EDGE Compact and EDGE Classic Packet Data Performance full report
#1

[attachment=3431]

EDGE Compact and EDGE Classic Packet Data Performance
Abstract
Even though cellular radio services have been extremely successful in providing untethered voice communications, wireless data services have captured only a limited market share so far. One obstacle for wireless data services is their limited peak bit rates. Existing wireless data rates, up to several tens of kb/s, may be over one order of magnitude short of what is required to make popular applications user-friendly. To accomplish these necessities we go for EDGE(Enhanced Data Rates for GSM Evolution) employs adaptation between a number of modulation and coding schemes (link adaptation) as a means for providing several hundred kb/s peak rates in a macro-cellular environment while supporting adequate robustness for impaired channels. In this paper we discuss the two phases of EDGE, classic and compact. We start our discussion with the page link adaptation and incremental tendency techniques. Secondly we go for discussing EDGE classic and compact systems and their deployment scenarios, followed by downlink performance comparison and some MAC layer enhancement techniques which improve the performance.
Key words
EDGE, GPRS, GSM, GMSK, MCS, SINR.

Presented By:
K. Naveen Krishna (Y1EC054): K.K.S. Anil Kumar (Y1ECO27):
V.R.Siddhartha Engineering College,
Vijayawada.


1. INTRODUCTION
The GSM system is the most popular second-generation wireless system today. It employs TDMA technology to support mobile users in different environments. This system, initially standardized and deployed in Europe, is currently deployed worldwide. The TDMA community adopted EDGE for high-speed data services in the third-generation, radio transmission technology proposal to ITU for IMT-2000 (UWC-136). Typically, current GSM service providerâ„¢s employ 3/9 or 4/12 reuse plans and they may not impose a different frequency reuse plan for EDGE, which is thus termed EDGE Classic. However, in North America, initial deployment using 1MHz in each direction is being considered due to limited spectrum and the potential need to re-deploy spectrum currently used for ANSI 136 systems. This implies very aggressive frequency reuse having a minimum of only three 200-kHz frequency carriers. This means allocating one frequency to each of the three sectors per base station and reusing the frequency set everywhere (1/3 reuse) and providing control signaling with extra reuse protection in the time domain, which is named EDGE Compact due to its compact spectrum requirement.
2. MODULATION AND CODING ADAPTATION
The basic concept of EDGE is to provide higher data rates per radio time slot than is possible with GMSK modulation. This allows the support of existing services with a lower number of time slots. In addition it allows the introduction of new services with up to 59.2 kb/s per timeslot or almost 480 kb/s per carrier in multi-slot operation, hence offering an evolution path for GSM to support multimedia applications.
2.1 Radio page link formats
Discussions in the ETSI workshops resulted in selection of 8PSK/GMSK to provide higher rates than the GMSK modulation with small envelope fluctuations and to provide backward compatibility to GSM and GPRS. The EDGE concept can be seen as an extension of GPRS for packet service, which is called EGPRS. ETSI has also combined EDGE with circuit switched data modes, and these modes are called ECSD. Efficient page link adaptation is a key feature for EDGE and has been jointly developed with EDGE enhanced modulation. With a high degree of compatibility in the bandwidth and symbol rates with GSM, EDGE provides higher rates for users with good signal to interference plus noise ratios (SINR). This is achieved by employing lower channel-coding redundancy and/or 8PSK, which carries 3 bits per symbol (as opposed to 1 bit per symbol achieved by GMSK). Table 1 shows the bit rate provided by different MCS (Modulation and Coding Scheme) modes. An EGPRS capable terminal will have 9 modulations and coding schemes available compared to 4 for GPRS.
Table 1: An overview of packet data services for EDGE
2.2 Radio page link control
Fig 1. Throughput as a function of SIR for different transmission modes
Radio page link control selects among the MCS options, in response to SINR or equivalent quality measures. Link adaptation explicitly changes MCS modes based on page link quality estimates, and is also called mode selection. Hybrid ARQ transmits additional redundancy bits after errors are observed. It is made possible by sending the packets with different puncturing patterns from the same mother code during retransmission. This allows data transmission to begin with low redundancy and increases redundancy only when errors occur, thus adaptively changing the effective date rates.
The criterion for selecting a particular data-rate as proposed is defined by
S=Rc(1-BLERc)
where Rc and BLERc are the data-rate and BLER (Block Error Rate, where a block is the RLC (radio page link control) block.) for the transmission mode chosen. Figure 1 shows the throughput as a function of SIR for different modes. It is found that this threshold criterion is generally effective in achieving a high aggregate system throughput, but the QoS (Quality of Service) for individual users can be degraded as system load increases. Furthermore, page link adaptation requires the receiver to continuously perform quality measurements and provide timely feedback to the transmitter, so typical operation may be with somewhat higher thresholds.
3. EDGE Compact and EDGE Classic
3.1 Classic
GSM systems are usually planned on the basis of 4/12 (4 base stations, 3 sectors each, per cluster) or 3/9 frequency arrangements. The carriers that contain broadcast control channels (BCCH carriers) are required to transmit continuously and without hopping on control time slots to facilitate handoff measurements, control channel acquisition, and so on. These carriers usually are arranged in a 4/12 reuse pattern. Traffic channels can frequency-hop and, on non-BCCH carriers, they can use discontinuous transmission (based on voice-activity detection), and if so, typically are arranged in a 3/9 reuse pattern. These arrangements provide the strong SIR protection typically required for delay-intolerant voice services and non-acknowledged control channels. EDGE Classic is defined to be a system using continuous BCCH carriers that are typically in a 4/12 or 3/9 reuse pattern and which requires at least 2.4 MHz bandwidth in each direction. Additional traffic carriers, if available with higher total bandwidth, can be deployed under a lower reuse factor. Some system operators, particularly those in North American where 3G spectrum has been partially allocated for PCS, have to re-allocate in-service spectrum to deploy EDGE.
3.2 Compact
In that case, EDGE Compact may be used for initial deployment using as little as 1 MHz in each direction allowing only three 200-KHz frequency carriers. This means allocating one frequency to each of the three sectors per base station, and the frequency set is reused at every base station (1/3 reuse for EDGE Compact mode). While good spectrum efficiency is achieved, the provisioning of common control functionality, such as system broadcast information, paging, packet access and packet grant, cannot be deployed with 1/3 reuse. 4/12 or 3/9 reuse is required for reliable control channels. In order to achieve adequate co-channel reuse protection for the control channels, reuse in the time domain is exploited, which requires frame synchronization of base stations. Figure 2 shows an example with 4 timing groups in addition to 1/3 frequency reuse to obtain 4/12 reuse for the control channels.
Fig 2. Example cell pattern for a 4/12 time and frequency reuse
EDGE Compact uses discontinuous transmission based on a 52 -multiframe2 (a multi-frame consisting of 52 frames) and designates different time slots and frames for sending control information. In blocks (blocks are non-overlapping and are each comprised of timeslots from the same timeslot number of 4 successive frames) when a sector belonging to one of the time-groups transmits or receives common control signaling (serving time-group), the sectors belonging to other (non-serving time-groups) are idle. This creates an effective reuse of 3/9 or 4/12, which is necessary for control signaling, while allowing 1/3 reuse for the traffic channels. Specifically, slots 1, 3, 5 and 7 are used for timing groups 0, 1, 2 and 3, respectively, to send common control information on frames 0-3, 21-24, 34-37 and 47-50. Frames 12, 25, 38 and 51 are also not allowed for traffic as they are reserved as idle frames or to send timing advance, frequency correction or synchronization information. More frames can be allocated for control signaling as needed. Therefore, up to 32 frames or 8 blocks per 52 multi-frame can be allocated for traffic channels on the designated control slots. This is 2/3 that of a regular slot capacity in which 48 frames in a 52 multi-frame are used for traffic in a given non-BCCH slot. When using 3 time-groups (i.e., effectively 3/9 reuse), one of the 4 time-groups is unused and it is instead used as a traffic channel. Figure 3 shows the control channel BLER distribution for both 3/9- and 4/12-reuse based systems. For about 90% of the cases, the BLER is better than 4% and 15% for 4/12 and 3/9 reuse, respectively, corresponding to the overall average BLER of about 2.4% and 5.2 % (not shown in the figure), respectively. The performance of the 3/9-reuse system may not be reliable enough at the tail end of the distribution. However, since the traffic channel performance is highly correlated with that of the control channel, i.e., a mobile station with poor control channel BLER most likely cannot support reliable traffic performance, the tail end performance may not be very crucial. The control channel performance for EDGE Classic is expected to be similar because the same reuse factor is employed for the control channel.
Fig 3. Distribution of BLER for control channels
4. DEPLOYMENT SCENARIOS
The minimum spectrum required for Compact deployment is 600 kHz and that for Classic is 2.4 MHz (neglecting guard bands in both cases). Therefore, at 2.4 MHz and above, there exists the option of either Compact or Classic deployment. The choice of system is partly dependent on the performance of the systems. The performance in turn is dependent on the reuse configuration employed in the deployment. For the purposes of this study and to eligible valid comparisons, the reuse configurations are such that control channels are always at 4/12 reuse while traffic channels are at 1/3 reuse whenever possible. The exceptions are the traffic channels of a Classic control (BCCH) carrier, which are at 4/12 reuse. We also consider the same control-channel capacity (one active slot of a carrier) for both cases under all scenarios. Table 2 and the text following describe the scenarios considered:
Table 2: Deployment Scenarios
a) 600 kHz deployment
i. Compact (Scenario 1)
There are three 200 kHz carriers, one per sector of a tri-sectored base station. A carrier in a given sector can use the even-numbered slots and the unused portion of the odd-numbered control slots for traffic in a 1/3 reuse. Here, we do not consider the unused portion of the odd numbered control slots.
b) 2.4 MHz deployment
i. Compact (Scenario 2)
There are twelve 200 kHz carriers. Three of the carriers are deployed in a configuration identical to that of the 600 kHz deployment. The remaining nine carriers are dedicated to traffic and deployed in a 1/3 reuse configuration. Therefore, any given sector of a tri-sectored base station has four carriers, three of which have eight traffic slots each and the fourth has four traffic slots, all in a 1/3 reuse pattern.
ii. Classic (Scenario 3)
There are twelve 200 kHz carriers, all continuous control carriers with one allocated per sector of a tri-sectored base station. Therefore, a given sector has one carrier of which one slot is dedicated for control and seven slots are dedicated for traffic. All control and traffic slots are in a 4/12 reuse configuration.
c) 4.2 MHz deployment
i. Compact (Scenario 4)
There are twenty-one 200 kHz carriers. Three of the carriers are deployed in a configuration identical to that of the 600 kHz deployment. The remaining eighteen carriers are dedicated to traffic and deployed in a 1/3 reuse configuration. Therefore any given sector of a tri-sectored base station has seven carriers, six of which have eight traffic slots each and the seventh has four traffic slots, all in a 1/3 reuse pattern.
ii. Classic (Scenario 5)
There are twenty-one 200 kHz carriers, twelve of which are in a 4/12 reuse pattern and the remaining nine in a 1/3 reuse pattern. Therefore, a given sector of a tri-sectored base station has four carriers. One of these is the continuous control carrier and it has seven slots dedicated for traffic in a 4/12 reuse pattern. The other three carriers have a total of twenty-four slots in a 1/3 reuse pattern.
5. PERFORMANCE COMPARISONS
Figures 4 and 5 show the average user-packet delay as the throughput per base station (in three sectors) increases for the 2.4 MHz and 4.2 MHz scenarios, respectively. Here we can clearly see the trade-off between QoS, as determined by the delay experienced by the web-browsing users, and the system capacity, as indicated by the total throughput that a typical base station can deliver to all users who are sharing the radio resources.
Note that with aggressive frequency reuse, EDGE Compact achieves higher efficiency due to additional traffic capacity that can be provided for the same bandwidth compared to EDGE Classic. It is therefore a viable option not only for an initial deployment but also for a system with higher available bandwidth. However, the requirement of synchronized base stations and other related issues must be carefully addressed in practical deployment.
Fig 4: Comparison of classic and compact performance for 2.4 MHz Scenario
Fig 5: Comparison of classic and compact performance for 4.2 MHz Scenario
6. PERFORMANCE ENHANCEMENTS
In this section we outline several enhancement techniques. All of these techniques can be implemented in the physical and MAC layers with little or no impact on standards. All the performance results shown below assume EDGE with one carrier per sector. Control channels are not explicitly simulated, so we assume 8 traffic slots are available per carrier. In addition, we consider 600 kHz total
bandwidth in Figures 7-8 and 2.4 MHz in Figure 9. Furthermore, the radio page link performance was based on an earlier proposal with rate adaptation among QPSK/16QAM modes, which were 15%-20% higher in the bit rates. However, the general performance trends for GMSM/8PSK mode adaptation are similar.
6.1 Simple diversity or interference suppression at terminals and smart antennas at base stations
These are techniques that can be implemented in the physical layer. Simple diversity selects the better one between two diversity branches available at a mobile terminal. Interference suppression uses the MMSE (Minimum Mean Square Error) algorithm to further suppress co-channel interference at a two-branch diversity receiver. Smart antennas are implemented by forming four fixed beams on the downlink, with the beam that provides the strongest signal to serve a given terminal. In Figure 6, the left curves show the improvement experienced by a user, in terms of the throughput at a moderate load, by using these methods, while the right curves indicate system capacity enhancement as traffic load increases. Clearly, all these methods are effective in improving user experience as well as system capacity.
Fig 6: Performance improvement by diversity, interference suppression, smart antennas
6.2 Improved resource assignment
All the results shown so far are obtained based on random slot assignment as demand arrives. Several possible enhancements in the MAC layer can be introduced, ranging from simple, autonomous processes performed at individual base stations to those involving more sophisticated coordination among base stations. The key enabler of intelligent resource assignment is signal quality measurements, which are inherently required to perform page link adaptation.
Fig7: Performance improvement by LI-DPA and its combination with mode-0
Here we show an example of LI-DPA. Mode-0 is an additional MCS mode for which no transmission is allowed if the signal quality is below a threshold. Using mode-0, transmissions that are likely to fail are eliminated; this reduces interference without causing reduction of total system throughput. In fact, since the radio resources are made available to users who are likely to succeed, system throughput is increased. LI-DPA, least interference dynamic packet assignment, selects the time slot with the lowest interference to deliver the packets. Figure 7, showing the message delay which is greater than what 90% of the users experience as the throughput per slot increases, clearly indicates that there is significant improvement that can be achieved by this method; a combination of LI-DPA with mode-0 can further enhance performance.
7. CONCLUSION
The TDMA and GSM systems have chosen the same EDGE radio-access and GPRS packet-switched core network technologies to provide third-generation services in existing spectrum. Accordingly, a common access for data services can be offered to more than 370 million mobile subscribers. EDGE can be deployed in two modes in TDMA systems: Classic and COMPACT. The Classic system requires only minimum extension to GSM EDGE and uses standard GSM/GPRS control channels, which facilitates global roaming.
The COMPACT system introduces a novel control channel configuration, synchronized base stations, and discontinuous transmission on the first carriers, which facilitates the deployment of EDGE control channels in a 1/3 frequency-reuse pattern. Thus, the initial deployment of COMPACT requires only a very limited amount of spectrum ”600 kHz plus guard bands. With fractional loading, excellent spectral efficiency can be attained with data rates of up to 384 kbit/s. COMPACT thus supports UWCC requirements for third-generation services with high spectral efficiency and initial deployment within less than 1 MHz of spectrum.
8. REFERENCES
[1] J. Cai and D. J. Goodman, General Packet Radio Service in GSM, IEEE Communications Magazine, October 1997, pp.122-131.
[2] ETSI Tdoc SMG2 1015/97, EDGE Feasibility Study, Work Item 184; Improved Data Rates through Optimised Modulation, version 1.0, December 1997.
Reply

Important Note..!

If you are not satisfied with above reply ,..Please

ASK HERE

So that we will collect data for you and will made reply to the request....OR try below "QUICK REPLY" box to add a reply to this page
Popular Searches: packet loss control using tokens at the network edge viva questions, top ten classic muscle cars, canny edge detection algorithm ppt, edge seminar reports, edge preservation index matlab code, edge technology seminar report doc mediafire, student edge apple,

[-]
Quick Reply
Message
Type your reply to this message here.

Image Verification
Please enter the text contained within the image into the text box below it. This process is used to prevent automated spam bots.
Image Verification
(case insensitive)

Possibly Related Threads...
Thread Author Replies Views Last Post
  Transparent electronics full report seminar surveyer 8 24,595 04-04-2018, 07:54 AM
Last Post: Kalyani Wadkar
  wireless charging through microwaves full report project report tiger 90 70,981 27-09-2016, 04:16 AM
Last Post: The icon
  Wireless Power Transmission via Solar Power Satellite full report project topics 32 50,446 30-03-2016, 03:27 PM
Last Post: dhanabhagya
  surge current protection using superconductors full report computer science technology 13 27,010 16-03-2016, 12:03 AM
Last Post: computer science crazy
  paper battery full report project report tiger 57 61,991 16-02-2016, 11:42 AM
Last Post: Guest
  IMOD-Interferometric modulator full report seminar presentation 3 11,460 18-07-2015, 10:14 AM
Last Post: [email protected]
  digital jewellery full report project report tiger 36 66,727 27-04-2015, 01:29 PM
Last Post: seminar report asees
  LOW POWER VLSI On CMOS full report project report tiger 15 22,307 09-12-2014, 06:31 PM
Last Post: seminar report asees
  gsm pdf and final seminars report suvendu9238 10 11,559 19-11-2014, 09:34 PM
Last Post: jaseela123d
  Performance and Advantages of HXJQ Jaw Crusher wanerjob 1 1,136 23-10-2014, 08:27 PM
Last Post: jaseela123d

Forum Jump: