Skip to main content

Quality on Demand API

API docs by Redocly

QoD for enhanced communication (0.10.1)

Download OpenAPI specification:Download

E-mail: contact@spryfoxnetworks.com License: Apache 2.0

The Quality-On-Demand (QoD) API provides programmable interface for developers and other users (capabilities consumers) to request stable latency or throughput managed by Telco networks without the necessity to have an in-depth knowledge of the 4G/5G system or the overall complexity of the Telecom Systems.

Product documentation at Camara

Introduction

Industrial (IoT), VR/Gaming, live video streaming, autonomous driving and many other scenarios demand network communication quality and are sensitive to any change in transmission conditions. Being able to request a stable latency (reduced jitter) or prioritized throughput from the network can improve user experience substantially.

The QoD API offers the application developers the capability to request for stable latency (reduced jitter) or throughput for some specified application data flows between application clients (within a user device) and Application Servers (backend services). The developer has a pre-defined set of Quality of Service (QoS) profiles which they could choose from depending on their latency or throughput requirements.

QoD API Overview

The usage of the API is based on QoS session resources, which can be created (based on available QoS profiles), queried and deleted. The deletion of a requested session can be triggered by the API consumer or can be triggered automatically. The automatic process is triggered either when the requested specified duration of a QoS session has reached its limit or the default session expiration time has been reached (within an example provider implementation it is set to 24hrs).

Relevant terms and definitions

  • QOD service endpoint: The URL pointing to the RESTful resource of the QoD API.

  • Authentication: Security access keys such as OAuth 2.0 client credentials used by client applications to invoke the QoD API.

  • QoS profiles and QoS profile labels: Latency or throughput requirements of the application mapped to relevant QoS profile class.

  • Identifier for the device: At least one identifier for the device (user equipment) out of four options: IPv4 address, IPv6 address, Phone number, or Network Access Identifier [5] assigned by the mobile network operator for the device.

  • Identifier for the application server: IPv4 and/or IPv6 address of the application server (application backend)

  • App-Flow (between the application client and application server): The precise application data flow the developer wants to prioritize and have stable latency or throughput for. This flow is in the current API version determined by the identifiers used for the device and the application server. And it can be further elaborated with details such as ports or port-ranges. Future version of the API might allow more detailed flow identification features.

  • Duration: Duration (in seconds) for which the QoS session (between application client and application server) should be created. This parameter is optional. When not specified, a default session duration (e.g. 24 hours) is applied. The user may request a termination before its expiration.

  • Notification URL and token: Developers may provide a callback URL on which notifications about all status change events of the session (eg. session termination) can be received from the service provider. This is an optional parameter.

API functionality

The usage of the QoD API is based on QoS profile classes and parameters which define App-Flows. Based on the API, QoS session resources can be created, queried, and deleted. Once an offered QoS profile class is requested, application users get a prioritized service with stable latency or throughput even in the case of congestion. The QoD API has the following characteristics:

  • A specified App-Flow is prioritized to ensure stable latency or throughput for that flow.
  • The prioritized App-Flow is described by providing information such as device IP address (or other device identifier) & application server IP addresses and port/port-ranges.
  • The developer can optionally specify the duration for which they need the prioritized App-flow.
  • Stable latency or throughput is requested by selecting from the list of QoS profiles made available by the service provider (e.g. QOS_E) to map latency and throughput requirements.
  • The developer can optionally also specify callback URL on which notifications for the session can be sent.

Following diagram shows the interaction between different components

QoD Management API

How QoS profiles are mapped to connectivity characteristics are subject to agreements between the communication service provider and the API invoker. Within the CAMARA project, you can find a sample for such a mapping of QoS profiles. CAMARA QoS Profiles Mapping Table (REFERENCE DRAFT)

Further info and support

(FAQs will be added in a later version of the documentation)

QoS Sessions

Manage QoS sessions

Creates a new session

Create QoS Session to manage latency/throughput priorities

If the qosStatus in the API response is "AVAILABLE" and a notification callback is provided the client will receive in addition to the response a QOS_STATUS_CHANGED event notification with qosStatus as AVAILABLE.

If the qosStatus in the API response is REQUESTED, the client will receive either

  • a QOS_STATUS_CHANGED event notification with qosStatus as AVAILABLE after the network notifies that it has created the requested session, or
  • a QOS_STATUS_CHANGED event notification with qosStatus as UNAVAILABLE and statusInfo as NETWORK_TERMINATED after the network notifies that it has failed to provide the requested session.

A QOS_STATUS_CHANGED event notification with qosStatus as UNAVAILABLE will also be send if the network terminates the session before the requested duration expired

NOTE: in case of a QOS_STATUS_CHANGED event with qosStatus as UNAVAILABLE and statusInfo as NETWORK_TERMINATED the resources of the QoS session are not directly released, but will get deleted automatically at earliest 360 seconds after the event. This behavior allows clients which are not receiving notification events but are polling to get the session information with the qosStatus UNAVAILABLE (the statusInfo parameter is not included in the current version but will be adding to SessionInfo in an upcoming release). Before a client can attempt to create a new QoD session for the same device and flow period they must release the session resources with an explicit delete operation if not yet automatically deleted.

Authorizations:
oAuth2ClientCredentialsthreeLegged
Request Body schema: application/json
required

Parameters to create a new session

required
object (Device) non-empty

End-user equipment able to connect to a mobile network. Examples of devices include smartphones or IoT sensors/actuators.

The developer can choose to provide the below specified device identifiers:

  • ipv4Address
  • ipv6Address
  • phoneNumber
  • networkAccessIdentifier

NOTE: the MNO might support only a subset of these options. The API invoker can provide multiple identifiers to be compatible across different MNOs. In this case the identifiers MUST belong to the same device

required
object (ApplicationServer) non-empty

A server hosting backend applications to deliver some business logic to clients.

The developer can choose to provide the below specified device identifiers:

  • ipv4Address
  • ipv6Address
object non-empty

The ports used locally by the device for flows to which the requested QoS profile should apply. If omitted, then the qosProfile will apply to all flows between the device and the specified application server address and ports

object non-empty

A list of single ports or port ranges on the application server

qosProfile
required
string <string> (QosProfileName) [ 3 .. 256 ] characters ^[a-zA-Z0-9_.-]+$

A unique name for identifying a specific QoS profile. This may follow different formats depending on the service providers implementation. Some options addresses:

  • A UUID style string
  • Support for predefined profiles QOS_S, QOS_M, QOS_L, and QOS_E
  • A searchable descriptive name
object
duration
integer <int32> [ 1 .. 86400 ]
Default: 86400

Session duration in seconds. Maximal value of 24 hours is used if not set. After session is expired the, client will receive a QOS_STATUS_CHANGED event with

  • qosStatus as UNAVAILABLE, and,
  • statusInfo as DURATION_EXPIRED. See notification callback.

Responses

Callbacks

Request samples

Content type
application/json
{
  • "device": {
    },
  • "applicationServer": {
    },
  • "devicePorts": {
    },
  • "applicationServerPorts": {
    },
  • "qosProfile": "QCI_1_voice",
  • "webhook": {},
  • "duration": 86400
}

Response samples

Content type
application/json
{
  • "device": {
    },
  • "applicationServer": {
    },
  • "devicePorts": {
    },
  • "applicationServerPorts": {
    },
  • "qosProfile": "QCI_1_voice",
  • "webhook": {},
  • "sessionId": "f6567dd8-e069-418e-8893-7d22fcf12459",
  • "duration": 86400,
  • "startedAt": 1639479600,
  • "expiresAt": 1639566000,
  • "qosStatus": "REQUESTED",
  • "messages": [
    ]
}

Callback payload samples

Callback
POST: Session notifications callback
Content type
application/cloudevents+json
{}

Get QoS session information

Querying for QoS session resource information details

Authorizations:
oAuth2ClientCredentialsthreeLegged
path Parameters
sessionId
required
string <uuid> (SessionId)

Session ID that was obtained from the createSession operation

Responses

Response samples

Content type
application/json
{
  • "device": {
    },
  • "applicationServer": {
    },
  • "devicePorts": {
    },
  • "applicationServerPorts": {
    },
  • "qosProfile": "QCI_1_voice",
  • "webhook": {},
  • "sessionId": "f6567dd8-e069-418e-8893-7d22fcf12459",
  • "duration": 86400,
  • "startedAt": 1639479600,
  • "expiresAt": 1639566000,
  • "qosStatus": "REQUESTED",
  • "messages": [
    ]
}

Delete a QoS session

Release resources related to QoS session

If the notification callback is provided and the qosStatus of the session was AVAILABLE the client will receive in addition to the response a QOS_STATUS_CHANGED event with

  • qosStatus as UNAVAILABLE and
  • statusInfo as DELETE_REQUESTED There will be no notification event if the qosStatus was already UNAVAILABLE.
Authorizations:
oAuth2ClientCredentialsthreeLegged
path Parameters
sessionId
required
string <uuid> (SessionId)

Session ID that was obtained from the createSession operation

Responses

Response samples

Content type
application/json
{
  • "status": 400,
  • "code": "INVALID_ARGUMENT",
  • "message": "Schema validation failed at ..."
}

Extend the duration of an active session

Extend the overall duration of an active QoS session. If this operation is executed successfully, the new duration of the target session will be the original duration plus the additionally requested duration. The new remaining duration of the QoS session shall not exceed the maximum remaining duration limit (currently fixed at 86,400 seconds) where the remaining duration is calculated as the difference between the expiresAt and current time when the request to extend the session duration is received. If this maximum limit would be exceeded, the overall duration shall be set such that the remaining duration is equal to this limit. An example: A QoD session was originally created with duration 80,000 seconds. 10,000 seconds later, the developer requested to extend the session by 20,000 seconds.

  • Original duration: 80,000 seconds
  • Elapsed time: 10,000 seconds
  • Remaining duration: 70,000 seconds
  • New remaining duration: 86,400 seconds (the maximum allowed)
  • New overall session duration: 96,400 seconds
Authorizations:
oAuth2ClientCredentialsthreeLegged
path Parameters
sessionId
required
string <uuid> (SessionId)

Session ID that was obtained from the createSession operation

Request Body schema: application/json
required

Parameters to extend the duration of an active session

requestedAdditionalDuration
required
integer <int32> [ 1 .. 86399 ]

Additional duration in seconds to be extended.

Responses

Request samples

Content type
application/json
{
  • "requestedAdditionalDuration": 60
}

Response samples

Content type
application/json
{
  • "device": {
    },
  • "applicationServer": {
    },
  • "devicePorts": {
    },
  • "applicationServerPorts": {
    },
  • "qosProfile": "QCI_1_voice",
  • "webhook": {},
  • "sessionId": "f6567dd8-e069-418e-8893-7d22fcf12459",
  • "duration": 86400,
  • "startedAt": 1639479600,
  • "expiresAt": 1639566000,
  • "qosStatus": "REQUESTED",
  • "messages": [
    ]
}

QoS Profiles

Manage QoS Profiles

Get All QoS Profiles

Returns all QoS Profiles that match the given criteria. If no criteria is given, all QoS Profiles are returned.

Authorizations:
oAuth2ClientCredentials
query Parameters
name
string

QoS Profile name

status
string (QosProfileStatusEnum)
Enum: "ACTIVE" "INACTIVE" "DEPRECATED"

The current status of the QoS Profile

  • ACTIVE- QoS Profile is available to be used
  • INACTIVE- QoS Profile is not currently available to be deployed
  • DEPRECATED- QoS profile is actively being used in a QoD session, but can not be deployed in new QoD sessions

Responses

Response samples

Content type
application/json
[
  • {
    }
]

Get QoS Profile for a given name

Returns a QoS Profile that matches the given name.

Authorizations:
oAuth2ClientCredentials
path Parameters
name
required
string <string> (QosProfileName) [ 3 .. 256 ] characters ^[a-zA-Z0-9_.-]+$
Example: QCI_1_voice

A unique name for identifying a specific QoS profile. This may follow different formats depending on the service providers implementation. Some options addresses:

  • A UUID style string
  • Support for predefined profiles QOS_S, QOS_M, QOS_L, and QOS_E
  • A searchable descriptive name

Responses

Response samples

Content type
application/json
{
  • "name": "QCI_1_voice",
  • "description": "QoS profile for video streaming",
  • "status": "ACTIVE",
  • "targetMinUpstreamRate": {
    },
  • "maxUpstreamRate": {
    },
  • "maxUpstreamBurstRate": {
    },
  • "targetMinDownstreamRate": {
    },
  • "maxDownstreamRate": {
    },
  • "maxDownstreamBurstRate": {
    },
  • "minDuration": {
    },
  • "maxDuration": {
    },
  • "priority": 20,
  • "packetDelayBudget": {
    },
  • "jitter": {
    },
  • "packetErrorLossRate": 3
}