Four scenarios with radio link failure
-DL physical layer failure (pdcch bler> 10%)
-Random Access problems (in connected state)
-Failure after RLC Retransmissions (specific PDU)
-Handover Failure-t304 expiry (acquisition or random access)
DL physical layer failure
Monitor detects physical layer issues, mainly because of the two-sided reference signal (RS) and the theoretical pdcch bler % (defined in 36.133 ). Based on DRC cycle and thresholds qout (10% )~ Qin (2%) generates in-Sync and out-of-sync.
-Assuming that drx is not used, the downlink channel quality is evaluated, assuming that less than qout within the last ms, ue's layer 1 will send a Out-of_Sync tag to a higher layer within Ms. The current row channel quality is better than Qin within Ms. Layer 1 on ue sends an in-sync message to the upper layer within Ms. There is a 10 ms interval between two consecutive tags. The transmission power will be disabled within 40 ms of t310 expiration.
-If drx is used, the interval between qout and Qin evaluation is determined by the drx configuration (the interval is shown in the table below ). In this case, the minimum interval between in-Sync and out-of-sync sending is 10 ms or determined by the drx cycle length.
DL physical layer failure-recovery
The figure shows the restoration process of UE before declaring Raido link failure.
Random Access problem
Assuming that uplink time alignment times out in connected mode, ue must run a random access process to obtain the new timing. In addition, new authorization is obtained before data transmission. Assume that ue sends the maximum number of preamble messages but does not receive a response, a radio link failure is declared.
In the figure, the maximum number of preamble values is 6,
Maximum RLC retransmissions
A radio link failure is declared when the maximum number of data duplicates is reached.
Handover Failure-t304 expiry
When ue receives the RRC connection reconfiugration command and prepares handover, t304 is started. UE is attached from the source cell, and the target cell is obtained successfully. The access is successful through the random access process. If the UE fails to access the target cell before t304 times out, the radio link failure is declared.
Recovery from RLF-Overview
Once RLF occurs, Timer 311 starts, srb1 hangs, and srb2 and all DRB instances are released.
Actual scenarios of two ue Raido LINK FAILURE
Recovery from the RLF in the same cell
When ue enters a coverage hole and RLF occurs, the recovery mechanism starts to start. UE selects the same cell and starts timer 301 to send messages, RRC connectin reestablishment, including C-RNTI, and finally serving cell C-rnti.
RRC connectin re-establishment in the same cell
After receiving the RRC connection reestablishment request message, the ENB will return the RRC connection reestablishment to restore the srb1 ENB. The UE context is determined because the request contains the PCI of the original (Same) residential area, c-rnti.
Handover Failure-t304 expiry
RRC reconfiguration is sent to ue to run handover. Switching from cell PCI = 400 to cell PCI = 250. ue will be assigned a new C-rnti for the target cell. At the same time, t304 will also be included in RRC connection reconfiguration. UE detach from the original cell (pCI = 400) by sending the Rach sequence attach to the target cell (pCI = 250)
In this example, ue does not receive the ENB response. t304 times out and an RLF occurs.
Recovery from Handover Failure in a different cell
After t304 times out, ue declares an RLF and starts the t311 timer. Before t311 times out, it successfully reselect the original target cell (pCI = 250 ). The t311 timer stops. UE started the t301 timer and sent the RRC connection reestablishment request message, which includes the C-RNTI and source zone PCI.
