cell RX level (Srxlev) greater than the sum of Qrxlevmin and PCompensation to be a serving cell. The Srxlev must be greater than Qrxlevmin because, in order for UE to correctly receive messages from its serving cell, the measured signal strength must be at least greater than Qrxlevmin. UE’s Tx power (PPowerClass) lower than the value allowed in the cell (PEMAX) will result in greater PCompensation value, making it hard to select the cell. When selecting a cell, UE’s transmission and reception conditions are considered.

In case of Release 9, in addition to Srxlev, Squal is added as a cell selection criterion. Qrxlevmeas is the cell’s Reference Signal Received Power (RSRP) while Qqualmeas is the cell’s Reference Signal Received Quality (RSRQ). Compared to RSRP which only indicates the strength of received signal, RSRQ provides more accurate information for radio link quality because it indicates the signal to interference and noise ratio (SINR).

 Cell Reselection Triggering
UE camping on the serving cell may continue to camp on there without having to measure other cells if the serving cell’s RX level fulfills the criterion (2-1) for Release 8 UE or (2-2) for Release 9 or later UE.    

 Release 8:
Srxlev > SintraSearch
(2-1)
 Release 9:
Srxlev > SintraSearchP  and  Squal > SintraSearchQ
(2-2)
   
where, Srxlev = Qrxlevmeas - Qrxlevmin - PCompensation [dB]
 
   
  Squal = Qqualmeas – Qqualmin [dB]
 

In the above inequalities, values of SintraSearch, SintraSearchP and SintraSearchP are given by SIB 3 (see Table 2). If the serving cell does not fulfill either of the foregoing criteria ((2-1) or (2-2)), i.e. if it fulfills the criterion (3-1) in case of Release 8 UE or (3-2) in case of Release 9 or later UE below, the UE begins measuring the neighbor cells for reselection.

Cell Reselection Triggering (Neighbor Cell Measurement Triggering)

 Release 8:
Srxlev ≤ SintraSearch
(3-1)
 Release 9:
Srxlev ≤ SintraSearchP  or  Squal ≤ SintraSearchQ
(3-2)

 Cell Reselection Criteria
Cell Ranking Criterion
If the measured Srxlev of the serving cell fulfills the foregoing criterion ((3-1) or (3-2)), the UE begins to measure the neighbor cell’s RSRP. Then based on the resulting measurements, it ranks all the cells by applying the criteria (4-1) and (4-2) below [5]. Table 4 shows the parameters used in cell ranking.
 
 Serving Cell Rank:
Rs = Qmeas,s + Qhyst
   (4-1)
 Neighbor Cell (non-serving cell) Rank:
Rn = Qmeas,n - Qoffset
   (4-2)

Table 4. Cell Reselection Criteria Parameters (TS 36.304 [5])

Cell Reselection
If there are multiple neighbor cells that fulfill the criterion (5) below, the UE selects the best cell and camps on it.
 

 Rn > Rs

 (5)

As seen in the foregoing criteria (4-1) and (4-2), Rs and Rn are calculated using different criteria. The greater Qhyst and Qoffset values are, the longer the UE can stay in the serving cell.

III. Procedure for Cell Reselection without TAU

Figure 3 illustrates the EMM Case 7. Cell Reselection without TAU () procedure. In Figure 1, the UE selected Cell 5 that fulfills the criterion (1-1) or (1-2), and was allocated a TA list of {TA1, TA2} by the MME after its initial attach to the network through Cell 5. Then later while being served in Cell 5, the UE transited to Idle state, and now is camping on Cell 5.

Figure 3 shows how UE camping on Cell 5 performs an intra-frequency cell reselection procedure as it switches to eNB 4, and camps on Cell 10. Here, the UE’s mobility state is “Normal”, and hence scaling factors are not considered. Neighbor cells to be measured are Cells 4, 6, 10 and 13, but Figure 3 displays Cells 10 and 13 only for the sake of convenience. Chapter III will explain the procedure for intra-frequency cell reselection that is successfully performed as seen in Figure 1, and thus satisfies the following conditions:
  • Camping on the serving cell: The UE is camping on Cell 5.
  • Cell Reselection Triggering: As UE moves away from the serving cell, cell reselection is triggered.
  1. Serving cell measurement: The serving cell is measured to decide whether to measure neighbor cells or not.
  2. Neighbor cells measurement: Neighbor cells (Cells 4, 6, 10 and 13) are measured for cell reselection.
  • Cell Reselection Criteria
  1. Cell ranking: Cells are ranked based on the measurement results for the serving cell and neighbor cells.
  2. Cell reselection: Cells that satisfy the criteria are identified, and the best satisfying cell (Cell 10) is selected.
  • Camping on the new cell: The UE camps on Cell 10.

Figure 3. Intra-frequency Cell Reselection Procedure (UE Moving to a registered TA)

   1)   [UE] UE Camping on the Serving Cell
The UE is camping on its serving cell (i.e. Cell 5 in eNB 2) while staying Idle. 

   2)   [UE] Obtaining SI from the Serving Cell
The UE obtains SI required for cell reselection from the serving cell. If each neighbor cell has different offset values with the serving cell, the serving cell then provides the UE with the list of neighbor cells through SIB 4. Then the UE acquires the following parameters through SIBs 3 and 4:
  • Parameters required for deciding on cell reselection triggering: q-RxLevMin, p-Max, s-IntraSearchP, s-IntraSearchQ, t-ReselectionEUTRA, q-QualMin (SIB 3)
  • Parameters required for ranking the serving cell: q-Hyst (SIB 3)
  • Parameters required for ranking the neighbor cells: q-OffsetCell (SIB 4)
   3)   [UE] Measuring the Signal Strength of the Serving Cell
At the end of the every DTX cycle, the UE wakes up and measure the signal of the serving cell (RSRP and RSRQ) to get Qrxlevmeas and Qqualmeas. Then, based on them, it computes the cell reselection received level (Srxlev) and cell reselection quality level (Squal). The UE, by applying the criterion (3-1) or (3-2), whichever is applicable depending on its release, checks whether it should reselect a new cell or it may continue to camp on the current serving cell. For example, if Srxlev and Squal does not satisfy the criterion, the UE continues to camp on the current cell. If either of them does, then it performs Step 4).

   4)   [UE] Measuring Neighbor Cells
The UE measures RSRP of the neighbor cells that are in the same frequency as the serving cell (Qmeas,n, n=4,6,10,13).

   5)   [UE] Cell-Ranking Criterion
Once the RSRPs are measured, the UE ranks the serving cell (Cell 5) and neighbor cells (Cells 4, 6, 10 and 13). The rank of the serving cell, Rank R5, can be computed by applying the criterion (4-1), and those of neighbor cells, R4/R6/R10/R12, can be determined by applying the criterion (4-2).

   6)   [UE] Cell Rank Comparison
Now, the UE compares Rank R5 and Rank Rn (n=4,6,10,13), and checks if the criterion (5) is satisfied. If no cell satisfies it, then the UE continues to camp on Cell 5. In the figure, the criterion (5) was satisfied by Cells 10 and 13.

   7)   [UE] Selecting a New Cell
The UE compares the two satisfying cells, R10 and R13, and selects R10, the best satisfying cell, as its new serving cell.

   8)   [UE] Camping on the New Cell
The UE now camps on Cell 10. After receiving SIB 1 broadcasted by Cell 10, it learns that TAI at Cell 10 is in the TAI list. Since the new serving cell belongs to the UE’s registered TA list, no TAU is performed. Thereafter, the UE wakes at the end of every DRX cycle, to monitor the system and paging information of Cell 10, and measure the signal of Cell 10 (RSRP, RSRQ).

IV. EPS Entity Information: Before/After Cell Reselection without TAU

This chapter will describe how information elements in the EPS entities are different before and after the cell reselection procedure. Since the UE stays in Idle state (EMM-Registered, ECM-Idle, RRC-Idle) before and after the procedure, MME also stays in Idle state (EMM-Registered, ECM-Idle). The UE moves from Cell 5 to Cell 10, and hence no TAU procedure was performed. Thus, after cell reselection, the information elements in the EPS entities remain unchanged, and will be the same as those stored after S1 release[10], as seen in Figure 4.

Figure 4. Information in EPS entity before/after Cell Reselection without TAU

V. Closing

We have learned how a UE in Idle state moves to TA where the UE is registered, and reselects a new cell without TAU. This document covers only the intra-frequency cell reselection procedure where cell reselection is performed within the same frequency. Most LTE operators might have more than one LTE carrier frequency in their commercial networks, and they usually operate their LTE network along with their 2G/3G networks. So, not only intra-frequency cell reselection that we discussed here, but also inter-frequency and inter-RAT cell reselections are considered in actual cell reselection. In the next document, we will discuss the procedure for cell reselection with TAU required when UE moves to TA where the UE is not registered

Saturday, May 9, 2015

MIMO Concepts: Antenna mapping process 2

Antenna mapping is the combination of layer mapping and pre-coding, which process the modulation symbols for one or two codewords to transmit them on different antenna ports. The stages of antenna mapping are illustrated in the next figure:


Where, RI is the Rank Indicator, for instance, a recommended number of layers by the User Equipment (UE) and PMI is the Pre-coding Matrix Indicator, and it could be a recommended pre-coder matrix by the UE.

The symbols for codewords, layers and antenna ports can be individually expressed as



In layer mapping, the modulation symbols for one or two codewords will be mapped onto one or several layers. Except the transmission on a single antenna port (in this case, the symbols for one code-word is directly mapped onto one layer), there are mainly two types of layer mapping: one for spatial multiplexing and the other for transmit diversity.
In case of spatial multiplexing, there may be one or two code-words. But the number of layers is restricted. On one hand, it should be equal to or more than the number of codewords. On the other hand, the number of layers cannot exceed the number of antenna ports. The most important concept is ‘layer’. The layers in spatial multiplexing have the same meaning as ‘streams’. They are used to transmit multiple data streams in parallel, so the number of layers here is often referred to as the transmission rank. In spatial multiplexing, the number of layers may be adapted to the transmission rank, by means of the feedback of a Rank Indicator (RI) to the layer mapping.


In case of transmit diversity, there is only one codeword and the number of layers is equal to the number of antenna ports. The number of layers in this case is not related to the transmission rank, because transmit-diversity schemes are always single-rank transmission schemes. The layers in transmit diversity are used to conveniently carry out the following precoding by some pre-defined matrices.



Transmit diversity for two antenna ports is based on Space Frequency Block Coding (SFBC), and transmit diversity for four antenna ports is based on a combination of SFBC and Frequency Shift Transmit Diversity (FSTD). According to the specifications, transmit diversity is implemented by a predefined matrix. It can be seen that comparing the case of two antenna ports, the four-antenna-port transmission has a reduced bandwidth. Note that unlike spatial multiplexing,                   M apsymb in transmit diversity equals M codewordsymb  rather than M clayersymb  , which just implies that the concept of layers for the two cases are basically different.



MIMO concepts: Antenna mapping process

User data and signaling messages are processed by the PDCP, RLC and MAC layers before being passed down to the PHY layer to be sent over the air. A lot happens to a data packet before PHY gets it, but for the moment, let's just treat the MAC PDU (Protocol Data Unit) that PHY receives from MAC as "data". To PHY, it's just a string of bits anyway. This will be our transport block.

Transport Blocks to Codewords

What does PHY do with a transport block? First, it converts the transport block into a codeword. There are a number of steps involved in this process, depending on the length of the transport block:
  • Append a 24 bit checksum (CRC) to the transport block. This CRC is used to determine whether the transmission was successful or not, and triggers Hybrid ARQ to send an ACK or NACK, as appropriate
  • Segment the transport block into code blocks. A code block must be between 40 and 6144 bits long. If the transport block is too small, it is padded up to 40 bits; if the TB is too big, it is divided into smaller pieces, each of which gets an additional 24 bit CRC.
  • Process each code block with a 1/3 turbo coder
  • Reassemble the resulting code blocks into a single codeword
A codeword, then, is essentially a transport block with error protection. Note that a UE may be configured to receive one or two transport blocks (and hence one or two codewords) in a single transmission interval.

Codewords to Layers

PHY then converts each codeword into modulation symbols. For each codeword, PHY must:
  • Scramble the contents of each codeword, using a sequence based on the UE's C-RNTI and the cell's Physical Cell ID (PCI)
  • Convert the bit sequences into the corresponding modulation symbols (using QPSK, 16QAM or 64QAM)
  • Assign the modulation symbols to one or more layers, depending on the specific transmission scheme being used
In the case of a single transmit antenna, the last step is pretty simple: the contents of the codeword are mapped to a single layer. For transmit diversity, it's almost as easy: the symbols from the codeword are distributed evenly across the 2 or 4 layers in a round-robin fashion.
In spatial multiplexing situations, things get a little more complicated, since one or two codewords may be distributed across 1, 2, 3 or 4 layers. In brief, here's how the mapping is handled:


The number of layers used in any particular transmission depends (at least in part) on the Rank Indication (RI) feedback from the UE, which identifies how many layers the UE can discern.

Layers to Antenna Ports

The final steps apply any required precoding adjustments and assign the modulation symbols to the physical resources:
  • Apply the required precoding factors to the modulation symbols in each layer
  • Map the precoded symbols to the appropriate antenna ports
  • Assign the modulation symbols to be transmitted on each antenna port to specific resource elements (the subcarriers and symbols within the resource blocks)
  • Generate the final time-domain OFDM signal for each antenna port
Note that the number of layers is always less than or equal to the number of antenna ports (transmit antennas). If there's only one antenna port, then it carries just a single layer. In multiple (2 or 4) antenna situations, though, each antenna port may end up carrying a complicated combination of the symbols from multiple layers. Check out spec 36.211, section 6.3.4 if you really want to dig into the details.

What's the answer in a nutshell? One transport block -> one codeword -> one or two layers -> one or more antenna ports.