By Steve Courtney and Silas Zhou

With tens of millions of residential broadband users projected by 2004, home networks will be required to enhance life styles. HomePNA can easily test these home systems for maximum performance. Demand and deployment of broadband access to the home is increasing rapidly. The market is expected to grow to 30 million subscribers by 2004 in the United States (Parks Associates), up from five million subscribers today. More than half of those subscribers are expected to also want home networking capabilities. Home networks will support simultaneously shared Internet access, printer/peripheral sharing, file and application sharing and networked gaming. New appliances such as entertainment and security equipment are being developed to take advantage of shared broadband access throughout the home. The market for home networking equipment is projected to grow to $16 billion by 2006 (Ref: Datacomm). Until now, the xDSL and cable modem industry focused on delivering high-speed (500 kbps or higher) broadband services over the “last mile” to the home and office. Several solutions have been developed to network broadband access throughout the home or to the “last inch” and involve several different media. These include using traditional local area networking (LAN) cables (CAT 5), existing power lines (HomePlug), airwaves (wireless networking – HomeRF) and existing in-home phone lines.

HomePNA in the Home Environment
HomePNA (the Home Phone Networking Alliance) is one of the easiest technologies for the end-user to install and maintain. It has highly developed standards, the support of many leading equipment manufacturers, an extensive list of compliant products and does not require expensive infrastructure upgrades. Driven by technical advances such as Microsoft Windows XP, an operating system that allows for “plug and play” additions to the network, consumers’ expectations have risen to expect out-of-the-box networked systems. radio frequency interference (RFI) and wireline environment for any potential vendor who wishes to deploy a residential gateway (RG) or other networking device. While HomePNA technology may be easy for the end-user to interconnect, the home contains a hostile radio frequency interference (RFI) and wireline environment for any potential vendor who wishes to deploy a residential gateway (RG) or other networking device . RG products have to co-exist with terminated or unterminated variable length phone lines, bridged taps and mismatches (stubs), fax and telephones (with on-hook, off-hook or ringing conditions), TV, radio and several other “noisy” devices such as air conditioners, ham radio, cordless phone sets, mobile phones and A/C motor appliances. This places the burden onto the manufacturers of home networking and RG products to perform thorough conformance testing against agreed standards with accurate, repeatable and industry standard test equipment.

HomePNA has been leading the way in defining conformance standards and procedures to ensure access equipment will operate in this environment. Tests can be performed manually or via automated test suites. The trend and preference is generally towards automated test suites that reduce testing and product development time and cost and allow for regression testing and repeatable result generation.

Testing Applications and Approaches
In general terms, whether the telephone lines are being used by HomePNA or xDSL equipment, maximum performance and interoperability within the home is the desired result. Figure 1 outlines the typical in-home networking environment and also highlights the primary factors affecting the performance of broadband gateway and access equipment. The following list shows some of the key factors affecting the performance of broadband systems networked by phone lines in a Small Office Home Office (SOHO) environment: Wireline loop topology, wire gauges, wire lengths, bridged taps and terminations Egress of noise from xDSL or HomePNA phone lines that affect other home systems (POTS phone, TV, radio, etc.)

RFI Noise Ingress and crosstalk from adjacent wire line systems (xDSL, fax, ISDN modem, etc.) This can occur from other houses that share the same external cable bundle Egress of noise from existing in-home power lines Wireless applications (cordless and mobile phones) IR (Infrared) applications (IR modems and Palm Pilot) Device interoperability (incompatible connections)

Vendors Can’t Test It All
Ensuring HomePNA Devices Work By Steve Courtney and Silas Zhou With tens of millions of residential broadband users projected by 2004, home networks will be required to enhance life styles. HomePNA can easily test these home systems for maximum performance. Demand and deployment of broadband access to the home is increasing rapidly. The market is expected to grow to 30 million subscribers by 2004 in the United States (Parks Associates), up from five million subscribers today. More than half of those subscribers are expected to also want home networking capabilities. Home networks will support simultaneously shared Internet access, printer/peripheral sharing, file and application sharing and networked gaming. New appliances such as entertainment and security equipment are being developed to take advantage of shared broadband access throughout the home. The market for home networking equipment is projected to grow to $16 billion by 2006 (Ref: Datacomm). Until now, the xDSL and cable modem industry focused on delivering high-speed (500 kbps or higher) broadband services over the “last mile” to the home and office. Several solutions have been developed to network broadband access throughout the home or to the “last inch” and involve several different media. These include using traditional local area networking (LAN) cables (CAT 5), existing power lines (HomePlug), airwaves (wireless networking – HomeRF) and existing in-home phone lines. HomePNA in the Home Environment HomePNA (the Home Phone Networking Alliance) is one of the easiest technologies for the end-user to install and maintain. It has highly developed standards, the support of many leading equipment manufacturers, an extensive list of compliant products and does not require expensive infrastructure upgrades. Driven by technical advances such as Microsoft Windows XP, an operating system that allows for “plug and play” additions to the network, consumers’ expectations have risen to expect out-of-the-box networked systems. radio frequency interference (RFI) and wireline environment for any potential vendor who wishes to deploy a residential gateway (RG) or other networking device. While HomePNA technology may be easy for the end-user to interconnect, the home contains a hostile radio frequency interference (RFI) and wireline environment for any potential vendor who wishes to deploy a residential gateway (RG) or other networking device . RG products have to co-exist with terminated or unterminated variable length phone lines, bridged taps and mismatches (stubs), fax and telephones (with on-hook, off-hook or ringing conditions), TV, radio and several other “noisy” devices such as air conditioners, ham radio, cordless phone sets, mobile phones and A/C motor appliances. This places the burden onto the manufacturers of home networking and RG products to perform thorough conformance testing against agreed standards with accurate, repeatable and industry standard test equipment. HomePNA has been leading the way in defining conformance standards and procedures to ensure access equipment will operate in this environment. Tests can be performed manually or via automated test suites. The trend and preference is generally towards automated test suites that reduce testing and product development time and cost and allow for regression testing and repeatable result generation. Testing Applications and Approaches In general terms, whether the telephone lines are being used by HomePNA or xDSL equipment, maximum performance and interoperability within the home is the desired result. Figure 1 outlines the typical in-home networking environment and also highlights the primary factors affecting the performance of broadband gateway and access equipment. The following list shows some of the key factors affecting the performance of broadband systems networked by phone lines in a Small Office Home Office (SOHO) environment: Wireline loop topology, wire gauges, wire lengths, bridged taps and terminations Egress of noise from xDSL or HomePNA phone lines that affect other home systems (POTS phone, TV, radio, etc.) RFI Noise Ingress and crosstalk from adjacent wire line systems (xDSL, fax, ISDN modem, etc.) This can occur from other houses that share the same external cable bundle Egress of noise from existing in-home power lines Wireless applications (cordless and mobile phones) IR (Infrared) applications (IR modems and Palm Pilot) Device interoperability (incompatible connections)

Wireline impacts on performance include:
Gauge – different wire gauge features different attenuation to the transmitted signal Reach – the longer the length, the more the transmitted signal is attenuated Impedance – the hook status (on-hook or off-hook) of a POTS phone device makes the input impedance different Bridged tap or mismatch – a bridged tap or mismatch (stub) causes reflection and re-reflection of the signal and affects the SNR Termination – an open (unterminated) line also causes reflection and re-reflection of the signal Topology – when different typical topology (bridge, daisy chain, star) is chosen, the loop attenuation and input impedance varies

Noise impairment impacts on performance include:
RFI
xDSL ingress
Appliance ingress
Crosstalk

Typical Test Setup to Allow for Conformance of a Residential Gateway (RG) Product
Figure 2 shows a typical test environment for residential gateway and other network access products. Generally, the test equipment is manually or PC GUI controlled during the product design and verification phase. Once the new product is stable and at the stage of subsequent releases and upgrades, the test environment is typically automated for regression and production testing. The key issue is to have accurate, repeatable and controllable test results that can be used to verify product conformance to standards like HomePNA. The Figure 2 setup consists of access devices such as PC HomePNA modems to be tested, together with the RG or Network Interface Device (NID), which are connected via the Spirent DLS 3200 series HomePNA wireline simulator system. The DLS 3200 can be used to selectively simulate each HomePNA loop configuration, as well as phone on/off hook characteristics and terminated and unterminated loops. The DLS 3200 unit is connected via the RG or NID and can be connected to a Traffic Generator and BER Tester. Alternatively, as Figure 2 shows in the more general end-to-end system testing configuration, a local loop wireline simulator such as Spirent’s DLS 400 (ADSL) or DLS 8000 (VDSL) series can be interfaced to an actual DSLAM or DSLAM simulator, which can then be connected to a CO traffic simulator and BER Tester. The DLS 5000 series noise impairment generators can be connected at various points in the system, and typically for HomePNA directly to the DLS 3200 series In-home wiring simulators to allow for injection of RFI and crosstalk noise profiles into the selected HomePNA/In-home wiring loop model. The HomePNA Testing process consists of selecting in a logical sequence as defined by the HomePNA testing and conformance specifications a particular loop model in the DLS 3200, with a particular noise profile as required on the DLS 5000 series and then initiating a series of data traffic through the RG to the HomePNA access device. Data throughput rates are measured and compared against the HomePNA standards to determine a pass or a fail. Test sequences are repeated for different HomePNA loop models and noise profiles. Additionally, voiceband effects on loop characteristics (phone on/off hook, unterminated loops) can also be selected using the DLS 3200 series to complete the testing. Testing requires the ability to select predefined test loops and models, as well as predefined noise impairments. These are set up while data traffic is passed through the system and measured as different combinations of equipment such as phones, faxes and terminations are applied, either automatically or manually to the system. A common PC or other computer system typically controls all the test equipment and devices under test via either an IEEE test bus or Ethernet. Figure 3 shows the typical complex attenuation characteristics for just one of the many HomePNA test loop models.

Creating the Repeatable Test Environment
A solid and repeatable test environment can be created by using test equipment like the Spirent DLS 3000 and DLS 5000 series systems. The DLS 3000 series is available as a DLS 3200 for HomePNA standard testing and as a DLS 3100 for xDSL In-Home Wiring testing. The DLS 5000 series, especially the DLS 5200 series allows for wide ranging arbitrary noise disturber combinations to be injected into the loops simulated by the DLS 3000 series. These products accurately simulate wireline loops, mismatches (stubs), terminations and noise impairments and can be used with Spirent BERT test systems like the SMB 6000 or AX/4000. RG device suppliers and chipset manufactures can conformance test their new products against myriad in-home wiring standards and environmental conditions. Written test results against defined conformance testing by the standards organizations (i.e., HomePNA) can be produced automatically and validate vendors’ claims. Defined test suites of equipment and test procedures can be used by all interested parties in the value chain, from chipset designers, RG product designers, vendor interoperability test houses, production facilities, service providers and others to ensure repeatability of test results. Often, standards organizations employ particular testing labs to perform periodic conformance and interoperability testing. Contrast this with using ad hoc rolls of wire by each vendor and custom produced noise impairments on arbitrary wave generators with substandard randomness or crest factor performance. The variability, accuracy, reliability and repeatability of these testing approaches are unacceptable in practice, and the confidence level of conformance to standards is severely compromised. Consumers will not be impressed with products that are not properly tested to worst case and standards-organization approved conformance tests, and hence fail to work in real life conditions.

Steve Courtney is vice president of marketing and Silas Zhou is a product manager for Spirent Communications.