Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

01 sibille

202 views

Published on

A key parameters based vision
of trends in Wireless systems

Published in: Engineering
  • Be the first to comment

  • Be the first to like this

01 sibille

  1. 1. A key parameters based vision of trends in Wireless systemsof trends in Wireless systems Alain Sibille Telecom ParisTech
  2. 2. Outline What do we speak about ? Tradeoff between key parameters Technology progress 18th EMF day, December 15, 2011, Parispage 1 / 20 From low-end to high-end technologies Conclusion
  3. 3. What do we speak about ? Key parameters: data rate – power – propagation loss • Data rate can be traded against radiated power: the energy/bit is a KPI • data rate ⇒⇒⇒⇒ power 18th EMF day, December 15, 2011, Parispage 2 / 20 • Data rate can be traded against propagation loss: by increasing the distance, there is less energy available/bit • Power can be traded against propagation loss: for a given data rate, less transmitted power reduces the correct reception distance (trivial)
  4. 4. What do we speak about ? The data rate – power – propagation loss tradeoff Data rate 4 GWimedia Total data rate = user data rate + overheads Algorithmic Complexity tuning 18th EMF day, December 15, 2011, Parispage 3 / 20 Power Propagation loss Sensors BAN RFID reader Deep space Consumed power = radiated power + processing power Propagation loss = free space loss + obstructions loss + antenna loss Electronics performance tuning Networking tuning
  5. 5. What do we speak about ? How do we use the resource • Wireless signals occupy a certain spectral band. This resource is expensive. Do we use it well ? Concept of spectral efficiency (bits/second/Hertz) Band width (Hz) 18th EMF day, December 15, 2011, Parispage 4 / 20 • Shannon maximum capacity (bits/s) The maximum SE is limited by the SNR (logarithmically)       +⋅= powerNoise P BC received 1log 2 Max. spectral efficiency       +⋅= 0 2 1log N PSD BC received receivedreceived PSDBP ⋅= 0NBpowerNoise ⋅= Band width (Hz)
  6. 6. What do we speak about ? Cell center vs. cell edge throughput • Since the received power is small far from the base station, the effective throughput is low ! Big lag between maximum spectral efficiency and mean spectral efficiency ! 18th EMF day, December 15, 2011, Parispage 5 / 20
  7. 7. What do we speak about ? Asymmetry • Most often the devices at both link ends differ in their capabilities/performance/complexity/power consumption • Most often spectral efficiency and data rate are higher in DL than UL • UL and DL spectral efficiency differ 18th EMF day, December 15, 2011, Parispage 6 / 20 • UL and DL spectral efficiency differ Duty cycle • The radiated signals are often bursty: low average power, high peak power
  8. 8. Tradeoff between key parameters The tradeoff is a matter of specifications / constraints • How much power is available ? (also a tradeoff vs. Battery capacity Power Propagation loss Data rate 18th EMF day, December 15, 2011, Parispage 7 / 20 • How much power is available ? (also a tradeoff vs. Battery capacity and required lifetime) • What is the needed data rate ? (application dependent)
  9. 9. Tradeoff between key parameters The tradeoff is a matter of specifications / constraints • What is the allowed transmission power ? (regulation spectral masks) Power Propagation loss Data rate 18th EMF day, December 15, 2011, Parispage 8 / 20 • What is the allowed transmission power ? (regulation spectral masks) • What is the acceptable device / access points density (max. distance)
  10. 10. Technology progress How to shift / turn around the tradeoff ? • Improve data rate under fixed power/propagation loss: increase the spectral efficiency (b/s/Hz) – closer to Shannon – : - OFDM (already in many standards): efficient use of the spectrum 1bps/Hz, 2bps/Hz… 12 bps/Hz, reduced by overhead 18th EMF day, December 15, 2011, Parispage 9 / 20 - Multiple antennas (MIMO): may multiply the spectral efficiency by the number of antennas … BUT : may consume processing power
  11. 11. Technology progress How to shift / turn around the tradeoff ? • Improve data rate under fixed power/propagation loss: increase the spectral efficiency (b/s/Hz) – closer to Shannon – : - Advanced coding schemes: to reduce the energy/bit at a given error rate – Turbo codes, LDPC… BUT : may consume processing power 18th EMF day, December 15, 2011, Parispage 10 / 20 - Flexible, adaptive techniques (crosslayer, link level…)
  12. 12. Technology progress How to shift / turn around the tradeoff ? • Make a better use of the spectral resource: fill the gaps, opportunely benefit of available spectrum: - Cognitive radio and dynamic spectrum access: future networks, cellular / local … 18th EMF day, December 15, 2011, Parispage 11 / 20 - Spectral bands agregation: use of several bands simultaneously - Use of higher frequencies: 60 GHz enormous band width, excellent frequency re-use
  13. 13. Technology progress How to shift / turn around the tradeoff ? • PHY/link/Network level advanced techniques: - Cooperative communications improves link robustness 18th EMF day, December 15, 2011, Parispage 12 / 20 - Multi-user communications improves agregated throughput - Relaying, BST cooperation reduces propagation loss
  14. 14. From low-end to high-end technologies Some examples / trends / upcoming technologies • « Low end »: sensors in houses, smart cities/grids, industrial premises... very low power & data rate (kb/day), UHF down to VHF - No battery, low processing power, low cost, green technologies… M2M 18th EMF day, December 15, 2011, Parispage 13 / 20 M2M
  15. 15. From low-end to high-end technologies Some examples / trends / upcoming technologies • « Low end »: RFID - UHF(865-960 MHz, < 2W radiated): active reader, passive tag: from kbps to ~100 kbps, ~5 m detection range, EPC Gen2 - Microwave (2.4 GHz, <1 mW radiated) : active reader, passive or active tag: from kbps to ~1000 kbps, ~10-100 m detection range, 6loPAN 18th EMF day, December 15, 2011, Parispage 14 / 20 from kbps to ~1000 kbps, ~10-100 m detection range, 6loPAN - UWB (~3-5, 6-8.5 GHz, <0.2 mW) active reader, passive or active tag: from kbps to ~250 kbps, ~10-100 m detection range, 802.15.4a, location enabled
  16. 16. From low-end to high-end technologies Some examples / trends / upcoming technologies • « Low-end »: the main target is low power (batteryless), low cost, interference management, multipath fading, NOT spectral efficiency as such very low duty cycle protocols, compressed sensing, energy harvesting… 18th EMF day, December 15, 2011, Parispage 15 / 20
  17. 17. From low-end to high-end technologies Some examples / trends / upcoming technologies • « medium-end »: GSM/GPRS/EDGE. The plain old GSM is not dead: still many virtues (cost, coverage…) for e.g. home & industrial automation - Spectral efficiency 0.5 bps/Hz 18th EMF day, December 15, 2011, Parispage 16 / 20
  18. 18. From low-end to high-end technologies Some examples / trends / upcoming technologies • « medium/high-end »: WIFI, femtocell (<100 mW): gateway to internet + home multimedia platform - Spectral efficiency : 43 bps/Hz agregated SE in 160 MHz (802.11ac, MU- MIMO) up to ~7 Gb/s ! - Cognitive WIFI in white spaces (802.11af <1 GHz) 18th EMF day, December 15, 2011, Parispage 17 / 20 - Cognitive WIFI in white spaces (802.11af <1 GHz) Major issue: how to deal with interference ?
  19. 19. From low-end to high-end technologies Some examples / trends / upcoming technologies • « high-end »: UMTS → HSDPA → LTE → LTE-Advanced - Peak data rate 0.38 Mb/s → 14 Mb/s → 300 Mb/s → 1000 Mb/s - Peak spectral efficiency 0.4 bps/Hz → 8.4 bps/Hz → 16 bps/Hz → 30 bps/Hz (8x8 MIMO) - Future massive MIMO ? 18th EMF day, December 15, 2011, Parispage 18 / 20 - Future massive MIMO ? A lot of work ahead on terminals, network architectures, green BST…
  20. 20. Conclusion Trends in wireless communications technologies: • Play around the tradeoff power/data rate/propagation loss • Depends on the applications: - Wireless technologies become « diffuse » - Many very low power devices, low data rate 18th EMF day, December 15, 2011, Parispage 19 / 20 - Many very low power devices, low data rate - High-end networks for interconnected media devices and internet gateway • Densification of the network: access points, femtocells: mostly (by far) contributes to the total b/s capacity ; lower radiated power, more smoothly distributed • Progress in the technologies: better use all wireless resources, less waste of the expensive frequencies: more Mb/s for fewer Watts. Short term & long term progress to come
  21. 21. 18th EMF day, December 15, 2011, Parispage 20 / 20

×