x
`
`1
`
`kl
`
`I
`
`h
`
`'
`
`Johan Sk6ld
`
`l
`LTE 1 LTE-Advaficed
`for Mobile Broadband
`
`;.|
`
`Erik Dahlman
`Stefan Parkvall
`
`7
`
`*
`
`..
`
`Samsung Exhibit 1015, Page 1
`
`Samsung Exhibit 1015, Page 1
`
`

`

`4G LTE/LTE-Advanced
`for Mobile Broadband
`
`Erik Dahlman, Stefan Parkvall, and
`Johan Sköld
`
`(cid:33)(cid:45)(cid:51)(cid:52)(cid:37)(cid:50)(cid:36)(cid:33)(cid:45)(cid:0)(cid:115)(cid:0)(cid:34)(cid:47)(cid:51)(cid:52)(cid:47)(cid:46)(cid:0)(cid:115)(cid:0)(cid:40)(cid:37)(cid:41)(cid:36)(cid:37)(cid:44)(cid:34)(cid:37)(cid:50)(cid:39)(cid:0)(cid:115)(cid:0)(cid:44)(cid:47)(cid:46)(cid:36)(cid:47)(cid:46)(cid:0)(cid:115)(cid:0)(cid:46)(cid:37)(cid:55)(cid:0)(cid:57)(cid:47)(cid:50)(cid:43)(cid:0)(cid:115)(cid:0)(cid:47)(cid:56)(cid:38)(cid:47)(cid:50)(cid:36)
`(cid:48)(cid:33)(cid:50)(cid:41)(cid:51)(cid:0)(cid:115)(cid:0)(cid:51)(cid:33)(cid:46)(cid:0)(cid:36)(cid:41)(cid:37)(cid:39)(cid:47)(cid:0)(cid:115)(cid:0)(cid:51)(cid:33)(cid:46)(cid:0)(cid:38)(cid:50)(cid:33)(cid:46)(cid:35)(cid:41)(cid:51)(cid:35)(cid:47)(cid:0)(cid:115)(cid:0)(cid:51)(cid:41)(cid:46)(cid:39)(cid:33)(cid:48)(cid:47)(cid:50)(cid:37)(cid:0)(cid:115)(cid:0)(cid:51)(cid:57)(cid:36)(cid:46)(cid:37)(cid:57)(cid:0)(cid:115)(cid:0)(cid:52)(cid:47)(cid:43)(cid:57)(cid:47)
`
`Academic Press is an imprint of Elsevier
`
`Samsung Exhibit 1015, Page 2
`
`

`

`Academic Press is an imprint of Elsevier
`The Boulevard, Langford Lane, Kidlington, Oxford, OX5 1GB, UK
`30 Corporate Drive, Suite 400, Burlington, MA 01803, USA
`
`First published 2011
`
`Copyright © 2011 Erik Dahlman, Stefan Parkvall & Johan Sköld. Published by Elsevier Ltd. All rights reserved
`
`The rights of Erik Dahlman, Stefan Parkvall & Johan Sköld to be identified as the authors of this work has been
`asserted in accordance with the Copyright, Designs and Patents Act 1988.
`
`No part of this publication may be reproduced or transmitted in any form or by any means, electronic or
`mechanical, including photocopying, recording, or any information storage and retrieval system, without
`permission in writing from the publisher. Details on how to seek permission, further information about the
`Publisher’s permissions policies and our arrangement with organizations such as the Copyright Clearance
`Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions
`
`This book and the individual contributions contained in it are protected under copyright by the Publisher
`(other than as may be noted herein).
`
`Notices
`Knowledge and best practice in this field are constantly changing. As new research and experience broaden our
`understanding, changes in research methods, professional practices, or medical treatment may become necessary.
`
`Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using
`any information, methods, compounds, or experiments described herein. In using such information or methods
`they should be mindful of their own safety and the safety of others, including parties for whom they have a
`professional responsibility.
`
`To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability
`for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or
`from any use or operation of any methods, products, instructions, or ideas contained in the material herein.
`
`British Library Cataloguing-in-Publication Data
`A catalogue record for this book is available from the British Library
`
`Library of Congress Control Number: 2011921244
`
`ISBN: 978-0-12-385489-6
`
`For information on all Academic Press publications
`visit our website at www.elsevierdirect.com
`
`Typeset by MPS Limited, a Macmillan Company, Chennai, India
`www.macmillansolutions.com
`
`Printed and bound in the UK
`
`11 12 13 14 10 9 8 7 6 5 4 3 2 1
`
`Samsung Exhibit 1015, Page 3
`
`

`

`Downlink Physical-Layer
`Processing
`
`CHAPTER
`
`10
`
`In Chapter 8, the LTE radio-interface architecture was discussed with an overview of the functions
`and characteristics of the different protocol layers. Chapter 9 then gave an overview of the basic
`time–frequency structure of LTE transmissions, including the structure of the OFDM time–frequency
`grid being the fundamental physical resource on both uplink and downlink.
`This chapter will provide a more detailed description of the downlink physical-layer functionality,
`including the transport-channel processing (Section 10.1), downlink reference signals (Section 10.2),
`details on downlink multi-antenna transmission (Section 10.3), and downlink L1/L2 control signal-
`ing (Section 10.4). Chapter 11 will provide a corresponding description for the uplink transmission
`direction. The later chapters will then go further into the details of some specific uplink and downlink
`functions and procedures.
`
`10.1 TRANSPORT-CHANNEL PROCESSING
`As described in Chapter 8, transport channels provide the interface between the MAC layer and the physi-
`cal layer. As also described, for the LTE downlink there are four different types of transport channels
`defined, the Downlink Shared Channel (DL-SCH), the Multicast Channel (MCH), the Paging Channel
`(PCH), and the Broadcast Channel (BCH). This section provides a description of the physical-layer
`processing applied to DL-SCH transport channels, including the mapping to the physical resource – that
`is, to the resource elements of the OFDM time–frequency grid. DL-SCH is the main downlink transport-
`channel type in LTE and is used for transmission of user-specific higher-layer information, both user data
`and dedicated control information, as well as part of the downlink system information. The DL-SCH
`physical-layer processing is to a large extent applicable also to MCH and PCH transport channels,
`although with some additional constraints. On the other hand, the physical-layer processing, and the trans-
`mission structure in general, for the BCH is quite different. BCH transmission is described in Chapter 14
`as part of the discussion on LTE system information.
`
`10.1.1 Processing Steps
`
`The different steps of the DL-SCH physical layer processing are outlined in Figure 10.1. In the case
`of carrier aggregation – that is, transmission on multiple component carriers in parallel to the same
`terminal – the transmissions on the different carriers correspond to separate transport channels with
`separate and more or less independent physical-layer processing. The transport-channel processing
`outlined in Figure 10.1 and the discussion below is thus valid also in the case of carrier aggregation.
`
`4G LTE/LTE-Advanced for Mobile Broadband.
`
`
`
`© 2011 Erik Dahlman, Stefan Parkvall & Johan Sköld. Published by Elsevier Ltd. All rights reserved.2011
`
`143
`
`Samsung Exhibit 1015, Page 4
`
`

`

`144
`
`CHAPTER 10 Downlink Physical-Layer Processing
`
`One or two transport block(s) of
`dynamic size delivered from the MAC layer
`
`CRC insertion per transport block
`
`CRC
`
`CRC
`
`Code-block segmentation including
`possible per-code-block CRC insertion
`
`Segmentation
`
`Segmentation
`
`Channel coding (Turbo coding)
`
`Coding
`
`Coding
`
`Rate matching and
`physical-layer hybrid ARQ functionality
`
`RM + HARQ
`
`RM + HARQ
`
`Bit-level scrambling
`
`Scrambling
`
`Scrambling
`
`Data modulation
`(QPSK, 16QAM, 64QAM)
`
`Modulation
`
`Modulation
`
`Antenna mapping
`
`Antenna mapping
`
`Up to eight antenna ports
`
`Mapping to OFDM time–frequency grid
`for each antenna port
`
`FIGURE 10.1
`
`Physical-layer processing for DL-SCH.
`
`Within each Transmission Time Interval (TTI), corresponding to one subframe of length 1 ms, up
`to two transport blocks of dynamic size are delivered to the physical layer and transmitted over the
`radio interface for each component carrier. The number of transport blocks transmitted within a TTI
`depends on the configuration of the multi-antenna transmission scheme, as described in Section 10.3.
`
`G
`G
`
`In the case of no spatial multiplexing there is at most a single transport block in a TTI.
`In the case of spatial multiplexing, with transmission on multiple layers in parallel to the same ter-
`minal, there are two transport blocks within a TTI.1
`
`10.1.1.1 CRC Insertion Per Transport Block
`In the first step of the physical-layer processing, a 24-bit CRC is calculated for and appended to each
`transport block. The CRC allows for receiver-side detection of errors in the decoded transport block.
`The corresponding error indication can, for example, be used by the downlink hybrid-ARQ protocol
`as a trigger for requesting retransmissions.
`
`1 This is true for initial transmissions. In the case of hybrid-ARQ retransmissions there may also be cases when a single
`transport block is transmitted over multiple layers, as discussed, for example, in Section 10.3.2.
`
`Samsung Exhibit 1015, Page 5
`
`

`

`10.1 Transport-Channel Processing
`
`145
`
`Transport block
`
`CRC
`
`Code block #1
`
`Code block #2
`
`Code block #M
`
`Code-block segmentation
`
`Insertion of filler bits
`in first code block
`
`Calculation and insertion
`of per-code-block CRC
`
`CRC
`
`CRC
`
`CRC
`
`FIGURE 10.2
`
`Code-block segmentation and per-code-block CRC insertion.
`
`To channel coding
`
`10.1.1.2 Code-Block Segmentation and Per-Code-Block CRC Insertion
`The LTE Turbo-coder internal interleaver is only defined for a limited number of code-block sizes,
`with a maximum block size of 6144 bits. If the transport block, including the transport-block CRC,
`exceeds this maximum code-block size, code-block segmentation, illustrated in Figure 10.2, is
`applied before the Turbo coding. Code-block segmentation implies that the transport block is seg-
`mented into smaller code blocks, the sizes of which should match the set of code-block sizes sup-
`ported by the Turbo coder.
`In order to ensure that a transport block of arbitrary size can be segmented into code blocks
`that match the set of available code-block sizes, the specification includes the possibility to insert
`“dummy” filler bits at the head of the first code block. However, the set of transport-block sizes cur-
`rently defined for LTE has been selected so that filler bits are not needed.
`As can be seen in Figure 10.2, code-block segmentation also implies that an additional CRC (also
`of length 24 bits, but different compared to the transport-block CRC described above) is calculated
`for and appended to each code block. Having a CRC per code block allows for early detection of cor-
`rectly decoded code blocks and correspondingly early termination of the iterative decoding of that
`code block. This can be used to reduce the terminal processing effort and corresponding energy con-
`sumption. In the case of a single code block no additional code-block CRC is applied.
`One could argue that, in case of code-block segmentation, the transport-block CRC is redundant
`and implies unnecessary overhead as the set of code-block CRCs should indirectly provide informa-
`tion about the correctness of the complete transport block. However, code-block segmentation is only
`applied to large transport blocks for which the relative extra overhead due to the additional transport-
`block CRC is small. The transport-block CRC also adds additional error-detection capabilities and
`thus further reduces the risk for undetected errors in the decoded transport block.
`Information about the transport-block size is provided to the terminal as part of the scheduling
`assignment transmitted on the PDCCH control channel, as described in Section 10.4.4. Based on this
`information, the terminal can determine the code-block size and number of code blocks. The terminal
`receiver can thus, based on the information provided in the scheduling assignment, straightforwardly
`undo the code-block segmentation and recover the decoded transport blocks.
`
`Samsung Exhibit 1015, Page 6
`
`

`

`146
`
`CHAPTER 10 Downlink Physical-Layer Processing
`
`First constituent encoder
`
`One code block
`
`D
`
`D
`
`D
`
`Code bits
`
`Systematic bits
`
`First parity bits
`
`QPP
`
`Inner
`interleaver Second constituent encoder
`
`Second parity bits
`
`D
`
`D
`
`D
`
`FIGURE 10.3
`
`LTE Turbo encoder.
`
`Input bits
`
`0
`
`1
`
`2
`
`c(i)
`
`K-1
`
`c(i)= f1 ⋅ i +f2 ⋅ i2 mod K
`
`One code block
`
`Output bits
`
`0
`
`1
`
`2
`
`i
`
`K-1
`
`FIGURE 10.4
`
`Principles of QPP-based interleaving.
`
`10.1.1.3 Channel Coding
`Channel coding for DL-SCH (as well as for PCH and MCH) is based on Turbo coding [53], with
`encoding according to Figure 10.3. The encoding consists of two rate-1/2, eight-state constituent
`encoders, implying an overall code rate of 1/3, in combination with QPP-based2 interleaving [69]. As
`illustrated in Figure 10.4, the QPP interleaver provides a mapping from the input (non-interleaved)
`bits to the output (interleaved) bits according to the function:
`
`( )
`c i
`
`f
`1
`
`⋅
`
`i
`
`2 mod ,
`K
`
`i
`
`⋅
`
`f
`
`2
`
`where i is the index of the bit at the output of the interleaver, c(i) is the index of the same bit at the
`input of the interleaver, and K is the code-block/interleaver size. The values of the parameters f 1 and
`f2 depend on the code-block size K. The LTE specification lists all supported code-block sizes, rang-
`ing from a minimum of 40 bits to a maximum of 6144 bits, together with the associated values for the
`parameters f1 and f2. Thus, once the code-block size is known, the Turbo-coder inner interleaving, as
`well as the corresponding de-interleaving at the receiver side, can straightforwardly be carried out.
`A QPP-based interleaver is maximum contention free [70], implying that the decoding can be
`parallelized without the risk for contention when the different parallel processes are accessing the
`
`2 QPP ⫽ Quadrature Permutation Polynomial.
`
`Samsung Exhibit 1015, Page 7
`
`

`

`10.1 Transport-Channel Processing
`
`147
`
`Circular buffer
`
`RV = 0
`
`)( 2
`Kp
`
`1s
`
`2s
`
`RV = 3
`
`)(1
`Kp
`
`Ks
`)(1
`1p
`)( 2
`)( 2
`2p
`1p
`)(1
`2p
`
`RV = 1
`
`Bit collection
`
`RV = 2
`
`Bit selection
`
`RV
`
`Systematic bits
`Ksss ...
`
`21
`
`First parity bits
`
`)(
`)(
`1
`1
`pp
`2
`1
`
`...
`
`)(
`1
`Kp
`
`Second parity bits
`)(
`)(
`)(
`2
`2
`2
`...
`Kp
`pp
`2
`1
`
`FIGURE 10.5
`
`Sub-block
`interleaving
`
`Sub-block
`interleaving
`
`Sub-block
`interleaving
`
`Rate matching and hybrid-ARQ functionality.
`
`interleaver memory. For the very high data rates supported by LTE, the improved possibilities for
`parallel processing offered by QPP-based interleaving can substantially simplify the Turbo-encoder/
`decoder implementation.
`
`10.1.1.4 Rate Matching and Physical-Layer Hybrid-ARQ Functionality
`The task of the rate-matching and physical-layer hybrid-ARQ functionality is to extract, from the
`blocks of code bits delivered by the channel encoder, the exact set of code bits to be transmitted
`within a given TTI/subframe.
`As illustrated in Figure 10.5, the outputs of the Turbo encoder (systematic bits, first parity bits,
`and second parity bits) are first separately interleaved. The interleaved bits are then inserted into what
`can be described as a circular buffer with the systematic bits inserted first, followed by alternating
`insertion of the first and second parity bits.
`The bit selection then extracts consecutive bits from the circular buffer to an extent that matches
`the number of available resource elements in the resource blocks assigned for the transmission. The
`exact set of bits to extract depends on the redundancy version (RV) corresponding to different starting
`points for the extraction of coded bits from the circular buffer. As can be seen, there are four different
`alternatives for the redundancy version. The transmitter/scheduler selects the redundancy version and
`provides information about the selection as part of the scheduling assignment (see Section 10.4.4).
`Note that the rate-matching and hybrid-ARQ functionality operates on the full set of code bits
`corresponding to one transport block and not separately on the code bits corresponding to a single
`code block.
`
`10.1.1.5 Bit-Level Scrambling
`LTE downlink scrambling implies that the block of code bits delivered by the hybrid-ARQ functionality
`is multiplied (exclusive-or operation) by a bit-level scrambling sequence. Without downlink scrambling,
`
`Samsung Exhibit 1015, Page 8
`
`

`

`148
`
`CHAPTER 10 Downlink Physical-Layer Processing
`
`the channel decoder at the terminal could, at least in principle, be equally matched to an interfering
`signal as to the target signal, thus being unable to properly suppress the interference. By applying dif-
`ferent scrambling sequences for neighboring cells, the interfering signal(s) after descrambling is (are)
`randomized, ensuring full utilization of the processing gain provided by the channel code. Thus, the
`bit scrambling essentially serves the same purpose as the scrambling applied at chip level after the
`direct-sequence spreading in DS-CDMA-based systems such as WCDMA/HSPA. Fundamentally, chan-
`nel coding can be seen as “advanced” spreading providing processing gain similar to direct-sequence
`spreading but also additional coding gain.
`In LTE, downlink scrambling is applied to all transport channels as well as to the downlink
`L1/L2 control signaling. For all downlink transport-channel types except MCH, as well as for the
`L1/L2 control signaling, the scrambling sequences differ between neighboring cells (cell-specific
`scrambling) to ensure interference randomization between the cells. This is achieved by having the
`scrambling sequences depend on the physical-layer cell identity (Chapter 14). In contrast, in the case
`of MBSFN-based transmission using MCH, the same scrambling should be applied to all cells taking
`part in the MBSFN transmission – that is, all cells within the so-called MBSFN area (see Chapter 15).
`
`10.1.1.6 Data Modulation
`The downlink data modulation transforms the block of scrambled bits to a corresponding block
`of complex modulation symbols. The set of modulation schemes supported for the LTE downlink
`includes QPSK, 16QAM, and 64QAM, corresponding to two, four, and six bits per modulation sym-
`bol respectively.
`
`10.1.1.7 Antenna Mapping
`The antenna mapping jointly processes the modulation symbols corresponding to the one or two
`transport blocks and maps the result to different antenna ports. The antenna mapping can be con-
`figured in different ways corresponding to different multi-antenna transmission schemes, includ-
`ing transmit diversity, beam-forming, and spatial multiplexing. As indicated in Figure 10.1, LTE
`supports transmission using up to eight antenna ports depending on the exact multi-antenna trans-
`mission scheme. More details about LTE downlink multi-antenna transmission are provided in
`Section 10.3.
`Note that the antenna ports referred to above do not necessarily correspond to specific physi-
`cal antennas. Rather, an antenna port is a more general concept introduced, for example, to allow
`for beam-forming using multiple physical antennas without the terminal being aware of the beam-
`forming carried out at the transmitter side.
`At least for the downlink, an antenna port can be seen as corresponding to the transmission of a
`reference signal (Section 10.2). Any data transmission from the antenna port can then rely on that ref-
`erence signal for channel estimation for coherent demodulation. Thus, if the same reference signal is
`transmitted from multiple physical antennas, these physical antennas correspond to a single antenna
`port. Similarly, if two different reference signals are transmitted from the same set of physical anten-
`nas, this corresponds to two separate antenna ports.
`It should be noted that the LTE specification actually has a somewhat more general definition of
`an antenna port, essentially just saying that two received signals can be assumed to have experienced
`the same overall channel, including any joint processing at the transmitter side, if and only if they
`have been transmitted on the same antenna port.
`
`Samsung Exhibit 1015, Page 9
`
`

`

`10.1 Transport-Channel Processing
`
`149
`
`10.1.1.8 Resource-Block Mapping
`The resource-block mapping takes the symbols to be transmitted on each antenna port and maps them
`to the resource elements of the set of resource blocks assigned by the MAC scheduler for the trans-
`mission. As described in Chapter 9, each resource block consists of 84 resource elements (twelve sub-
`carriers during seven OFDM symbols).3 However, some of the resource elements within a resource
`block will not be available for the transport-channel transmission as they are occupied by:
`
`G different types of downlink reference signals, as described in Section 10.2;
`G downlink L1/L2 control signaling (one, two, or three OFDM symbols at the head of each sub-
`frame), as described in Section 10.44.
`
`Furthermore, as will be described in Chapter 14, within some resource blocks, additional resource
`elements are reserved for the transmission of synchronization signals as well as for the transmission
`of the BCH transport channel.
`In the TDD special subframe (Section 9.5.2), mapping is limited to the DwPTS.
`
`10.1.2 Localized and Distributed Resource Mapping
`
`As already discussed in Chapter 7, when deciding what set of resource blocks to use for transmission
`to a specific terminal, the network may take the downlink channel conditions in both the time and
`frequency domains into account. Such time/frequency-domain channel-dependent scheduling, taking
`channel variations – for example, due to frequency-selective fading – into account, may significantly
`improve system performance in terms of achievable data rates and overall cell throughput.
`However, in some cases downlink channel-dependent scheduling is not suitable to use or is not
`practically possible:
`
`G For low-rate services such as voice, the feedback signaling associated with channel-dependent
`scheduling may lead to extensive relative overhead.
`G At high mobility (high terminal speed), it may be difficult or even practically impossible to track
`the instantaneous channel conditions to the accuracy required for channel-dependent scheduling to
`be efficient.
`
`In such situations, an alternative means to handle radio-channel frequency selectivity is to achieve
`frequency diversity by distributing a downlink transmission in the frequency domain.
`One way to distribute a downlink transmission in the frequency domain, and thereby achieve fre-
`quency diversity, is to assign multiple non-frequency-contiguous resource blocks for the transmis-
`sion to a terminal. LTE allows for such distributed resource-block allocation by means of resource
`allocation types 0 and 1 (see Section 10.4.4). However, although sufficient in many cases, distributed
`resource-block allocation by means of these resource-allocation types has certain drawbacks:
`
`G For both types of resource allocations, the minimum size of the allocated resource can be as large
`as four resource-block pairs and may thus not be suitable when resource allocations of smaller
`sizes are needed.
`In general, both these resource-allocation methods are associated with a relatively large PDCCH
`payload.
`
`G
`
`3 72 resource elements in the case of extended cyclic prefix.
`4 In MBSFN subframes, the control region is limited to a maximum of two OFDM symbols.
`
`Samsung Exhibit 1015, Page 10
`
`

`

`150
`
`CHAPTER 10 Downlink Physical-Layer Processing
`
`In contrast, resource-allocation type 2 (Section 10.4.4) always allows for the allocation of a sin-
`gle resource-block pair and is also associated with a relatively small PDCCH payload size. However,
`resource allocation type 2 only allows for the allocation of resource blocks that are contiguous in the
`frequency domain. In addition, regardless of the type of resource allocation, frequency diversity by
`means of distributed resource-block allocation will only be achieved in the case of resource alloca-
`tions larger than one resource-block pair.
`In order to provide the possibility for distributed resource-block allocation in the case of
`resource-allocation type 2, as well as to allow for distributing the transmission of a single resource-
`block pair in the frequency domain, the notion of a Virtual Resource Block (VRB) has been intro-
`duced for LTE.
`What is being provided in the resource allocation is the resource allocation in terms of VRB pairs.
`The key to distributed transmission then lies in the mapping from VRB pairs to Physical Resource
`Block (PRB) pairs – that is, to the actual physical resource used for transmission.
`The LTE specification defines two types of VRBs: localized VRBs and distributed VRBs. In the
`case of localized VRBs, there is a direct mapping from VRB pairs to PRB pairs, as illustrated in
`Figure 10.6. However, in the case of distributed VRBs, the mapping from VRB pairs to PRB pairs is
`more elaborate in the sense that:
`
`G consecutive VRBs are not mapped to PRBs that are consecutive in the frequency domain; and
`G even a single VRB pair is distributed in the frequency domain.
`
`The basic principle of distributed transmission is outlined in Figure 10.7 and consists of two
`steps:
`
`G A mapping from VRB pairs to PRB pairs such that consecutive VRB pairs are not mapped to
`frequency-consecutive PRB pairs (first step of Figure 10.7). This provides frequency diversity
`between consecutive VRB pairs. The spreading in the frequency domain is done by means of a
`block-based “interleaver” operating on resource-block pairs.
`
`One resource-block pair
`
`0
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
`
`VRBs
`
`PRBs
`
`FIGURE 10.6
`
`Direct VRB-to-PRB mapping
`
`VRB-to-PRB mapping in the case of localized VRBs. A cell bandwidth corresponding to 25
`resource blocks is assumed.
`
`Samsung Exhibit 1015, Page 11
`
`

`

`10.1 Transport-Channel Processing
`
`151
`
`G A split of each resource-block pair such that the two resource blocks of the resource-block pair
`are transmitted with a certain frequency gap in between (second step of Figure 10.7). This also
`provides frequency diversity for a single VRB pair. This step can be seen as the introduction of
`frequency hopping on a slot basis.
`
`Whether the VRBs are localized (and thus mapped according to Figure 10.6) or distributed (mapped
`according to Figure 10.7) is indicated on the associated PDCCH in the case of type 2 resource alloca-
`tion. Thus, it is possible to dynamically switch between distributed and localized transmission and also
`mix distributed and localized transmission for different terminals within the same subframe.
`The exact size of the frequency gap in Figure 10.7 depends on the overall downlink cell band-
`width according to Table 10.1. These gaps have been chosen based on two criteria:
`
`1. The gap should be of the order of half the downlink cell bandwidth in order to provide good fre-
`quency diversity also in the case of a single VRB pair.
`2. The gap should be a multiple of P2, where P is the size of a resource-block group as defined in
`Section 10.4.4 and used for resource allocation types 0 and 1. The reason for this constraint is to
`ensure a smooth coexistence in the same subframe between distributed transmission as described
`above and transmissions based on downlink allocation types 0 and 1.
`
`Due to the constraint that the gap size should be a multiple of P2, the gap size will in most cases
`deviate from exactly half the cell bandwidth. In these cases, not all resource blocks within the cell
`bandwidth can be used for distributed transmission. As an example, for a cell bandwidth correspond-
`ing to 25 resource blocks (the example in Figure 10.7) and a corresponding gap size equal to 12
`according to Table 10.1, the 25th resource-block pair cannot be used for distributed transmission.
`
`One resource-block pair
`
`0
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
`
`VRBs
`
`PRBs
`
`FIGURE 10.7
`
`RB pair interleaving
`
`RB distribution
`
`Gap (12 for 25 RB bandwidth)
`
`VRB-to-PRB mapping in the case of distributed VRBs. A cell bandwidth corresponding to
`25 resource blocks is assumed.
`
`Samsung Exhibit 1015, Page 12
`
`

`

`152
`
`CHAPTER 10 Downlink Physical-Layer Processing
`
`Table 10.1 Gap Size for Different Cell Bandwidths (Number of Resource Blocks)
`
`Bandwidth
`
`P
`
`Gap size
`
`6
`
`1
`
`3
`
`7–8
`
`9–10
`
`11
`
`12–19
`
`20–26
`
`27–44
`
`45–63
`
`64–79
`
`80–110
`
`1
`
`4
`
`1
`
`5
`
`2
`
`4
`
`2
`
`8
`
` 2
`
`12
`
` 3
`
`18
`
` 3
`
`27
`
` 4
`
`32
`
` 4
`
`48
`
`Table 10.2 Second Gap Size for Different Cell
`Bandwidths (Only Applicable for Cell Bandwidths of
`50 RBs and Beyond)
`
`Gap size
`
`Bandwidth
`
`50–63
`
`9
`
`64–110
`
`16
`
`As another example, for a cell bandwidth corresponding to 50 resource blocks (gap size equal to 27
`according to Table 10.1), only 46 resource blocks would be available for distributed transmission.
`In addition to the gap size outlined in Table 10.1, for wider cell bandwidths (50 RBs and beyond),
`there is a possibility to use a second, smaller frequency gap with a size of the order of one-fourth of
`the cell bandwidth (see Table 10.2). The use of the smaller gap enables restriction of the distributed
`transmission to only a part of the overall cell bandwidth. Selection between the larger gap accord-
`ing to Table 10.1 and the smaller gap according to Table 10.2 is indicated by an additional bit in the
`resource allocation on PDCCH.
`
`10.2 DOWNLINK REFERENCE SIGNALS
`Downlink reference signals are predefined signals occupying specific resource elements within the
`downlink time–frequency grid. The LTE specification includes several types of downlink refer-
`ence signals that are transmitted in different ways and used for different purposes by the receiving
`terminal:
`
`G Cell-specific reference signals (CRS) are transmitted in every downlink subframe and in every
`resource block in the frequency domain, thus covering the entire cell bandwidth. The cell-specific
`reference signals can be used by the terminal for channel estimation for coherent demodulation
`of any downlink physical channel except for PMCH and for PDSCH in the case of transmission
`modes 7, 8, or 9. As described in Section 10.3, these transmission modes correspond to so-called
`non-codebook-based precoding. The cell-specific reference signals can also be used by the termi-
`nal to acquire channel-state information (CSI). Finally, terminal measurements on cell-specific
`reference signals are used as the basis for cell-selection and handover decisions.
`
`Samsung Exhibit 1015, Page 13
`
`

`

`10.2 Downlink Reference Signals
`
`153
`
`G Demodulation reference signals (DM-RS), also sometimes referred to as UE-specific reference
`signals, are specifically intended to be used by terminals for channel estimation for PDSCH in
`the case of transmission modes 7, 8, or 9. The label “UE-specific” relates to the fact that each
`demodulation reference signal is intended for channel estimation by a single terminal. That spe-
`cific reference signal is then only transmitted within the resource blocks assigned for PDSCH
`transmission to that terminal.
`G CSI reference signals (CSI-RS) are specifically intended to be used by terminals to acquire
`channel-state information (CSI) in the case when demodulation reference signals are used for
`channel estimation.5 CSI-RS have a significantly lower time/frequency density, thus implying less
`overhead, compared to the cell-specific reference signals.
`G MBSFN reference signals are intended to be used for channel estimation for coherent demodula-
`tion in the case of MCH transmission using so-called MBSFN (see Chapter 15 for more details on
`MCH transmission).
`G Positioning reference signals were introduced in LTE release 9 to enhance LTE positioning func-
`tionality, more specifically to support the use of terminal measurements on multiple LTE cells to
`estimate the geographical position of the terminal. The positioning reference symbols of a certain
`cell can be configured to correspond to empty resource elements in neighboring cells, thus ena-
`bling high-SIR conditions when receiving neighbor-cell positioning reference signals.
`
`10.2.1 Cell-Specific Reference Signals
`
`Cell-specific reference signals, introduced in the first release of LTE (release 8), are the most basic
`downlink reference signals in LTE. There can be one, two, or four cell-specific reference signals in a
`cell, defining one, two, or four corresponding antenna ports.
`
`10.2.1.1 Structure of a Single Reference Signal
`Figure 10.8 illustrates the structure of a single cell-specific reference signal. As can be seen, it con-
`sists of reference symbols of predefined values inserted within the first and third last6 OFDM symbol
`of each slot and with a frequency-domain spacing of six subcarriers. Furthermore, there is a fre-
`quency-domain staggering of three subcarriers for the reference symbols within the third last OFDM
`symbol. Within each resource-block pair, consisting of 12 subcarriers during one 1 ms subframe, there
`are thus eight reference symbols.
`In general, the values of the reference symbols vary between different reference-symbol posi-
`tions and also between different cells. Thus, a cell-specific reference signal can be seen as a tw