RFCellular Network vs. Cell-Free Network: Key Differences
Advertisement
This article compares cellular networks and cell-free networks, highlighting their key differences.
Introduction
In traditional cellular networks, the coverage area is divided into smaller regions called “cells.” Base Stations (BSs) are deployed within each of these cells to provide network access.
- Each BS serves multiple users within its coverage area.
- All BSs are interconnected and connected to the Core Network (CN) using backhaul infrastructure.
In contrast, cell-free networks cover an entire area with numerous Access Points (APs), eliminating the concept of distinct cells.
- All APs are connected to a Central Processing Unit (CPU).
- The CPUs are then connected to the Core Network (CN).
- Cell-free networks consist of antenna units, Antenna Processing Units (APUs), and CPUs.
- Multiple APs serve all users within their range.
- In a cell-free network, each user is typically served by multiple APs.
Cellular Network
Figure 1: Cellular Network Architecture
As depicted in Figure 1, BSs are interconnected in various configurations (e.g., star, mesh). They connect to the Public Switched Telephone Network (PSTN), Mobile Switching Centres (MSCs), and Public Switched Data Network (PSDN) via the backhaul. Common cellular network technologies include GSM, CDMA, LTE, and 5G.
Cell-Free Network
Figure 2: Cell-Free Massive MIMO Architecture
Figure 2 illustrates the cell-free network architecture. Cell-Free Massive MIMO comprises Access Points (APs) and a central processor connected via a fronthaul network.
- APs are distributed throughout the coverage area.
- Each AP is connected to a central processing unit using fronthaul cables.
- The central processor manages all APs.
- All APs serve the surrounding users, and there are no cell boundaries, hence the name “cell-free.”
In this architecture, multiple antennas are geographically distributed rather than co-located at a single Base Station (BS). These distributed antennas are the Access Points (APs).
- APs operate jointly to serve users in the same time/frequency resources.
- APs function synchronously and coherently, acting as a massive MIMO BS coordinated by one or more Central Processors via the fronthaul network.
Key Differences Between Cellular and Cell-Free Networks
The following table summarizes the key differences between cellular and cell-free networks:
| Parameters | Cellular Network | Cell-Free Network |
|---|---|---|
| Definition | The area to be served is divided into sub-areas called cells. APs are deployed in each of these cells and communicate via backhaul. APs are interfaced with the core network (CN) to serve users. | No concept of cells. The entire area is covered by APs spread throughout. APs are connected to a Central Processing Unit (CPU) using cables (fronthaul). |
| Interference | Interference exists between neighboring cells. Mitigation techniques include fractional spectrum reuse or reduction in transmit power. | No cells, hence no inter-cell interference. |
| Visibility | APs, in the form of cell tower antennas, are visible. | APs are generally not visible to users. |
| Distance | The distance between APs (cell tower antennas) and users is generally larger. | The distance between APs and users is smaller. |
| Signal Strength | Signal variation is significant, potentially leading to weak signals for some users. | Signal variation is reduced due to the proximity of users to APs. Signals are typically strong for most users. |
| Signal Blockage | Sensitive to signal blocking. | Less sensitive to signal blocking due to the use of metasurfaces (Intelligent reflecting surfaces) and the deployment of APs throughout the coverage area. |
| Service Quality | Non-uniform service quality. Users at the edges of cells may experience lower performance compared to users near the cell tower. May be adequate for voice but not ideal for data. | Uniform service quality across the entire network. |
| Antenna Design | Directive antennas (~ three numbers) with higher directivity are used in 4G LTE networks. 5G uses massive MIMO antennas (~ 64) which provides strong and adaptive directivity. Nearby users get good performance, but edge users experience poor performance. | All users receive good performance due to the proximity of distributed (omni-directional) antennas (radio stripes) everywhere. |
| Cooling | Required due to high power transmission to cover longer distances and address significant signal strength variations. | Not typically required. |
| Architecture Design | Network-centric. | User-centric. Each user is served by all APs within its area of influence. |
| Joint Signal Decoding | Not present. | Present. |
| Performance | Generally poorer compared to cell-free networks. | 90% improvement in performance (CDF vs. spectral efficiency) in centralized versions. |
| Cost | More expensive due to the higher cost of Base Stations. | Cheaper due to less costly APs. Radio stripes are being developed to further reduce costs. |
| Control Complexity | Lesser. | Higher. Minimized using radio stripes based new architecture in which APUs (Antenna Processing units) take care of antenna units rather than single CPU (Central Processing Unit). |
| Wireless Connectivity | Good. | Better, and potentially more reliable than existing cellular networks. |
| Example Setup | 1. Massive MIMO using 4 APs with 100 antennas each 2. OR small cells having same AP locations as in cell free scenario | 400 APs on a square grid (e.g. 1km by 1km) using either centralized or distributed processing. |
