Brief introduction to the design and implementation of carrier communication center

takeaway:

With the continuous upsurge of global mobile traffic and the continuous extension of communication cloud in vertical industry, Internet communication cloud has gradually become the mainstream communication service form. The COVID-19 pandemic has also stimulated users’ online demands. Whether it is personal life or corporate office, it has become a trend for communication activities to move from offline to online, thereby deriving new consumption habits. In this context, China’s Internet communication cloud industry is facing new opportunities and challenges. In addition to meeting users’ more stringent requirements for interactive experience and exploring innovative scenarios, it is also necessary to quickly provide enterprise customers with “thin and complete” solutions to deeply integrate communication capabilities with application scenarios.

Yang Xianjun Architect and Technical manager of NetEase Yunxin

preface

Amid the rise of compelling videoconferencing, artificial intelligence and robotic automation, there is a silent, emerging disrupter: the humble phone call and SMS messaging. 2022 is a year of rapid growth in the communications business, with the volume of calls and text messages already increasing significantly. Messaging is the most basic communication business and the entry point for services and marketing in many industries. In the ERA of 5G, SMS is upgraded to 5G messages, bringing new opportunities to various industries, among which the financial industry will be one of the most typical application fields of 5G messages. Through the centralization of carrier communication infrastructure, voice calls, smart calls, text messages, video messages, smart messages, 5G messages, and RTC and IM integration are realized. With an open and agile architecture, enterprises can improve service efficiency and achieve integrated precision marketing. Combine with line/media management platform, intelligent marketing system, all media intelligent customer service system, etc., expand service application, build a complete communication ecosystem and communication chain closed loop. Comprehensive management of carrier communication, reduction of carrier communication cost and establishment of carrier communication ecosystem will be realized.

One, see through the fog: uncover the veil of carrier communication center

For the field of Internet communication, zhongtai is more a subject involved in the construction of 2B products, because the abstract reuse of software system is more a problem faced in the construction of complex B-end system. Therefore, mid-stage design is a topic that all B-side system architects should pay close attention to. How to design the carrier communication center? What are the characteristics? What is the nature of design? What are the challenges? This article will review the construction of communication center from a new perspective, so that you can understand the essence of communication center more deeply.

In the field of software development, there is a saying that “any problem in software design and development can be solved by adding a layer”. Here we do not discuss its right or wrong and scope of application, but it can be determined that the emergence of the middle stage is to solve the contradiction between the front and back stage operation efficiency, through the middle stage of the change gear connection front stage and back stage, eliminate the difference in efficiency between the two, so as to achieve the overall balance of the system.

I very agree with NetEase vice president Wang Yuan’s concept: “all are business in Taiwan.” In a broad sense, the so-called middle stage is for business services, in order to enable enterprises to quickly respond to business needs and launch new products with lower cost and higher efficiency. Carrier Communication Center is no exception. It belongs to a subset of carrier communication Center. It mainly processes carrier basic services such as voice call, VoLTE video call, short message, video message, AND 5G message, and re-processes the output results for services. Essentially, carrier communication ability is the source of business, communication operators China send a DE of each SKU is the carrier of the business, can let the enterprise fast and efficient solution of copy, so that case reuse, but the same principle come in different forms, thus realize the improvement of production efficiency, achieve the goal of authors efficiency.

(a) what is the operator communication center?

Jian Wang from ThoughtWorks has a succinct definition: China Platform is an “enterprise capability reuse platform.” In searching for various materials, the author found that there are many related concepts of “Zhongtai”, but this definition is more inclusive and clearly delineates the scope and key values. ** “enterprise platform” represents the platform is multi-faceted coverage, overall planning of the top mechanism, rather than a single system or service system; “Capability” is a highly abstract description of different technologies, and various capabilities such as business, technology, algorithm, data and AI are combined into enterprise multi-dimensional capability network. “Reuse” is the core value of zhongtai, the core essence of more agile and lower cost operation of enterprises.

Operator communication in Taiwan, I think it is unprecedented, here I also make a definition for its birth: Is the result of basic operators offer line voice, video and other media resources, text messages, RCS, such as text and rich media news resources, combined with the RTC real-time audio and video communication, IM instant message communication, coupled with infrastructure and services integration and scheduling, and multi-tenancy system with distributed cluster management, follow the design concept of domain driven model It provides the integration and precipitation of various commercial solutions, and ultimately serves various concrete requirements of enterprises, thus forming a unified platform of reusable capabilities.

(2) Characteristics of carrier communication center

1. Reusability

What the center provides is necessarily the general function that different products or different enterprises can use. If it’s a product-specific feature, it doesn’t make sense to reuse it as a feature. In terms of reuse, it will also comprehensively evaluate function reuse, data reuse, industry reuse and so on. The existing solutions of carrier Communication Center, such as the high-connection fusion solution launched by RTC, can be applied to entertainment and social networking scenarios. Cloud call center SKUs can be provided to various customer service SaaS platforms as call center core scenarios. Intelligent voice robot capability integrates ASR, TTS, NLP and other general intelligent technologies of major manufacturers, which can be provided for all kinds of marketing call and call voice robot SaaS platform to support capabilities. Voice notification, voice verification code, 5G messages, video text messages can provide enterprises with a variety of access services and information exchange satisfaction evaluation services. General outbound/inbound call can be integrated into SCRM, ERP, HIS, IT systems and so on as a general value-added service for government and enterprise office to help traditional enterprise Internet + upgrade and transformation. All the cases can be batch copied and applied in various scenarios.

2. Platform

In order to play an effective role, zhongtai needs to empower multiple product lines, so it emphasizes platformization and serves the whole product ecology. Therefore, in the design of Taiwan products to try to consider facing a product ecology, your product can meet how many common needs, and for different industry ecology, your product can meet what common scenarios. At present, carrier Communication Center mainly discusses the open platform scenario of the combination of the core switching network of the three major carriers and innovative services; Cooperate with medical enterprises to explore scenarios such as medical consultation, chronic disease tracking, patient return visit and doctor-patient communication; Provide customer service system capability and office system upgrade for financial industry; Provide high touch rate for entertainment and social industry, complete the last mile of communication; For the e-commerce industry to provide business and consumer communication, marketing promotion, customer relationship management and other scenarios. A series of common application scenarios have been extracted, standardized and platformized for government and enterprise, medical care, finance, e-commerce and other industries.

3. Business

In the past period of time, it has been widely discussed whether the products in Taiwan should highlight the business, but in fact, the concept of Taiwan itself is put forward to reduce the business cost, business product resource sharing, data interchange. Therefore, the mid-taiwan products that serve business must have their own business, that is, business functions can realize different business scenarios, and data can be analyzed based on different business dimensions. Carrier communication medium platform is not only a kind of communication capability, but also a layer of abstraction based on the actual customer needs by combining the above four industries, with certain common service capabilities. Through separation, layered model on basic ability can not interfere with each other to encapsulate the various types of industry related application scenario for the aggregation ability, such access will greatly simplify the complexity of the logic in the enterprise, can be used directly for common ability, thus to reach the company can focus on business needs, without having to master the specific technical details, reduce the coupling and understand the business cost, Achieve the goal of raising productivity, reducing cost and increasing efficiency.

4. Standardization

The medium platform needs to provide standardized functionality, and if your communication module functionality only serves a single industry, as a medium platform functionality, there will be very few reusable scenarios. This is not to say that your middle platform business module can only be unchanged, but can have custom extended functions, to meet the needs of multiple foreground support for this function in different scenarios. Carrier communication center has extracted a set of standardized capabilities and at the same time refined a set of standardized methods.

  • Access and egress interface standards: Provides a standard set of access standards for backbone networks such as E1 cable, NGN core network, and IMS core network, as well as standard protocols such as SIP, H323, CMPP, SMGP, and SGIP, and non-standard hot-swappable component standards for the access of carriers or secondary agents. Supports standard codecs such as OPUS, G711, iLBC, G729, H264, VP8, and VP9. Supports standard push-pull flow protocols, such as RTMP, HLS, and FLV. Supports standard pluggable components, such as ASR and TTS, and supports standard MRCP. For the forward open interface, it also outputs the standard RESTFUL network open interface by referring to the general specification of NetEase cloud communication open capability. For the standard system, it also provides seamless connection between SIP and CMPP and other general carrier interface standards. A set of docking process specifications and standards are defined, which are followed by the implementation personnel, thus reducing the cost of front-to-back docking and improving the docking efficiency.
  • Standard of business areas: follow the micro ecology and DDD domain driven design of ability to service module division, made the model and the principle of architectural layering, defines the responsibilities and role of function module, rely on the hierarchical relationships and depth, reverse dependence of decouple the tactics and micro service governance norms, the construction of the infrastructure, technology components of application and extension, etc.
  • Standards in the R&D process: also align the development collaboration specification, deployment specification, logging specification with existing standards, and explore further. The main purpose is to solve the contradiction between the stability and flexibility of the system, the fuzzy business boundary between the center and the forward business, the distance, the differential contradiction of the r&d life cycle and so on. As a mediator of conflicts and contradictions to solve problems and improve productivity.

(3) The nature of the operator’s communication medium

Enterprise-level, abstract, sinking, reuse, these keywords represent the essence of the construction of carrier communication center. At the same time, it is also a problem to be considered deeply in the design of enterprise application architecture. (enterprise application architecture, refers to the enterprise internal each software system, should be in the form of what construction, combination, so as to efficiently support the enterprise the management operation), therefore, if you want to deep thinking about the enterprise abstract, sinking, reuse of software products, can undertake new look from the following three angles, respectively is: Based on the perspective of abstract reuse, based on the perspective of architecture rationality, based on the perspective of business unified management. I don’t want to go into that here, but we’ll talk about it in an example.

Two, the dream: interpretation of the carrier communication platform design scheme

(I) System technical characteristics

1. Follow the micro-service ecosystem architecture and provide open and standard network application programming interfaces

  • The feature of it is that the third party users can develop a variety of services without knowing the specific technologies and protocols adopted by the basic network as long as they know what capabilities the basic network can provide and make use of these capabilities, so as to provide colorful services by using the operator’s basic network. For example, the service logic of location-based services, call center services, and switchboard services resides on application servers.
  • Open API interfaces are provided to facilitate the secondary development of services and the introduction of third-party application developers to provide services.
  • Equipment suppliers and operators can work together to develop new services that are needed by the market.
  • Operators can develop new services independently and quickly according to market demand.
  • Operators and third parties can cooperate to develop new services with market prospects.

2. Unified communication platform across PSTN, GSM, NGN, IMS, and IP networks

  • To meet the integration of traditional services and data services, CTCS provides multiple protocol adapters, such as SIP and SMPP, and supports the access of multiple network devices. CTCS can build a unified service platform across PSTN, GSM, NGN, IMS, and IP networks to provide services for users on these networks.
  • The carrier communication center shields the details of the underlying bearer network and provides a flexible networking mode, which can keep the original access mode unchanged, reduce operation and construction costs, and facilitate the development of comprehensive services.
  • Carrier communication center platform is developed on the basis of Internet application and IP network. At the beginning of the design, seamless integration with Internet application was taken into account, and the unified business platform can perfectly connect with the carrier’s basic network.

3. Richer business

  • CTC takes full advantage of the IP to user features of NGN network to provide value-added services combining voice and data, which is more competitive than pure voice services in PSTN network and supports voice, data, and multimedia services in a more comprehensive manner.
  • It provides enterprise users with point-to-point voice and SMS services, call centers, contact centers, voice robots, multi-party calls, conference calls, and enterprise switchboard services that integrate with the Internet, voice verification codes, and message notifications. Become an important guarantee for operators to attract customers and win profits.

4. Flexible and powerful billing function

  • With flexible and powerful billing function, it can provide a variety of billing packages to meet the needs of various consumer groups and complete billing data generation, storage and transmission functions. Charging by session duration, charging by item, charging by calling number, called number.
  • Implement charging functions based on charging information and charging rules of specific services, and generate charging records of fixed content based on service requirements.
  • You can also generate charging events or charging reconciliation files. A third party can directly collect accounting reconciliation or perform real-time charging.

5. High reliability and performance

  • Carrier communication center adopts advanced micro-service ecosystem to realize service discovery, automatic failover, fuse, downgrade, cure, etc.
  • Unlimited horizontal expansion performance without attenuation. Compared with hot and cold backup solutions, resources can be fully utilized and the requirements on the hardware host are lower, ensuring high reliability and high performance of the system.
  • The system is divided by service functional domain and runs on different application servers in distributed mode to reduce fault coupling between systems. When a service fails or the number of calls is heavy, other services are not affected.

6. Mature and reliable products meet the needs of operation

  • At the present stage, all the capabilities provided by CARRIER Communication Center are independently developed and run stably, with obvious advantages in maintenance, modification, customization and response time, which can fully guarantee the smooth operation of services.
  • Since July 2017, the system has been running online, carrying tens of billions of short messages sent online, billions of calls, millions of users online, tens of thousands of concurrent sessions, and thousands of enterprises in normal operation.

(2) Overall structure interpretation

​

This block diagram covers the overall picture of the platform in operator communication. The mode direction is divided into five levels, and the vertical service governance is divided into two latitudes. The five horizontal layers are access layer, gateway layer, application layer, capability layer and base layer respectively. Three steps of progressive polymerization were carried out.

The first layer is a fundamental resource operators, cloud based communication letter RTC, AI related providers and integrating research ability and so on, through the line business platform for the distribution of resources to form a resource collection to solve a single resource stability, concurrent ability, blind area, and the optimal path selection, resource isolation, to do a unified scheduling and distribution management.

The second layer is to encapsulate these resources into discrete standard capabilities with the smallest granularity, highest composability and weakest correlation, which do not reflect the upper business and application performance. The purpose is to hide the technical details of operators, RTC, AI, etc., while providing atomic basic capabilities.

The third layer relies on the atomic ability of the second layer to carry out polymerization packaging according to business requirements. The emphasis of this layer is to be flexible and follow product requirements. It is heavily involved in the design, extraction, and encapsulation of business applications. Generate a set of application capability, can not interfere with each other, or can be linked to each other, to support rapid product trial and error and agile iteration.

  • Access layer

    1. It mainly refers to the SDK of various platforms. At present, it is mainly based on WebRTC development for terminal SDK.

    2. Another is the aPaaS call toolbar, which can be easily embedded to reduce development.

    3. The resources of the three directly connected carriers are also connected to SIP through the access layer.

    4. The front and back ends of the enterprise can also access the gateway directly to receive cc events or initiate calls or short messages.

  • The gateway layer

    – OpenAPI is mainly an external capability open interface, with degraded fusing frequency control and other strategies. OpenAPI can be packaged in accordance with cloud open specifications and standards.

    – * -gatewayrestAPI This API is a series of application gateways that encapsulate the internal Dubbo microservice interfaces and present rest interfaces externally, providing authentication and interface aggregation functions.

    – WssLink is a WebSocket long connection. It pushes tenant notifications or signaling to the SDK and improves server performance through push-pull.

    – Zuul and Eureka are Gateway service managers. Zuul and Eureka manage route distribution of various gateways, implement visual configuration, reduce configuration on Nginx, and dynamically distribute subdomain routes, focusing on fuses limiting traffic frequency control and unified authentication.

    – The other layer is the SIP to carrier access layer. Part of this layer is the proxy forwarding of SIP signaling and the other part is the proxy of media (depending on the actual network architecture).

  • The application layer

    1. Application domains are divided into domains to logically isolate each other. Currently, five domains are expanded: cloud call capability domain, universal voice service domain, AI-related intelligent call domain, short message service, and 5G message.

    2. The cloud call service domain provides common inbound, outbound, call control, interactive voice response, automatic call allocation, call queue management, call routing and other related capabilities, which can be expanded according to the features of the cloud service platform.

    1. General Voice Service Domain: This field mainly implements simple IVR broadcasting. VoIP point-to-point calls connect the telecom network to the Internet. Two-way callback The server sends instructions to control the carrier-to-carrier callback mode, and encapsulates implicit number logic through number pool functions. Encapsulate the voice authentication service logic and other simple point-to-point service capabilities by delivering, reporting, and verifying the verification code.

    4. Intelligent call service domain: notification robot, single-wheel and multi-wheel, mainly relying on NLP and ASR and TTS capabilities of communication capability domain. ASR is currently connected to Ali Cloud ASR and NetEase Hangyan ASR. TTS is connected to Alibaba Cloud TTS, NetEase Hangyan TTS and local TTS of our own research, realizing man-machine assistance, intelligent quality inspection and other business functions.

    5. Other business capability domains: the parallel domains are independent and decoupled from each other, and the communication capabilities of the lower domains are shared. Emerging capabilities can be expanded according to the development dynamics of business capabilities.

  • Ability to layer

    It mainly provides a set of communication capabilities for operators. Internally, it can be divided into two layers from the perspective of application. The first layer is the basic capabilities, and the second layer is the resource integration platform for line operators.

    To differentiate from the Angle of function module can be divided into: tenant specific account system, rights management and dispatching center dispatching load balancing, many regional scheduling, etc all kinds of media, media center, mainly is the edge of the media link signaling routing, relay agent distribution, basic ability mainly include all kinds of system level among components and plug-ins or module on the media services.

    – Tenant Management: This is mainly about the construction of SIP account system. SIP inbound or linkage operations require an account that can be logged in and registered and a heartbeat link to keep alive. Therefore, it will establish its own account system for communication, and there is no need to enter the account system for two-way callback or point-to-point simple calls. Tenant management account allocation is a logical tenant concept. Media allocation supports logical isolation and physical location isolation of media services. It also includes the management of the trunk number and the management of the black and white list to do part of the risk control processing. The online billing system is mainly used for settlement reconciliation with the line operators or operators, as well as the tenant enterprise prepaid mode.

    – Scheduling center: media-service resource scheduling is implemented in two scenarios. One: Media groups used by different service domains are scheduled between application clusters based on the association between tenants and application clusters. Second, the same service capability is used for load balancing according to the cluster survival status or health status. Capability signaling agent layer is the interface layer between capability domain and application domain, which is mainly responsible for signaling packaging, conversion and transfer or callback. The media signaling management and control service is mainly the assembly of the event notification from the translation capability signaling to the softswitch signaling and the signaling generation and aggregation to the softswitch service.

    – Media Center: The current architecture of the media center is the same as that used by NetEase Qiyu. The media center is divided into tenants, application functions, and clusters, and can be supported to the tenant application level. Media, routing, and relay services require the development of a visual cluster management interface to manage various local configuration files, with the main design remaining unchanged.

    – Basic capability: includes components or basic capability services such as configuration center, distributed task, distributed timer, and various self-developed AI capability modules of softswitch, supporting the basic capability layer or each capability domain and application domain.

    – Line management: mainly refers to the resource scheduling management of line No., which will be specially explained below.

  • Base layer

    It mainly includes data access layer, relational data storage, K-V data storage, object storage, data retrieval and so on. In fact, the base layer also contains some line resources and the hardware gateway equipment of hybrid cloud, which together constitute a basic resource network for the line operator platform to carry out resource scheduling and data storage of the capability layer.

  • Run domain

    This area governs application functionality and capabilities at a macro level. It is mainly a variety of operation support systems to support colleagues in the forward department, such as operator management platform, billing and settlement platform, and ability opening management platform. The data analysis platform provides forward analysis and statistics of relevant business data as reference data for decision-making.

  • Operational domain

    This area governs the availability and ease of operation of service clusters on a micro level. Convenient main research and development, technical support, testing of students management visualization system, online problems, performance optimization provide monitoring data or performance indicators, quality of service functions such as monitoring management operations, also built a set of micro service governance management tools, including service management, configuration management, data query and alarm management, link tracking, target center, etc.

(3) Interpretation of network structure

The figure below mainly shows the network hierarchy distribution of operator communication medium in a privatization project. The combination of three networks is as follows:

  • Operator communication backbone network: It refers to the operator’s core network access point. This layer of network basically belongs to the operator’s internal private network that is different from the Internet, which generally refers to GSM, NGN and IMS. The end office is basically closed to the outside, and all external resources through DC1 and DC2 must be strictly controlled by SBC edge control node before opening to the outside. Small customers open resources through analog lines, digital E1 lines, IMS lines, and VoIP lines.
  • Intranet communication area: generally refers to the internal network of an enterprise. It stores service clusters at the application layer, capability layer, and gateway layer. If there is a network requirement, the media layer is also placed at this layer and the SBC is used as a media proxy to access the public network.
  • Enterprise DMZ mixed zone: Generally stores application clusters related to the access layer. If the enterprise does not have requirements, the media layer can be directly stored in the mixed zone without the media proxy process of the SBC.

With multiple equipment rooms and multiple activities, the whole equipment room can be failover. Global distribution of media endpoints CAN be achieved through we-CAN global communication network. Through the weak network resistance capability of YUNxin RTC, the voice quality of each application scenario is 40% higher than that of the traditional architecture model, and the weaker the network, the more obvious it is.

The importance and importance level of the application are defined for product, r&d, and test by layering the application and defining the application level. The call dependency and concurrent performance requirements of the application are determined.

The details of how the three networks get through media exchange can be continued.

​

(4) Media exchange layer interpretation

The core of this center is the construction of media layer, signaling and media scheduling management based on abstract multiplexing. Rely on repetitive functions and modules to abstract and sink one layer. What is abstract? What is sinking? You can refer to the picture below.

​

There are multiple subsystems that access, access, and edit media. Have rely on operating exchange, after analysis found that media processing functions highly similar to repeat, can completely abstract merged, avoid redundant construction, so there will be a core signaling and media processing high availability cluster, and then according to the different business divisions subset, have unified routing scheduling policy, which has reached the consolidated unified abstraction, and can do moderate operations.

  • The media layer of communication capability platform construction is divided into five levels to carry out transmission, including access terminal, WE-CAN communication network, converged communication gateway, audio and video media exchange capability domain and line supplier platform.

  • The access terminal mainly provides a carrier communication capability SDK, which encapsulates a layer of services on top of the existing RTC-SDK of cloud communication. The media transmission adopts the existing way, packaging the transmission capability based on the existing interface. Business signaling USES independent link connection on the encapsulation of call center, login, online management, session management, call management, called event management, event management, sending DTMF, keep call, mute, call forward, multi-party conference related basic module, these modules connected directly to the back-end call ability domain related application services, To drive the overall operation of the application layer and media layer.

  • We-can communication network is mainly responsible for the network quality optimization of RTC audio and video. Including weak network confrontation correlation, packet loss and retransmission mechanism, QoS dynamic detection and optimization adjustment, supporting point-to-point media negotiation and transmission. Key control of media quality, and signaling bridge with the fusion gateway layer through WSS + JSONRPC.

  • The layer of fusion communication gateway is the key to fusion. It is a bridge between THE WE-CAN communication network and the calling capability domain. It is only responsible for point-to-point signaling negotiation, and some specific interaction flows will be described in the detailed design later. For the sake of stability and capacity, converged communication gateway does not do media processing and media proxy. It only does signaling negotiation and transformation, and then exchanges RTP ports negotiated bilaterally to each other to start media transmission. Another use of this gateway is to address and route the WE-CAN communication network to discover available MediaServer.

  • Audio and video media switching capability domain is the main body of the carrier communication medium. This architecture is composed of distributed, high availability, high concurrency media cluster and extensible business layer.

    – The media cluster consists of the access layer, routing layer, and media processing layer. It uses SIP-proxy as a Sip signaling Proxy to receive events or messages such as Register and Invite sent by the converged communication gateway. It uses the POLICY service to discover and Register available nodes in the routing cluster. The available media nodes are discovered by addressing the policy service, and the RTP ports of the WE-CAN communication network are exchanged with the RTP ports of the media node, and then point-to-point audio and video calls are made bilevically.

    • All events generated during access and call control are converted by the event control service and transmitted to the upper-layer application service domain. The service domain processes its own status, records data it cares about, changes its own status, or initiates control on the platform side.

    – Through signaling interconnection of the SIP-Trunk service, available nodes at the media layer are connected to the carrier platform, outbound media lands, incoming requests from carrier platform 400 are received, and the media node is connected. The agent addresses the media node through the control of the application service domain, and finally an RTC terminal is connected to the called party.

  • Line is the primary responsibility of the business platform: direct backbone of trunk line and operators and enterprise docking pull line, number of risk control management, billing line side, and the incoming breath process bridge consultation, circuit platform will manage and control circuit of concurrent and optimal path selection, as well as the connection of detection and abnormal line and avoid blind spots. The maintenance and management of resource pools can solve single points of failure and line bottlenecks through various scheduling policies to meet the performance and stability requirements of single access operators.

(v) Interpretation of terminal access layer

As shown in the figure below, the NECCC SDK architecture is also a good hierarchical model design. It mainly directly connects with the signaling service and media service of the carrier communication medium platform, encapsulates and reduces the detailed process, and provides the functional interface of closure externally. It is divided into six levels: base layer, engine layer, cross-platform encapsulation layer, SDK interface layer, platform terminal layer and easy to use system. Through these layers of assembly, a lightweight, extensible application terminal SDK library is formed. Mainly responsible for audio and video collection, editing, optimization, transmission, signaling negotiation and interaction and part of the core business logic processing.

​

  • The basic layer mainly consists of basic network library, encryption and decryption algorithm library, management modules such as thread concurrent processing, log, and resource layer unrelated to business.
  • Cross-platform encapsulation includes framework and component layers such as Uniapp, Ajax, WebSocket, and Signalling to enhance compatibility with different platforms.
  • The SDK interface layer mainly includes the end-to-end capability interface and signaling interface encapsulation of the carrier communication medium, the establishment and destruction of the user agent layer, the session management of RTC, and the processor of some asynchronous events, etc., which provides the standard interface for the upper platform terminal.
  • Platform terminal is mainly adapted to Web, small programs and other related performance layer applications, and can be extended to more other terminals in the future.
  • The ease-of-use architecture includes sample code, common components, solution demos for enterprise developers, easier access, and functional demonstrations.
  • The engine layer is shown below, the core of SDK, which mainly includes audio engine, video engine and QoS engine. A variety of audio and video algorithms are integrated to carry out weak network countermeasures such as packet loss, delay and jitter. Some AI technologies are also extended to collect sample data from SDK to build learning models, so as to achieve the best effect of self-regulation of network fluctuations.

​

(6) Interpretation of line resource layer

Based on the perspective of unified business management, the line operator platform mainly plays a centralized resource management and scheduling, so as to avoid the physical failure of a single line, improve the stability, improve the concurrency ability, and break the concurrency limitation and regional limitation of a single operator. Access as many line resources as possible backward. Forward to provide internal business lines, external to provide agents and enterprises to provide optimal line resources, the more important point is to provide line resources on the open platform of forward operator ability application to construct applications and solutions that fit the needs of enterprises, so how does it do? Let’s move on.

​

  • The input and output of the platform are UDP/SIP/RTP protocol as a standard protocol, and whether IMS or VoIP lines between operators are SIP standard, in the room to the room network performance is good. At present, analog lines and E1 digital lines are converted into standard SIP in the remote machine room through the hardware trunk gateway, and then connected to the line supplier platform on the cloud to achieve unification.
  • This platform mainly includes the management configuration visualization platform, softswitch self-developed module to achieve the synchronization of management configuration application, proxy service to do load balancing and high availability cluster management.
  • Load balancing SIP302 forwards traffic in jump mode, allocates traffic based on actual load, and supports quantitative weighted random allocation of a single softswitch host, random allocation, and sequential allocation. For details, refer to the microservice discovery mode.
  • Softswitch service developed module to synchronize and apply dialling plan, number change, access gateway and the creation and use of the gateway. Supports active push synchronization after configuration changes, supports hot loading of configurations, supports media and media transparent transmission modes, and supports configurable SIP header parameters.
  • The configuration management service provides visual display, configuration editing, billing logic, tenant management, number pool management, and intelligent routing rule management. As shown below, packet capture analysis, data statistics and other functions are being improved in the future.

​

In the industry, such a line centralized management platform is relatively rare. It may be highly professional, but the command line of the crowd concerned about this area is sufficient, so it has not been developed. At present, only VOS system has a more comprehensive visual line management docking platform, after cloud letter is also counted as one, welcome to consult and build.

(7) Application ability layer interpretation

Based on the perspective of architecture rationality, part of the previous theory of application capability layer has been mentioned, which is mainly divided into atomic basic function encapsulation, general basic application capability, and part of the solution suitable for business. Spent a lot of space above all in the part of an exchange of audio-visual media circuit part of resource management, on the two parts is the ability to apply layer, application layer are follow the service of ecological system, micro service related about the internal structure is not here, here mainly take 5 g news platform related applications for an interpretation of the diagram below:

​

​

NetEase Intelligence has obtained the QUALIFICATION of 5G messaging CSP of the three major operators in 2021. The 5G messaging system architecture mainly includes 5G messaging center, MaaP platform and 5G messaging gateway, and is connected with user data management (HSS/UDM), SMS center, security control system and business support system, etc.

The 5G message center is responsible for processing 5G messages and connecting with THE MaaP platform to provide industry message functions. The MaaP platform connects to the 5G messaging center to provide industry messaging functions, and is responsible for Chatbot and message access management, authentication, and multimedia content uploading and storage in industry messages. The 5G messaging gateway provides inter-carrier COMMUNICATION between 5G messaging networks.

The core functions are as follows: CSP and RCS, contract client management, rich media message template editing, material management, message sending. I won’t go into more details here.

​

​

The difficulties of 5G messaging platform are mainly the various audit mechanisms, agent operation mechanisms, terminal equipment support and message degradation strategies, which have been defined in a complex way in the early stage, and state flow control needs to be handled according to standards. According to users’ operations on non-direct customers and Chatbot, we changed the status of corresponding entities inside our platform when synchronizing with operators, so as to help users understand the current information more intuitively. After performing an operation, you need to wait for the platform review and determine the operation that can be performed based on the review result. The process will not be described in detail, but the main purpose of this is to intercept any possible disturbance in the early stages of message sending. The following figure shows the internal state flow mode.

​

​

​

5G messaging has very strong application advantages. First, it has both the chat, group chat and small programs of OTT applications, as well as many functions of various service apps. Second, the traffic entrance, 5G messages directly occupy the mobile phone SMS this strong entrance; The third is the native application, no download, registration, login and other pre-operations, no binding mobile phone number, verification code collection and other actions, ready to use; Fourth, it is safe and reliable. Based on the strong association of mobile phone number and real name authentication, personal data can be interconnected among different applications to ensure data security and reliability. Fifth, cross-platform connection, users do not need to switch multiple apps, the same interface, cross-application interaction, inter-application connectivity.

Three,Floating world qinghuan: a brief taste of carrier communication in Taiwan practice case

Due to the length, the examples here only list a few innovation points related to integrated communication, and the details of the specific implementation of the case will not be expanded here. If you are interested in sustainable attention to NetEase Intelligent enterprise technology + public account, the following will share each case from the source to thinking, to design details, and finally the whole process of practice interpretation.

Cloud call center applications

Implementation effect: intelligent route allocation, accurate service; Service process monitoring, efficient management; Insight into customer needs, active marketing; System to assist manual, improve efficiency; Sales process monitoring and control.

Application scenario: Applies to customer service scenarios and telemarketing scenarios.

Refer to NetEase Qiyu:

qi.163.com/callcenter

Outbound robot application

Implementation effect: In response to massive scenarios, instead of manual autonomous call multi-round interaction, enabling enterprises to accurately identify target users, achieve new customer growth, improve retention, and recover loss.

Application scenarios: shoes and clothing, maternal and child, e-commerce, medical care, government affairs, public security. Refer to NetEase Qiyu:

qi.163.com/excallrobot

Converged call high-connection application

Effect: When making a call, a user can select an INTRA-application RTC voice call, a mobile phone number for direct calls, and a mobile phone number for transferring an RTC call when the call cannot be answered.

Application scenario: mainly used in entertainment industry line, as the last road call, improve the contact rate.

Reference the Demo:

Res.qiyukf.net/voip/7b6734…https://res.qiyukf.net/voip/7b6734f6d3c15a736611e50e1336894e

Converged call voice notification application

Effect: Users can receive voice notifications and listen to short voice messages after answering calls.

Application scenarios: It can be used for medical treatment, finance, enterprise administration, electric marketing and other industries, as a single point of one-time notification application, simple access and easy to use.

Reference the Demo:

Res.qiyukf.net/voip/ea4469…https://res.qiyukf.net/voip/ea4469d81d3fb35ea7a05b50204c8e17

Converged call VoLTE application

Effect: When making a video call, a user can select RTC video call or VoLTE video call. When the called phone supports VoLTE, the called phone can answer the call using video.

Application scenario: It provides a new possibility as the contact point of G2 audio and video calls. It is currently in the experimental stage.

Reference the Demo:

Res.qiyukf.net/voip/2e672c…https://res.qiyukf.net/voip/2e672c4056e670745577d2726a651ea0

VoLTE + Virtual human application

Effect: After the RTC is connected to the video of the mobile phone user, the action of the RTC user can be captured directly and displayed to the mobile phone user through the virtual image.

Application scenario: It can be used as a virtual reception or interview scene when the host cannot appear on the screen, such as virtual customer service.

Reference the Demo:

nullhttps://res.qiyukf.net/voip/359245ca2522154abe7538af0d6728f1

Iv. Dadao Futu: Imagine the application development of carrier communication in Taiwan

Carrier Communications China Is open to the outside world as a cloud PaaS communication capability, and will also be a long-term strategic partner of operators, committed to new technology research and development, aiming to help global operators meet new challenges and opportunities.

Carrier communication center is not only the construction of capability but also the construction of ecology. We will take openness, cooperation and win-win as the purpose, work with operators and partners to actively participate in the enterprise Internet + transformation construction. The development of the Central Taiwan is not achieved overnight, but a gradual process in which values are reflected. Just as Chen Li, general manager of Yunxin business, said: Every system and solution we do is to extract, translate, sort out and abstract the requirements from the enterprise, and then carry out the concrete packaging and integration of functions, and finally land in the enterprise to reflect its value. With the development of the system and in-depth understanding of enterprise applications, the following will be further discussed in the future:

  • Fine communication capability: Risk control, (video, IVVR, ringback detection such as refinement capacity expansion and application, in today’s information, information security will become one of the most important link, we will continue to explore risk control system, the construction of risk control system in the text have relatively mature, the risk control to explore in the field of audio and video we have a good direction, However, in the construction of voice call risk control, operators will continue to try to reduce costs and improve identification efficiency. Self-service interaction such as IVVR is introduced from the traditional CALL center IVR and applied to the video field to enable online streaming editing. And so on and so forth and there are many other functional points that enable capabilities to be refined and more closely aligned with the business shape of the enterprise.

  • All-channel Contact Center: Of course we are not doing a SaaS application of the contact center, we mainly provide channels needed to contact center endpoint resources and application solutions, the future will application scenario for the SMS, email, fax, telephone, APP, weibo, WeChat, video, IM messages, 5 g global unicom, a touch namely, channel integration.

  • Cognitive intelligence and human-machine collaboration: customer service assistance, ASR, TTS, NLP access standardization, process engine and other AI-related basic capabilities are integrated to provide a platform for the application scenarios of ARTIFICIAL intelligence.

  • Auxiliary G2 capability expansion: Provides touchability for audio and video calls or conferences. In the future, functions such as IVVR, ACD, and ROUTER can be directly applied to G2 audio and video media editing to expand G2 application capabilities.

  • Application componentization, reduce enterprise access cost: recently the development of low code has some substantial progress, take the official website for example, simple information management system can be built in tens of minutes, but can you quickly and conveniently build a medium complexity application system? There are challenges. Below for netease canoe low code platform of an idea, can be based on components and modular programming, all function points can be abstract into small components, users can selectively assembled at random, after the completion of the formation of one of the available communication platform, can be used direct manipulation can also be secondary abstracts interface to call PaaS capabilities. It can be very convenient to choose on demand, reducing the cost of enterprise purchase. Moreover, the application capability can be accessed with a lower amount of coding, which improves the access efficiency. Subsequent solutions can be built directly on low-code platforms.

    ​

​

  • Converged communication medium platform: Carrier communication medium platform is a subset of communication medium platform, and we will practice through it in the early stage. It can generate a variety of unrelated component libraries like building blocks, which can be assembled at will and become controllable platforms that can be large or small. Enterprise customers can assemble and use public cloud capabilities on demand, or they can purchase the required modules to form a customized private platform for deployment within the enterprise. Later, more modules will be integrated in the same mode, and finally integrated into the converged communication center.

reference

  • The B-Side
  • Transformation of Enterprise IT Architecture
  • Global Internet Communication Cloud Industry Research Report 2021
  • Phoenix Architecture: Building Reliable Large-scale Distributed Systems
  • White Paper on 5G Information

Related Reading recommendations

  • Technology of dry goods | real-time communication services reverberation voice solution algorithm in practice
  • Depth profiling system design | “circle group” news “circle group” technology series
  • Technology of dry goods | large data insight into portrait automation practice