What is IMS?

IP Multimedia Subsystem defines a framework for multimedia services over IP networks. It has been adopted by 3GPP as the standard framework for IP based services in 3G networks. IMS defines signaling mechanisms based on SIP and ensures end to end service delivery by defining different roles and functions to entities. The architecture has scope for

User authentication and authorization

Hiding the identity of the user from the network & other users

Roaming services and re-routing to the correct network

Support for different kinds of media and re-routing to appropriate media sources

Incorporation of new application services

Network resource utilization, management and control

Support for different kinds of access networks – both wireless and fixed

Support for multiple user endpoints and devices

Uniform charging and billing architecture

Hooks for network and service management

IMS has also been adopted by ETSI and ITU as a uniform service framework. The diagram below has origins at ETSI and defines their interpration of the IMS framework:

Although IMS has a history of over a decade in standardization bodies, the deployment has been rather slow owing to

complexity

need to upgrade network elements and the CAPEX involved

integration into existing OSS/BSS systems (this is supposed to give an OPEX benefit in the long term, but is a cost in the short term)

lack of clear standards in certain areas (ex) charging for roaming services

incompatibility between vendor implementations

ISPs/telcos improving existing services via incremental add-ons rather than full system upgrades

However, as ISPs/telcos plan their upgrade cycles, they are slowly bringing in IMS capable elements with a view to increase IMS based services over time. It remains to be seen how fast or slow the service uptake will be.
Wikipedia on IMS: IP Multimedia System

IMS receives a lot of attention from publications and analysts –  bringing forth the old saying about “you can love it, you can hate it; but you cannot ignore it”. Some samples are:

Fierce Telecom

Frost & Sullivan

ABI research

Infonetics on IMS marketshare

Techworld

Network World

Tutorials and white papers are available in plenty and the official standards can be obtained from 3gpp

What is IPv6?

IPv6 is the next generation internet protocol, designed at the IETF and defined in RFC 2460 in 1998. The goals of the new protocol were to provide expanded addressing capabilities with simple headers and provide direct support for features like address auto-configuration, prioritization via flow labels, security etc, to make for a better internet layer protocol.

By increasing the address field from 32 to 128 bytes, IPv6 offers the biggest advantage of large address space. As mobile devices and embedded devices proliferate, it is possible to interconnect more devices over the internet than ever before. The address space available with IPv6 seems to be adequate for a significant part of the conceivable future.

The header simplification and improved options processing, makes it easier for intermediate routers to forward IPv6 packets. Traffic flow identification to allow prioritization of different flows and handle them differently has been inbuilt into the protocol as has been the support for security via usage of IPSec.
In order to route IPv6, the routing protocols require upgrade and we have versions of OSPFv3, RIP-NG and enhanced BGP available.

Although the protocol supports auto-configuration via route-advertisements to the nearest router at startup (the address can be divided into a network specific portion that can be provided by the router and a host specific portion that is dependent on the MAC of the datalink interface on the host), support for DHCPv6 is also available to distribute addresses from a preconfigured pool.

Transitioning a network from IPv4 to IPv6 is not a simple job, even with small LANs – the complexity grows for larger network (enterprise-wide or ISP networks). The first issue is to obtain IPv6-ready hardware (ie) deploy hosts, routers and switches that have IPv6 implemented, even if the current network runs on IPv4. This can easily be done as part of the normal upgrade cycle for any organization (hence we find mandates for suppliers of network and host hardware/software to provide support for IPv6 as a criteria for purchase for the last few years). However, since IPv6 is yet to see large scale deployment, the extent of support and interoperability across vendors has to be tested in lab scenarios before live network deployments. For instance, certain network management features that are taken for granted with the current IPv4 systems, are not yet supported by all IPv6 implementations.

As the deployment level of IPv6 on the internet is not yet very significant, the current implementations are yet to undergo the test of time. The CAIDA (Co-operative Association for Internet Data Analysis) charts for the comparison of IPv4 and IPv6 deployment indicate very sparse usage currently for IPv6, although it is expected to pick up with the APAC countries leading the deployment.

At the application level, rewriting the application to use the underlying IPv6 infrastructure is necessary and hence the upgrade is an expensive process. Over time we see many product vendors announcing support for IPv6 and many standard applications today can run over IPv6 with configuration changes.

Transition technologies to ease IPv6 deployment have been in vogue as it is envisaged that IPv6 deployment is likely to grow only in pockets while the main internet backbones continue to run on Ipv4 for a long time. Some of the transition technologies have been controversial and these continue to evolve over time.

The v6ops work group at IETF continues to focus on operational issues with Ipv6 and current items for standardization include Ipv6 CPE routers and issues with router advertisements

As Ipv4 address exhaustion looms before us with a projected date early in the next decade , it is necessary for all technologists involved with IP networks and applications that run over IP to understand IPv6 and prepare for its deployment.

Wikipedia reference: IPv6

What is IP?

The Internet Protocol is used for transferring information across any network, predominantly the internet. IP version 4 or IPv4 is the current version that is in use. The primary identification mechanism for any entity on the internet is its IP address. The IPv4 packet structure (defining the header and payload format) as well as the functions of the IP layer are defined in RFC 791 and have been used since 1981 . The functions of the IP layer include encapsulating higher layer packets into IP datagram formats defined by the RFC and then transfering them across the network using IP routing mechanisms.

IP runs over several different datalink technologies like ethernet, WiFi, WiMAX, 3G RAN protocols etc and each of these protocol definitions include mechanisms for encapsulating IP datagrams into their packet structure.

IP is a connection-less protocol and does not require any kind of circuit set up. If an application requires an end-to-end circuit, it has to be handled at the transport layer or any of the higher layers. Consequently, IP does not guarantee delivery of packets at the destination, datagrams could travel by different routes and hence reach at different times. Missing datagrams would need to be redespatched and duplicate delivery is also possible. Packet reassembly is done at the destination and the complete packet is handed to the upper layer applications. IP is called an unreliable protocol (since there are no built-in mechanisms to ensure delivery of datagrams) and performs best effort delivery. However, the protocol is simple and robust enough to carry across the bulk of the internet traffic today.

As IP addresses are the main mechanism for identification of entities on the internet, they are an important resource and are governed by ICANN (Internet Corporation for Assigned Names and Numbers). Distribution of IP addresses to telcos, ISPs and enterprises are handled by regional registries in different parts of the globe. These providers in turn, hand over individual IP addresses (static or dynamic) to users and other entities in the network.

IP routing or transferring of packets from a known source address to an identified destination address is a complex process and there are several control protocols for performing IP routing with inter-domain and intra-domain routing handled differently. BGP is the main inter-domain routing protocol and there are several intra-domain protocols like OSPF, RIP and IS-IS.

IPv6 is the next generation of the internet protocol, but more about that in a separate discussion. Vint Cerf, who is one of the original authors of the IP protocol has also come out with a concept of interplanatary internet.

Wikipedia reference: IP

What is green telecom?

In a world where “green is the new black” can telecom stay away from green shades? A telco is almost always grappling with ways of reducing opex and as such focus on power efficiencies and recycling or extending the lifetime of products is something that happens on an ongoing basis. With “green” becoming the flavour of the month (or year), it is fitting to add more focus on something already in progress and make your presence felt the environment friendly bandwagon.

It should also be noted that telecom networks touch the lives of people and have become “lifestyle altering” systems. As such, it is a right place to build green and environment friendly initiatives into their business as well as actively promote it with their customers, so as to make “green” a part of the lifestyle alterations.

There are many different aspects to the “greening” of telecom, a few are listed here:

• Use of ROHS compliant hardware: EU has had focus on Reduction Of Hazardous Substances (ROHS) for over half a decade and equipment deployed in Europe requires to be free from substances like lead, mercury, cadmium etc. The NEPs (Network Equipment Providers) have been focusing on this for all new engineering since 2006 so as to be able to meet the certification requirements governed by the EU 2003 directive on ROHS and to be able to sell to telcos. Additional info at wikipedia: ROHS

• Hardware / software upgrades to reduce power requirements: As concerns on power requirements to run telecom networks increase, NEPs have been focusing on improved power management in their new engineering cycles – for instance Huawei’s ASIC used in its mobile BTS claims energy efficiency improvements of 40% . Automated switchoffs via software when equipment is not in use is another mechanism that reduces power consumption – for example, Nortel has SPM software available that turns off the radio network when there are no calls in progress

• Use of alternate “green” energy sources: Alternate energy sources like wind power and solar power have gained ground as viable and cheaper options against conventional sources of power. NOCs and datacentres are resorting to renewable energy to cover at least a part of their power requirements – for example: aiso.net. At least one vendor has come out with a solar powered BTS for GSM networks that consumes just 50-120 watts as against the conventional BTS requiring 3000 watts. The product is called WorldGSM from VNL and is being deployed in rural India. It also boasts of very simple installation requirements as well as next to no maintenance requirements.

• Energy efficiency measurement: While improvements in power consumption and heating requirements have been targeted, there has not been ways and means of quantifying such efficiencies and measuring which product is having maximum efficiencies. To this end, the Energy Consumption Rating (ECR) initiative has started work on test procedure specification for network and telecom equipment, the first set of equipment being considered being traditional routers and switches.

• Green data centres: Telco datacentres have major power and cooling requirements and have become a focus area for green initiatives. Datacentre design has become a  focus area with power budgets and air-conditioning budgets being attended to in minute detail. Even the location of datacentres is being reconsidered and choice of location is where natural cooling environments are available like near rivers or towards polar regions or even on the sea. More on datacentres here.

• Network infrastructure sharing: The installation and configuration of radio networks for mobile services is an expensive proposition. In recent times, mobile telcos have realized that this is a necessary activity that is sufficiently different from the business of offering mobile services to customers. This has given rise to network infrastructure companies that completely own and manage this infrastructure. The obvious next step was to reuse the infrastructure across multiple competing mobile telcos so as to better utilize capacity and reduce costs to individual networks. The Indian market had embarked on this fairly early, driven primarily by the cost reduction aspects and the need to expand to new areas. TRAI has recommended infrastructure sharing since 2007.

• Improved videoconference services: Business travel is a major source of generating carbon footprints and the telecom industry plays an important role in reducing this via improved videoconferencing services that no longer necessitate travel. While several lower cost solutions are also available, Cisco’s telepresence solution has garnered maximum publicity with Cisco itself claiming a travel cost reduction of $62 million since they started internal usage of the telepresence services.

• E-waste reduction and refurbishing equipment: With the current focus on attempting for maximum utilization of existing deployed equipment, the refurbished equipment market is thriving as never before – need citation. E-waste processing is another area of focus for telcos – and the awareness from both the telco and IT perspective is increasing. Vendors like Nokia have been running mobile phone take back programs since 2006.

• Paperless billing: One of the early focus areas was the generation and distribution of hard copy bills. Almost all telcos offer bills via email or their dedicated customer portal. In fact, this is the default option with the hard copy bill being generated and despatched only in specific instances.

As can be seen there are different approaches to green telecom, and not all have been enumerated above. Indeed, ITU has set up a Focus Groups on ICT and Climate Change that is charged with developing guidelinesand standards in this area.

A recent LR article on this topic: Recent moves in green telecom

What is 3G?

3G refers to the third generation of mobile communication standards and is the evolution path for GSM which is considered as the second generation standard. 3G offers advantages in terms of better spectral efficiency, higher datarates and sophisticated services.

Although circuit switching is defined as part of the 3G standards, it is considered as the first step to full packet switched radio networks – moving from the ATM core towards an all-IP network in the core.

3G networks also bring in the first attempt to separate the radio network, the core transmission network and the services architecture so as to allow each area to evolve separately.

High data rates are expected from 3G networks: a minimum speed of 2Mbit/s and maximum of 14.4Mbit/s for stationary users, and 348 kbit/s in a moving vehicle, although this is not mandated by the standards. With HSPA (High Speed Packet Access) support, datarates are expected to stabilize at 14.4 Mbit/s for downloads with 5.8 Mbit/s for uploads.

The spectrum bands defined by the UMTS Forum for 3G are 1900-1980 MHz, 2010-2025 MHz, and 2110-2170 MHz bands for terrestrial UMTS and this is largely accepted in Europe and Asia. Each country has chosen to auction its 3G spectrum to mobile providers at different timelines (the telecom downturn in the early part of this decade is attributed to overspending to acquire the 3G spectrum licenses). Japan(2001) and Korea (2002) were among the initial deployers and services were deployed pre-3G standards. Subsequently 3G services are available in most parts of Europe and Asia and pockets of USA. India has been postponing the spectrum auction (expected later this year), however, the state owned operator has been allowed to commence services pre-auction and hence services are to be rolled out this year.

While 3G represents the next level of sophistication in services available to the consumer, there are several stumbling blocks in the form of spectrum availability and costs, handset costs, infrastructure upgrade costs, technology stability and interoperability (especially important for roaming services), in-progress services infrastructure standardization.

The IP Multimedia Subsystem is an important aspect of 3G networks that provide the framework for complex multimedia services. The topic is important enough to afford a separate discussion subsequently.

The 3G standards are available for download from the 3GPP website.
Wikipedia is your friend: 3G

What is GSM?

GSM or Global System for Mobile communications is the dominant standard for mobiles as against CDMA which is popular in specific regions. From an international roaming and coverage perspective, it is easy to obtain services via GSM in most regions of the world, although it started life as an European standard.

GSM uses time division multiplexing (TDM) to allow multiple users to access the radio spectrum and has multiple frequency bands assigned – 900 MHz or 1800 MHz bands are common.

One of the advantages of GSM is the availability of the detachable SIM (Subscriber Identity Module) that is a smart card storing basic information on the subscriber. The detachable SIM permits end users to switch handsets easily, and the same handset can be used across different GSM providers.

For mobile data services, the GSM world offers GPRS (General Packet Radio Services)that offer upto 60 kbps (realistic measures, not theoretic limit from standards) and the subsequent enhancements via EDGE (Enhanced Data rates for GSM Evolution) that allows upto 150 kbps.

The biggest advantage of GSM is that it is a well deployed technology that has seen years of use, subsequently the costs of deployment are cheap in comparison to other newer radio technologies.

GSM standards were defined by ETSI and are now available for free download from the 3GPP website: The table indicates which standards pertain to GSM
The GSM Association is very active in supporting the deployment and usage of GSM networks and focuses on interoperability with subsequent families of standards.

Several technical books on this topic are available like: The GSM system for mobile communications by Michel Mouly, Marie-Bernadette Pautet, published in 1992 by Telecom Publishing ISBN:0945592159

Wikipedia is your friend: GSM

Bharti & MTN

The media has been full of info on the newly revived interest between Bharti and MTN. The interest in these discussions is due to the following:

• Valued at $23 billion it is the largest telecom M&A deal this year (of interest in a “recession” economy)
• Both operate in emerging markets and have strong growth and expansion plans – the merger would probably lead to a telecom powerhouse in their regions of operation
• The financials of the deal are rather unique in their usage of stock - ET

The discussions will be on till end July, and the world watches to see if the deal will be clinched this time around.  Meantime, there is objection to the deal from some investors and employee unions of MTN.

What is firewall?

A firewall is a primary security mechanism employed at the border of a network (could be any kind of network – home network or enterprise network or service provider network etc) to keep out and regulate unwanted traffic. The external network is considered an “untrusted” domain and the firewall is normally part of a demilitarized zone (DMZ) behind which is the “trusted” internal network. The DMZ is an intermediate zone where all external points of interface with the internet reside.

In the simplest form a firewall denies admission to all traffic, except for those that meet special criteria:

ACLs or Access Control Lists provide the acceptable list of external sources that are allowed to access the trusted network or servers within the DMZ. ACLs also provide the authorization privileges for the external users to determine what they are allowed to do within the network.

Port based access – allowing or denying specific types of application traffic

Stateful packet filters that take different actions depending on whether the packet is at the beginning, middle or end of a flow

Firewalls also are configured for dynamic update of the rules to prevent DoS (Denial of Service) attacks in situations when there is a similarity in an attack pattern that can be determined. The kind of updates that can be done is only limited by the amount of intelligence programmed into the software.

Modern day firewalls incorporate malware and virus protection techniques by rejecting traffic based on specific data in the contents. However, these are generally specialized software that are used independent of a classic firewall.

Firewalls can be implemented via simple configurations on linux boxes or gateway routers. At the other extreme sophisticated hardware based firewalls can be obtained from speciality vendors.

In the modern network, the firewall forms a first line of defense against external attacks and is a necessary perimeter security mechanism.

Wikipedia is your friend: Firewall

What is fibre (telecom perspective)?

Fibre in the telecom world refers to optical fibre communication technologies which offer a high bandwidth alternative to traditional metal based cabling.

Optical fibres are made of glass or plastic and can carry light signals of many different wavelengths in multiple channels and achieve datarates of 10-40 Gbps. Since the laying of an optical fibre and its interconnects with network elements are both expensive, fibre technologies initially saw deployment in long-distance cabling where the benefit of low attenuation of signal with distance overrode the cost factor. More details here: Optical fibres

In comparison with electrical transmission, optical fibre technologies are less susceptible to interferences, and require less number of repeaters to be used. Additional info here: Fibre optic communications

Over time, with the user requiring higher and higher bandwidth levels, it has become necessary to bring the fibre closer to the user and the various FTTx concepts have come into existence. Wikipedia offers a good illustration: Fiber-to-the-x. The basic concept is that of continuing to use existing metallic cables closer to the user while using fibre upto a point somewhere near the user, hence we have:

• Fibre to the Node – FTTN – where the fibre is terminated at the telco’s cabinet or node, which would be at a little distance from the end user, and existing coax or twisted pair infrastructure is used beyond that point upto the user.

• Fibre to the Curb – FTTC – where the fibre comes upto a junction box serving a bunch of customers, probably on the curb of the street

• Fibre to the building – FTTB – where the fibre comes upto the distribution point at the building but not into the residences, ideally suited for multi-tenanted buildings.

• Fibre to the home – FTTH – where the fibre comes into the home and fibreoptic outlets are used for connecting the networked devices.

Obviously, the cost factor goes up as the fibre comes nearer and nearer the house but the seeming advantage is to do away with the last mile inconsistencies and bottlenecks for bandwidth delivery.

Optical networking technologies can use active or passive mechanisms to split the optical signal from the telco’s central office to individual customers:

• AON or active optical network uses electrically powered switches, routers etc to achieve this with active Ethernet or Ethernet in the first mile being one of the common mechanisms used.

• PON or passive optical network uses unpowered optical splitters and hence is considered a cheaper option. Downstream signals are encrypted and broadcast and decrypted at the destination. Upstream signals are typically collated using TDMA (Time Division Multiple Access) where each residence transmits at a different timeslot.

The physical media standards are defined by ITU as part of its G-series: Transmission systems and media, digital systems and networks (too many to be quoted individually) while the Broadband forum includes fibre as a network deployment technology in its specifications since its 3.0 release (approximately from 2007)

What is ethernet?

Ethernet is a LAN (Local Area Network) networking standard originally defined in 1975 which defines signalling interchange, physical characteristics, data formats for the physical and data-link layer of the network. Over time it has morphed from being a simple in-building interworking technology to a robust communications standard offering high bandwidth with specific features to support higher layer applications as well as covering various physical media.

Ethernet originally operated on the concept of a shared broadcast medium where every participant heard everyone else, but accepted only the traffic directed towards them specifically. Collision detection and retransmission were an integral part of the protocol as was required by a shared medium. This was the basic definition of a CSMA/CD protocol – Carrier Sense Multiple Access with Collision Detection.

The overview at Wikipedia is here: Ethernet

The IEEE standards define Ethernet and it is part of the 802.1 family of standards for LAN/MAN architectures. 802.1 defines the common aspects of all LAN/MAN standards while 802.3 defines the specifics for Ethernet.

The basic physical layer characteristics are defined in the basic 802.3 standard and ethernet works over different physical media like UTP (Universal Twisted Pair), Fiber etc.