Wireless Future - Telenor

More documents

Recommendations

Info

12 Switched lobe Dynamically phased array Adaptive antenna Figure 7 Smart antenna realisations (from [30]) Figure 8 Bell Labs’ BLAST concept (from [31]) Signal Interference adapt to the changing channel, using high-order modulation when the channel is good and low order modulation when the channel is bad, considerably higher average data rates are achievable. Adaptive modulation is already being introduced with systems like HIPERLAN/2, where the modulation can be changed from one OFDM symbol to another. In [26], it is claimed that even better performance is achievable if one also allows different modulation of each subcarrier in one OFDM symbol depending on the frequency response of the channel. Developments are also done on the coding side, with turbo-code being shown great interest recently. This topic is comprehensively covered in [27] in this issue of Telektronikk. Smart and Smarter Antenna Concepts Smart antennas have great potential in increasing the spectrum utilisation in radio communication. In a traditional smart antenna, phase and ampli- TX data Vector Encoder: Demux TX data info 4 QAM symbol streams Substreams α 1 α 2 α 3 α 4 TX TX TX TX Notation: Vector symbol a = (α 1 , α 2 , α 3 , α 4 ,) T Number of xmtrs = M Number of rcvrs = N Rich scattering environment tude of the elements of an antenna array are adjusted to form a beam towards the desired receiver/transmitter. This improves the link budget and at the same time reduces the interference transmitted to or received from the environment. Such smart antennas can be divided into three classes: switched beam, phased array and adaptive antenna (Figure 7). In the switched lobe case, one of a predefined set of lobes is selected based on maximum received signal power. With a phased array, the lobe is adjusted to point in the direction of the strongest transmitter. The full adaptive antenna seeks to maximize received power and at the same time minimize interference, i.e. optimise C/I. This is achieved by directing lobes towards desired signal(s) and nulls towards interferers. Comprehensive tutorials on smart antennas can be found in [28] and [29]. The performances of the different concepts were evaluated in [30] based on real channel measurements at 2.1 GHz, assuming an 8-element linear array. The results were compared to a conventional sector antenna, and indicated a gain of around 6 dB in C/I for the switched lobe and phased array, and 12 – 16 dB for the adaptive antenna. A new type of smart antenna technology has gained considerable attention the last couple of years. It is popularly termed MIMO – multiple input multiple output – and was first proposed by Bell Labs in the so-called BLAST concept [31] (Figure 8). In traditional smart antennas, the same information symbol is sent on all antenna elements, with the phase and amplitude of each element being adjusted to direct the power in the desired direc- RX RX RX RX RX RX V-BLAST signal processing: Estimate α and decode RX data Telektronikk 1.2001
tion. In MIMO systems, different symbols are sent on the different antenna elements. By using antenna arrays on both the transmit and receive sides and assuming a radio channel characterised by rich scattering, a number of uncorrelated radio channels can be realised. In the ideal case this number is equal to the number of elements of the “smallest” array (4 in Figure 8). Compared to the case with one antenna at both transmitter and receiver, the capacity of the link is increased by a factor equal to the number of uncorrelated channels. For MIMO to work, an accurate channel estimate has to be made by the receiver. In laboratory experiments, Bell Labs have demonstrated spectrum efficiency in the order of 20 – 40 bit/s/Hz. The principles and possibilities related to communications with multiple antennas are treated in [32] in this issue. Packet Based MAC Protocols In the future, it is expected that most of the traffic on wireless systems will be packet based data, as opposed to circuit-switched voice in today’s cellular systems. In order to optimise the use of the radio spectrum, the MAC protocols should be designed to reflect this change. In [33], AT&T proposes an OFDM based cellular system using 5 MHz channel raster, the same as being used by UMTS. By optimising the MAC protocol for packet based data transfer, a significantly higher capacity is achieved than in UMTS. AT&T claim that the system will be able to deliver 2 – 5 Mbit/s in macro-cells, while the goal for UMTS is 384 kbit/s. Relay protocols In most radio systems, there is only one radio hop. However, from a pure link-budget viewpoint, it is beneficial to split a single, long hop into several smaller hops. This is one of the main ideas behind radio relay protocols. In the ODMA concept, originally proposed as a candidate for UMTS, the idea is to use other mobile terminals as relays between a given terminal and the base station. By doing this, the amount of power used for transmission – and thereby the total interference level in the system – can be decreased, and the range of a base station increased. The concept is shown in Figure 9. ODMA is adopted by 3GPP as an option in the TDD mode, but the standardisation is only in the initial stage and has shown little progress lately. Preliminary results documented in [34] indicate that ODMA has a great potential for increasing the capacity in UMTS, although there are still unresolved issues before the system can be implemented – not the least regarding routing protocols. Telektronikk 1.2001 Trends on the Network Side The trend is towards convergence between the telecom world and the Internet. IP technology is to a large degree introduced in UMTS and the UMTS vision is to standardise an “All IP” core network in the future. Organisations like the Internet Engineering Task Force (IETF) and the IST BRAIN project [9][11] develop IP based mobility algorithms to be used in wireless access networks and mobile networks. Why Strive towards Integration? The customers want seamless use of services across different access networks. Mobility is a major driver in the industry. With IP it is possible to make a generic mobility handling mechanism across different networks. The All IP Vision The “All IP” vision is a vision shared by many. This architecture (Figure 10) shall allow operators to deliver services, both real-time and non real-time services for speech, data and multimedia over a common service platform based on IP. The user shall be given full service and terminal portability. Personalised services shall be offered across different access networks, both wireless and wireline. Eventually all services shall be handled by the IP core network. Evolution from 3rd Generation Systems 4th Generation mobile networks will most likely be based on evolution from 3G. For the core network and service platforms we see a migration from UMTS and IP/Internet towards a common core network/service platform handling all services across different access networks. Figure 11 describes the migration path foreseen in 3GPP. Eventually other access networks than the ones specified in 3GPP will be connected to the All IP platform. ETSI BRAN works to specify inter- High Bit Rate Data TDD Coverage TDD ODMA High Bit Rate ODMA TERMINAL Low Bit Rate Data TDD Coverage TDD ODMA High Bit Rate Layer 1 Synch Infor4mation Figure 9 ODMA – Opportunity driven multiple access (from [34]) Border Region NO Coverage 13
Page 2 and 3: Telektronikk Volume 97 No. 1 - 2001
Page 4 and 5: Per Hjalmar Lehne (42) obtained his
Page 6 and 7: Stein Svaet (41) is Senior Research
Page 8 and 9: Pbit/day 6 150 125 100 75 50 25 0 F
Page 10 and 11: 8 Video communication as a utility
Page 12 and 13: 10 The idea of reconfigurable mobil
Page 16 and 17: Figure 10 The All IP vision Figure
Page 18 and 19: 16 The IP world has some problems t
Page 20 and 21: 18 Abbreviations and acronyms less
Page 22 and 23: Jorge Pereira (40) obtained his Eng
Page 24 and 25: (Millions) Figure 1 Exponential gro
Page 26 and 27: 24 Defining Generation based upon D
Page 28 and 29: Multi-mode Terminal by Software Rad
Page 30 and 31: distribution layer possible return
Page 32 and 33: 30 A Co-Ordinated Approach to 4G As
Page 34 and 35: Figure 2 Applications will grow fas
Page 36 and 37: Figure 5 Evolution from multi-mode
Page 38 and 39: Operator/Provider download librarie
Page 40 and 41: Radio network subsystem - service c
Page 42 and 43: Dr. Techn. Jørgen Bach Andersen (6
Page 44 and 45: Figure 3 Two arrays with M and N el
Page 46 and 47: Figure 5 Cumulative probability dis
Page 48 and 49: Figure 8 As in Figure 5, but with F
Page 50 and 51: 48 Conclusion The challenge of prov
Page 52 and 53: 50 models are described simply by a
Page 54 and 55: 52 ping intervals or “quantizatio
Page 56 and 57: Figure 3 ASE as a function of the a
Page 58 and 59: 56 7 Conclusion and Further Researc
Page 60 and 61: Dave Wisely (39) graduated in 1982
Page 62 and 63: Figure 2 Business case study Figure
Page 64 and 65:
Figure 6 BRAIN QoS architecture 62
Page 66 and 67:
64 4 BRAIN homepage. (2001, March 6
Page 68 and 69:
Figure 2 The Bluetooth protocol sta
Page 70 and 71:
Figure 5 A Bluetooth radio module i
Page 72 and 73:
70 • Instant postcard: A Bluetoot
Page 74 and 75:
72 work visions come true, it is ne
Page 76 and 77:
Table 1 Allocated HiperLAN bands (i
Page 78 and 79:
Figure 4 Basic MAC frame structure
Page 80 and 81:
78 Preamble BCH FCH ACHs Preamble S
Page 82 and 83:
throughput bit/s 80 3 2.5 2 1.5 1 0
Page 84 and 85:
82 5 References 1 CEPT. ERC Decisio
Page 86 and 87:
84 IETF Internet Engineering Task F
Page 88 and 89:
Figure 2-3 Mobile IPv6 architecture
Page 90 and 91:
88 domain, it attaches to a new lea
Page 92 and 93:
90 Domain GFA/MAP Figure 3-3 Archit
Page 94 and 95:
Geir Gylterud (35) is Research Scie
Page 96 and 97:
94 Framework Resource Resource Int
Page 98 and 99:
Figure 6 SCSs and network functiona
Page 100 and 101:
Figure 9 The new business model. Fr
Page 102 and 103:
100 A strong consortium backs up th
Page 104 and 105:
102 Music Company Figure 3 Service
Page 106 and 107:
Figure 4 OSA architecture service c
Page 108 and 109:
Do van Thanh (43) obtained his MSc
Page 110 and 111:
Figure 2 The first use case diagram
Page 112 and 113:
110 Box 1 - The Open Service Archit
Page 114 and 115:
112 X:User 1:Device 2:Device 3:Devi
Page 116 and 117:
Figure 10 Initiating call from PSTN
Page 118 and 119:
Figure 13 Splitting the incoming st
Page 120 and 121:
Figure 17 A set of communication de
Page 122 and 123:
120 Box 3 continued Session Descrip
Page 124 and 125:
Figure 21 A laptop and peripheral d
Page 126 and 127:
124 Box 4 - Bluetooth Bluetooth is
Page 128 and 129:
126 6 The ICEBERG project. (2001, M
Page 130 and 131:
128 new door towards the upcoming f
Page 132 and 133:
? Figure 1 The Mobile Service Platf
Page 134 and 135:
Figure 2 Needs and Related Services
Page 136 and 137:
134 Telektronikk 1.2001
Page 138 and 139:
136 Telektronikk 1.2001
Page 140 and 141:
Anne Lise Lillebø (52) is Adviser
Page 142 and 143:
140 with other SDOs (Standards Deve
Page 144 and 145:
142 Box 2 - ETNO Declaration We, th
Page 146 and 147:
Arve Meisinset (52) is Senior Resea
Page 148 and 149:
146 In ITU-T we have seen needs to
Page 150 and 151:
148 Each viewpoint constitutes a se
Page 152 and 153:
Figure 2 UML class diagrams for tra
Page 154 and 155:
152 The functionality describing th
Page 156 and 157:
154 9 Potter, B et al. An introduct
show all

Wireless Future - Telenor

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?