GSM Cellular No. 1 ãSeattle Pacific University

Description
1. GSM Cellular Standards: A look at the world’s most common digital cellular system Kevin Bolding Electrical Engineering Seattle Pacific University 2. GSM…

Please download to get full document.

View again

of 12
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Information
Category:

Business

Publish on:

Views: 2 | Pages: 12

Extension: PDF | Download: 0

Share
Transcript
  • 1. GSM Cellular Standards: A look at the world’s most common digital cellular system Kevin Bolding Electrical Engineering Seattle Pacific University
  • 2. GSM <ul><li>GSM is the world’s most popular standard for cellular </li></ul><ul><ul><li>82% of market </li></ul></ul><ul><li>Digital, TDMA/WDMA </li></ul><ul><li>Details on </li></ul><ul><ul><li>Voice data format, encoding </li></ul></ul><ul><ul><li>Error-control coding </li></ul></ul><ul><ul><li>Channel sharing (TDMA) </li></ul></ul><ul><ul><li>Channel allocation (WDMA) </li></ul></ul><ul><ul><li>Other things... </li></ul></ul>H Base H H H H
  • 3. Wireless Channel Uplink Architecture Base Station Mobile Handset Voice Encoder Modulator Channel Encoder Demodulator Channel Decoder Channel Noise
  • 4. Voice Coding The output is a 13kbps signal (in 260-bit units) that can be implemented with a cheap ADC, requires only a small amount of memory and sounds close to the actual analog voice signal. <ul><li>Analog signal is sampled using PCM at 64kbps. </li></ul><ul><li>The signal is broken into 20 ms samples, which contain 1280 bits each </li></ul><ul><li>A Regular Pulse Excited - Linear Predictive Coder (RPE-LPC) is used to compress the audio data, which outputs a 260 bit sample that represents 20 ms of analog voice signal. </li></ul>IA – 50 bits IB – 132 bits II – 78 bits Most critical Very Important Icing Voice Encoder 260 bits
  • 5. Channel Coding - Blocks <ul><li>The 260 bit (20ms) sample is divided into class IA, IB and II, based on how important the bits are in determining the sound quality. </li></ul><ul><li>IA uses a 3 bit CRC . If the CRC fails, the whole sample is thrown out. </li></ul><ul><li>IA and IB together have a 4-bit trailer. This is then put into a 1/2 convolutional coder of length 4 that doubles the number of bits. </li></ul><ul><li>II bits are appended unencoded , giving an overall sample of 456 bits. </li></ul>IA – 50 IB – 132 bits II – 78 bits IB – 132 bits Block - 57 Block - 57 Block - 57 Block - 57 Block - 57 Block - 57 Block - 57 Block - 57 <ul><li>The 456 bit encoded sample is divided into 8 blocks of 57 bits each (each contains the equivalent of 2.5 ms of speech) – these are the basic units of transmission. </li></ul>One sample is 20ms of speech --> 456 bits --> 8 blocks One block is 2.5ms of speech --> 57 bits Channel Encoder 456 bits
  • 6. TDMA Bursts <ul><li>Blocks are gathered together to form a TDMA burst </li></ul><ul><li>2 separate speech sample blocks are gathered together </li></ul><ul><ul><li>Interleaved to protect against burst errors </li></ul></ul><ul><li>26-bit training sequence </li></ul><ul><ul><li>To characterize multipath and filter it out </li></ul></ul><ul><li>16.25 tail/guard bits </li></ul>Block - 57 Block - 57 Training - 26 T G T G T G T/G <ul><li>Total Burst is 156.25 bits </li></ul>One burst is two blocks --> Two 2.5ms samples of speech from same source 156.25 bits IA – 50 IB – 132 bits II – 78 bits IB – 132 bits Block - 57 Block - 57 Block - 57 Block - 57 Block - 57 Block - 57 Block - 57 Block - 57 First sample (20ms) IA – 50 IB – 132 bits II – 78 bits IB – 132 bits Block - 57 Block - 57 Block - 57 Block - 57 Block - 57 Block - 57 Block - 57 Block - 57 Second sample (20ms) Channel Encoder
  • 7. Sharing the channel – TDMA Frames <ul><li>Eight bursts (from different sources) make up a TDMA frame </li></ul>One TDMA frame is eight bursts --> 8 sources x (2 x 2.5ms sample of speech) <ul><li>This allows eight sources to share a channel </li></ul>Burst Burst Burst Burst Burst Burst Burst Burst Each burst comes from a different source (phone) Eight phones share a channel using TDM. Channel Block - 57 Block - 57 Training - 26 T G T G T G T/G TDMA Frame - 8 bursts - 8 x 2 x 2.5ms sample of speech - 1250 bits - 4.62ms
  • 8. Sharing the channel <ul><li>26 TDMA frames make up one Multi-frame </li></ul>One TDMA MultiFrame is 26 Frames (24 data) --> 8 sources x (24 x (2 x 2.5ms sample of speech)) --> 8 sources x 2 x 60ms sample of speech --> 8 sources, Two 60ms samples of speech <ul><li>24 are for data (speech) </li></ul>F F F One TDMA MultiFrame takes 120ms --> 8 sources, Two 60ms samples of speech each --> Each of eight sources can transmit 2 60ms samples of speech every 120 ms <ul><li>1 is for control , 1 is unused </li></ul><ul><li>8 Bursts per TDMA frame (2 x 2.5ms sample each) </li></ul>Channel Burst Burst Burst Burst Burst Burst Burst Burst TDMA MultiFrame - 26 Frames - 24 x 8 x 2 x 2.5ms sample of speech - 32500 bits - 120ms F F F F F F F F F F F F F F F F F F F F F F F
  • 9. GSM: Modulation <ul><li>Each Multiframe has 32500 bits and lasts 120ms </li></ul><ul><ul><li>270833 bps </li></ul></ul><ul><li>Transmitted using a channel 200kHz wide </li></ul><ul><li>Uplink in 25MHz wide band from 890-915MHz </li></ul><ul><ul><li>Divided into 124 200kHz wide channels </li></ul></ul><ul><li>Downlink from 935-960MHz </li></ul><ul><li>Cell towers arranged in a hexagonal grid </li></ul><ul><ul><li>Keep spacing between channels on same frequency </li></ul></ul><ul><ul><li>Common to tile in a pattern of 7 cells, each assigned 1/7 of the channels </li></ul></ul><ul><ul><ul><li>Each cell gets 124/7 = 17 channels </li></ul></ul></ul><ul><ul><ul><li>Cell capacity = 17 * 8 = 136 conversations </li></ul></ul></ul>Modulator
  • 10. GSM: Modulation <ul><li>GSM uses Gaussian-filtered Minimum Shift Keying (GMSK). </li></ul><ul><ul><li>MSK is a minimum-shift form of FSK </li></ul></ul><ul><ul><li>Gaussian pre-filter reduces bandwidth </li></ul></ul><ul><li>MSK gives the best spectral efficiency of any digital bandpass signal set. </li></ul><ul><li>FSK only has one amplitude level, allowing for a simpler amplifier in the handset </li></ul>Modulator
  • 11. Discontinuous Transmission <ul><li>Discontinuous transmission (DTX) allows for the transmitter to be turned off 60% of the time. </li></ul><ul><ul><li>Saves power </li></ul></ul><ul><ul><li>Reduces the overall ambient noise in the cell sector. </li></ul></ul><ul><li>DTX requires voice detection, so that the handset knows when to restart transmission. </li></ul><ul><li>DTX also requires a synch signal , so that the receiver can differentiate between silence and a dropped connection. </li></ul>
  • 12. Dynamic Power Compensation <ul><li>The power between the handset and the tower can be dynamically adjusted in response to the channel BER. </li></ul><ul><li>This allows the channel to start at a minimum power level, and only increase when the signal requires a greater SNR. </li></ul><ul><li>For CDMA , Dynamic Power Compensation is a necessity </li></ul><ul><ul><li>All transmit on same band at the same time </li></ul></ul><ul><ul><li>Power must be adjusted so that all signals are received at the same strength </li></ul></ul><ul><ul><ul><li>Otherwise, one channel would overpower all others </li></ul></ul></ul>
  • Related Search
    Similar documents
    View more...
    We Need Your Support
    Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

    Thanks to everyone for your continued support.

    No, Thanks