Low Cost Mod Final

7/28/2019 Low Cost Mod Final

1/36

NSN ATF

Girish Kumar P R

Krishnan Unni P

NSN ODC, Cochin

Possibility for low cost NE


2/36

Agenda

Exploring possibilities of Lowering cost for BTS used inexisting popular technologies like GSM,WCDMA

Business and Market directions for GSM and othertechnologies

10 different Research directions / latest trends to lowerthe cost for BTS

Software Defined Radio opening possibilities for COTSusage

Conclusion


3/36

Business Need to reduce cost for BTS

Heavy Competition and dwindling Revenue from calls forcing operators tolook for cheaper options

Rising Energy costs directly making a big hole in the operators pockets.

High energy consuming BTS tower and related electrical and cooling equipmentsare a big maintenance burden

Area/tower required for heavy equipment against spiraling rental cost Carbon footprint against global warming

Installation and site maintenance costs against rising labor costs

A deployed hardware has to be in service for a long duration of time tomake decent returns for the operator

But ever changing technology and rising consumer expectations requires too

frequent up gradations Network planning could go wrong and overcapacity in

wrong area makes a big loss, dynamic change of

deployment plans a necessity


4/36

GSM Not dying out yet

The highly predicted 3G revolution has not happened, instead 3Gand GSM are coexisting now also.

No Dispute that GSM is Worldwide success, still dominating over85% market share.

Over 100 million subscribers added in the last 3 years alone in thewestern world. And 3.5 billion customers worldwide.

GSM-EDGE along with UMTS is here to stay for another 10 yearsdue to seamless evolution possibility


5/36

Continuous voice growth, and non voice services, operational costkeeping GSM alive.

By 2012, the global 3G wireless market will include a total of 1.7billion subscriptions, of which 1.3 billion will be UMTS/HSPA with a

78% share of market

GSM/UMTS/HSPA customers will comprise the overwhelmingmajority of subscribers

In a country like Austria, over 35% of all broadband connections are

now mobile and the number for Slovakia is 30%.


6/36

Market Scenario

6


7/36

New Subscriber Market share


8/36

Trends in Deployment scenario


9/36

Mobile Broadband Technology Capabilites


10/36

Cost reduction Methods

10


11/36

Cost Reduction Solution 1: Single RAN

Multi Mode solutions

Huawei in 2008 released their first BTS with 2G/3G single RAN solution using SDR(software Defined Radio).

operators able to seamlessly switch from 2G to 3G or use both simultaneously.

Similarly NSN has its own product Flexi Edge BTS which has 2G,3G,4G built into onebox

cost efficiencies for operators in the areas of power consumption, transmissionexpense, footprint and maintenance costs

All these Revolutionary development is made possible by SDR approach.

Pros:- Change Deployment Configuration dynamically as per demand oftechnology/capacity

Cons:- usual size around 20Lts whereas single technology BTS are available at aslow as 10Lts


12/36

Software Defined Radio change

Started as Military application for adhoc changing encryption,

bandwidth, data rate etc.

New systems like 3G not aggressive due to hardware replacement

cost, within 3G regional differences in SCTDMA, WCDMA.

SOME BASIC details on SDR to be added.


13/36

Solution 2: Multi RAN

Virtual Base Station

SDR allows multiple virtual base stations (VBTS) running on a single BTShardware platform.

The expense of antennas, BTS electronics, and backhaul can all be shared.

Advantages for Mobile Operators

Separate Logical BSCs: Separate logical BSCs allow MultiRAN to usestandard BSS interfaces and enable shared active infrastructure roll-outs

Shared Antennas Between Operators: MultiRAN enables antenna sharingbetween multiple mobile operators, minimizing site costs.

Shared Backhaul Between Operators

Independent Management Systems: Each mobile operator has direct

access to network key performance indicators Power Consumption Savings: Reduced power consumption

versus running multiple non-shared networks.

Pros: Cost Reduction possibility at multiple fronts

Cons: Security Risk


14/36


15/36

Solution 3: By outsourcing (not really to India)

New BTS interface standardization likeOBSAI and CPRI opening newpossiblities.

The interface seprates Waveformgenaration and waveform transmission

Modular approch allows DSP,diplexer,,PA cards to be seprate andbaseband digital hardware, diplexerand high power RF tranciever to beoutsourced

The RF hardware elements requiresskilled (and expensive) engineering

effort and carry substantialdevelopment risk, yet yield minimalbenefit in the form of saleable features


16/36

Solution 3-Contd

The adoption of an SDRarchitecture for a BTS moves theinterface to the digital domain andleads to the concept of an RFblack box containing all of the RFaspects of both the transmitterand receiver(s),

Pros: Allows Venders toconcentrate on Applicationsoftware and compatativefeatures, rather than putting efforton costly hardware inovations.

Cons: Old operators losemonopoly New Players withstrong software side can easilyenter this highly elctronic market


17/36

Logic Behind Outsoursing

The radio parts of a BTS consist of two main section:

The radio frequency and analogue electronics, including the analoguebaseband processsing (eg., anti-alias filtering), IF components, localoscillators, low-noise amplification, power amplification, and so forth.These elements are mostly housed within the transceiver unit(sometimes known as the TRx), with the RF power amplifier anddiplexer ofen being seprate components.

The digital signal processing hardware, firmware, and software. Thisoften appears as one or more seprate cards (eg, one per RF carrier)plugged into a card frame, with a common bus. This part of the BTScontains the DSP devices, application-specific integrated circuit(ASICs), FPGAs, memory and clock oscillators necessary to generatethe modulation, coding, framing, and so forth required for the systemor systems which it is designed to support.

There is a significant difference in the rate and type of developmentbetween these two parts of a BTS system.

The RF PA is intended to work on say a band eg 900 MHz. The newdesigns are unlikely add any new functionalty, It is therefore realistic todesign the RF elements of the system to be future-proof in somemeaningful way


18/36

Logic Behind Outsourcing-Contd

Moores law

The most sensible option for future-proofing maytherefore be to invest significant design effort in future-proofing the RF and analogue baseband elements of thesystem

make the baseband digital cards in the form of current-technology throwaway items, much like the motherboardin a PC.

the RF power amplifier alone is likely to account for 50%or more of the cost of a third generation BTS installation(excluding site costs) within the next few years

The baseband, crest-factor reduction, DAC, andupconversion architectures are reuseable across a rangeof frequency bands


19/36

Solution 4: RRH and New BTS topology

In a conventional base station, the baseband and RF sections of thetransceiver are usually physically close to each other, and in many cases inthe same box.

The RF power is not, therefore, being generated close to its intended pointof use (the antenna), and a significant amount is wasted in getting it there

Separation of Digital and RF

The advent of OBSAI and CPRI, together with that of the digital-inputtransceiver systems means both digital and RF sections can be physicallyseparated.

A single location could be used to house the base-station digital andnetwork interfacing hardware for a number of base sites, required, forexample, to provide coverage in a large building (e.g., a shopping mall or anairport). The individual base sites would then merely consist of a number ofRF sections [orremote RF heads (RRH)], with each comprising an RF blackbox and an antenna and could upto 20kms away


20/36

RRH and New BTS topology-contd


21/36

RRH and New BTS topology-contd

Advantages of BTS Hoteling

Simplification of maintenance/upgrades

Reduction/elimination of base-site huts and cabins. Reduced power consumption.

Lower deployment costs..

Lower operating costs..

Higher reliability..

Ease-of-network expansion.


22/36


23/36

Solution 5: Smart antenna systems

A smart antenna system can bring significant benefits to both 2G and 3G networks, principally in the areas ofimproved interference cancellation and in enhanced system capacity.

DisAdv: Higher costs due to need for multifple RF PA, armoured feed cables and calibration systems

But smart use of SDR technology can mitgate the above costs and use baseband beamforming to make smartantenna systems acceptable to operators

There are two main methods of realising a baseband beamforming smart antenna system:

1. Utilise conventional (e.g., feedforward) RF-input/output multicarrier power amplifiers (MCPAs) in conjunctionwith a traditional transmitter/ upconverter. In this configuration, the power amplifiers are mounted in the cabinet at

the base of the mast and are cooled by fans and an overall air conditioning system for the cabinet itself. Therelative inefficiency of this type of amplifier and the requirement to overcome feeder losses in transmitting thesignals to the top of the mast, both lead to the requirement for large amounts of heat to be dissipated.

2. Utilise a digital-input SDR transmitter. The base-station cabinet no longer contains the RF elements of thetransmit chain, as these have now been moved to the masthead, along with the MCPA.

The Second approch also reduces power req by 60%

The first approch requires extra calibration equipment and feedback cables.

With an SDR transmitter-based solution, it is possible to lock each of the high-power transmitter units together(with one acting as the master), hence preserving their relative gain and phase properties without repeated

calibration. This saves both RF signal processing hardware and multiple RF cablesboth adding to cost and also,potentially, to reduced reliability.


24/36

Solution 5: Smart antenna systems-figure


25/36

Solution 6: superconducting today's fiction

tomorrow reality

Impact of Superconducting Technologies on Future SDR Systems

The main advantages claimed for a superconducting SDR solution include:

Very high-speed digital logic (~50 times faster than silicon LSI);

Very low power dissipation (10,000 times lower than for conventionalsemiconductor technologies). This figure does not take account of thepower consumption of the cryocooler and vacuum pump (if required).

Very high accuracy (5 parts per billion accuracy at 10V); Very high SFDR for both ADCs and DACs (due to the fundamentally different

way that quantisation takes place in a superconducting converter);

Very high sensitivity (claimed to be 60 dB better than a conventionalsemiconductor front end);

Very low noise (system is essentially thermal noise-free);

Ideal digital interconnects within an LSI chip (no R-C delay,hence, speed of light transmission);


26/36

Solution 6-contd

Large feature size and hence low mask costs, simple fabrication, and so forth . As anexample, the 40-GSPS ADC discussed above was fabricated using 3-m lithography(current silicon ADCs are fabricated using, typically, 0.25-m lithography or less).

There are, however a number of obvious disadvantages (at least at present):

Requirement for cryocoolers and (often) vacuum pumps. These are expensive items(particularly the cryocooler) and are also mechanical devices, leading to a reliability-cost compromise.

Packaging costs. The requirement to maintain a temperature close to absolute zero(4.25k, in some case]) or around 70 or 80k (for high-temperature superconductors])leads to a requirement for both airtight seals and good thermal insulation. Both ofthese are expensive to achieve.

Size. This has also, traditionally, been an issue, however the size of the mechanicalcomponents involved has come down significantly in recent years, to the point whereit is now close to being comparable with some larger base-station installations. Sizemay therefore become less of an issue for superconductig wireless solutions.

Essentially, the above disadvantages can be summarized as a major cost issue. Atpresent, superconducting solutions are a very long way from the cost of equivalent (atleast in terms of functionality) conventional solutions


27/36

Solution 7: Digital active antena

All Digital Active Antenna System another practical trend

Ubidyne latest product uB900 (demonstrated last month) is a unique, pure digital radio systemwith an Active Antenna Array that integrates a digital radio frequency unit directly into the antennahousing.

full featured digital radio system that is completely embedded into the antenna, thereby eliminatingthe need for motors, mechanical tilt hardware, coaxial feeders and other bulky equipment.

Multi-Standard Capability

The uB900 is a realization based on Ubidynes active antenna platform. The platform has anintegrated multi-antenna (MIMO) transmission and reception-ready RF unit for radio systemswithin the carrier frequency range from 400MHz to 3.8GHz that covers an array of radio standardssuch as GSM, UMTS (including the evolution to LTE) and WiMax.

The products includes remote radio head (RRH) functionality without the need for additionalantenna equipment and covers solutions for pico, micro and macro cellular communicationssystems.

Designed on a programmable hardware platform to maximise

system flexibility and scalability while optimising performance

and power (upto 50% less than legacy system)


28/36

Solution 8: Distributed Antenna System

The Distributed Antenna System (DAS) is defined as a multi-airlink/multi-frequency/multi-WSP (Wireless Service Provider), fiber-optic based,distributed antenna system. The concept is very similar to the RRH one.

greater flexibility and capacity that can be dynamically allocated andoptimized for varying traffic needs.

A DAS enables the independent operation of multiple frequency-bands and

multiple protocols across a single access network. Consequently multipleoperator network sharing can be supported by a DAS system.

Different link delays calibirated for different cellular technologies on digitalDAS using same optical fibre interconnections

DAS will demand in general an overall link data rate much higher than theone of a RRH unit, namely because DAS links could transport data relevantto multiple RF carriers of different standards and multiple WSPs


29/36

Medium term outlooks

The Micro- and Pico-Basestations have always represented a nichesegment and even in future they will likeky continue to take up about 10 to20% of the basestations market.

In a 2 to 4 year time frame the market outlooks for RRH and DAS basedsystems should become quite attractive for the following basic

considerations.The installation cost of a RRH Unit or of a RF subsystem of a DAS systemis pratically the same as the one of a Micro or Pico basestation, but theremotization of just the RF front-ends can reduce the O&M costs withrespect to a Micro or Pico solution.

Moreover, a DAS system allows also to share the same RF hardwareamong different WSPs even if using different standards. This last feature

will bring ultimately the equipment manufacturer to directly supply the WSPs, this way changing the current market scenario.Finally , an important technical characteristics of both the DAS and theRRH-based systems is offered by their suitability to optionally implementstatic or dynamic beam forming of an antenna array (smart antennas) andto support simulcasting.


30/36

Solution 9: Open BTS & Voip

Open Source initiative to develop GSM BTS uses SDR

Highly active group unperturbed by non-standard by BTS-BSC ABiSinterface

Current Abis interface are operator dependant stopping independentdevelopment of BTS

The Open BTS is not interfaced to a BSC but to Asterisk.

New Communication architecture using GSM BTS on air interface andVoIP on the Network side

Capable of reducing per call costs to operator by 1/10


31/36

Solution 10: Commonly Off The Shelf

Servers COTS

Requirements for COTS

Use reusable hardware platform to support multiple wireless services andstandards simultaneously.

Software should be cable of being deployed on individual servers or blades

for scalability while allowing broad range of RAN configuration IP based BTS will be the most easiest to implement, simplifyingconnectivity/ redundancy and highly developed protocol stacks are alsoavailable

Implementation in high level languages on high performance COTSpossible allowing operators to chose the best and most cost effectiveprocessors


32/36

Advantages for Mobile Operators

Upgradeability: Allows for the addition of new wireless standards quicklyvia remote software downloads, eliminating costly hardware duplicationwhen adding new services in an existing deployed frequency band. Speedstime to market through ease of use and cost minimization.

On Demand Capacity Allocations: The addition of system capacity can beexpedited simply by remotely downloading software. Traffic channels canalso be added to multiple sites within hours instead of weeks, thus ensuringyour RAN is flexible enough to meet immediate needs of end userperformance.

Remote Management: The reduction of the number of site visits throughremote network diagnostics and software downloads helps operators control

administration and maintenance costs. Cost Reduction: The reduction of site rental and electrical utility costs due

to a smaller base station footprint with fewer hardware requirements

enables a greener solution.


33/36

Wipro and COTS strategy

33


34/36

Wipro and COTS strategy

Wipro with its strong Software team can easily develop a BTS usingSDR based COTS

BTS on COTS + the best of the previous solution can be made intoa product i.e. software BTS with only core features can be developed in-house

Procure RRH and other electronic hardware

GSM BTS-BSC interface is a big hurdle but another open sourceproject Open BSC has demonstrated a standard BSC which can beused

3G BTS (node B) does not have non-standard interface issues andcan be easily developed on COTS using above approaches.


35/36


36/36

GirishkumarPandalaRamachandran

Software Engineer

[email protected]

Thank You

Low Cost Mod Final

Documents