Commit 6d9200f4 authored by Yann Garcia's avatar Yann Garcia
Browse files

Apply review from Giovanni (UNIPI)

parent edb1c2bd
Loading
Loading
Loading
Loading
Loading
+14 −12
Original line number Original line Diff line number Diff line
@@ -504,7 +504,7 @@ The current architecture comes with some limitations:


The Simu5G/SimuLTE simulators [\[i.5\]](#_ref_i.5) are open-source, system-level network simulators built on the OMNeT++ [\[i.4\]](#_ref_i.4) discrete event simulation framework. They provide comprehensive simulation of 3GPP-compliant 4G (LTE/LTE-A) and 5G (NR) networks.
The Simu5G/SimuLTE simulators [\[i.5\]](#_ref_i.5) are open-source, system-level network simulators built on the OMNeT++ [\[i.4\]](#_ref_i.4) discrete event simulation framework. They provide comprehensive simulation of 3GPP-compliant 4G (LTE/LTE-A) and 5G (NR) networks.
The key characteristics are:
The key characteristics are:
- Framework: OMNeT++ 6.0.x with INET Framework 4.5.x;
- Framework: OMNeT++ 6.3.x with INET Framework 4.5.x;
- License: LGPL v2.1;
- License: LGPL v2.1;
- Language: C++;
- Language: C++;
- Architecture: Modular, event-driven simulation.
- Architecture: Modular, event-driven simulation.
@@ -522,10 +522,10 @@ Simu5G provides a full 4G/LTE/LTE-A (E‑UTRAN) data‑plane stack alongside 5G
### A.3.3 Simu5G (5G NR) features
### A.3.3 Simu5G (5G NR) features


The Simu5G (5G NR) models the full 5G NR user‑plane stack (Rel‑16‑oriented) with gNB, UE, and UPF entities, supporting both Standalone NR and LTE–NR Dual Connectivity deployments over FDD and TDD, with heterogeneous cells and realistic PHY/channel models:
The Simu5G (5G NR) models the full 5G NR user‑plane stack (Rel‑16‑oriented) with gNB, UE, and UPF entities, supporting both Standalone NR and LTE–NR Dual Connectivity deployments over FDD and TDD, with heterogeneous cells and realistic PHY/channel models:
- 5G NR protocol stack (RRC, SDAP, PDCP, RLC, MAC, PHY);
- 5G NR protocol stack (e.g. RLC);
- 5GC (5G Core) simulation;
- 5GC (5G Core) simulation;
- Support for gNB and UE entities;
- Support for gNB and UE entities;
- Advanced features: Network Slicing, MEC integration;
- Advanced features: MEC integration;
- Real-time emulation support;
- Real-time emulation support;
- Integration with real applications.
- Integration with real applications.


@@ -542,7 +542,8 @@ The Simu5G (5G NR) models the full 5G NR user‑plane stack (Rel‑16‑oriented


- Complete Stack: All layers from PHY to application
- Complete Stack: All layers from PHY to application
- 3GPP Compliance: Implements 3GPP specifications
- 3GPP Compliance: Implements 3GPP specifications
- Dynamic Behavior: Handovers, connection establishment, bearer management
- Dynamic Behavior: Handovers, connection establishment, dynamic radio resource scheduling
- Support of Wi-Fi, based on INET framework [\[i.4\]](#_ref_i.4)


#### A.3.4.3 MEC Integration
#### A.3.4.3 MEC Integration


@@ -618,7 +619,7 @@ The fugure below shows a simplified architecture of Simu5G.


<mark>Not sure this clause is useful at this stage</mark>
<mark>Not sure this clause is useful at this stage</mark>


### A.3.8 Advantages of Simu5G/Simu4G simulators
### A.3.8 Advantages of Simu5G/SimuLTE simulators


- Realistic Protocol Simulation: Complete 3GPP protocol stack
- Realistic Protocol Simulation: Complete 3GPP protocol stack
- Dynamic Network Behavior: Handovers, connection management, bearer setup
- Dynamic Network Behavior: Handovers, connection management, bearer setup
@@ -627,13 +628,12 @@ The fugure below shows a simplified architecture of Simu5G.
- MEC Integration: Built-in MEC support
- MEC Integration: Built-in MEC support
- Research-Grade: Widely used in academic and industry research
- Research-Grade: Widely used in academic and industry research


### A.3.9 Limitations of Simu5G/Simu4G simulators
### A.3.9 Limitations of Simu5G/SimuLTE simulators


- Performance: Discrete event simulation can be slower than simple TC-based approach
- Performance: Discrete event simulation can be slower than simple TC-based approach
- Complexity: Requires understanding of OMNeT++ and simulation concepts;
- Complexity: Requires understanding of OMNeT++ and simulation concepts;
- Integration: Need to bridge simulation with Kubernetes/Docker environment ([\[i.9\]](#_ref_i.9), [\[i.8\]](#_ref_i.8));
- Integration: Need to bridge simulation with Kubernetes/Docker environment ([\[i.9\]](#_ref_i.9), [\[i.8\]](#_ref_i.8));
- Real-Time: Real-time emulation requires careful synchronization;
- Real-Time: Real-time emulation requires careful synchronization.
- WiFi: Limited WiFi support (additional modules are required, particularly ).


### A.3.8 Conclusions
### A.3.8 Conclusions


@@ -673,11 +673,13 @@ The replacement will be considered successful if:


> **⚠️ IMPORTANT CONSIDERATIONS**
> **⚠️ IMPORTANT CONSIDERATIONS**
>
>
> The counterpart is a huge changes in the current ETSI MEC Sandbox architecture, and a higher complexity. The Simu5G simulator is implemented as a single-process, single-threaded application. As a result, executing simulations that involve a large number of User Equipments (UEs) or intricate scenarios may result in diminished performance, which could potentially introduce operational challenges.
> The counterpart is a huge changes in the current ETSI MEC Sandbox architecture, and a higher complexity. The Simu5G simulator is implemented as a single-process, single-threaded application. As a result:
> 1. Executing simulations that involve a large number of User Equipments (UEs) or intricate scenarios may result in diminished performance, which could potentially introduce operational challenges;
> 2. A multi-core machine can be exploited by running multiple independent simulations on different cores.
>
>
> In addition, two warnings shall be raised regarding performance and the objectives to achieve:
> In addition, two warnings shall be raised regarding performance and the objectives to achieve:
>
>
> 1. A network scenario containing a lot of PoAs and UEs can drastically reduce the performance of the Simu5G/4G simulators. The objective of 10 simultaneous connections on the MEC Sandbox could be achieved.
> 1. A network scenario containing a lot of PoAs and UEs could drastically reduce the performance of the Simu5G/4G simulators. The objective of 10 simultaneous connections on the MEC Sandbox could be achieved. **Note** that a paper (https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9591605) demonstrates the capacity of Simu5G to  emulate a few base stations and tens of UEs.
>
>
> 2. If the network scenario is to complex, the real-time emulation mode cannot be apply and the simulation will not match with real-tme clock.
> 2. If the network scenario is to complex, the real-time emulation mode cannot be apply and the simulation will not match with real-tme clock.
>
>
@@ -687,7 +689,7 @@ The replacement will be considered successful if:


---
---


> NOTE: In addition, a phase of learning is mandatory before to start the replacement of the current network similation by Simu5G/Simu4G simulators [\[i.5\]](#_ref_i.5).
> NOTE: In addition, a phase of learning is mandatory before to start the replacement of the current network similation by Simu5G/SimuLTE simulators [\[i.5\]](#_ref_i.5).


## A.4. Replacement procedure
## A.4. Replacement procedure


@@ -874,7 +876,7 @@ The HTTP REST API mechanism is too slow to be used. The proposal is to use a TCP


The Real-Time Emulation Bridge enables real applications to connect to the Simu5G/SimuLTE simulations, allowing them to send and receive packets through the simulated network as if connected to a real 5G/4G radio access network.
The Real-Time Emulation Bridge enables real applications to connect to the Simu5G/SimuLTE simulations, allowing them to send and receive packets through the simulated network as if connected to a real 5G/4G radio access network.


Simu5G/Simu4G simulators use TUN (level 3) integration:
Simu5G/SimuLTE simulators use TUN (level 3) integration:
- Create TUN interfaces for each UE pod
- Create TUN interfaces for each UE pod
- Route traffic through simulation
- Route traffic through simulation
- Apply simulation delays/losses to real traffic
- Apply simulation delays/losses to real traffic