Entre el realismo, lo abstracto y el hiperrealismo

El otro día, aprovechando las "vacaciones" en Madrid, fui a visitar el Museo de Arte Reina Sofía. La entrada me salió gratis, gracias a una invitación por ser mutualista. Soy, soy soy soy... soy soy soy...

Mi prioridad era ver de nuevo el Guernica, ya que la única y última vez fue en el Casón del Buen Retiro, aislado con cristales de alta seguridad. Ahora no está cubierto por nada, aunque si está bien protegido con sendas vigilantes que ladraban a cualquiera que se acercase a escasos metros. Las fotos debían realizarse desde un pequeño hueco, no entiendo muy bien el por qué, pero bueno, al menos pude capar esta instantánea.

En la sala de los alrededores había otros Picasos, Dalís, etc que hacían que valiese la pena estar allí unas horas.

Pero el motivo de este post no era hablar de esto, sino de una de las exposiciones temporales que había ese día. Trataba sobre el hiperrealismo y la verdad es que aún sigo intentado pensar si todo lo que allí ví es una vacilada, si soy demasiado viejo para comprender este nuevo arte o si hay algo que se me escapa de mi corta cultura.

Para empezar, en los pasillos del Museo, como bien puede apreciarse en la siguiente foto, me encontré con cientos de ruedas de desecho. Lo mismo que hace unos meses me encontré en un desgüace a las afueras de Madrid. Aún no he encontrado la diferencia entre el arte de esta foto y el arte de desgüaces La Torre.

En las siguientes salas de esta exposición temporal me encontré con infinidad de cosas incomprensibles para mí. Trapos tirados por los suelos, papeles estrujados, cartones pintados similares a los que hacen los niños, etc.

Pero al igual que el ejemplo de las ruedas, encontré en una de las salas, varias peceras de cristal llenas de bombillas, chapas y desperdicios varios. Esto me llamó mucho la atención, ya que en mi salón dispongo de algo similar, pero repleto de canicas. Este invento sirve de sujeción para un tablero de mi mesa de trabajo, que junto con unas luces y las canicas de colores, dan un color "diferente" a mi salón.

Entonces, mi reflexión es... soy un artista hiperrealista? por qué no se apreciar el hiperrealismo de otros? será que tengo un artista encerrado en una mente obtusa? Ser o no ser, esa es la cuestión.

REGRESO DE ALCALA

Después de un viaje de reconocimiento al nuevo trabajo de mi mujer y de un pequeño baño de cultura por los alrededores de Alcalá de Henares. El viaje de regreso en transporte público fue genial. Aquí se puede ver:

La verdad es que me lo pasé genial, algunos de mis colegas son unos fueras de serie. Que pena que no pueda contar las anécdotas de Cuenca. Si un día me dan permiso los personajes en cuestión, las publicaré. Son anécdotas dignas de hacer una película del tipo Airbag.

SATURADO

Me veo en la obligación de escribir algo en el blog porque lo tengo abandonado desde hace tiempo. El problema es que he pasado de ser una basura incompetente para la sociedad, a tener dos trabajos que me consumen las 26 horas del día.

El caso es que tengo bastantes cosas de las que escribir, pero mi tiempo últimamente es limitado. De hecho estoy probando alguna aplicación para el Iphone y aprovechar los pequeños momentos libres mientras me muevo en transporte público. De esa forma sacaría tiempo para escribir cosillas en el blog.

¿Lo haré o terminaré cerrando el blog?

Semejanzas

El otro día tuve una reunión en unas oficinas que casualmente tenían las mismas ideas de decoración que las que he utilizado en mi despacho.

Ahí van unas fotos comparativas que así lo reflejan:

No es sólo el uso de frases célebres, sino el estilo utilizado. Sobre todo el predominio del color blanco.

Iphone versus Blackberrry

Foto sacada desde un Iphone:

Anecdotario

En el andén del metro a primera hora de la mañana:

Una señora y dos señores y mayores exaltados hablando sobre los grandes problemas del país.

Fíjate lo del 11M, si ya te dije yo que era un golpe de estado encubierto contra Aznar, si estaba clarísimo. Y mira los de ayer manifestándose (se refería a la manifestación de los sindicatos contra la ampliación de la edad de jubilación a los 67 años), menudos sirvengüenzas, si estuvieran trabajando en lugar de manifestándose por las calles, otro gallo nos cantaría.

Entre tanto, mi mujer se les quedó mirando, un poco por lo que decían y otro poco por la exaltación extrema de estas personas. Y parece ser que no les gustó que les mirasen, porque la mujer mayor fue directa a mi esposa y la dijo acaloradamente: !!Qué miras!! !!Tú también eres una de esos!!

Mi mujer, educadamente, la pidió perdón por si le había causado alguna molestia el haberla mirado. Finalmente, uno de los señores mayores se llevó a la mujer y pidió disculpas a mi mujer por la exaltación de esta.

Y yo me pregunto... ¿Por qué se intenta crear este clima de malestar y de mala leche desde los medios de comunicación, desde los partidos políticos, etc? ¿Por qué nos utilizan para azuzarnos entre nosotros?

Por favor, respetémonos los unos a los otros, empecemos por lo básico, por la EDUCACIÓN, por el RESPETO, por los valores esenciales de cualquier convivencia. No vale con mandar unos euros de ayuda a Haití, o donar unos euros en la gala de Inocente Inocente. Estas cosas se tienen que demostrar desde abajo, forjando los pilares de una correcta EDUCACIÓN y RESPETO.

Arcos de Seguridad

En estas dos últimas semanas me he fijado en la poca utilidad que tienen los arcos de seguridad que podemos encontrar en supermercados, administraciones públicas, etc.

Mis últimas compras en los supermercados Mercadona han coincidido con que llevaba puesta un abrigo que recientemente me he comprado en el Decathlon. Por alguna razón, alguna etiqueta debe estar aún magnetizada y suena cuando paso por algún arco de seguridad del Mercadona.

Lo más curioso de todo es que tanto cuando entro, como cuando salgo, siempre pita el arco y nadie de seguridad me dice nada. Esto también me ha sucedido en una última visita al Decathlon. Entonces, ¿para qué ponen estos arcos? y ¿para qué están los guardias de seguridad?

Lo mejor de todo, hablando de arcos de seguridad, me ha sucedido hoy mismo en una visita a unas oficinas del Ayuntamiento. Resulta, que cuando iba a entrar, había una cola bastante grande de personas esperando a pasar por el arco de seguridad. La gente que pitaba al pasar, tenía que vaciarse los bolsillos y volver a pasar. De esta forma, la espera era más larga, pero todo sea por una mayor seguridad de todos.

Cuando terminé mis gestiones y al salir por la misma puerta que entré, donde se encontraba el arco de seguridad, el guarda ya no estaba y la gente pasaba sin problemas. Entonces, ¿para qué nos hacían esperar y vaciarnos los bolsillos diez minutos antes?

¿Por qué intentan vendernos todos los días la seguridad, los cacheos, las normas estrictas para subir a los aviones, etc, si luego nos saltamos las normas a la torera?

¿Algún día despertaremos de esta farsa? ¿o quizás no interesa a muchas empresas de seguridad que despertemos?

Atardecer en Madrid

El otro día tuve la ocasión de visitar uno de los edificios más antigüos y altos de Madrid. Aproveché para sacar algunas fotos desde las diferentes vistas que encontré.

Tenía ganas de hacerlo, ya que siempre había publicado fotos de otras ciudades y países, pero nunca había tenido la oportunidad de sacar unos fotos de Madrid desde algún sitio peculiar.

Aquí van unas cuantas fotos, me imagino que sabréis desde donde se han hecho.

45 Millones de Españoles, 45 Millones de Informáticos

Llevo tiempo queriendo escribir sobre la profesión de informático. Hacer una reivindicación de una profesión que quizás no sea tal.

En cualquier otra profesión se respeta bastante los consejos o indicaciones que son dadas por profesionales cualificados. Pero esto ocurre con los informáticos, nunca se recurre a ellos, ni siquiera cuando tenemos un problema con nuestro ordenador, con nuestra red o con nuestra impresora, por poner algunos ejemplos. Siempre se recurre al amigo, al primo, al cuñado o alguien que conocemos "que sabe mucho de ordenadores".

¿Por qué cuando tenemos un problema informático nunca acudimos a un profesional? ¿Por qué confiamos antes en un primo, un amigo o en el cuñado? ¿Por qué estamos tan seguro de que nos van a arreglar el problema? Encima, lo más gracioso, es que nos pensamos que nos los han solucionado, cuando a lo mejor lo que han hecho es crearnos otro, pero no llegaremos a darnos cuenta o si lo hacemos, nos percataremos cuando pase mucho tiempo.

Sin embargo, si algún día tenemos un problema legal, lo primero que se nos viene a la cabeza es ir a un abogado y vamos predispuestos a pagar por esa consulta. O si tenemos un problema fiscal, acudimos a un asesor, previo pago por la consulta.

Creo que el problema radica en la confusión entre saber de informática a nivel de usuario, conocer el manejo de ciertas aplicaciones informáticas, etc, y entre saber que hay por debajo de todo eso y comprender de verdad cuál es el problema y qué lo ha ocasionado.

Me gustaría que al menos, se respetase la profesión y que la gente comience a entender que un problema informático es similar a tener un problema de salud o tener un problema legal. Hay que aprender a pedir ayuda a profesionales del sector y además, pagar por ello.

Pd.- Doy las gracias a quién me regaló una camiseta con la frase estampada que dice así: "No, no te voy a arreglar el ordenador". Desde que me la pongo, ya no me molestan con problemas informáticos. Se acabó la caridad.

LA SEGURIDAD EN LOS COMERCIOS

El otro día, dando un paseo por un centro comercial, me fijé en dos detalles muy llamativos. El primero de ellos es ver como en uno de los ordenadores de atención al cliente del FNAC se podía ver perfectamente la contraseña de acceso.

El segundo caso que me encontré, se podía ver en la misma caja un montón de CV's apilados donde se el acceso a ellos era tan simple como leerlos o incluso cogerlos y llevártelos debajo del brazo. Esto sucedió en la tienda ADIDAS.

Se puede decir, que incluso siendo tiendas comerciales con renombre, su seguridad y la protección de datos, deja bastante que desear.

6 Conclusions

Risks and Recommendations

Risks for IPTV Operators

According with my experience, I have analyzed different in-home networking technologies than can be used for IPTV supporting and content sharing into the house. In spite of each technology has their own advantages and disadvantages, we have found out some risks than can be taken into consideration at the choice time. Some of this tricks can be on hand immediately and other ones could be revealed in a later moment: they will be called current and future risks.

CURRENT RISKS

• Short term most important problem is scalability. In each of the analyzed technologies, this could be solved:

1. WiFi - 802.11n
2. PLC (UPA) – Automatic management of transmitted power

Each of these technological evolutions will offer less interference by improving physical layer sharing.

• There is no technology mature enough, thus, it is necessary to use different technologies. This will result on the need to perform an on site test to determine if the user is suitable for a specific technology.

FUTURE RISKS

• Using 802.11n wifi pre-standard nowadays could be very risky if the final standard requires any hardware or software change in current wifi 802.11n devices.
• The existence of two PLC standards (UPA and HomePlug) may foresee that market chooses the option fitting the best, eliminating the other one. It is very likely that one of the technologies will disappear or its use will be much reduced.

Recommendations for IPTV Operators

Using an end-to-end video transport network. The operator approach will include three independent subnetworks: video, data and management.

Until a completely mature technology is available, it is highly recommendable to use different inhome technologies which offer the required alternatives for homes where problems are observed.

Using a management platform based in TR-069 will allow managing any inhome networking device, regardless of the technology used.

Usage of PLC technology, regardless of the chosen standard, will allow choosing among different equipment providers including the chosen chipset. This will ease negotiation, obtaining better prices due to simultaneous contacts with different vendors. For analyzed WiFi technology, Ruckus is the only manufacturer, which reduces negotiation options.

How to select the best solution for customers

Improving technologies of wireless networks, increases in hard disk-drive sizes and the increasing number of flat-screen TVs in households, makes the home network inevitable in the near future. Unfortunately the home network still remains more of promise than reality for high quality broadcast TV transmission, mainly because the standards and interoperability are some way behind.

The number of home networking options continues to grow with solutions offered using technologies such as GigE, Wifi IEEE 802.1n, PLC, HomePNA 3.0, and MoCA, IPTV end users can create high speed multimedia home networks that are able to carry various types of services at home.

As consumer demand for distributing, sharing, and engaging with digital content grows, so too does the need for a home networking middleware platform that provides these features.

Unfortunately for operators there is not a “one-size-fits-all” solution. No two households are the same, including the placement and number of outlets, or the type of medium that is connected to them. Additionally, the applications, services and traffic patterns will vary substantially from household to household. To compound matters, there are multiple home-networking standards (many proprietary) available, even over the same medium, that makes it difficult to determine the best solution.

Each technology has some type of trade off – whether it is cost to implement, interference issues, or coexistence issues. Therefore, operators must consider a number of factors when selecting a home networking solution. These include the types of services and number of devices to support, reach and available throughput, as well as ease of use and installation. As such, many operators are choosing to evaluate and trial multiple home-networking options.

Service providers must be concern about technology that could adapt their requirements and all stage marketing opportunities, from near-term services to long-term personalized services to the customer.

Also, device management is a key factor that needs to be biddable in order to offer different support levels to the customer. Indeed, Service Providers would be in advantage if they had some important design, monitoring and troubleshooting tools at any moment.

In the end, many operators might just find that the upfront expense of installing “clean cabling” may be the ultimate solution to meeting their requirements related to reach, rate, QoS and most importantly, Quality of Experience (QoE) for the end user.

So, the user-perceived experience of what is being presented by a communication service or application user interface defined by QoE, must be assured. This is highly subjective and takes into accounts many different factors beyond the quality of the service, such as service pricing, viewing environment, stress level and so on. In an IP network, given the diversity and multiplicity of the network, this is more difficult and therefore more critical to success than in other transports.

5 Managing home networking devices with TR-069

TR-069 Overview

TR-069, which stands for Technical Report 069, is a technical specification developed and published by the ADSL Forum (now called Broadband Forum). This report is entitled CPE WAN Management Protocol, and defines a protocol for remote management of end-user devices.

As shown in the next figure, together with TR-069, a number of protocols and specifications have been developed to fully cover the remote management and configuration of any end-user networked device as well as integrating this management system into the operator’s backend. However, since TR-069 was the first one of these standards, “TR-069” is commonly the term used to refer to the set of functionalities provided by many of these protocols.

TR-069 and associated protocols overview

The motivation for this family of standards comes from the current complexity of CPE device configurations, where more and more services are converging (broadband data, VoIP, IPTV…). This fact sets configuration complexity too high for the end user, making it necessary for the operator to manage the device to guarantee high QoS for the end user.

TR-069 is access and device type agnostic. It is currently widely used in DSL broadband market for CPE activation, while it starts being adopted by other bodies as Home Gateway Initiative (HGI) and DVB (for example, for managing IPTV STBs).

The goal of TR-069 and its associated protocols is offering the operator the remote management (including configuration, software/firmware updates, activation, troubleshooting…) of not only the CPE (understood as the border device), but also the inhome networked devices. The main protocols and standards are:

• TR-069. Remote management protocol and routing gateway configuration and management.
• TR-106 and TR-111. Combine to allow the remote management of devices on a LAN, even those using the private IP space behind a NAT gateway
• TR-98. Services differentiation (QoS) in the home network.
• TR-135 and TR-140. Object model for the STB and a network storage respectively. Goals of TR-135:

1. Support various types of STB, e.g. satellite, cable, terrestrial and IP STBs, with or without PVR
2. STB may be embedded in an Internet Gateway
3. Enable troubleshooting and remote configuration

• Note that, unless the STB or network storage are edge devices, TR-106 and TR-111 support will also be required.
• TR-104. VoIP Object model.

TR-069. Components and operation

TR-069. Components

To implement the TR-069 family of protocols, there are two main components required for the operator to deploy, the Auto Configuration Server (ACS), and the TR-069 client.

The ACS is located in the operator premises and it is ideally connected to both the operator OSS/BSS and the operator call center. Connection to the OSS/BSS will allow the ACS to access firmware/software/and configuration files in order to send them to the managed devices while connection to the call center will ease troubleshooting.

The TR-069 client is embedded in the different end-user devices (DSL CPE, home gateway, IPTV STB, VoIP phone…) supporting TR-069.

If both components are fully compliant with the standard specifications, any vendor ACS should work with any vendor client, making it possible for the operator to reuse any existing component and just include the complementary one.

TR-069 Operation

Connection

• TR-069 is based on sending Remote Procedure Calls (RPC) methods in SOAP messages (standard XML-based syntax), transported over a SSL/TLS, over HTTP, over TCP/IP connection, between a CPE and a Management Server.
• TR-069 is running on top of IP, and is transparent to any access technology (example TR-142: TR-069 for PON and fiber access).

Session

• A TR-069 session is always started from the side of the CPE, after the CPE has successfully authenticated. However, it is also possible to have asynchronous ACS initiated notifications.

Data structure

• In TR-069, parameters are arranged in an object model. The object model is build up as a tree like structure.
• TR-069: use of XML-open standard and human readable, easy to share info with other applications.
• Under a root object (a CPE) different service objects can be grouped. A service object can contain basic parameters, an object or multiple instances of an object. Those are building blocks that can be used to create a full object model for a complex device with different services.
• Notion of profiles, the object models are generic.
• TR-106 (template for TR-069 devices), TR-104: VoIP, TR-98: QoS differentiation, TR-135: STB, …

Security

• Security is enhanced with TR-069 by the CPE initiating all communication
• TR-069 utilizes a SSL/TLS connection to provide a single secure connection via x.509 certificate with Public / Private key technology to provide high security encryption and Certificate based authentication
• Moreover, the CPE can use the same x.509 certificate to provide encryption. Client devices are authenticated via the widely implemented HTTP authentication.

Software upgrade\Configuration\Download\File upload

• TR-069 defines separate methods to be used for download and upload of files and software (https). By using the TransferComplete RPC, the CPE can report the results of the download or upload action to the ACS.

Eventing

• The TR-069 eventing mechanism is based on providing active or passive notification on parameter changes. The notification mechanism can be activated by setting, with the SetParameterAttributes RPC, the appropriate Notification attribute on a parameter, an object or a path in the object model.

Functionalities and advantages for the operator

TR-069 family of standards is now described mentioning the functionality offered together with the advantages and added value offered to the operator as a result of these implemented functionalities.

TR-069. Main functionalities

• Auto-configuration and dynamic service provisioning

1. Initial CPE configuration
2. Re-provisioning at any subsequent time
3. Allow vendor-specific parameter configuration

• Software/firmware image management

1. Version identification
2. File download initiation
3. Notification of the success or failure of a file download

• Status and performance monitoring
• Log file and dynamic notification
• Diagnostics

1. Connectivity and service issues

• End-users devices (RG, STB, VoIP,…) management

1. TR-069 can go directly to end-users devices
2. TR-111 defines a mechanism for the ACS to learn which Gateway is associated with a given device.
3. TR-111 defines a standard mechanism to ensure the ACS can initiate contact with the Device using the TR-069 Connection Request mechanism (CPE maintains the NAT binding alive using the STUN process)
4. Or TR-069 can manage services on the remote gateway for other end-user devices (for example, manage the VoIP client for a telephone)

TR-069 / TR-135. Use cases for IPTV operators

A number of remote management use cases can be considered for remote management of TV platform devices. Some of them are presented here. The STB data model supports at least the functionality that is implied by these use cases. Classification of remote management activities can be done by making reference to the FCAPS model for systems management, where FCAPS stands for:

• Fault
• Configuration
• Accounting
• Performance
• Security

The STB data model does not need to account for all of the FCAPS functions. Usually, the FCAPS functions supported are Fault, Configuration and Performance. Accounting and Security functions usually take advantage of pre-existing infrastructure, so the relevant use cases are not considered here.

Configuration of the STB is done both by the IPTV Service Platform and by higher layer OSSs via the ACS. It can also be done by a trained technician, usually as a reaction to an end user complaint. This latter activity is here referred to as Trouble Management.

Performance Management can be performed by the ACS on a regular basis, to try and identify a malfunction as soon as it occurs, or by trouble management personnel on a limited set of STBs for special purposes.

Fault Management is generally driven by fault notifications from the STB. It is usually carried out automatically by higher layer OSS systems, based on signaling from the ACS.

Configuration

The ACS may perform some initial configuration of a newly installed STB. For example, it might initiate a channel scan in order to populate a DTT service list database, or it might set some user preferences such as audio and subtitling languages. During the initial configuration, the ACS can also update the STB firmware. Most of the initial configuration will be performed by the IPTV Service Platform.

Trouble Management

A trained technician may take control of the STB, generally in response to a customer complaint. The STB malfunction may be the result of improper customer settings, or may be due to network or hardware problems. Access to the STB data model allows the technician to carry out a number of tasks, namely:

• Verify/Restore the STB configuration. The STB data model parameters under the ACS control can be re-configured to the correct values contained in the ACS.
• Verify/Update software version. Incorrect software version (e.g. the STB was switched off for a long time and was not included in the last software upgrade campaign) can cause improper operation. In this case the operator can force an upgrade of the STB software to the latest release.
• Perform diagnostics. The technician can run diagnostic tests to identify whether the trouble is in the network (and at which point) or the STB and try to classify the trouble.

Depending on the cases, the technician can carry out actions on specific subsets of STBs (identified e.g. by a range of serial numbers, by a specific software/hardware version, by the geographical area they are in) or on single devices.

Performance Management

The ACS carries out automatic monitoring of STB performance. Performance reports can include QoS parameters (e.g. network parameters such as average bit rate, jitter and packet loss ratio), QoE parameters (e.g. visual quality indicators or indicators of how fast on the average the channel change is), usage statistics (e.g. how many STBs were on at a certain time, or for how long each of them remained tuned to a certain channel). Monitoring campaigns may be performed:

• Periodically on all STB devices to check that network and devices are working properly.
• On subsets of STB devices, for instance after identifying problems by means of periodic tests. Criteria to select subsets can be geographical or tied to specific characteristics of the STBs (manufacturer, hardware and/or software version).
• Periodically on specific STB devices. The problem here could be the management of a SLA (Service Level Agreement) with subscribers to premium services. Performance management could be used to identify problems on these lines as soon as they show up. Trouble management technicians could then act to (try to) solve them.

STB QoS and QoE reporting capabilities allow for “in service” “passive” measurements done at the service level. These are of fundamental importance to an operator in a number of cases, a number of which are listed hereinafter. Other cases are possible beyond those listed here:

• Understand and measure the QoE delivered to individual end users, via collection and aggregation of STB reports across the user base.

• Troubleshoot the service delivered: STB reporting allows near real time processing of collected reports and correlation of indicators that let the operator determine where the fault lies: in the head end, in the network, in the local loop, in the home network, or in the STB itself.
• Assess and measure the IPTV service as delivered in the mid to long term, and define and control whether performance objectives are being met.
• Pro-actively catch some hidden behavior which is increasing, and is reducing service performance, but has not yet been noticed by the end user.
• Pro-actively manage certain end users who are receiving a poor level of service but who have not yet called customer care.
• Configure and define operations management service quality thresholds on aggregated reports that can be tuned in order to take action before problems are noticed or reported by the end users.
• Understand loop and end-to-end behavior in order to design and assess error correction strategies for the IPTV service.
• Manage service maintenance and understand the impact on the IPTV service of any changes in the network, device upgrades or new device insertion.

Fault Management

The ACS automatically collects events from the STB for various reasons, including detection of faults. A way to detect faults taking advantage of TR-069 notification features could be the following: the data model contains parameters describing the operational status of specific functional blocks in the STB, and Active Notification is enabled for these parameters. In case of an STB error these parameters change their values and the Active Notification mechanism delivers information about the STB fault to the ACS. The ACS recognizes the fault and consequently notifies the OSS in charge of the End-to-End Fault Management operations.

4.8 References

The next figure shows a summary of main market references of different technologies. WiFi references come just from one provider, Ruckus Wireless and include operators in Europe and South America. PLC references come from both UPA and HomePlug associations and include operators in both South America and Europe. HPNA and MoCa, which have not been analyzed in this report are primarily used in North America (USA and Canada).

Market references for different technologies and vendors

4.7 Comparative

After individual tests performed to all technologies, this section pretends to summarize main conclusions obtained. The next figure shows a comparative of the main parameters evaluated.

*Data collect tests in scenarios mediums.

Inhome Technologies main parameters comparison

With the current analyzed technologies:

• Physical Rate. This is a parameter where all technologies are the same performance. While current PLC, MoCA, HPNA and POF technologies is able to reach 200 Mbps, WiFi limit is 54 Mbps with both 802.11a and 802.11g standards and 200Mbps with 802.11n standard. Next steps and standards will change PLC primacy since PLC will reach 400 Mbps with next generation chipsets while next WiFi standard, 802.11n, will achieve a physical bandwidth of 600 Mbps. This is shown in the next figure where next generation WiFi is over next generation PLC.
• Throughput. This parameter is based in scenarios standards with interferences on the spectrum radio electric, obstacles, electricity interrupters, etc.
• Distance. Wired technologies as PLC have larger range than wireless inhome technologies as WiFi. While range is not a problem for PLC in almost any scenario, it is more common to find coverage problems for WiFi. 802.11n is expected to improve distance and coverage by using MIMO and combining 2.4 GHz and 5 GHz bands.
• QoS. Ruckus Wireless and PLC offer more QoS options, including layer 4 while MoCA and HPNA QoS related features are limited to layer 3. POF doesn’t support QoS.
• Maximum number of devices. Ruckus Wireless and HPNA allow networks with up to 64 nodes while PLC and MoCA networks are limited to 32 nodes. However, in terms of TV distribution, this figure will never be reached since the throughput will limit the maximum number of STBs to be connected.
• Management. Management options are similar, and Ruckus, PLC, MoCA and HPNA support TR-069, which is the current best option for the operator to manage inhome devices. POF is unmanageable.
• Security. All technologies implement different security features since in all cases medium is shared. However, security is still an issue to be solved in WiFi technologies.

Roadmap for inhome technologies

References

The next figure shows a summary of main market references of different technologies. WiFi references come just from one provider, Ruckus Wireless and include operators in Europe and South America. PLC references come from both UPA and HomePlug associations and include operators in both South America and Europe. HPNA and MoCa, which have not been analyzed in this document are primarily used in North America (USA and Canada).

Market references for different technologies and vendors

4.6 Wired Solutions - Other Technologies

MoCA, HPNA and POF are other available options for the in-home IPTV market. These Technologies are under testing in my lab nowadays; due to the use coax technology has a marginal use with a low penetration level in Europe.

In the other hand, we have found some problems under odd circumstances that may imply not to recommend using them at this moment.

MoCA:

In spite of this technology has been introduced well into USA market due to historical cable operators and some other few American countries, there is no reference for great deployment into European Market. At glance, we can guess some issues that involve this technology:

• Marginal installation into European market.

• More suitable with a high cable networks penetration rate.

• Security is a problem: It does not support AES encryption.

• Service Providers must be careful with old pre-existing cable installations. Usually the work into a Do It Yourself environment, with defective existing installations.

• Complex Troubleshooting rises up due to unknown problems with amplifiers and active elements into the cable network.

• Restricted market to coax installations.

HPNA:

HPNA bring up with Telephony network to deploy an IPTV infrastructure at far end, but it introduced the Coax cable network with latest releases into the Alliance. Working with telephony lines, it carries only a few Mbps, so it is not a valid technology for HD IPTV support. By the other hand, with coax cabling support, it shares the same problems and network issues than MoCA:

• Marginal deployment into European market.

• More suitable with a high cable networks penetration rate.

• Service Providers must be careful with old pre-existing cable installations. Usually the work into a Do It Yourself environment, with defective existing installations.

• A few chipset manufactures available.

• Doesn't coexist with DOCSIS.

POF:

• Two options:

1. POF - USB adapter (without AC power adapter)
2. POF – Ethernet (with AC power adapter)

• Up to 30 meters.
• Easy installation, but high cost if technical installer is needed.
• Unmanageable.

4.5 Wired Solutions - PLC Technology

Power Line Communications comprises all technologies which attempt to transmit data over the power lines. The tests performed assess the technology for home networking, considering specifically IPTV data transmission.

The tests have been performed using both current existing standards: the one from the HomePlug Powerline Alliance (HomePlug) and the one from the Universal Powerline Association (UPA).

For the UPA standard, the tested equipment has been Comtrend devices with DS2 chipsets. For the HomePlug, Intellon chipsets inside Thomson and Devolo devices have been tested.

Performance

Both technologies have reached a 200 Mbps performance at the physical level. This means that actual throughput in a typical scenario could be up to 100 Mbps.

Performance in PLC is always conditioned by:

• Noise

Any domestic appliance (air conditioning systems, blender …) may alter the PLC signal. PLC equipment and signal is prone to be interfered by certain types of noise. Depending on the chipset used (DS2 or Intellon), certain types of noise are more harmful, creating a dependency chipset-noise type. PLC vendors solve this problem by using adaptation.

• Interference

Devices working in surrounding bands may affect PLC signal transmission. For example, interferences with bands used by amateur radio, cordless keyboard/mouse … prevent an optimal transmission. PLC vendors solve this problem by using adjustable notches.

• Electrical installation

There are several factors connected to the home electrical installation which have an influence on the performance:

1. Installation quality: cabling type, shielding, installation age …
2. Distance
3. Intermediate elements: electricity interrupter or electricity meter …
4. External elements: it could be a problem to performance if PLCs would be placed just to an electrical building infrastructure, another PLCs devices sharing the same electrical wiring, etc…

Solving these problems is not easy since it would involve a lot of effort and a high cost, which would discard this technology as an inhome networking solution.

During the lab tests, it has been observed that Intellon chipsets perform slightly better in non optimal scenarios, obtaining a higher throughput.

In optimal conditions, PLC technology can provide up to 4-6 SD MPEG2 channels. However, as shown in the next figure, actual throughput depends on several factors.

Throughput results on different scenarios (UPA)

*Bad. With electricity interrupter and poor quality electrical infrastructure.

**Intermediate. With electricity interrupter and standard quality electrical infrastructure.

***Optimal. Without electricity interrupter and standard quality electrical infrastructure.

Throughput results on different scenarios (HomePlug)

*Bad. With electricity interrupter and poor quality electrical infrastructure.

**Intermediate. With electricity interrupter and standard quality electrical infrastructure.

***Optimal. Without electricity interrupter and standard quality electrical infrastructure.

Quality of Service

Class of Service (CoS)

Class of Service (CoS) is a way of managing traffic in a network by grouping similar types of traffic (for example, e-mail, streaming video, voice, large document file transfer) together and treating each type as a class with its own level of service priority.

CoS woks with 3 bits within a Layer2 Ethernet frame header of the IEEE 802.1Q protocol. These bits specify a priority value of between 0 (signifying best-effort) and 7 (signifying priority real-time data) that can be used by Quality of Service disciplines to differentiate traffic.

Unlike Quality of Service (QoS) traffic management, Class of Service technologies do not guarantee a level of service in terms of bandwidth and delivery time; they offer a "best-effort." On the other hand, CoS technology is simpler to manage and more scalable as a network grows in structure and traffic volume.

One can think of CoS as "coarsely-grained" traffic control and QoS as "finely-grained" traffic control.

Type of Service (ToS)

The Type of Service is used to indicate the quality of the service desired. The type of service is an abstract or generalized set of parameters which characterize the service choices provided in the networks that make up the internet. This type of service indication is to be used by gateways to select the actual transmission parameters for a particular network, the network to be used for the next hop, or the next gateway when routing an internet datagram.

The TOS facility is one of the features of the Type of Service octet in the IP datagram header. The Type of Service octet consists of three fields:

Type of Service (ToS)

The first field, labeled "PRECEDENCE" above, is intended to denote the importance or priority of the datagram. The second field, labeled "TOS" above, denotes how the network should make tradeoffs between throughput, delay, reliability, and cost. The last field, labeled "MBZ" (for "must be zero") above, is currently unused.

The originator of a datagram sets this field to zero Routers and recipients of datagrams ignore the value of this field. This field is copied on fragmentation. The TOS field as a four bit field defined as following values expressed as binary numbers):

1000 -- minimize delay

0100 -- maximize throughput

0010 -- maximize reliability

0001 -- minimize monetary cost

0000 -- normal service

The TOS field value 0000 is referred to as the "default TOS". Although the semantics of values other than the five listed above are not expressly defined, they are perfectly legal TOS values, and hosts and routers must not preclude their use in any way.

The equipments will accommodate this TOS value according the service Level need into the home network.

Layer 4 QoS

Layer 4 QoS is a method of QoS classification that allows data streams to be classified based upon IP-Address and Port number. However, it does bring challenges in the real world in implementing this kind of classification reliably.

Security

Hardware-supported encryption is a very important feature, especially when using AES 128-bit or 256-bit. This feature is supported by DS2 chipsets while Intellon chipsets perform software encryption. This results in performance lowering when a robust encryption is used in Intellon chipsets.

For splitting networks belonging to different homes, a network ID and a unique key are used for data transmission. This way, any neighbor node which is able to physically detect another network won’t be able to decrypt data since the key is only known by the devices which the network comprises.

Management

There is considerable difference between features offered by DS2 and Intellon chipset for device management. The next figure shows main options available in both chipset equipments.

Management options available in PLC equipment

Advanced Features

DS2 offers a set of advanced functionalities which are not available in Intellon chipset equipment:

• Advanced tools which allow to collect data about the physical layer and to change settings in the device to improve the device performance. Among these tools, we have:

1. SNR Viewer. It allows to graphically controlling physical layer behavior when traffic is being sent between PLC devices.
2. Notcher. It allows enabling spectral notches in the desired frequencies which may be interfering with other devices (amateur radio, cordless keyboards…)

• It is also possible to develop applications at operator’s request. These applications specifically developed will allow for more flexibility during the deployment stage, for example, helping the installer when performing the installation at the customer premises. Portugal Telecom has developed a tool for the installers where a small piece of software tells them if electrical installation is suitable or not for PLC deployment. It just shows a red or green light.
• Another available feature is Automatic Multicast, which allows the automatic configuration of the multicast for the inhome network. IGMP snooping is configured at the same time that PLC receivers search automatically the emitter PLC device. This allows automatic provisioning during deployment stage or when substituting damaged devices.
• Smart Routing functionality optimizes throughput between devices since it chooses the best physical path based on the link quality. It is also possible to use a PLC device as a repeater in scenarios where interference or distance makes it impossible that two devices communicate directly in the network.
• Automatic Power Saving Mode switches the device off when no traffic is being sent through the devices. Once the device detects link activity, it switches on automatically. Now, new products based Intellon chipset are arising with that functionality.
• New DS2 PLC devices are fully backwards compatible with existing ones, allowing joint use, and being managed with the same tools.
• Devices include an embedded filter in the chassis, whose main advantages are:

1. There is no loss of an electric socket, indirectly avoiding the use of a multiple outlet for the connection of all nearby devices. This reduces noise caused by power supplies.
2. The use of the socket included in the PLC device for connecting the associated STB assures that the embedded filter eliminates any possible existing noise.

Roadmap

DS2 roadmap plans to launch low cost chipsets (100 Mbps) by the end of 2008. During 2009, launching of the 400 Mbps chipset is planned, which will be an evolution of the existing 200 Mbps Aitana chipset.

DS2 Roadmap

Also, Intellon roadmap plans to launch 400 Mbps chipset during 2009.

All planned DS2 chipsets to be launched will be backwards compatible with existing ones.

Backward Comapibility

Actually, Intellon chipsets are not backward compatibility.

In year 2009, DS2 will offer a functionality which will allow PLC devices the automatic management of transmitted power. This will prevent devices in neighboring networks from interfere among them. Each network will transmit with a certain power so that not to flood the neighbor physical layer, improving medium sharing and obtaining a higher performance.

4.4 Wireless solutions - Wireless HD

Wireless HD is the first specification aiming to define an interface for the transmission of high definition uncompressed video within the home environment. Its main characteristics are:

• Uncompressed HD video, audio and control data transmission.
• Operation in the unlicensed 60 GHz band.
• Beam forming and smart antenna technologies to overcome the intrinsic limitations of 60 GHz.
• Maximum range around 10 meters.
• Rates up to 4 Gbps, though the limitation of the selected technology would be set around 25 Gbps.
• Error protection, framing and timing control techniques.
• Secure communications with DTCP (Digital Transmission Content Protection).
• Intelligent device control system, for easy device discovery and control within the wireless video area network (WVAN).

This technology is backed by the most important consumer electronics device manufacturers: Intel, LG, Panasonic, NEC, Samsung, Sony, Toshiba, Broadcom, Sibeam… Nevertheless, Sony and Samsung have recently joined the group of interest for the WHDI standardization.

Wireless HD shows higher data rates, implements content protection techniques and has the support of major CE manufacturers. However, it has the important drawback of operating in the 60 GHz unlicensed band, what significantly limits its maximum range.

Wireless HD consists of one Coordinator and zero or more Stations. The Coordinator is

normally, but not always, a device that is a sink for audio or video data, e.g., a display,

but also potentially a media storage device like a PVR. A Station, on the other hand, is a

device that either has media that it can source and/or sink, potentially at the same time.

The device that is the Coordinator also acts as a Station in the wireless network. An example is illustrated in the following figure.

Wireless HD Inhome Scenary.

In the following table, WHDI and Wireless HD characteristics are compared:

In short, WHDI and Wireless HD are competing for CE manufacturers sector. Although WHDI is newer technology, it has a slight advantage over Wireless HD.

4.3 Wireless solutions - WHDI (Wireless Home Digital Interface)

WHDI (Wireless Home Digital Interface) is presented as a wireless alternative to state-of-the-art interfaces for the transmission of uncompressed video between video sources, players and displays (e.g. HDMI).

The objective of WHDI technology is to enable wireless streaming of uncompressed HD video and audio between CE devices such as LCD and plasma HDTVs, multimedia projectors, A/V receivers, DVD and BD players, set-top boxes (STBs), game consoles and PCs.

The main characteristics of this incoming standard are:

• Use of 40 MHz channel within the 5 GHz band
• 30 meters of maximum coverage
• Very low latency (<1ms)
• Up to 3 Gbps video transmission (enough for 1080p HD)

WHDI was defined so as to fulfill two main requirements: (i) it would be optimized for video transmission, and (ii) transmitted video contents would not be compressed.

In general, communication between video sources, players and displays (both in home or production environments) is based on the transmission of uncompressed video. This way it is possible to guarantee a maximum image quality, as well as diminishing latency due to codec selection and decoding processes, while avoiding illegitimate signal captures and enabling the implementation of content protection strategies (e.g. HDCP).

Because of the high bandwidth requirements of uncompressed video transmission, these connections have been wired so far. Traditional wireless networks support, in their best performance, data rates of only around a few hundred Mbps. Besides, these networks are in general packet transmission oriented, and use retransmissions and other link-level strategies in order to overcome the unreliability of the wireless channel.

Therefore, today the most extended approach for the distribution of digital video is the compression, then packetization of video contents for their transmission through wireless data-oriented networks. In this scenario, optimization consists in the application of strategies at different levels, aiming to mitigate the severe impact that packet losses and delays have for the perceived video quality.

On the contrary, WHDI solution proposed by the company AMIMOM takes a completely different approach. WHDI starts up from the premise that distributed video will not be compressed, and then adopts different optimization strategies from there, aiming to add capacity and robustness to the wireless transmission.

WHDI modulators analyze the structure of the uncompressed digital video inputs and split them in separate fragments, depending on their relative importance for video quality in the receiver. This way it is possible to transmit the most significant bits in a more robust manner, and devoting less channel resources to that information with minor importance. This additional robustness is obtained by tuning parameters such as power levels, spectrum assignment or channel coding. Hence, WHDI exploits aprioristical knowledge about what type of content is being distributed and applies Joint Source Channel Coding (JSCC) techniques in order to achieve the most efficient transmission for the most significant information, then boosting perceived quality in noisy environments.

At the physical level, WHDI shows many common points with other popular wireless technologies. Like Wi-Fi systems, WHDI is based on OFDM modulation, operates in the license-free 5 GHz band spectrum, takes benefit from MIMO (Multiple Input Multiple Output) antenna schemes and shares many functional blocks with 802.11n. However, WHDI uses a specific modem optimized for video transmission instead of using a data modem. WHDI also standardizes mechanisms for the transmission of audio and control information, but no support is given for the transmission of other data types.

As a conclusion, the following additional considerations can be made:

• Once the standardization process (directed form WHDI Special Interest Group) is done, WHDI will appear in the market as one of the most interesting alternatives for the wireless transmission of video contents within the Home Network. In this way, some WHDI modulators and receivers are recently starting to be shown in international forums.
• When evaluating this technology, the way the unavoidable signal degradation in the wireless channel (due to losses, delays, fading…) affects the received quality would have to be carefully analyzed.
• WHDI aims to take a strong benefit from the possibilities of multicast transmission. Many WHDI receivers would be able to connect to the same source device, after following a simple search-and-associate process (in a similar way to Bluetooth systems) with no additional bandwidth requirements. However, performance in this scenario must be observed since, unlike IEEE 802.11 standards, WHDI does not support Medium Access Control schemes in order to avoid collisions among several devices which can be interchanging requests at time.
• Finally, the content protection technique that will be implemented in WHDI interfaces has not yet been defined, although this feature has been taken as a requirement for the WHDI standardization group.

WHDI technology complements other wireless and wired standards with a new class of connectivity within the home. WHDI enable robust wireless delivery of uncompressed HD video between CE devices taking out HDMI cables.

WHDI Inhome Scenary.

For all the above, WHDI is recommended according to:

• Sharing High Definition content among all devices in the home.
• Eliminating unsightly cables.
• Hassle-free, low-cost installation.

4.2 Wireless solutions - WiFi

WiFi

Traditional WiFi based technologies have difficulties delivering multimedia traffic due to range limitations, unpredictable performance, inadequate quality of service (QoS), and for IPTV in particular, the gratuitous handling of multicast traffic.

As an alternative to standard WiFi solutions, I have tested Ruckus Wireless multimedia equipment. This has been the chosen option according to actual performance and new features (QoS and multicast managing, directional antenna system) of the devices and its current deployment in some IPTV operators both in Europe and America.

Ruckus Wireless covers the next Wireless bands; 802.11a/b/g/n. Each one of them has been tested like IPTV inhome networking solution. The test and results are explained below.

Performance

Performance is mainly conditioned by two factors:

• Technology limitations

WiFi solutions are not ready for a multicast environment. Packets are not retransmitted since link layer ACKs are not supported. Ruckus solves this problem by using a proprietary technology (patent pending) where ACKs which are needed for retransmission of lost packet at link level are sent. By doing this, anytime a problem at physical layer stops the transmission, the packets will be resent. This functionality is called SmartCast and has been developed by Ruckus, achieving good performance when using the WiFi transmission.

Ruckus equipment may reach a throughput up to 20 Mbps in 802.11a/b/g and up to 60 Mbps in 802.11n when used in optimal scenarios. This is equivalent to eight MPEG4 SD (2,5 Mbps) channels or one MPEG4 HD channel (8-10 Mbps) plus four MPEG4 SD channels in 802.11a/b/g and overhead of twenty MPEG4 SD channels or 6 MPEG4 HD channels.

• Environment limitations

The widespread use of 802.11b/g networks is flooding the WiFi radio band. In terms of data transmission (Internet traffic), the problem is affordable, but in video transmission, this is a tough problem.

There are two approaches in order to solve this issue, a short term one and a mid term one. The first one would involve using 802.11a since the 5 GHz band is much less used and comprises more available channels to be used (8 channels vs. 3 channels in 802.11b/g). The second approach will use 802.11n, which will unite the advantages of both 802.11b/g and 802.11a, resulting in better performance and less interference. Using 802.11n (once the standardization is finished) will allow to automatically regulate the transmission power and antenna direction (MIMO) in order to obtain a better coverage range and reduce neighboring interference.

Some problems have been found in some functionalities regarding channel selection. Both automatic channel selection (when the device is started) and CCS (Continuous Channel Selection) are not working as expected. This issue prevents a performance improvement.

Both in 802.11b/g and 802.11a, physical data rate is 54 Mbps; in 802.11n the physical rate is 200Mbps actually. However, actual throughput is scenario dependent. The next figure shows the results obtained from performance tests over different scenarios.

Throughput results on different scenarios (802.11a/b/g)

*Bad. High level of interference

**Intermediate. Mid-low level of interference

***Optimal. No interference

Throughput results on different scenarios (802.11n)

*Bad. High level of interference

**Intermediate. Mid-low level of interference

***Optimal. No interference

¹Only optimal scenarios have been tested with 802.11a since this is considered to be a “clean” band. If interference happens on the band, the performance would be as with 802.11b/g.

²Only optimal scenarios have been tested with 802.11n since this is considered to be a “clean” band. If interference happens on the band, the performance will decrease.

It is possible to limit bandwidth in each WLAN (Rate Limiting) defining these limits as Uplink and Downlink. This allows small improvements in certain scenarios where interference is a problem due to saturated radio band. It is not a solution but it may rather be considered as a temporary fine tuning.

Quality of Service

The rapid expansion of wireless networks is making it possible for new services and applications with digital equipments (PDAs, PCs, STBx, etc) to access information anywhere and at any time. The number of wireless terminals with 3P applications is growing rapidly, and therefore, it is important that the quality model in wireless networks be consistent with requirements.

As next generation for IPTV services will include a price tag for premium services, each application will deal with the proper level agreement. The service contract will fix the Quality of Service (QoS) expected by the application and the price the user is willing to pay for the service.

Four QoS modules are available in this technology: CoS, ToS, Layer4 and heuristic. The devices have the option to respect incoming QoS marking, transmitting packets through the Ruckus to the receiver with no change. This is useful if packets get the CPE with QoS preclassification.

Next section will describe different Wifi QoS technologies

QoS Preclasification inside Wifi

Class of Service (CoS)

Class of Service (CoS) is a way of managing traffic in a network by grouping similar types of traffic (for example, e-mail, streaming video, voice, large document file transfer) together and treating each type as a class with its own level of service priority.

CoS woks with 3 bits within a Layer2 Ethernet frame header of the IEEE 802.1Q protocol. These bits specify a priority value of between 0 (signifying best-effort) and 7 (signifying priority real-time data) that can be used by Quality of Service disciplines to differentiate traffic.

Unlike Quality of Service (QoS) traffic management, Class of Service technologies do not guarantee a level of service in terms of bandwidth and delivery time; they offer a "best-effort." On the other hand, CoS technology is simpler to manage and more scalable as a network grows in structure and traffic volume.

One can think of CoS as "coarsely-grained" traffic control and QoS as "finely-grained" traffic control.

Type of Service (ToS)

The Type of Service is used to indicate the quality of the service desired. The type of service is an abstract or generalized set of parameters which characterize the service choices provided in the networks that make up the internet. This type of service indication is to be used by gateways to select the actual transmission parameters for a particular network, the network to be used for the next hop, or the next gateway when routing an internet datagram.

The TOS facility is one of the features of the Type of Service octet in the IP datagram header. The Type of Service octet consists of three fields:

Type of Service (ToS)

The first field, labeled "PRECEDENCE" above, is intended to denote the importance or priority of the datagram. The second field, labeled "TOS" above, denotes how the network should make tradeoffs between throughput, delay, reliability, and cost.The last field, labeled "MBZ" (for "must be zero") above, is currently unused.

The originator of a datagram sets this field to zero Routers and recipients of datagrams ignore the value of this field. This field is copied on fragmentation. The TOS field as a four bit field defined as following values expressed as binary numbers):

1000 -- minimize delay

0100 -- maximize throughput

0010 -- maximize reliability

0001 -- minimize monetary cost

0000 -- normal service

The TOS field value 0000 is referred to as the "default TOS". Although the semantics of values other than the five listed above are not expressly defined, they are perfectly legal TOS values, and hosts and routers must not preclude their use in any way.

The equipments will accommodate this TOS value according the service Level need into the home network

Layer 4 QoS

Layer 4 QoS is a method of QoS classification that allows data streams to be classified based upon IP-Address and Port number. However, it does bring challenges in the real world in implementing this kind of classification reliably.

Heuristic QoS

This QoS classification method allows systems to inspect the size of the packet crossing the equipment and they would be redirect to the proper queue to process the packet. This method is based in packet inspection.

It is recommended to use CoS or ToS preclassification (packets are already marked properly when get the Ruckus) in an IPTV. It is also recommended to use a specific home network just for video transmission. Both recommendations will allow that all the traffic marked as video by the service provider is retransmitted with the highest priority within the home network.

In many cases, the most common practice would be prioritizing video traffic once it reaches the home network by applying advanced classification based on protocol type, origin/destination port or even packet size (heuristic). This is a not reliable way of prioritizing since there is a lot of IPTV traffic which is not UDP or which does not use the typical ports for TV reception. For example, communication traffic between the STB and the middleware for control/management, which may be standard web traffic (TCP).

Security

As main security features, Ruckus devices implement:

• Traffic transmission using WPA2.
• Two different networks and SSIDs, one for data and one for the IPTV service.

Management

Management protocols allowed are:

• Device access: HTTP, HTTPS, Telnet, SSH.
• O&M: TR-069, SNMP, Automatic & Manual Firmware Upgrade and Syslog.

An interesting implemented feature is the option to manage the any Adapter device from the Access Point with no need to access directly into the adapters.

A desirable feature would be an independent network for management where management traffic is actually is prioritized, which involves guaranteeing a high priority small bandwidth since this traffic amounts small traffic transmission.

Another desirable feature is the option for an automatic configuration of QoS parameters in every device in the home network once it is changed, for example, on the access point.

Advanced Features

The device is able to provide information about the WiFi band where the devices are transmitting. This feature, combined with an event management system will allow the O&M module to collect all the necessary data in order to know if the physical level is optimal for video transmission.

Roadmap

The most expected feature will be the fully standardization of 802.11n. This will permit to know if WiFi may become a mature enough technology to support massive IPTV deployments. The standard is currently approved in draft 4.0, but a full standardization is expected in end of 2009.