Preface
Technologies continue to evolve to manage diversified market needs. With the advent of 5G, the telco industry digitalization has already entered into a new transformation. Now, telcos focus is on customer experience rather than merely providing connectivity. Data processing near the edge becomes the core behind this implementation. Edge computing is not a new approach. However, the inclusion of network functions virtualization (NFV) and related distributed cloud radio access network (RAN) architecture are new. These technology features can enable low-latency services on a level not even considered five years ago. Edge enables 5G ultra-reliable low-latency connectivity (URLLC), helping to transform telcos from “Digital Service Provider” (DSPs) to “Experience Service Providers” (ESPs).
MEC is designed to enable flexible and rapid deployment of new applications and services for customers, as it is introduced to run the applications and performance-related processing tasks closer to the cellular customer. MEC allows telco’s to open their network to authorized third parties, such as application developers and content providers for developing use cases and vertical-specific applications.
Some of the drivers of MEC are given below:
Diverse bandwidth emerging needs: With more devices moving on and off networks, and with more diverse bandwidth needs emerging, network providers are facing mounting management challenges. They are looking for a solution that can provide efficient bandwidth management and greater network visibility to support low latency/near-real-time requirement management, as well as high bandwidth management that not only provides support for conserving device battery life but also improves the overall experience.
Lower operational costs: 5G and other emerging networks need to support a large amount of data to be transported. So, if they follow the traditional solution approach, then they need to put additional infrastructure. Hence, they are looking for a solution which can offload the traffic and get better control over data triage, normalization, residency, and privacy through the ability to keep more data local, rather than requiring to transmit it across the network to the core. Besides that, the solution should support unlicensed spectrum as well as multiple access technologies to lower the operational cost.
Network simplification: This large amount of data processing also requires greater automation, autonomy, and support for the following features:
- Hosted model, cloud-like experience
- Multi-workload with multiple layers of control
- Centralized control & telemetry consolidation through a hosted controller
Support for emerging players: CDN, V2X providers, drone providers and many other vertical players are emerging. Earlier, they were looking to invest in MEC infrastructure, but now they are following different strategies. Some are relying on telcos to support their specific data communication & processing needs while others are partnering with silicon vendors.
MEC promises to enable a plethora of vertical and horizontal use cases. Some of the below use cases explain how these drivers can be managed in actual business scenarios.
- Network Optimized Video on Demand
Sending video streaming from cloud-based centralized servers to end-user devices always had challenges. Within a geographic area, when many subscribers are watching the same news feed, TV show or movie, then one can imagine how much load is generated over the transport network. Now the same scenario can be optimized by utilizing MEC. The video content can be sent only once from the centralized video server, cached, and distributed locally to all the subscribers. Management and control information can be exchanged between the subscriber device, the MEC application, and the centralized application for authorization, settings, and content requests.
The benefits for the telco industry firms are reduction in traffic at the transport and core networks. For the subscriber, the Quality of Experience (QoE) is enhanced because of the shorter path from the server to the device, thus improving the latency, jitter, and reliability.
- Assist in Intensive Computation
One of the challenges faced by evolving devices (smartphones) is improving UE battery life. Many evolving scenarios/use cases require a huge amount of processing to be performed in real-time, such as mobile gaming, robotic image recognition, security applications, etc., which are typically performed by end devices.
Assist in Intensive Computation
This increases the cost and complexity of the device, which hinders the device's marketability. MEC can provide an instant fix to these scenarios. This can be done by outsourcing the processing need of these applications to the edge server, which can then send back the result. For example, a robot identifies a ball, image recognition software is run in the MEC, and the final result, the information that ‘it is a ball’ is sent back to the robot. Similarly, in smart surveillance systems, if an abnormal sound like bomb explosion or gunshot is detected, then a video recording of that specific duration will be immediately analyzed, and important evidences can be sent back to the server. Also, the current AGV works with preconfigured scripts and follows a static path. With the help of MEC, dynamic routing and auto stop/start features can be implemented. In this way, MEC reduces the complexity and cost of the device, while improving battery life.
- Augmented Reality & Virtual Reality
To function properly, both augmented and virtual reality require the fastest possible response times and low-latency communications, making multi-access edge computing an ideal partner for them. This, alongside further development in the applications of augmented reality and virtual reality technologies could bring about the next generation of remote maintenance, gaming, and entertainment systems. Augmented reality can help in a maintenance site, art gallery, sporting event, aquarium, zoo, or any other instance where additional information relevant to the context adds value to the quality of experience. In this case, the camera captures the point of interest, and the backend processing needs to arrive at the precise location and direction the person is facing through positioning techniques, or the camera, or both. After analyzing this information, additional information relevant to the context may be provided in real time to the user.
Augmented Reality & Virtual Reality
Considering the pandemic situation, where virtual tour needs to be inducted or remote maintenance support is required, then virtual reality solutions require real-time streaming support. These require very fast response times in near-real-time to provide the user with a realistic experience. Hence, in any AR/VR scenario, many MEC characteristics like low latency, large quantity of data processing, access to location details, bandwidth savings, and improving QoE of the user are required.
- Enterprise-Centric IoT Use Case
The Internet of Things (IoT) unleashes a plethora of messages. To streamline these messages according to latency, security, etc., different gateways with a grouping of sensors are needed for efficient service. These devices may be connected using any access technologies — Wi-Fi, BLE/Zigbee or LPWAN, etc. Typically, the messages are small, encrypted, and employ different protocols. The MEC server provides capabilities such as processor, memory capacity, etc., to resolve the challenges. These servers can perform aggregation and distribution, analytics, decision logic based on analytics, database logging, remote provisioning, and access control to the end devices.
Enterprise-Centric IoT Use Case
MEC technologies can also improve safety levels in industrial environments and provide analysts with real-time information regarding tools, equipment, machinery, vehicles, and environmental factors to keep workers safe. When it comes to emergency services, access to real-time data and information, as well as a reliable means of communication can be the difference between safely responding to a situation or the unfolding of a disaster. MEC also provides advanced surveillance and video analytics solution that gather and collect data closer to the source in the industrial environment.
- Connected Vehicles
MEC platform services can provide information to the V2X applications — for example, location updates to the advanced navigation system, thereby providing the car with updated data on traffic situations. One of the major developments across the world in the transport sector, is autonomous driving and connected vehicles. Connected vehicles may be connected using Dedicated Short Range Communications (DSRC) for short distances or via 4G/5G wireless for long-distance connectivity. Connected vehicles increase road safety, reduce traffic congestion, provide value-added services such as car finders, parking location identifiers, etc. As the need for these applications is real-time communications, MEC's low latency characteristic can be put to good use.
Connected Vehicles
The V2X apps provided updated information to the car, such as advanced routing for traffic avoidance. When the MEC applications start to experience high processing rates (e.g., high CPU usage), it is reported to the MEC Application Manager. These KPI reports can be sent to the ME orchestrator. The ME orchestrator uses policy rules to determine that more application capacity is required and instructs the ME system to instantiate additional VNFs/CNFs.
- Video Acceleration
Most current video streaming services depend on Transmission Control Protocol (TCP) congestion control to evaluate the video coding options that may be employed for the best delivery of video content. With access to information about the quality of the radio link and radio measurement reports made available on MEC, applications may use this information to assist in TCP congestion control decisions like deciding the initial window size, setting the value of the congestion window during the congestion avoidance phase, and adjusting the size of the congestion window when the conditions on the radio link deteriorate.
Video Acceleration
Guide the streaming server to better estimate the quality of the radio links available to the UE and thereby make better judgments on coding the video, so that it matches the estimated capacity at the radio downlink.This enhances the quality of experience for the end-user and provides an opportunity to maximize the usage of available radio resources.
Conclusion
To summarize, MEC use cases and applications are capable of generating value for all stakeholders. Telco industry can extend MEC use cases implementation as a service. End enterprises are getting reliable & robust network infrastructure to implement their next-generations solutions. Similarly, it helps players like AR/VR devices or AGV providers to offload the computation hardware, which will result in cost reduction and additional feature implementation. This will increase the adoption rate, and ultimately result into productivity gain. The end goal of MEC is to provide an optimized and low-latency computing infrastructure with deployment agility that can scale horizontally or vertically based on requirements. With MEC, we can move services and content closer to end-users and get more QoE, QoS while reducing backhaul congestion and optimizing gateway interconnectivity costs.
