Microsoft® Response Point™
Microsoft Response Point Guidance
©2007 Microsoft Corporation. This work is licensed under the Creative Commons Attribution-Noncommercial License. To view a copy of this license, visit
http://creativecommons.org/licenses/by-nc/2.5/
Introduction
Microsoft® Response Point™ is a small-business phone solution that is designed to be easy to use and manage. While easy to use, the variety of choices available for connecting a Response Point phone system to a telephony service provider network can be slightly confusing. This paper will help to simplify the decision-making process by explaining the considerations involved in choosing the type of service and the actual provider for public telephony services that will work in conjunction with a Response Point solution.
Overview
This paper will help Microsoft Response Point Value Added Resellers who have little telephony experience understand how Response Point can be connected to service provider networks and the considerations involved in choosing a provider for a Response Point deployment.
This paper also discusses some common troubleshooting scenarios that will help VAR technologists identify and resolve issues related to provider connectivity and services.
Definitions
The following list defines key terms and acronyms used in this document:
ATA An ATA (Analog Telephony Adapter) converts standard analog RJ-11 connections to Ethernet RJ-45 connections so that analog devices can be used on VoIP networks.
Automated Receptionist The Automated Receptionist, or Auto Attendant, is a Response Point voice-recognition technology that can prompt callers for information and respond to their requests.
DID Direct Inward Dialing (DID) is a virtual phone number that can be dialed directly from outside of the business and will be forwarded to a specific VoIP phone. This feature will be available with the release of Response Point Service Pack 1 (SP1).
FXO A Foreign Exchange Office (FXO) is the customer side of the connection between an external phone service provider (FXS or Foreign Exchange Station) that generates a ring signal and the customer (FXO) that receives a ring signal.
IAD An Integrated Access Device (IAD) enables the conversion of analog and digital signals for convergence of network services.
IP Telephony This is the general term used to describe VoIP.
ITSP An Internet Telephony Service Provider (ITSP) is a company that offers Internet-based data service for VoIP telephony.
PBX A PBX (Private Branch Exchange) is a private telephone switch that provides full switching features for an office or campus.
POTS Plain Old Telephone Service (POTS) means simple analog telephone service. POTS is sometimes used interchangeably with PSTN.
PSTN Public Switched Telephone Network (PSTN) is the switched analog voice network that most people use for telephony today.
QoS Quality of Service (QoS) is a method used to mark and prioritize different types of network traffic passing through a gateway or router to enable time-sensitive traffic, such as voice conversations, to have transmission priority over data packets.
RTP Real-time Transport Protocol (RTP) is a standardized packet format for streaming multimedia content and VoIP communications. RTP utilizes Session Initiation Protocol (SIP) to initiate and terminate sessions.
SIP Session Initiation Protocol (SIP) is a transport-independent application layer control protocol often used by multimedia or VoIP applications to initiate, maintain, and terminate connection sessions with one or more participants.
VoIP Voice over Internet Protocol (VoIP) is the practice of transmitting voice communications over data networks by using Internet protocols.
Response Point Connection Options
Response Point phone system software is designed to work best when connecting the Response Point VoIP PBX unit to a PSTN provider network via analog lines. Response Point PBX units, also referred to as base units, will be shipped with either an analog FXO or, after the release of SP1, native VoIP. This means that after the release of Response Point SP1, there will be four ways to connect Response Point:
- Connected directly to analog lines provided by a traditional telephone company.
- Connected to a VoIP service via analog connections to an IAD
- Connected to another PBX via analog lines.
- Connected directly to a VoIP over Internet ITSP via Ethernet.
While an analog connection is the default method of connecting Response Point to external telephone networks, Response Point can be employed as part of an end-to-end VoIP solution by connecting base units to an IAD that is compatible with the one used by the IP telephony provider's network.
Advantages and Limitations
Each connectivity method presents its own considerations for small-business telephony needs. Generally, traditional phone service providers offer greater stability than VoIP providers, especially when the possibility of electrical outages is taken into consideration. PSTN networks utilize time-tested, reliable technologies that are powered from PSTN provider stations that often have backup generators. In most cases, a PSTN phone system will continue functioning even when the power goes out locally. PSTN providers also boast better average voice quality, and sometimes offer more features, like call waiting, than VoIP providers.
VoIP providers, on the other hand, can be much cheaper, since they allow businesses to converge their existing data communications systems with their voice communications systems. Another cost benefit can be realized if a business places a lot of long-distance calls. Even though long-distance calls over an IP telephony network are not necessarily free, long-distance calls are usually priced below the average cost charged by PSTN providers. While VoIP providers can offer the same services that PSTN providers can, there often are limitations. For example, some VoIP providers offer long-distance services in limited geographic areas, or are limited in where their discount rates apply.
Connection Considerations
In addition to the advantages and limitations of each provider type, each connection type has unique considerations and configurations that need to be addressed before and during a Response Point installation. This section will review some of the main considerations for each connection type in order to highlight the differences among them.
Traditional Telephone Services
Response Point is currently designed to use PSTN analog connections as the default method of connecting to external telephony services. This type of installation will be most common when deploying Response Point solutions in small-business environments. While Response Point can be used as part of an end-to-end VoIP solution, at this time the only supported method of using Response Point is to connect it to a PSTN provider.
When selecting a PSTN provider, there are several things that you need to do to ensure that the Response Point system functions correctly with the PSTN provider network.
The following tasks should be completed prior to connecting a Response Point system to a PSTN network:
- Stick with well-known providers that use Bell-standard line configurations. Some smaller PSTN providers use incompatible standards that may prevent Response Point from functioning properly.
- Calculate the number of phone lines needed to support the business by counting the number of users and determining how many peak calls are normally expected. If the number of peak calls means that "X" connections will occur simultaneously, you must have "X" lines to ensure uninterrupted service.
- Calculate the number of dedicated lines needed to support analog devices such as fax machines and security systems that depend on analog phone service to function correctly. While some devices might be capable of using an ATA device to connect to a VoIP network, you will need to check with the manufacturer to make certain that ATA connections are supported.
- When planning a deployment date, check with the PSTN provider to determine how long it takes to provision and activate the necessary lines. You will need to cushion the deployment date in order to have time to test that the additional lines will be fully functional when deployment occurs.
- In addition to checking lines for activation, also ensure that all requested features have been enabled on all lines. Some Response Point services depend on caller ID, so it is important to make sure that all lines have caller ID enabled.
- Response Point's Automated Receptionist feature reliably supports only 8 simultaneous speech-recognition connections. If the business expects more than 8 simultaneous inbound calls, it may want to designate an employee to answer calls rather than rely on the Automated Receptionist feature.
End-to-End VoIP
While PSTN providers are currently the only supported method of connecting Response Point to telephony provider services, it is possible to use Response Point in conjunction with an IP telephony provider as part of a comprehensive, end-to-end VoIP solution. There are some obvious advantages to this approach, especially when connecting multiple sites to each other when a high-bandwidth WAN connection is in place, or when long-distance utilization is high.
While Response Point does not yet directly support this type of connection, this section will list some of the considerations when attempting this type of deployment after Response Point SP1 is released in early 2008.
- Work with an established and trusted VoIP network provider to help determine the network connectivity requirements that must be met to ensure proper functionality.
- Response Point will require the following information from the ITSP:
- Service provider name
- SIP proxy or registration server address
- Registration interval
- User Address of Record (AOR) or Uniform Resource Identifier (URI)
- Default domain
- Authentication ID or SIP user ID
- Password
- Caller ID display name
- For best results, use a DSL or faster connection.
- It is a common misconception that VoIP long-distance services are free or can be connected to any location. Ensure that your customers carefully review VoIP provider information to determine long-distance costs and capabilities. Before choosing a provider, consider the customer's typical outbound long-distance usage patterns and the service levels the customer will require.
- Ask the provider how many simultaneous calls its service can support; some providers can support only a limited number of simultaneous calls.
- Confirm that the provider can deliver caller ID services. Some Response Point features depend on caller ID to function correctly.
- Confirm that the ITSP uses the SIPconnect interface specification. You will also need to determine whether the customer's firewall and router can allow and prioritize the ITSP's traffic.
- Confirm that the current router supports QoS for voice traffic. If the router that will be used for VoIP traffic does not support QoS, the customer may need to purchase additional network equipment.
- Determine whether the current firewall can allow inbound and outbound VoIP traffic to pass through. Some less expensive or bundled firewalls may not allow you to specify the types of packets that may be allowed through in both directions. If the firewall does not support the provider's VoIP standard, the customer may need to purchase a new firewall. Ask the service provider to specify the firewall settings it will require for its service.
- Keep in mind that, until the release of Response Point SP1, Response Point base units do not directly support IP telephony. You will need to check with the service provider to determine if it can supply or recommend an IAD for use with its solution, and factor that additional cost into the total cost of an end-to-end VoIP solution.
Troubleshooting Provider Issues
The first step in resolving problems in a Response Point environment is to determine the source of the problem. Typically, there are a few possible sources of problematic behavior in any VoIP environment, including Response Point environments. Generally, these problems can be broken into the following categories:
- Line quality
- Phone units
- Internal network cabling and patch cables
- Local Area Network traffic congestion
- PBX units
- Switch settings or hardware
- Router settings or hardware
- Firewall settings or hardware
- Link Access problems
- External network traffic congestion
- VoIP call establishment delays
- Provider-based connectivity or setting problems
While it is possible for you to directly isolate, troubleshoot, and resolve most of these problems, some of them are not due to anything that occurs on the Response Point system itself. This section will give you information that can be used to determine when problems are not a part of the Response Point system and may require the assistance of a service provider to resolve.
Troubleshooting PSTN Issues
Generally speaking, the Response Point network effectively ends at the connection between the VoIP PBX and the PSTN analog lines. At this point the VoIP traffic has been converted to an analog signal and is carried by the PSTN service provider to its destination. Any issues that originate outside the base unit will likely require the assistance of the PSTN service provider to resolve.
Some of the more common issues that originate with PSTN service providers include:
- Dead lines
Symptoms can include no dial tone, customers reporting frequent busy signals, nobody answering when the business places outgoing calls, and the inability to make as many simultaneous calls as the user should be able to make based on the unit's configuration.
The Response Point administrative interface can be used in some cases to help spot problems caused by dead lines. However, these issues usually require the intervention of a service provider. You can reduce the risk of dead lines by testing all lines prior to Response Point deployment to ensure that they have been activated.
- Features not working
Feature functionality issues are usually due to the provider failing to enable caller ID or rollover features on all lines connected to the Response Point base unit. Even seemingly unrelated features, including Response Point's direct-dial features, can be dependent on carrier services like caller ID. To prevent these problems, test the features available on all lines prior to deploying Response Point and whenever a new line is added to the Response Point system.
Troubleshooting VoIP Issues
Because Response Point does not officially support connections to IP telephony networks yet, you must carefully identify the source of any problem prior to requesting support from Microsoft or the VoIP carrier. This section provides some general guidelines for identifying VoIP issues that may reside outside of the Response Point system.
Access Link Issues
Access link issues are problems that occur between local area networks and lower bandwidth networks like Internet connections and WAN links. Access link congestion issues can cause a number of symptoms, including:
- Pauses or delays in conversations, or gaps in speech. When the router's buffer is full and it cannot send packets correctly, it can cause excessive pauses or delays in voice traffic that can, in turn, cause conversational difficulties.
- Random popping sounds or garbled speech. This can be caused by access link congestion as well, since routers will invoke random early detection when their buffers fill, dropping packets to prompt the sender to resend packets. When this occurs, voice packets can be lost or sent out of order, causing voice quality issues or popping sounds.
- Occasional ticking sounds, sometimes at regular intervals. This is usually caused by routers switching routes. This causes timing issues that result in a ticking-like background sound.
- Excessive echo or "tunnel" voice quality. While this problem can also occur because of poor analog line quality, an excessive echo during conversations or a hollow sounding voice can be caused by improper phone volume settings.
Jitter
Jitter describes problems associated with packet timing issues. Users will notice garbled voices or dropped portions of words. This is usually caused by problems along the network path, either due to congestion, route changes, or similar issues. These occur most frequently outside of the local area network, so troubleshooting should begin at the router and may involve the service provider's network.
Latency
Latency refers to delays caused by the distance packets must travel between the router and the service provider's network, or when buffers are used to compensate for excessive jitter. Most major service providers have multiple network access points that are positioned to reduce the effect of latency. When the ping time from the local network to the service provider exceed 250ms, users may notice a significant delay between the time they speak and the time the person on the other end hears what was said. This can also result in excessive echoing or even tunnel voice quality.
Packet Loss
While the effect of lost packets or delayed packets on data traffic is usually unnoticed, it can cause major quality issues over voice networks. When packet loss is minor, users will notice an increase in echo or robot-like voice quality. High levels of packet loss or significant delays can cause speech to break up or become drastically distorted. Packet loss or delayed delivery is generally caused by network saturation anywhere along the network path, but generally only occurs on the local network when hubs are used or when computers are attached to the network through the phone's built-in switch.
Outages
Outages can be caused by hardware failures along the network path or by failed links along the network path. To isolate the source of network failures, perform pings or trace routes to locate the network failure. If the failure occurs outside of the local area network, you will need to contact the service provider to resolve the issue.
Additional Resources