@NTUST

# RAN Slice SLA Assurance

- [x] => Sentence with a check mark is the key points.

:::info
**Network Slicing：**
- **Network Slicing** is expected to play **a critical role** in 5G networks because of **various use cases and services** 5G will support.
- [x] It allows a network operator to provide **services tailored to customers’ requirements** (e.g. the application for very high data rates, traffic densities, service availability and very low latency).
- Network slice is defined as a logical network with **a bundle of specified network services** over a common network infrastructure.
- [x] A single physical network is sliced into **multiple virtual networks** that can **support different service types over a single RAN**.
- 3GPP has standardized 4 different service types: **eMBB, URLLC, MIoT and V2X**.
:::

:::success
**What is the SLA ?**
- [x] The SLA is **a contract** between the **Network Service Provider and the customer**, which underlines **each party’s responsibilities** while at the same time defining the **performance standards** that are to be met by the provider.
- SLAs establish customer expectations regarding the **service provider's performance and overall quality**.
- Over the past few years, SLAs set the expectations for a service provider performance, establish **penalties for missing the targets** and, in some cases, bonuses for exceeding them.
:::

## 1. Background

According to 5G standardization efforts, a 5G system should **support the needs of the business through** the specification of **several service KPIs** such as data rate, traffic capacity, user density, latency, reliability, and availability. These capabilities are **specified based on a Service Level Agreement (SLA)** between the mobile operator and the business customer, which has resulted in increased interest in mechanisms to **ensure slice SLAs** and **prevent its possible violations**.

### 1.1 Why to use it?

- Although network slicing support is started to be defined with 3GPP Release 15, slice assurance mechanisms in RAN needs to be further addressed to achieve deployable network slicing in an open RAN environment. It is necessary to assure the SLAs by dynamically controlling slice configurations based on slice specific performance information.
- [x] **Existing RAN** performance measurements and information model definitions are **not enough to support RAN slice SLA assurance use cases**.
- Considering the **dynamic nature of RAN**, providing **desired levels** of service quality **for each RAN slice** is a **challenging topic** that requires further investigation and standardization efforts for a multi-vendor open RAN environment.

- **This use case** is intended to **clarify necessary mechanisms and parameters** for RAN slice SLA assurance.

### 1.2 How do O-RAN overcome the challenge of slice SLA assurance ?

- [x] O-RAN’s **open interfaces** and **AI/ML based architecture** will **enable such challenging** mechanisms to be implemented and help pave the way for operators to **realize the opportunities of network slicing**.
- That could potentially change the way network operators do their business and also enable new business models. For example, O-RAN architecture and interfaces can enable operators to **manage spectrum resource allocation across slices more efficiently and dynamically** in response to usage patterns, thereby allowing more efficient use of spectrum resources.

## 2. Required data
- The measurement counters and **KPIs** (as defined by 3GPP and will be extended for O-RAN use cases) should be appropriately aggregated by cell, QoS type, slice, etc.
- **Per-UE Channel State Information (CSI)**
- **Per slice performance statistics** such as PDCP throughput, PRB usage

## 3. Components & Resources involved in the use case


### 3.1 Non-RT RIC：
- [x] Retrieve **RAN slice SLA target** from respective entities such as SMO, NSSMF.
- Long term monitoring of RAN slice **performance measurements**.
- [x] **Training of potential ML models** that will be deployed in Non-RT RIC for slow loop optimization and/or Near-RT RIC for fast loop optimization.
- Support deployment and update of AI/ML models into Near-RT RIC.
- Receive slice control/slice SLA assurance xApps from SMO.
- Create A1 policies based on RAN intent A1 feedback.
- [x] Send A1 policies and enrichment information to Near-RT RIC to **drive slice assurance**.
- Send O1 reconfiguration requests to SMO for slow-loop slice assurance.
### 3.2 Near-RT RIC：
- Near real-time monitoring of slice specific RAN performance measurements.
- [x] Support **deployment and execution** of the AI/ML models from Non-RT RIC.
- Receive slice SLA assurance xApps from SMO.
- [x] Support interpretation and execution of policies from Non-RT RIC.
- Perform optimized RAN (E2) actions to achieve RAN slice requirements based on O1 configuration, A1 policy, and E2 reports.
### 3.3 RAN (E2 Node)：
- [x] Support slice assurance actions such as slice-aware **resource allocation**, **prioritization**, etc.
- Support slice specific **performance measurements** through O1.
- Support slice specific **performance reports** through E2.

## 4. Detailed Flow of Use Case

### 4.1 Creation and deployment of RAN slice SLA assurance models and control apps：
- Goal：Training and distribution of the model, or distribution of control apps.


### 4.2 Slow loop RAN Slice SLA optimization
- Goal：RAN Slice SLA optimization (Slow)


### 4.3 Fast loop RAN Slice SLA optimization
- Goal：RAN Slice SLA optimization (Fast)