In cellular networks, the A5 handover is a method used for maintaining signal quality as mobile devices move across different sectors. It differs from A3 handover by using signal strength thresholds from both the source and target base stations for handover decisions.

A3 Handover:

• Depends on a handover margin.
• Handover occurs when the target base station's signal exceeds the source's by a predefined margin.

A5 Handover:

• Utilizes signal strength thresholds from both serving and target base stations.
• Handover happens when the serving station's signal is below one threshold while the target's signal is above another.

Essentially, A3 handover is based on to relative signal strength differences, whereas A5 is based on absolute signal strength thresholds.

Implementing the A5 Handover Algorithm in NetSim

NetSim implements the A3 handover. Minor code changes are required to implement the A5 handover in NetSim. 

We substitute the A3 handover condition for checking the handover margin and added a call to function that first checks the handover mechanism; if the Handover Algorithm is set to A3, it uses the current SINR-based Handover Margin condition, whereas if A5 is set, it compares the source and target signal strength to the source and target thresholds, as previously explained. If the handover condition is satisfied based on the output from the function, the current code is executed. The source and target thresholds are user-input values obtained from the GUI.

Additional variables are added in LTENR_GNBRRC.h to accommodate A5 Handover Algorithm. This includes handover_type (an enum to get the handover type A3 or A5) and Source and Target thresholds, in case of A5 handover.

In function, fn_NetSim_LTENR_RRC_UE_MEASUREMENT_RECV in file LTENR_GNBRRC.c replace the handover margin check condition to a function call checking if the condition satisfies or not, passing the necessary parameters.

The function fn_NetSim_check_handover_condition is added in file LTENR_GNBRRC.c as shown below:

The LTENR Handover Logs are also updated according to whether the handover algorithm is A3 Handover, or A5 Handover. The A5 handover log provides information about the source and target SSB Downlink RSRP, based on which handover decisions are taken.

NetSim GUI XML files are modified to allow users to choose between A3 and A5 handover algorithms and to configure the thresholds. 

Example 1: A simple scenario to illustrate the A5 Handover 

We consider a scenario similar to the current handover algorithm and configure mobility such that the UE_9 moves 200 meters rightwards every second across the simulation time, with a TTT of 100 milliseconds. 

 The thresholds for A5 Handover Algorithm are set as shown below:

The workspace containing the revised source codes and updated XMLs is accessible in the provided link. Additionally, there are experiment files for a handover scenario without time-to-trigger (TTT) and another scenario depicting handover failure within the TTT duration.

https://github.com/NetSim-TETCOS/5G_A5_Handover_v14/archive/refs/heads/main.zip

Follow the instructions specified in the following link to download and setup the Project in NetSim:

https://support.tetcos.com/support/solutions/articles/14000128666-downloading-and-setting-up-netsim-file-exchange-projects

Additional steps for setting up the project involves the following steps:

1. Right-click on the NetSim icon and select "Open file location." Navigate to the directory: Current Netsim Install Directory\Docs\UI_xml\ExternalFiles.
2. Replace the existing LTE-RAN-HANDOVER.xml file in that directory with the downloaded file.

Use 5G A5 Handover with TTT, experiment file for the current experiment. In accordance with the configured scenario, the handover condition is met at 15 seconds, and the handover process is initiated at 15.1 seconds (with a 100-millisecond TTT duration). The corresponding handover logs are available.

Example 2: impact of A5 handovers in a 5G Heterogeneous Network

This example is based on the article https://support.tetcos.com/a/solutions/articles/14000136902, which explains the impact of time-to-trigger and handover margin on handover counts and throughputs in a 5G heterogenous network when using the A3 handover algorithm. We now analyze the same the A5 handover algorithm.

For additional details on the scenario parameters, you can refer to the A3 event-based handover approach, accessible through the link: https://github.com/NetSim-TETCOS/5g-Heterogeneous-Networkv14.0/archive/refs/heads/main.zip

Results and discussion 

We tabulate below the handover count and sum throughput for various values of Time to trigger (ms) and Source and Target Thresholds. 

Table 1: Sum Throughput and Handover Count Value for various Time-to-Trigger and Source, Target Handover Thresholds

Figure 1: Handover Count vs. Target Threshold for Different TTTs

It is evident from the plot that the handover count tends to decrease as the target threshold increases for a fixed source threshold. This trend is consistent across different TTT values, suggesting that a higher target threshold generally results in fewer handovers. The rationale behind this trend it that increased target threshold leads to more stringent conditions for handover and thereby reduces the frequency of handover occurrences. Shorter TTT values lead to quicker responses to signal changes, resulting in more frequent handovers, while longer TTT values delay the handover process, thereby reducing the handover count. The plot highlights the effects of both the target threshold and the TTT on handover count.

Figure 2: Sum Throughput vs. Target Threshold for Different TTTs

The chart's depicts the variation in sum throughput which initially rises and then falls with increasing handover margin. Initially, with a higher handover margin unnecessary and frequent handovers between cells are avoided. This leads to better throughput, but only to a certain extent. Beyond a point, a high handover margin causes delayed handovers. Users may stay connected to a weaker cell longer, despite being closer to a stronger cell, leading to poorer signal quality and thus lowering throughput.

It is important to note that the presented results are obtained from a simulation run with a single seed. To achieve statistically valid conclusions, it is essential to conduct simulations with multiple seeds.