Tdoc R1-02-0354
`
`
`
`
`
`
`
`
`
`6.7
`Philips
`Coding of Channel Quality Information (rev1)
`Discussion and Approval
`
`
`
`
`3GPP TSG RAN WG1#24
`Orlando, USA, February 18th-22nd, 2002
`
`Agenda Item:
`Source:
`
`Title:
`
`
`Document for:
`
`
`
`Introduction
`
`Previous contributions which have considered the coding of channel quality information in HSDPA FDD
`mode are summarised in [1]. The current working assumption is that 5 information bits will be transmitted
`in a 20 bit data field. The most promising proposals are
`• To take the (16,5) bi-orthogonal TFCI code and append 4 of the 5 data bits to a make a 20 bit code
`word [2]. The same result could be achieved by suitably extending the basis vectors to 20 bits.
`• To take 5 basis vectors from a (32,10) Reed-Muller code and puncture to 20 bits [3]
`Both these proposals achieve minimum distance of 9 in the resulting (20,5) code and have the advantage of
`re-using existing system components.
`
`As well as maximising minimum distance it is also desirable that if possible the most significant bits of the
`data are better protected than the least significant bits. This would reduce the probability that transmission
`errors would result in large errors in the received channel quality value. As a performance metric, we
`propose the expectation value of the rms error between the transmitted quality value and the decoded value
`at the receiver. This metric can be computed for any code (of reasonable size!) from the pdf of the analogue
`values and the probability of each possible error vector. Here we assume that the analogue data is
`uniformly quantized over 32 levels and has uniform pdf. We also assume that the channel is binary
`symmetric with an uncorrelated error process with uniform bit-error probability.
`
`Proposal
`
`Therefore we propose to extend the basis vectors of the (16,5) TFCI code as shown in Table 1. Here Mi,4 is
`the most significant bit. This arrangement gives significant extra protection to the MSB, and a little more
`robustness to the next most significant bit.
`
`IPR2021-00908 Honeywell Exh. 1010 - Page 1 of 5
`(Honeywell International, Inc., et al. v. 3G Licensing S.A.)
`
`

`

`Table 1: Basis sequences for (20,5) code as an extended (16,5) TFCI code
`i
`Mi,0 Mi,1 Mi,2 Mi,3 Mi,4
`0
`1
`0
`0
`0
`1
`1
`0
`1
`0
`0
`1
`2
`1
`1
`0
`0
`1
`3
`0
`0
`1
`0
`1
`4
`1
`0
`1
`0
`1
`5
`0
`1
`1
`0
`1
`6
`1
`1
`1
`0
`1
`7
`0
`0
`0
`1
`1
`8
`1
`0
`0
`1
`1
`9
`0
`1
`0
`1
`1
`10
`1
`1
`0
`1
`1
`11
`0
`0
`1
`1
`1
`12
`1
`0
`1
`1
`1
`13
`0
`1
`1
`1
`1
`14
`1
`1
`1
`1
`1
`15
`0
`0
`0
`0
`1
`16
`0
`0
`0
`0
`1
`17
`0
`0
`0
`0
`1
`18
`0
`0
`0
`0
`1
`19
`0
`0
`0
`1
`0
`
`
`
`The performance of this code, in terms of rms error in the channel quality metric is shown in Figure 1 as a
`function of bit error rate. The rms error has been normalised to a value of unity at a BER of 0.5.
`
`Results are also given for the earlier proposals in [2] and [3]. It turns out that these were so close that they
`are both represented by the same curve.
`
`The size of the error due to quantization to 5 bits is also shown for comparison.
`
`
`IPR2021-00908 Honeywell Exh. 1010 - Page 2 of 5
`(Honeywell International, Inc., et al. v. 3G Licensing S.A.)
`
`

`

` Extended (16,5) and Punctured (32,10) (see [2,3])
` Extended (16,5) code (New Proposal)
` Quantisation error
`
`0.05
`
`0.10
`Bit Error Rate
`
`0.15
`
`0.20
`
`Figure 1: Performance of Channel Quality Coding Schemes
`
`
`
`0.5
`
`0.4
`
`0.3
`
`0.2
`
`0.1
`
`RMS Error in Quality Metric (Normalised)
`
`0.0
`0.00
`
`
`Although the minimum distance of the proposed scheme is only 8, the distance for the MSB is 11. This
`extra protection for the MSB results in the improved performance shown in Figure 1.
`
`Conclusion
`
`The (20,5) code produced by extending the (16,5) TFCI code should be adopted for the coding of channel
`quality information. The small improvement in performance is achieved with no additional complexity
`compared with other proposed schemes.
`
`If the current working assumption of 5 information bits is changed so that fewer bits are sent, this should be
`done by setting the appropriate number of LSB’s in the data word to zero.
`
` A
`
` code with similar performance could also be designed by puncturing a (32,10) code [3]. In this case it
`may be possible to optimise the puncturing pattern, depending on the number of information bits.
`
`References
`
`[1] R2-01-0019, “Channel coding and error detection for uplink QI signalling”, Nokia
`
`[2] R1-01-1144, “Coding for channel quality related information”, Ericsson.
`
`[3] R1-01-1324, “Coding scheme for Quality Indicator (rev 1)”, Samsung.
`
`
`IPR2021-00908 Honeywell Exh. 1010 - Page 3 of 5
`(Honeywell International, Inc., et al. v. 3G Licensing S.A.)
`
`

`

`
`Text Proposal
`
`-------------------------New cross-reference in TR 25.858, section 8.2.2.1-----------------------------
`
`8.2.2.1 HS-DSCH Associated Uplink Dedicated Control Channel
`The following information is carried on the HS-DSCH associated uplink dedicated control channel
`(DPCCH-HS):
`- H-ARQ acknowlegemnt
`A 1-bit Ack/Nack indication is used for a H-ARQ acknowledgement. The acknowlegement bit is
`repetition coded to 10 bits and transmitted in one slot. H-ARQ acknowledgement field is DTX’ed when
`there is no ACK/NACK inforamtion being sent.
`- Measurement feedback information
`Measurement feedback information contains channel quality indicator that may be used to select transport
`format and resource by HS-DSCH serving Node-B. A [5]-bit channel quality indicator is coded as
`described in section 8.2.2.2.5 and transmitted over two slots. The transmission cycle and timing for
`channel quality indicator is determined by UTRAN and signalled by higher layer. (Note: It is to be
`determined whether the transmission cycle is influenced by HS-DSCH activity or downlink channel
`quality) Details for measurement feedback procedure is described in section 8.2.2.2.
`The channel quality indicator consists of a recommended TFRC provided by the UE to Node-B. The
`recommended TFRC is chosen by the UE from a TFRC reference list. An example of TFRC reference list
`is provided in section 8.2.2.2.3.
`The format for the additional DPCCH is shown in Figure 10. The ACK/NACK message is transmitted
`with a power offset ΔPAN relative to the Release '99 uplink DPCCH. The power offset ΔPAN is a higher-
`layer parameter.
`
`
`n bits
`
`Channel Coder
`
`20 coded bits
`over 2 slots
`
`10 bits/slot
`
`10 bits
`On one slot
`
`Figure 10. Format for additional DPCCH for HS-DSCH related uplink signalling
`
`Repeat
`
`1 bit
`
`For TDD, the structure is TBD.
`
`
`
`-------------------------New sub section in TR 25.858, section 8.2.2.2-----------------------------
`
`8.2.2.2.5 Coding of Channel Quality information
`
`The channel quality information is coded using a (16,5) bi-orthogonal code, extended by 4 bits to form a
`(20,5) code. The code words of the (20,5) code are a linear combination of the 5 basis sequences defined in
`Table xxx.
`
`IPR2021-00908 Honeywell Exh. 1010 - Page 4 of 5
`(Honeywell International, Inc., et al. v. 3G Licensing S.A.)
`
`

`

`Table xxx: Basis sequences for (20,5) code
`i
`Mi,0 Mi,1 Mi,2 Mi,3 Mi,4
`0
`1
`0
`0
`0
`1
`1
`0
`1
`0
`0
`1
`2
`1
`1
`0
`0
`1
`3
`0
`0
`1
`0
`1
`4
`1
`0
`1
`0
`1
`5
`0
`1
`1
`0
`1
`6
`1
`1
`1
`0
`1
`7
`0
`0
`0
`1
`1
`8
`1
`0
`0
`1
`1
`9
`0
`1
`0
`1
`1
`10
`1
`1
`0
`1
`1
`11
`0
`0
`1
`1
`1
`12
`1
`0
`1
`1
`1
`13
`0
`1
`1
`1
`1
`14
`1
`1
`1
`1
`1
`15
`0
`0
`0
`0
`1
`16
`0
`0
`0
`0
`1
`17
`0
`0
`0
`0
`1
`18
`0
`0
`0
`0
`1
`19
`0
`0
`0
`1
`0
`
`
`
`
`The channel quality information bits are a0 , a1 , a2 , a3 , a4 (where a0 is LSB and a4 is MSB). The output code
`word bits bi are given by:
`4
`= ∑
`Ma
`b
`(
`2mod
`)
`×
`
`,
`ni
`
`i
`
`n
`
`n
`
`=
`
`0
`
`where i = 0, …, 19.
`
`The output bits are denoted by bk, k = 0, 1, 2, …, 19.
`
`
`-- End of text proposal --
`
`
`IPR2021-00908 Honeywell Exh. 1010 - Page 5 of 5
`(Honeywell International, Inc., et al. v. 3G Licensing S.A.)
`
`