INDEX

Downlink Throughput Troubleshooting. Physical layer cell identity planning Uplink signalling load.

Downlink Throughput Troubleshooting

Downlink Throughput Troubleshooting flow chart

The general troubleshooting strategy is described below and the covered reasons for bad throughput are shown in the figure below.

 Step 1: Identify cell with low DL (downlink) throughput a) The first thing is to identify those cells with low throughput. This threshold is defined by your network policies and practices (it also depends on your design parameters). Reports should be run for a significant number of days so that data is statistically valid. Step 2: Identify Downlink interference a) Cells with downlink interference are those whose CQI values are low (an exception to this rule is when most traffic is at the cell edge bad cell location-). Analyze the CQI values reported by the UE for Transmit Diversity MIMO one layer MIMO two layers Typical values for transmit diversity oscillate between 7 and 8. Typical values for MIMO one and two layers oscillate between 10 and 12. b) If low CQI values are found after a CQI report is obtained, then downlink interference might be the cause of low throughput. c) Common sources of interference in the 700 MHz band (LTE deployment in the USA) are: inter-modulation interference, cell jammers and wireless microphones

 Step 3: BLER Values a) Run a report for BLER in the cells identified. The BLER should be smaller or equal than 10%. If the value is larger, then, there is an indication of bad RF environment. b) Typical causes of bad BLER are downlink interference, bad coverage (holes in the network, etc.) Step 4: MIMO Parameters a) Identify the transmission mode of your network. There are seven transmission modes as shown in the table below.

b) Adjust the SINR thresholds for transition of transmission modes as recommended by the OEM. Request the Link Level simulations they used to set these thresholds and see if the conditions under which the values were calculated apply to your network. Otherwise, update them if the parameters are settable and not restricted. Step 5: Low Demand a) Run a report using the counters to find 1. Maximum number of RRC connections supported per cell (parameter or feature) 2. Maximum number of RRC connections active per cell 3. Average number of RRC connections active per cell 4. Maximum number of users per TTI supported per cell (parameter or feature) 5. Maximum number of users scheduled per TTI in the cell(s) of interest 6. Average number users scheduled per TTI in the cell(s) of interest

b) If the maximum number of RRC connections active per cell is close or equal to the maximum number of RRC connections supported, then. The cause for low throughput is load. c) A high number of scheduled users per TTI does not necessarily mean that demand is the cause for low throughput. Step 6: Scheduler Type a) Find the scheduler types. b) Select the one that is more convenient for the type of cell you are investigating. Examples of schedulers are: round robin, proportional fairness, maximum C/I, equal opportunity, etc. OEMs allow you to switch the scheduler in your network but recommend one in particular. c) The wrong scheduler may be the reason for bad throughput. Step 7: CQI reporting parameters a) Check if network is using periodic or aperiodic CQI reporting (or both). b) Verify the frequency in which the CQI reporting is carried out for periodic reporting as well as the maximum number of users supported per second.

c) If the value is too small compared with the maximum number of RRC active connections, then, increase the values of the parameters CQIConfigIndex as well as RIConfigIndex d) If network is not using aperiodic CQI reporting, then enable it. e) Slow frequencies of CQI reporting might yield bad channel estimations that prevent the eNodeB from scheduling the right amount of data and Modulation and Coding Schemes to UE. Step 7: Other a) Run a VSWR report. b) High values of VSWR result in low throughput due to losses. c) Check backhaul capacity. Often times, the backhaul links are shared among multiple RATs. Make sure backhaul is properly dimensioned.

Physical layer cell identity planning

Physical cell identity

Only a maximum of 504 PCIs have satisfactory orthogonal performance. Therefore, they must be numbered to prevent PCI confusion. Though all cells have different PCIs the PCI reuse distance is insufficient for UEs to prevent interference between non-correlated pilot signals. Consequently, errors occur when the UE trances pilot signals. If the errors occur during eNodeB identification, the UE may be unexpectedly handed over to a different cell, which may cause service drop. The problems when these 504 PCIs are reused. Collision If two neighbouring cells are allocated with the same PCI in an intra-frequency network, a maximum of one cell can be detected by the UE, and only one cell can be synchronized during initial cell search. If the synchronized cell does not meet the handover requirements, a collision occurs, as shown in figure 3-1. Figure 3-1 Collision

Confusion If neighbouring cells have the same PCI (ID A in Figure 3-2) and UEs are to be handed over to a neighbouring cell, the eNodeB cannot decide which neighbouring cell is the target cell. Consequently, confusion occurs. Figure 3-2 Confusion

Therefore, PCI planning must ensure that the PCI is free from confusion and collision. In addition, PCI planning must comply with the following principles: If a serving cell is configured with intra-frequency neighbouring cells with strong interference, the neighbouring cells cannot use the same PCI as the serving cell.

NSN recommendations The isolation between cells which are assigned the same physical layer cell identity should be maximised and should be sufficiently great to ensure that UE never simultaneously receive the same identity from more than a single cell Whenever possible, cells belonging to the same eNodeB should be allocated identities from within the same group Specific physical layer cell identities should be excluded from the plan to allow for future network expansion There should be some level of co-ordination across international borders when allocating physical layer cell identities Better to avoid Cell IDs with identical values mod 3 among neighbors to distinguish PSS. PCI=PSS+3*SSS PSS={0,1,2} SSS={0 to 167} Groups

if the 2 cell are transmitting in the same carrier, there is a high probability that inter-cell RS to RS interference will occur in the overlapping coverage area. One way to work around this is to shift the neighbour cell RS symbol by one RE (aka RS planning PCI mod 3). Hence, PCI planning should consider PCI mod 3 planning as well to reduce inter-cell RS-RS interference. The position of an LTE pilot symbol is associated with the PCI code assigned by the cell. To prevent interference between pilot symbols and improve overall network performance, the pilot symbol of the serving cell cannot be located side by side with those of neighboring cells. The position of pilot symbols in the frequency domain is determined by PCI MOD 3 in two- and four-antenna scenarios and by PCI MOD 6 in the singleantenna scenario.

Reducing uplink signalling load

Effect of Open Loop Power Control on the UL RSSI

The Received Signal Strength Indicator (RSSI) in the uplink is affected by the parameter settings that govern open loop power control in LTE. Open loop power control is used during Random Access. The random access is often the first transmission from the UE, and it is a short transmission (less than 3 ms at most). Consequently, the network does not have an opportunity to power control the PRACH transmitted by the UE. Instead, the UE must estimate the minimum amount of power it needs to send the access request without causing excessive interference. The UE receives a number of key parameters for PRACH power control in SIB 2, including: Preamble Initial Received Target Power: The power level the eNB would like to receive for a random access. The default value is -104 dBm. Power Ramping Step: The amount of additional power to be used every time the random access is attempted again. This can be 0, 2, 4 or 6 dB. Preamble Trans Max: The maximum number of times a random access can be attempted before the UE gives up, to a maximum of 10 tries. RA Response Window Size: The number of subframes the UE will wait for a response after a random access, between two and 10 subframes. T300: the time the UE has to receive the RRC connection Setup message from the eNodeB. The UE will determine the initial power level based on the Preamble Initial Received Target Power value and an estimate of the uplink Path Loss (PL) as follows:

Pinitial = min (Pmax, Preamble Initial Received Target Power + PL); where Pmax is the maximum transmit power of the UE, based on its category. If the eNB fails to respond to the random access in the designated time window (RA Response Window Size), then it can repeat the random access (after waiting at least four more subframes), increasing its power level by the Power Ramping Step value. If no response is received after Preamble Trans Max attempts, then the UE will return an access failed indication. The values of Preamble Initial Target Power and Power Ramping Step directly affect RSSI. High values of both parameters may result in high RSSI, particularly in indoor environments (i.e.: Airports, convention centers, etc.) and events (i.e.: foot ball games at stadiums, concerts, etc.). In this type of environments, a high concentration of UEs exists and often times, many of them try accessing the network at the same time. Even when a different UE has selected a different preamble and has calculated the right amount of power, the eNodeB often times will only answer to ONE of them (this is vendor implementation dependent) per sub-frame, leading to the rest of the UE to increase their transmit power. If this condition prevails, quickly, all UEs will be transmitting at is maximum transmit power in less than a second, producing a high RSSI at the eNodeB. Values of -110 to -112 dBm are recommended for Preamble Initial Target Power and 2dB for Power Ramping step for events and indoor environments. For outdoor environments, depending on the load and RSSI, values of -104 dBm and 4dB could be used, respectively.

Reducing LTE Uplink Transmission Energy by Allocating Resources

Physical Resource Block (PRB) allocation effects on LTE UE transmission power and energy consumption were examined. The simulation results, based on a mapping from transmission power to energy consumption, show that it is more energy efficient to allocate as many PRBs as possible to a single user instead of assigning several users less PRBs. On average at least 24 % energy can be saved if a user is allocated an entire 10 MHz channel (48 PRBs) instead of 8 PRBs. LTEs Uplink Power Control entails that users with more PRBs will transmit with higher power, but the throughput increases concurrently and therefore energy can be saved. Further more the applied power consumption model entails that the UEs efficiency increases when the transmit power increases. An equal opportunity turn-based PRB scheduler was implemented to evaluate how scheduling of maximum 6, 8, and 10simultaneous users affect the energy consumption. The results show scheduling maximum 10 users instead of 6 increases the average transmission time with4 % and the average energy consumption with 6 %. Yet there is no incentive to allow more than 6 users because the cell throughput is independent of the number of users. The conclusion is that one user should be allocated as many PRBs as possible, while limiting the number of simultaneous users to reduce the average waiting time.

LTE473: Extended DRX settings Introduction to the feature The Flexi Multiradio BTS supports with this feature an extended range of 3GPP settings for the long DRX cycle, two additional operator configurable DRX profiles, and uplink Out-of-Sync handling. Benefits The extension to support longer settings for the long DRX cycle leads to a lower UE power consumption mainly for UEs with only occasional data transmission. System Requirements eNodeB MME SAE GW UE NetAct release Release RL30 LBTS3.0 Software requirements Table 37 Software requirements lists the software required for this feature. Hardware requirements This feature requires no new or additional hardware. Functional description Feature scope Extended DRX settings feature improves power savings for the UE by supporting also settings of the long DRX cycle beyond 80ms from the 3GPP defined range. Those additional savings in power are feasible for bursty traffic patterns (i.e. short phases with data transmission followed by long phases of idle period). The potential additional power savings come, however, at the cost of increased latency for DL transmission whenever the UE is in DRX sleep mode.

Basic characteristics and limitations LTE473 is an extension to feature LTE 42:Support of DRX in RRC Connected. LTE473: Extended DRX settings feature comprises three subfeatures: support of an extended value range of the long DRX Cycle two additional operator configurable DRX profiles uplink Out-of-Sync handling
Support of an extended value range of the long DRX Cycle Two profiles which support extended 3GPP value range for the long DRX cycle (160, 320ms in Profile4; 640, 1280, 2560ms in Profile5). Two additional operator configurable DRX profiles Two additional operator configurable DRX profiles are introduced with this feature in order to allow more flexible definition of different DRX use cases, e.g. Out-of-Sync handling. drxProfile4: non-GBR (<500ms, e.g. QCI 5) The profile is optimized for non-GBR bearers with specific latency requirements that allow setting medium DRX cycle length (e.g. IMS signaling) drxProfile5: non-GBR (>=500ms) The profile is appropriate for non-GBR bearers without specific latency requirements that allow setting long DRX cycle length (e.g. web browsing) Additional profiles will only yield some gains if gaps in UE data transmission are sufficiently large (e.g., always-on devices doing only an occasional data transmission with gaps of at least several tens of seconds). The UEs are kept DRX Active always during phases of data transmission and dropped to UL out-of-sync afterwards (the UE benefits from the extended DRX when in UL Out-of-Sync state). Any kind of traffic mapped to QCIs using such profiles should be latency tolerant as to match at least the long DRX cycle setting. Extended DRX is supported for QCI 5-9 (i.e., QCI types without any delay guarantees). Uplink Out-of-Sync handling The uplink Out-of-Sync handling comprises the following two subfeatures: Uplink Out-of-Sync enforcement By using very long settings for the DRX cycle, the UE may go to or is even actively sent to the uplink Out-of-Sync status. For this timing alignment is stopped some time after UE has finished data transmission. Transition to uplink In-Sync DL data transmission trigger The eNodeB initiates a random access procedure for UEs which are in uplink Out-of-Sync and have data for downlink transmission. The eNodeB provides in this case the RACH parameters via the PDCCH order to the UE. UL data transmission trigger During the UL Out-of-Sync state, UE may start a contention based random access procedure in order to transmit data on uplink Following the transition to In-Sync, UEs are reconfigured with any resources released during UL Out-of-Sync transition (resources on PUCCH and for SRS, if applicable).

Transition to uplink In-Sync DL data transmission trigger The eNodeB initiates a random access procedure for UEs which are in uplink Out-of-Sync and have data for downlink transmission. The eNodeB provides in this case the RACH parameters via the PDCCH order to the UE. UL data transmission trigger During the UL Out-of-Sync state, UE may start a contention based random access procedure in order to transmit data on uplink Following the transition to In-Sync, UEs are reconfigured with any resources released during UL Out-of-Sync transition (resources on PUCCH and for SRS, if applicable). Related parameters Apply UL Out-of-Sync Determines which UEs is actively sent to UL Out-of-Sync state provided that bearer combination and applied DRX profile allows this. Three settings are possible: extendedDrxonly: only UEs being configured with extended settings for the long DRX cycle allDrx: all UEs being configured for DRX provided that applied DRX profile allows allUEs: all UEs independently of DRX configuration provided that bearer combination allows Short Term Inactivity Factor Short Term Inactivity Timer determines when to trigger sending the UE to UL out-of-sync by stopping timing alignment to the UE after a data transmission phase of the UE has ended. Short Term Inactivity Timer is configurable in multiples of the DRX Inactivity Timer setting applied by the parameter Short Term Inactivity Factor.